Cannabis and Man

Psychological and Clinical Aspects and Patterns of Use, Edited by P.H. Connell and N. Dorn 1975

Foreword
Sir Harry Greenfield, C.S.I., C.I.E., Chairman, International Narcotics Control Board, Geneva. This Symposium is abundantly welcome for a variety of reasons: first as yet another example of constructive research by the Institute for the Study of Drug Dependence in natural and logical succession to its two earlier Symposia on The Botany and Chemistry of Cannabis (organised by Dr Joyce) and on Cannabis and its Derivatives, Pharmacology and Experimental Psychology (organised by Professor Paton). It is even more welcome for the solid contribution which it can be expected to make to the growing international harvest of accurate information regarding this particular drug, the more so since, like its predecessors, the Symposium has the special merit of being directed to a limited and clearly defined area of study and aims at reaching positive and sustainable conclusions. A rigorous scientific approach of this kind is, I need hardly say, essential in studies directed to all aspects of such a subject, in which public interest is closely engaged, yet where public understanding is clouded by insufficient knowledge of what is, let us frankly admit, a complex subject and by consequent inability to appreciate the implications which it may hold for the individual and for the community at large. The flow of agreed findings emanating from symposia such as this, which assemble and analyse in conjoint discussion the results of researches by the individual participants themselves and by other scientists, is of great assistance to the International Organs whose duty it is to formulate international policy in this field. Moreover, as you may be aware, the World Health Organisation has lately instituted a series of studies into the long-term effects of cannabis consumption. Not only are these scientific interchanges of great service to the International Organs, they also provide a valuable source of authentic information for inclusion in health education programmes, - a vitally important sector to which the I.S.D.D. is now in process of making a significant contribution. The Report of the International Narcotics Control Board makes appreciative reference to the rapid growth of basic research on cannabis in a number of countries and to the fact that the detailed data yielded thereby is gradually enhancing scientific knowledge of this substance. The Board emphasises, however, that much has still to be learnt, especially in regard to effects of long-term consumption; and it points out that the need for fuller knowledge is rendered even more urgent by the newly-manifested availability of cannabis concentrate in liquid form, which opens up disquieting new perspectives. It was doubtless with these

perspectives in mind that the WHO decided to launch the series of studies on long-term effects of cannabis which I mentioned above. I hope therefore that the talented scientists whom Dr Connell has assembled here in the scholarly setting so generously provided by the CIBA Foundation will not only feel that the Symposium has an immediate and worldwide relevance but that they will hereafter pursue their researches in the knowledge that they will be assisting authorities, national and international, in coping with one of the great social problems of the present age. 1975 H.G.

Participants
Ms K, Berntsen Ungdomskliniken, 2200 Kobenhavn N, Laessoesgade 8A, Denmark. Ms J. Blackwell 27 Mildmay Grove, London Ni. Dr D.A. Cahal Dept Health & Social Security, York House, 37 Queen Square, London WC1N DBJ. Dr D0C, Cameron Medical Officer, Office of Mental Health, WHO, 1211 Geneva 27, Dr P,H. Connell (Conference Chairman) Director, Drug Dependence Clinical Research & Treatment Unit, The Maudsley Hospital, Denmark Hill, London S.E.5. (and the Bethlem Royal Hospital, Monks Orchard Road, Beckenham, Kent). Mr N. Dorn (Conference Secretary) Head, Evaluation and Educational Research Unit, ISDD, 3 Blackburn Road,London NW6. Dr G. Edwards Director, Addiction Research Unit, Institute of Psychiatry, 101 Denmark Hill, London SE5. Sir Harry Greenfield Chairman, International Narcotics Control Board, Geneva. Mr S. Hasleton Dept of Psychology, University of Sydney, Sydney, Australia. Dr D.V. Hawks Whitchurch Hospital, Whitchurch, Cardiff. Dr I. Hindmarch Dept of Psychology, University of Leeds. Ms A. Kosviner Addiction Research Unit, Institute of Psychiatry, 101 Denmark Hill., London SE5. Dr E. Leuw Foundation for the Study of Alcohol and Drugs, de Lairessestraat 39, Amsterdam. Mr F. Logan Director, ISDD, 3 Blackburn Road, London NW6. Dr C.G. Miles Addiction Research Foundation, 33 Russell Street, Toronto, Ontario.

Dr S. Martens Director, Division of Narcotic Drugs, UN Office at Geneva,. Palais des Nations, CH-1211 Geneva 10. C.G. Miles Director, Marijuana Study, Addiction Research Foundation, 33 Russell Street, Toronto, Ontario. Dr R.D. Miller Drug Research Consulting, PO Box 1327, Station B, Ottawa, Ontario. Ms J. Mott Home Office, Romney House, Marsham Street, London SW1. Prof. W. Paton University Dept Pharmacology, South Park Road, Oxford. Dr R. Porter Deputy Director, The Ciba Foundation, 41 Portland Place, London WiN 4BN. Dr N. Rathod St Christopher's Day Hospital, 52 Hurst Road, Horsham, Surrey. Dr V. Rubin Research Institute for the Study of Man, 162 East 78 Street, New York 10021 NY. Dr R.G. Smart Addiction Research Foundation, 33 Russell Street, Toronto. Dr D Somekh Dept of Psychology, Bedford College, Regents Park, London NW1. Prof. M.I. Soueif Professor of Psychology, 5 Sayed-el-Bakri Str, Gesira, Zamalek, Cairo, Egypt. Dr J.R. Tinklenberg Stanford University Medical Center, Stanford, Calif 94305, USA. Mr J. Woodcock Assistant Director, ISDD, 3 Blackburn Road, London NW6.

Contents
Foreword Participants Introduction P.H. CONNELL and NICK DORN PART ONE: EFFECTS OF USE 1. Cannabis and Driving Risk R.G. SMART Discussion 2. Psychomotor and Cognitive Deficits Associated with Long- and Short-term Cannabis Consumption: Comparison of Research Findings and Discussion of Selected Extrapolations M.I. SOUEIF

Discussion 3. Psychological and Cognitive Effects of Cannabis J.R. TINKLENBERG and C.F. DARLEY Discussion 4, A Selective Review of Studies of Long Term Use of Cannabis on Behaviour, Personality and Cognitive Functioning C.G. MILES Discussion 5. Cannabis Psychosis N.H. RATHOD Discussion PART TWO: PATTERNS OF USE AND SOCIAL ATTITUDES 6, The Uses and Implications of the Log-normal Distribution of Drug Use W.D.M. PATON Discussion 7. The Psychology of Cannabis - Techniques for Investigating the Frequency and Patterns of Use of Cannabis in Groups of Drug Users I. HINDMARCH. 8. Use of Drugs Other Than Cannabis and Attitudes to Drug Use in UK Student Populations D. SOMEKH Discussion 9. Prevalence, Characteristics and Correlates of Cannabis Use in the UK Student Population A. KOSVINER 10. Drug Use in the Netherlands (brief presentation) E. LEUW 11. Social Attitudes of Users (brief presentation) K. BERNTSEN Discussion of papers by Kosviner, Leuw and Berntsen 12. Cannabis, 'Permissivism' and Social Response in Australia S. HASLETON Discussion Addendum to Hasleton's paper PART THREE: PAST AND FUTURE RESEARCH 13. Some Suggested Research Priorities R.D. MILLER Closing Discussion 14. Asking Better Questions in Cannabis Research P.H. CONNELL and N. DORN

Introduction
P. H. Connell, Bethlem Royal and Maudsley Hospitals, London. N. Dorn, Evaluation and Research Unit, ISDD, London.

This third symposium, organised by the Institute for the Study of Drug Dependence, follows naturally from the first symposium (The Botany and Chemistry of Cannabis)* and the second symposium (Cannabis and its Derivatives: Pharmacology and Experimental Psychology)** in that the theme 'Cannabis and Man: Psychological and Clinical Aspects and Patterns of Use' extends discussion from the laboratory and the basic sciences into the, in many ways, much more difficult and less precise area of Man, his personality and behaviour and the community in which he lives. As before, the Conference was by invitation only and comprised 28 experts. The invitations to participants from the Planning Committee note that 'It is recognised that the proposed field of discussion has been the subject of review in many quarters, including the Wootton Report (United Kingdom, 1968), the Le Dain Report (Canada 1970) and the Schafer Report (U.S.A. 1972). The aim of the Conference is, therefore, to attempt an overview of the present state of knowledge; to discuss conceptual and methodological problems; to bury dying issues and to point the way to future avenues of worthwhile research.' The majority of those invited to give papers were given a brief to review the particular aspect concerned but a small number of papers dealing with specific researches were accepted later. The format of the meeting was changed from that of previous symposia in that the Planning Committee, taking cognizance of the less precise nature of the field to be covered, arranged for summaries of papers to be presented in thirty minutes and for forty-five minutes of discussion after each paper. Thus free interchange with less pressure of time, could take place. There was, of course, some anxiety that with such a liberal period for discussion, the proceedings might dry up. In the event, it was interesting and gratifying to observe that throughout the whole proceedings, stimulating and lively discussion took place and the Chairman often had to bring discussion to a close some fifty minutes or more after the presentation of the papers. The Planning Committee deliberately refrained from arranging matters in such a way that participants were steered or forced into producing a formal blue-print summarizing the proceedings (as is perforce the case in many national and international meetings of experts). The Editors have attempted to draw together the lessons learned from the Conference in a final chapter and take full responsibility for this! The discussions were all tape-recorded and the reports in the text are much summarized versions of the discussions. They are therefore condensed and the Editors are grateful to participants for allowing them the freedom to undertake this difficult task and for permission to go ahead with publication without further reference to contributors, in the interest of rapid publication of the proceedings. - The meeting was organized by a sub-committee consisting of Dr David Hawks, Mr Frank Logan, Professor W.D.M. Paton and ourselves. The Editors wish to express their gratitude on behalf of the Committee and all those participating, to the Ciba Foundation and in particular to Dr G.E.W. Wolstenholme and Dr Ruth Porter for their help and advice and for the use of the efficient, comfortable and excellent facilities, both in terms of the technical facilities and also in terms of the hospitality offered to participants. Joyce, C.R.B., and Curry, S.H. (Eds.) "Botany and Chemistry of Cannabis" Churchill, London, 1972

** Paton, W.D.M., and Crown, J. (Eds.) "Cannabis and its Derivatives: Pharmacology and Experimental Psychology". Oxford University Press, 1972 PERSONAL APPRECIATION P.H. Connell It was my privilege to be Chairman of the Planning Committee of this Conference and to Chair the proceedings throughout. I would like, therefore, to thank all members of the Committee and all participants for their work and contributions. In particular I would like to express my gratitude to Nick Dorn (Secretary of the Planning Committee, Conference Secretary and Joint Editor for his energy, efficiency and expertise and also to Dr David Hawks who was invaluable in the planning stages in relation to a balanced selection of contributors. Donations towards the cost of this conference were received from: Beecham Group Glaxo Group Pfizer Ltd. Smith, Kline & French Sterling-Winthrop Group Wellcome Trust Department of Health and Social Security. The Ciba Foundation generously allowed the conference to be held at its headquarters in Portland Place, and provided many valuable facilities.

1. Cannabis and Driving Risk

Reginald G. Smart, Evaluation Studies Department, Addiction Research Foundation, Toronto, Ontario. Evidence accumulates to suggest that the use of cannabis and other hallucinogens is increasing in North America. For the most part these drugs are not replacing the older, socially acceptable drugs such as alcohol and tranquillizers. Naturally, drug use of all types leads to an interest in the driving risks they might represent. Research in this area has been of several types: 1) laboratory, simulator and closed course studies of the impairing effects of various drugs and combinations: 2) surveys of the prevalence of drug use in driving populations: 3) studies of drug use among various persons involved in accidents: and 4) studies of accident rates among drug using or drug abusing populations. In general, research in these areas is far less adequately developed for cannabis than for alcohol. Some of the reasons involve the recency of interest in cannabis, the lack of easy methods for cannabis detection in body fluids, and the generally held view that drugs in total represent a less important factor in accidents than alcohol. The aims of this paper are: (1) To review research on cannabis each of the areas listed above; (2) To suggest what research remains to be done in assigning driving risk- and (3) To discuss the need for countermeasures related to cannabis use and driving. Previous reviews of drugs and driving problems (eg. Kibrick and Smart, 1970; Nichols, 1971) have tended to emphasize psychoactive drugs rather than hallucinogenic drugs but the present review attempts to consider only cannabis. Before examining the relevant laboratory and epidemiological work certain methodological problems need also be considered.

METHODOLOGICAL DIFFICULTIES IN ASSESSING CANNABIS EFFECTS ON DRIVING There are several complicating factors in assessing the effects of drugs such as cannabis on driving. One is the question of drug interaction. A bewildering number of drugs interact with one another or with alcohol, consequently, assessing the effects of each possible combination is a difficult task. Research on some of these combinations has begun, but not all combinations have been tested. It is known that cannabis users are more often drinkers of alcohol than are non-users (Smart and Fejer, 1973) and that cannabis and alcohol are often used on the same occasion. The interaction between alcohol and drug impairments may not be easily discovered. In some studies impairment appears greatest for naive cannabis users and least or even nonexistent for experienced users. Additive detrimental effects of alcohol and cannabis have been found for some complex tracking behaviours but not for simple ones where low doses of alcohol are used (B.A.C. = 0.03 per cent). These additive effects could be important when it is realized that marijuana is often taken with alcohol. Low blood alcohol levels in drivers who have taken cannabis may suggest less behavioural or driving impairment than actually exists. The combination of experience, multi-drug use and task complexity make clear generalization difficult. Soehring and Wolters (1968) have reviewed the literature on relationships between blood (or urine) levels of drugs and driving performance. They state (in translation) that '(1) The ideal conditions prevailing in ethanol decomposition to carbon dioxide and water, distribution in an organ corresponding to water content in the tissues, etc., cannot be transferred to drugs- (2) In general, it appears that the 'blood level' cannot be regarded as a reliable criterion for the effect of different drugs...; (3) Many active drugs can only be separated with difficulty from their inactivated decomposed forms...; (4) Some drugs in therapeutic doses can be found in the urine many days after ingestion...; (5) Retrogressive calculation that can be done with ethanol cannot be done with most drugs, as their decomposition rates are affected with chronic use of certain drugs and by other factors.' Most of this applies to cannabis and in addition, there are also problems of achieving rapid body fluid analyses which will allow even a positivenegative judgment. Certain facts about the metabolism and effects of cannabis make it difficult to study in relation to driving and accidents. Cannabis and its metabolites become widely distributed in many body tissues and fluids soon after ingestion. However, they do not appear in the lungs for long after smoking and hence breath tests on the alcohol model are impossible. It is also important to note that cannabis users may be able to 'come down' or decrease their behavioural impairment voluntarily; experimental demonstrations of this have been made (Cappell and Pliner, 1973). Cannabis may sometimes be detected in breath samples immediately after use. However, no easy breath or blood or urine testing method has been developed which would rival the alcohol testing methods in assessing accident drivers or victims. Methods for measuring metabolites (eg. King and Forney, 1967) will assess the remains of doses taken days or weeks prior when the major impairing effects have long disappeared. Sometimes skin swabs with alcohol or chloroform washings have been used (eg. Robinson, 1971), to retrieve cannabis samples from saliva. Fortunately, Stones and Stevens (1969) have demonstrated a method for retrieving cannabis from the fingers and the mouth of users. It is sensitive to exposure to marijuana within the past hour, however, it may mean only that the positive case has handled marijuana or been in a room where it was used.

This latter method has been used in a study of fatally injured accident drivers (Benjamin, 1972) however some difficulties with delayed analyses of skin swabs have been experienced with this method. As well, the method is not quantitative but qualitative in a vague manner. Clearly, the appearance of a reliable, trouble free method for relating recent cannabis use to THC or metabolite levels in body fluids would greatly facilitate accident research. However its measurements should also relate to actual behavioural impairment. A further difficulty with cannabis and driving studies is the selection of relevant laboratory or simulator methods for assessing driving skill and its detriments. Some studies with instrumented cars are able to depart somewhat from the artificiality of the laboratory situation in studying alcohol effects (eg. Huntley, 1973). However, many aspects of driving and careless driving are difficult to directly study for drug effects in a real way. Perhaps one example would be the high speed passing manoeuver which has been found to lead to so many fatal crashes. Reluctant to ride with drunk or doped drivers in real situations, researchers have turned naturally enough to psychomotor simulation or closed course studies of alcohol and cannabis impairments. This decision has its own problems. It is also the case that most laboratory, instrumented car and closed course studies of drug effects are made under ideal conditions. That is there is, no rain, snow or fog, the driver is not fatigued, the sessions are short and there is no threatening traffic. A major difficulty is that one cannot be sure what simple or complex skills interact to produce what may be known as 'safe' or non-accident driving. Nearly all human behaviours may sometimes be involved in driving skills at one time or another including aspects of personality, social behaviour, and cognitive function, as well as the more obvious visual and psychomotor skills eg. vision, reaction time etc. In fact there seems to be no generally accepted set of necessary and unique driving behaviours; the tendency has been to assume from face validity that certain behaviours are involved in safe driving. Some anomalies are obvious in this area. For example, Fergenson (1971) found that drivers with high accident rates did more poorly on a choice reaction test than did non-accident drivers. However, those with high violation rates (and low accidents rates) had the fastest reaction times. The validity of 'simulators' is also in doubt. Edwards, Hahn, and Fleischman (1969) found almost no correspondence between simulator behaviour and actual driving. However Crancer (1968) found that simulator driving was related to 5 year accident records, with the best performers having fewer accidents. Unfortunately, behind the wheel tests such as those that are done in licensing examinations showed no relation to accident rates. These difficulties have led to a decision to see driving as does Moskowitz (1973) i.e. 'primarily a time-shared activity between a visual search and recognition task and a compensatory tracking task'. Because of the difficulties of interpreting laboratory and simulator studies more emphasis in assessing driving risk and cannabis should be placed upon epidemiological studies of drivers and accident victims. Laboratory and simulator studies are perhaps most valuable for indicating possibly dangerous drugs and what behaviours they might most affect. How ever, real risk studies will most likely come from studies of accident involved populations compared to those not so involved. EFFECTS OF CANNABIS ON PERSONALITY AND COGNITIVE FUNCTIONS Theoretically, any number of personality or cognitive effects of cannabis could relate to driving risk. So little information exists about how such behaviours as aggression, risk taking, memory, intellectual performance and the like relate to driving that a complete review of this

area is not attempted. The reader is directed to other reviews (eg. Le Dain, 1972) and to other sections of the symposium proceedings for this material. An idiosyncratic selection of relevant research may be of some interest. In general, many inconsistent results have been found for the cognitive or intellectual effects of marijuana. The value of work on memory and intellectual functions for assessing driving risk is uncertain. However, many studies (eg. Le Dain, 1972; Clark, Hughes and Nakashima, 1970) plus subjective reports indicate a reduction in vigilance, attention and concentration. One of the major clinical effects of cannabis is a tranquillizing or calming one. Users sometimes report difficulties in concentrating or attending to intellectual tasks. In the Le Dain Commission's experiments 'sustained attention was assessed in a 40 minute visual signal detection task which was subjectively boring, and made precise demands on the-subject'. In this study cannabis 'consistently reduced accuracy' as might be expected from subjective reports and clinical studies. However, in a short, 5 minute, Continuous Performance Test Weil et al. (1968) found no effect of two doses of cannabis; it may be that sustained, boring attention tasks will be most affected by cannabis. Personality studies are also difficult to relate to driving or accident rates. One generally accepted finding is that cannabis does not increase aggressiveness as has sometimes been found for alcohol. Most laboratory studies including those of the Le Dain Commission (1972) and the Addiction Research Foundations long term study (Miles et al, 1972) have found no increase in aggressive behaviour. Most studies (eg. Soueif, 1971; Goode, 1970) have also failed to find that cannabis users engage in aggressive criminal behaviour but that most of their crimes relate directly to drug use. Almost no interest has been taken concerning the effect of cannabis on risk taking but this might be particularly important in driving. Benjamin has stated (1972) that an unpublished study showed an increase in driving related risk after cannabis use, however the author has not seen this report. Studies of the effects of cannabis on personality and cognitive variables can do no more than suggest that driving hazards do exist. In general, the available data suggest that cannabis would be most likely to cause accidents (if at all) by reducing attentiveness to traffic hazards and least likely by increasing aggressiveness, or speed. It may be that cannabis would contribute to running off the road accidents more than to those involving high speed or passing. EFFECTS OF CANNABIS ON DRIVING-RELATED SKILLS It is not easy to decide which of the many effects of cannabis are most relevant to driving skill or accident involvement. As stated above emphasis should probably be placed upon visual attention and recognition tasks, compensatory tracking and the more complex simulator and closed-course studies. Numerous studies have been made of the effects of cannabis on vision. Allentuck and Bowman (1942) reported nystagmus after use of marijuana. However Caldwell et al. (1968) found no effect upon brightness thresholds at low doses. Effects upon depth perception seem to be small (Clark and Nakashima, 1968; Le Dain Commission, 1972). Sharma and Moskowitz (1972) have demonstrated a dose dependent effect of marijuana on visual autokinetic phenomena. The lights were seen to have complex paths and the authors concluded that 'The results imply that hazards are involved in operating vehicles (driving or flying, for example) at night when under the influence of marijuana'. This effect was also

found in the Le Dain Commission studies (1972). A more important effect may be upon peripheral vision in a divided attention task. Moskowitz et al. (1972) found a substantial effect of two cannabis cigarettes on a task requiring fixation on a central task as well as vigilance for peripheral signals. This task would appear to have more relation to driving than simple visual acuity tasks. Some controversy surrounds the effects of cannabis on glare recovery. The original study by Frank et al. (1971) found a delay of several seconds in glare recovery under cannabis. However studies by the Le Dain Commission (1972) have apparently found no cannabis effect on glare recovery after moderate and large doses. Cannabis effects on pursuit rotor and tracking skills are somewhat more consistent Weil et al. (1968) found a decrement in pursuit-rotor performance for 9 cannabis naive subjects but no decrement for 8 heavy users. A dose response relationship was found. Manno et al. (1970) also found dose-related impairment on a tracking task. As well, they tested alcohol and cannabis together and found a greater impairment than from either above. The Le Dain Commission tested 26 subjects on simple and complex tracking tasks with two doses of alcohol (0.03 and 0.07 per cent B.A.C.) and two doses of marijuana (1.6 and 6.8 mg 9THC). Results were complex, however, alcohol and the higher cannabis dose produced impairment in both simple and complex tracking tasks. The drugs together produced higher impairments than either separately. These data generally suggest that driving impairments may be greatest amongst naive users and on occasions when alcohol has also been used. There is also a possibility that blood alcohol levels will underestimate the amount of behavioural impairment if cannabis has also been used. Therefore breath testing alone is liable to miss some impaired drivers. It might also be noted that at least one study (Soueif, 1971) found a suggestion of chronic psychomotor impairments. Soueif found that prisoners in Egypt arrested for hashish (chiefly very heavy users) performed more poorly than did prisoners arrested on other charges. Long term effects of this type could however have a number of explanations and more chronic studies should be done. Of great interest to those concerned with driving risk are the studies of cannabis and simulator or instrumented car performance. The first of these studies was done by Crancer et al. (1969) and it unfortunately has been taken in journalistic reports to indicate no cannabis impairment. Briefly, the authors attempted to determine the effect of a 'normal social high' on driving simulator performance. All 36 subjects were experienced users and drivers. Subjects smoked 2 marijuana cigarettes of approximately 0.09 per cent each but apparently Ti-IC content was not assessed (it was assumed to be about 22mg THC in total) or received the equivalent of 6 ounces of 86 proof liquor in 30 minutes. This was intended to achieve a B.A.C. of about 0.10 per cent at the time of testing. It was found that accelerator, brake, signal, steering and total errors under marijuana were not greater than under control conditions but that those for alcohol were much higher. Total speedometer errors under marijuana were higher than in the control condition. Kalant (1969) and Rafaelson et al. (1973) have pointed out several problems with the Crancer et al. study. Kalant criticized the lack of a dose-response design and he has argued that the high alcohol dose should have achieved 0.13 per cent B.A.C. rather than 0.10 and was therefore not a 'normal social high'. Rafaelson et al. (1973) have argued that tests of the same batch of marijuana as used by Crancer have produced estimates of 3 to 8mg THC content not 22mg. If this were the case then the effects may have worn off by the time testing was done (30 to 60 minutes after smoking).

Rafaelson et al. (1973) also studied cannabis effects on a driving simulator. However, they used 'cannabis cakes' which were eaten. The rationale for oral administration was that it is a 'more reliable and reproducible method than smoking and that it reduces variations in dosage between subjects'. Eating of(annabis is rare in western countries and results in slower absorption and more extended effects, so the practical importance of Rafaelson's study is uncertain. He also used 70g of 96 - per cent ethanol which generated blood levels of about lg per litre or 0.10 per cent at testing. Both cannabis and alcohol increased the time required to brake (in response to red lights) and start. Alcohol increased and cannabis decreased the number of gear changes but neither affected speed. A more interesting study of cannabis and driving was conducted by the Le Dain Commission (1972). They investigated two cannabis doses and a single alcohol dose (0.07 per cent B.A.C.) in 16 regular users of both drugs. The high cannabis dose was similar to the regular 'high' of users and the 0.07 per cent level is that at which accident risk seems to exceed the sober level (Borkenstein et al., 1964). Subjects were tested under all conditions including non-alcoholic drinks and a 'placebo' cigarette. Testing was done on a 1.1 mile track and a driving course with poles and plastic cones. Tasks were lap driving, parking and manoeuvring. Subjects with the low cannabis dose did not make more manoeuvering errors than under placebo. However, the higher doses of both drugs produced more errors. Driving speeds were not affected, however, awkward or 'rough' handling occurred under both drug conditions. The available data would appear to suggest that 'highness' under marijuana is liable to decrease sensory, motor and tracking performances associated with driving. The effects of 'sub-high' doses are probably smaller and not significant. Effects are probably greatest with naive users and when even small amounts of alcohol have been taken (0.03 B.A.C.). It should be noted here that a number of useful studies have not yet been done in this area. Subjects are almost always driving and drug-experienced, well-adjusted college students (or soldiers) in excellent health. Their performance may be more refractory to cannabis effects than that of young, driving and drug naive high school students perhaps having their first cannabis and liquor on the same occasion. It is also the case that simulator and instrumented car studies have not been done in dense traffic situations, at high speeds or under fatigue conditions. It could be argued that laboratory trials are established to find simple and large drug effects and not drug-situation interactions which may be rare but fatal in real driving situations. CANNABIS USE AMONG DRIVERS, ACCIDENT DRIVERS AND VICTIMS Anecdotal reports of cannabis use among drivers and accident drivers are not difficult to assemble. However epidemiological studies of cannabis use in accident drivers or victims are scarce. Many reports have been made of driving under cannabis. McGlothlin et al. (1970) in their study of 247 persons who had also used LSD found one case of accident attributed to a cannabis 'flashback'. Hollister (1971) asked 'high' subjects in his experimental studies whether they could drive in the 'high' condition, but none said they could. However, Klein et al. (1971) reported that two experienced marijuana smokers drove through Miami while 'stoned' on marijuana. Many reports collected by the Le Dain Commission in Canada show that about 50 per cent of cannabis users sometimes drive after cannabis but nearly the same number refuse to drive until a 'high' is well past. Certainly sufficient cannabis users do drive while high, to expect some cannabis-related accidents.

Studies of how much and how confidently cannabis users drive have not been very frequent. However, this problem was studied incidentally by Haines and Green (1970). They studied marijuana use among 131 heavy and moderate cannabis users, almost all of whom (94 per cent) used marijuana at least once a week. Of this number 81 stated that they drove while high on marijuana. Most of those who did not were recent users. Of the 81, 63 said they drove frequently while high and 67 said they drive as well or better than when 'straight'. Some mentioned that their concentration was improved, fewer distractions were found and reactions were improved. Another study by Klein, Davis and Blackbourne (1971) enquired about marijuana and driving among some 571 college students. It is not stated what proportion drove while high and the focus was on subjective effects on driving skills. More users found effects on time judgement than on any other skill. However, reaction time, speed estimation, distance perception and emergency handling capability were all said to be impaired. Generally, 2 to 3 times as many infrequent users as frequent users said that these skills were impaired. It is unfortunate that so few studies have been made of cannabis use among accident drivers or victims. The only study found so far has been reported by Berg et al. (1971) but it has a number of important limitations. Berg et al. studied drug use among accident victims seen in a university health service and comparable groups of students seen for non-accidental illnesses and for routine physical examinations. The experimental sample was small (n=24) and it is unlikely that marijuana use would often be detected in so few people. In any case no difference in reported drug use was found and no case of cannabis use was chemically detected in the experimental group. However results were based on blood analyses rather than the more sensitive swab methods suitable for cannabis detection. Laboratory tests were done for all commonly used psychoactive drugs including cannabis. Benjamin (1972) has described a large scale study financed by the National Highway Traffic Safety Administration which attempted to determine the incidence of cannabis use among fatally injured drivers. A skin swab method was used which has a lower reliability for analyses done some time after collection. It appears that since swab samples were mailed to a single centre for analysis there was a considerable delay and consequent unreliability in the data collected. Results from this study have apparently not been reported. Presently, there is almost no knowledge of how many drivers, accident drivers or victims have recently been under the influence of cannabis. A study of a large sample of accidents where drivers and/or victims were asked to report recent cannabis use and allow skin swabs, would reduce our ignorance considerably. This, of course, should include a comparable group of controls not involved in accidents, eg. passing the scene of a prior accident. INVOLVEMENT OF CANNABIS USERS IN ACCIDENTS Although cannabis use in accident drivers has rarely been studied the accident and violation rates of cannabis users have excited more interest and at least 6 studies are available. Somewhat contradictory results have been found concerning the involvement of marijuana users in accidents. One of the first and best studies in this series was reported by Waller (1965). For this study records of 231 drivers convicted for illegal possession or use of 'addicting drugs were searched. There were cases under review by the California Dept. of Motor Vehicles for possible license suspension. The sample probably contained many heroin and marijuana users and few users of prescription drugs. Drug users had only the

expected accident rate per mile driven but 1.8 times the violation rate. In a later paper Waller (1971) noted that the high violation rates of drug users commonly preceded their use of drugs, thereby suggesting that drug use could not have caused the excessive violations. Waller and Goo (1969) studied the type of accident for the drug user groups from 1965. They showed that the accidents and violations of drug users were similar to those of younger drivers in general and not peculiar to drug users. They found that drug users (as were persons of their age) were more often at fault and involved in more crashes due to excessive speed, weaving, and being on the wrong side of the road. Similar results to Waller's were found by McGlothlin, Arnold and Rowan (1970) in a study of adult marijuana users who volunteered for LSD. They did not have elevated accident experiences when high. Similarly, Haines and Green (1970) studied moderate to heavy users: none of the marijuana users who drove when Ilign reported accidents when 'stoned'. A study by Moser et al. (1971) of some 1889 arrestees for serious crimes also investigated accident and violation rates for drug users of various types. In this study hashish and marijuana users appeared not to differ in accident or violation rate from non-drug users. However, exposure for miles driven was not controlled in this study. Klein et al. (1971) found that marijuana users (especially heavy users) admitted more violations and license revocations than non-users but is not known whether the driving errors occurred while 'high'. They found that 18 per cent of infrequent users and 53 per cent of heavy users had been stopped by police while under the influence of marijuana. However being 'high' may not be the reason they were stopped. This finding is of some interest in that frequent users were more confident of their abilities under marijuana, but managed to attract much more police attention than did infrequent users. Excessive driving confidence under marijuana may be worth some investigation. Klein et al. also presented sufficient case history material relating marijuana use to accidents to suggest the need for further research. However, accident rates were not presented in terms of miles driven, nor were there comparable accident or violation data for non-users. Somewhat different results for accident rates have been found by Crancer and Quiring (1968). They studied 160 marijuana users drivers known to the Seattle Police Dept. The marijuana users had accident rates 39.2 per cent and violation rates 180.4 per cent higher than a comparable sample of non-users in Washington. Unfortunately, no control for exposure in terms of miles driven was performed. A study of high school students in Virginia (Ferguson and Howard, 1971) reported that 2.93 per cent of students admitted involvement in an accident as driver or victim in which marijuana 'may have been a cause'. This would appear to suggest considerable driving risk but without more complete knowledge about the contribution of marijuana it is impossible to interpret. Although interesting and suggestive, these studies do not provide unambiguous information about the role of drugs in accident causation. The frequency of accidents and driving offences after marijuana use is - still not known for any group. Drug users are typically also alcohol users and assigning their accident experience to the effects of a particular drug is often

difficult to do. There is no indication of how many accidents or violations occur per unit of exposure, eg. number of 'stoned' or intoxicated occasions. Lastly, no study has looked in detail at accident rates both before and after drug use began, consequently, drug use may be only tangentially related to high accident involvement. At present strong evidence that marijuana users have higher accident rates than non-users does not exist. Most studies have used small, rather limited samples (eg. illegal users known to the police) and studies of larger or more varied groups are needed before clear statement could be made. A paper by Smart (1974) reported a survey of 293 college students which attempted to answer the following questions: 1. How often do students drive after marijuana use? 2. How often do students drive after having used both alcohol and marijuana? 3. How frequent are driving charges and accidents after marijuana use? 4. How frequent are driving charges and accidents after alcohol use? 5. What is the relative risk potential of marijuana and alcohol in relation to driving? About 42 per cent of college student drivers in Smart's study reported marijuana use but only 61 per cent of those reported driving soon after that use. For most users the number of occasions was small, much smaller than the number of occasions for drinking and driving. Very few students reported that their accidents or moving violations involved prior marijuana use. Far more (3 times as many) reported violations or accidents after alcohol use. However, the frequency of marijuana-driving occasions in the past year was reported to be about 30 per cent of the frequency of alcohol-driving occasions. The results do suggest, however, that for this population, marijuana use contributes to very few accidents and charges. It probably contributes at most to only about one-third of those for alcohol. However, marijuana and driving occasions are much less numerous than alcohol and driving occasions. Problems arise when actual contribution of each drug state to accidents has to be assessed. On some marijuana and driving occasions users were also under the influence of alcohol and perhaps this was also true at the time of their accidents. There is no guarantee that for the respondents the levels of impairment after alcohol and marijuana were similar on the average. A further problem is that measures of exposure could involve comparisons of miles driven when 'marijuana influenced' and when 'alcohol intoxicated'. Smart's study did not inquire about this comparison but it would be worth doing so later. The data do suggest, however, the need to study the infrequency of marijuana-related accidents in relation to the low level of actual exposure. If driving occasions after marijuana use were increased, as a result of legalization or increased popularity, it may not turn out to be safer than driving after alcohol use. Conclusions The following tentative conclusions can be supported about cannabis and driving risk: 1. Methodological problems concerning both easy cannabis testing and laboratory analysis of the driving task have hindered the development of knowledge in this area. 2. The effects of cannabis on cognitive functions suggest that major decrements in attention may occur in driving situations, particularly if it is a long or boring journey. 3. Cannabis effects on personality and attention suggest that cannabis related accidents may more often be rear-enders or running-off-the-road than high speed passing accidents. 4. Laboratory studies of vision indicate a major cannabis effect on peripheral vision or divided

visual scanning tasks and less effect on depth perception, glare recovery, or visual acuity. This is a major concern since these scanning tasks together with tracking seem most important in driving. 5. Compensatory tracking tasks are impaired by 'high' dose levels of cannabis, particularly for naive users and where even small amounts of alcohol are also given. 6. Simulator and instrumented car performance decrements occur for braking and starting but effects on speed seem minor. Some studies have shown a lowering of speed after cannabis. 7. Numerous artificialities exist in the typical laboratory or simulator study which mitigate against too much confidence in the results as applied to actual driving. 8. At least half of cannabis users sometimes drive after cannabis use. Driving after cannabis use is more frequent among frequent users. 9. Absolutely nothing is known of the extent or recent (within an hour?), cannabis use among accident drivers, victims or pedestrians. 10. Some samples of cannabis users would appear to have higher violation rates than do nonusers. 11. It is uncertain whether cannabis users have higher accident rates (per unit of exposure) than do non-users. The importance of social and personality factors exclusive of drug use have not been assessed. 12. It would appear that marijuana on its own contributes to very few accidents or charges among college students as most such occasions also involve prior alcohol use. 13. Part of the apparently low frequency of marijuana accidents may be due to the infrequency of marijuana and driving occasions. 14. Countermeasures against cannabis and driving are probably not needed at this time. Impaired driving after cannabis use is illegal in Many western countries at present under general 'drugs and driving' legislation. Without an easy roadside testing device a credible countermeasure is not likely to be developed. SUGGESTIONS FOR FURTHER RESEARCH These suggestions are almost limitless but a few seem particularly important: 1. Developmental research on reliable roadside cannabis testers whose results correlate with impairment will be needed for many driving studies. 2. It would be of interest to study the type of accident occurring under cannabis effects somewhat fewer high speed passing accidents would be expected for instance. 3. A large sample study should be made of cannabis swabs from a variety of accident victims including drivers, passengers, pedestrians and non-accident drivers. Some assessment of driving errors would also be needed in this study. 4. Interview studies of large groups of young people would be useful in assessing the frequency of driving-marijuana risk and the frequency of accidents. 5. Simulator and closed course studies of less experienced drivers under various conditions and drug combinations (especially with higher alcohol levels) would be of interest. 6. It would be of interest to see some studies of cannabis and driving risk in traditional cannabis using countries, eg. Jamaica, India, North Africa. There are suggestions of rather different effects of cannabis in different countries and most driving related studies have been done in Western countries. 7. Much cannabis research is concerned with decrements in performance. It would be of interest to overcome this bias in some driving research, perhaps on the facilitating effects of cannabis-induced anxiety reduction. It should be possible to look for beneficial cannabis and driving effects on some drivers under some circumstances.

REFERENCES Allentuck, S. and Bowman, K.M. The psychiatric aspects of marijuana intoxication. Amer. J. Psychiat., 1942,99, 248-251. Benjamin, F.B. The effect of marijuana on driving performance. In M.F. Lewis (Ed.) Current Research in Marijuana. Academic Press, New York, 1972. Berg, S.W., Fryback, J.T., Goldenbaum, D.M., Jones, R.K., Joscelyn, K.B., Michael, R.P., Potter, W.Z., Zabik, J. The study of possible influences of licit and illicit drugs on driver behaviour. Washington, DOT/HS,800,613, 1971, Borkenstein, R.F., Crowther, R.F., Shumate, R.P., Ziel, W .B., and Zylman, R. The Role of the Drinking Driver in Traffic Accidents. Bloomington, Indiana, 1964Caldwell, D.F., Myers, S.A., Domino, E.F., and Merriam, P.E.-Experimental studies of marijuana. Amer. J. Psychiat., 1968,125,379-384. Cappell, M.D. and Pliner, P.L., Volitional control of marijuana intoxication: a study of the ability to 'come down' on command. J. Abnorm. Psychol., 1973,82, 428-434. Clark, L.D. and Nakashima, E.N. Experimental studies of marijuana. Amer. J. Psychiat., 1968,125,379-384. Clark, L.D., Hughes, R. and Nakashima, E.N. Behavioural effects of marijuana. Arch. Gen. Psychiat., 1970,23, 193-198. Crancer, A. Predicting Driving Performance with a Driver Simulator Test. Washington Dept. of Motor Vehicles, Olympia, 1968. Crancer, A. and Quiring, D.L. Driving Records of Persons Arrested for Illegal Drug Use. Administrative Services, Report Oil, State of Washington, Dept. of Motor Vehicles, May 1968. Crancer, A., Dille, J.C., Wallace, J.E., Haykin, M.D. Comparison of the effects of marijuana and alcohol on simulated driving performance. Science, 1969, 164, 851-854. Edwards, D.W., Hahn, C.P. and Fleischman, E.A. Evaluation of Laboratory Methods for the Study of Driving Behaviour: the Relationship between Simulator and . Street Performance. American Institutes for Research, Washington, 1969. Fergenson, P.E. The relationship between information processing and driving accident and violation record. Human Factors, 1971, 13, 173-176. Ferguson, W.S. and Howard, W.L. Marijuana and drug use and highway safety - a survey of high school students in Virginia. Virginia Highway Research Council, Charlottesville, 1971. Frank, I.M., Hepler, R.S., Stier, S., Rickles, W.H., and Ungerleider, J.T. Marijuana, tobacco and functions affecting driving. Paper presented at American Psychiatric Assoc., Washington, 1971. Goode, E. The Marijuana Smokers, New York, Basic Books, 1970. Haines, L. and Green, W. Marijuana use patterns. Brit. J. Addict., 1970, 65, 347-362. Hollis ter, L.E. Marijuana in man: Three years later. Science, 1971, 21-29. Huntley, M.S. Alcohol influences upon closed-course driving performance. J. Safety Res., 1973, 3, 149-164. Kalant, M. Marijuana and simulated driving. Science, 1969, 166, 640. Kibrick, E. and Smart, R.G. Psychotropic drug use and driving risk. J. Safety Res., 1970, 2, 73-84. King, L.J. and Forney, R.B. The absorption and excretion of the marijuana constituents, cannabinol and tetrahydrocannabinol. Fed. Proc., 1967, 260, 540. Klein, A.W. Davis, J.H. and Blackbourne, B.D. Marijuana and automobile crashes. J. Drug Issues, 1971, 1, 18-26. Le Dain Commission. Cannabis: A Report of the Commission of Inquiry Into the NonMedical Use of Drugs, Information Canada, Ottawa, 1972.

Manno, J.E., Kiplinger, G.F., Scholz, N. and Forney, R.B. The influence of alcohol and marijuana on motor and mental performance. Olin. Pharmacol. and Therap., 1970, 11, 808815. Miles, C.G., Congreve, G., Devenyi, P., Gibbins, R., Marshman, J., and Rankin, J. A preliminary and tentative progress report on the marijuana project. in Le Dain, G., Cannabis, Information Canada, 1972. McGlothlin, N.H., Arnold, D.O. and Rowan, P.K. Marijuana use among adults. Psychiatry, 1970, 33, 433-443. Moser, B.A., Bressler, L.D. and Williams, R.B. Collection Analysis and Interpretation of Data on Relationships Between Drugs and Driving. DOT/HS, 800-648. National Highway Traffic Safety Administration, Washington, 1971. Moskowitz, H., Sharma, S. and McGlothlin, W. Effect of marijuana upon peripheral vision as a function of the information processing demands in central vision. Percept. Mot. Skills, 1972, 35, 875-882. Moskowitz, H. Laboratory studies of the effects of alcohol on some variables related to driving. J. Safety Res., 1973, 3, 185-199. Nichols, J. Drug Use and Highway Safety. DOT/HS-800-580, Washington, D.C., July, 1971. Rafaelson, 0.J., Bech, P., Christiansen, J., Christup, H., Nyboe, J., and Rafaelson, L. Cannabis and alcohol: Effects on simulated car driving. Science, 1973, 179, 920-923. Robinson, A.E. Recovery of cannabis constituents from the hands at autopsy. Bull. on Narcotics, 1971, 23, 37-40. Sharma, S. and Moskowitz, H. Effect of marijuana on the visual autokinetic phenomenon. Percept. Mot. Skills, 1972, 35, 891-894. Smart, R.G. Marijuana and driving risk among college students. In Press. J. Safety Res., 1974. Smart, R.G. and Fejer, D. Marijuana use among adults in Toronto. Brit. J. Addict., 1973, 68, 117-128. Soehring, K. and Wolters, M.G. Pharmakologische Grundlagender Wirking von Arzneimittelnauf die Verkehrstuechtigkeit. (Pharmacological Principles of the Effects of Drugs on Driving Ability). In Wagner, K., and Wagner, J.H., (eds) Handbuch der Verkerhrswissenschaften, Berlin: Springer, pp. 854-883, 1968. Soueif, M.I. The use of cannabis in Egypt: a behavioural study. Bull. on Narcotics,1971, 23, 17-28. Stones, H.M. and Stevens, H.M. The detection of marijuana in the mouth and fingers of smokers. J. Forensic ScieRce Soc., 1969, 9, 31-34. Wallace, J.E., and Crancer, A. Licensing Examinations and Their Relation to Subsequent Driving Record. Washington Dept. of Motor Vehicles, 1968. Waller, J.A. Chronic medical conditions and traffic safety: Review of the California experience. New Engl. J.Med., 1965, 273, 1413-1420. Waller, J.A. and Goo, J.T. Highway crash and citation patterns and chronic medical conditions. J. Safety Res., 1969, 1, 13-27. Waller, J.A. Drugs and highway crashes: can we separate fact from fancy? J. Amer. Med. Assoc., 1971, 215, 1477-1482. Weil, A.T., Zinberg, N.E. and Nelsen, J.M. Clinical and psychological effects of marijuana in man. Science, 1968, 162, 1234-1242. DISCUSSION RESEARCH TOOLS

Dr Edwards suggested that the reason why the epidemological approach paid off in the Grand Rapids studies of drinking and driving was that they related level of dose to likelihood of accident. 'The fact that you could get beyond saying that people who had had alcohol were more likely to have an accident was very important, since the average drinker may in important respects be different from the non-drinker, and this personality difference may itself bear on accident proneness. The important finding in the alcohol studies was that the level of blood alcohol was related to the chance of an accident. But I can't see how this can be met in the forseeable future in the case of cannabis, because we lack a test of levels of cannabis in the body.' Much discussion centred on the likelihood of development of a roadside test for cannabis. Professor Paton reported that work in Oxford had established that in small animals, the levels of THC and its first metabolites were similar in the brain and in the blood. Because the vapour pressure of cannabis is so low, a test for cannabis would have to involve body fluid in distinction to the breath test for alcohol. Dr Miller pointed out that saliva could not form the basis of a test, since saliva would be directly affected by cannabis smoke. Dr Tinklenberg reported that body swabs and nasal swabs did not at present provide reliable evidence of recent use, let alone providing a test of levels in the body. Professor Paton thought that with the use of gas liquid chromatography, mass spectrometry and other specialised techniques, a test might be developed in a year or two for use in research laboratories but that it might well be ten years before a roadside test was available. It was agreed that a test would have to use biological fluids and there was some discussion about the difficulties of gaining driver's consent to blood tests. Dr Cahal suggested that if an appropriate reagent could be developed, then a spin assay technique would require only microlitres, or a pinprick. He agreed that a roadside test might be available in ten years.

Dr Smart pointed out that 'blood alcohol tests were available for a very long time before breath alcohol tests and they contributed almost nothing to accidents tests, simply because they were blood tests'. This experimental work may be more useful than roadside testing of drivers in determining the relationship between amount of cannabis consumed and accident proneness. TYPES OF ACCIDENT There was agreement that accidents in which cannabis was involved were characteristically low speed shunts and cross-road accidents, associated with lapses of attention, rather than being high speed passing accidents, as are some alcohol accidents. There is some experimental evidence that drivers drive slightly more slowly (one study found a 7 per cent lowering of speed) under the influence of cannabis. Also, cannabis users themselves state that they tend to drive more slowly when stoned. Dr Miller noted the question of speedometer errors and wondered whether illegality itself would make users more careful. Dr Somekh pointed out that some investigators had found a direct relation between level of dose of cannabis and ability to detect peripheral stimuli. This has been shown to be due to central processes, that is, a funnelling of attention rather than a tunnelling of vision. The importance of these findings, Somekh said, 'is that deficits may be more clearly shown in those tasks where processing capacity is being used up to its maximum.'

Smart pointed out that driving is not a very difficult task for most of the time. 'Most drivers have an enormous spare capacity for most of the time. Where you need your spare capacity is where someone comes out of a side street.' He also suggested that the cannabis user seems to be 'tuning in and out: he sees a lot of the stimuli, and doesn't see others.' OTHER DRUGS Dr Tinklinberg reported that in California there is an increase in accidents in seasons when there is a high pollen count, when many people are using antihistamines. Not only the antihistamines, but also their interactions with other drugs may be involved. He also reported that in the prison study conducted by the Stanford group, 'We have asked a question about driving under the influence of cannabis, and found that in this deviant group, most said that they did drive under the influence of cannabis. Most said they have had accidents, mostly under alcohol and barbiturates rather than under cannabis alone. But it is difficult to extrapolate from prison to other populations.' GENERAL Studies of different geographical areas at different times when the availability of cannabis differed would also be helpful. Epidemiological studies of motor vehicles in relation to bumps, scratches, etc might also furnish helpful data. Editorial footnote The World Health Organisation convened a meeting - of Experts in Geneva late 1973 on the subject of the detection of drugs in biological fluids. This report is shortly to be published, and includes consideration of cannabis.

2. Psychomotor and Cognitive Deficits Associated with Long- and Short-term Cannabis Consumption: Comparison of Research Findings and Discussion of Selected Extrapolations
M. I. Soueif, Professor and Chairman, Psychology Department, Cairo University. About four centuries ago the Arabic pharmacopoeia of Daoud Al-Entakui underlined, among other effects of cannabis, lethargy and sensory debilitation (Soueif, 1972). In the fall of 1957 we were just commencing our exploratory studies on the psychosocial aspects of chronic cannabis consumption. We were, then, struck by the frequency of remarks made by takers and non-takers as well, and by the abundance of hints in various kinds of folklore (eg. Arabian Nights) and anecdotes, emphasizing the psychomotor and cognitive dysfunctions usually shown by hashish users. In the pilot study we conducted from 1960 to 1962 administering a completely standardized interview (comprising about 400 items whose retake reliabilities were established) to 204 regular takers and 115 non-takers, we obtained a number of suggestive pieces of information. In Figure 2.1 we present a sample of

this information about the subjective effects of hashish, immediately after smoking and 24 hours later. A few points should be made clear here. 1. The users we interviewed for this pilot work were ordinary citizens. The average frequency they took the drug was 3 times a week, 1.08 gm. each time. Route of administration was usually smoking. When the kind of hashish they took was assayed for its THC content it was found to contain approximately 0.3 per cent by weight (Soueif, 1967). 2. Discrepancies in the same direction, though smaller in size, have been revealed on most of the items mentioned in Figure 2.1 when we interviewed about 350 chronic takers, who were all prison inmates, as part of our main study. This stability of the profile of findings emerged in spite of the fact that the latter group took the drug about 30 times a month. The findings of the pilot study induced us to include among our tools of investigation, a number of objective psychological tests for the quantitative assessment of certain aspects of performance (Soueif, 1971). With the sampled background of exploratory information in mind, and being aware of the main aim of our project, viz, the study of the psychology of chronic cannabis consumption, the problem we were to tackle was, then, formulated as follows:

Fig.2.1. Under the immediate effect of cannabis and 24 hours later: subjective reports. Are there significant differences between chronic takers and non-takers on the specified test variables? In this paper my intention is to present part of our data on the performance of chronic cannabis takers on objective tests, compare such data with recently published information on analogous test performance displayed by subjects who were under the immediate effect of the drug, integrate both kinds of reports within a tentative theoretical framework and discuss extrapolations for future research. I We administered 16 test variables to 850 chronic takers and 839 non-takers. All were males ranging in age between 15 and slightly over 50 years. Our takers comprised the whole population of prison inmates that were exclusively convicted for cannabis use during the

period from mid-1967 to about March 1968. Controls were selected from among inmates of the same prisons. The two groups were reasonably comparable regarding the relevant independent variables such as literacy, urbanism, socio-economic status and history of psychiatric complaint. Analysis of variance based on 3 x 3 x 2 factorial design was carried out, to supply 3 comparisons and 4 interactions for each test variable. Out of this complex matrix I am presenting in Table 1 F ratios between users and non-users irrespective of literacy and residence.

Chronic takers were definitely slow on simple psychomotor tasks as shown on tests 2,3 and 5. But when the speeded task required fine finger movements combined with change of orientation in the visual field, as was the case in test 4 (Mark Making) takers did not differ significantly from non-takers, though the latter still tended to show superiority. Again users were much below the average for controls on the Bender Gestalt (Lofvinger's method) which requires psychomotor activity scored for accuracy but not for speed. Tests 1 and 6 involve speed of some psychic processes without appreciable motor involvement. Tool Matching measures speed of perceptual discrimination and it sorts out takers from non-takers at a high level of confidence. Initial Reaction Time requires speed of

verbal reaction. This test failed to differentiate between the two groups though inspection of the cells showed that takers tended to react slowly compared with controls. The remaining tests in Table 2.1 are rather cognitive. We found that cannabis takers tended to over-estimate moderate distances (a few centimeters), while non-takers tended to underestimate such distances. When we analysed the data regardless of direction of the error the two inclinations cancelled each other. The result on time estimation was contrary to common expectation. Average time estimation for a period of 3 minutes (unoccupied by any overt behaviour) did not differentiate between users and non-users. When we analysed the differences between time as estimated by our subjects and geophysical time we found that the size of errors made by takers was smaller than that made by non-takers. The Digits Forward, as a test for the span of immediate memory for numbers did not differentiate between users and controls. But the Bender Gestalt Recall which measures the subjects' span of memory for simple geometric designs discriminated between the two groups significantly. The Digits Backward, which gauges the capacity for mentally holding and reorganizing a number of items marked a highly significant difference between users and non-users. This is all about Table 2.1. To summarize these findings in one comprehensive formula: The failures of chronic takers become more pronounced the more speeded motor activity combined with visual discrimination is required for the task they perform. It appears, on the basis of our observation, that within this paradigm the motor component is assigned more weight than the visual-perceptual. II The question now is, how do these findings on the chronics compare with the rather corresponding results recently reported on the acute effects of the drug. In a series of experimental studies carried out and reported between 1968 and 1973 a number of findings were presented which make such comparison possible though conclusions should be drawn with due caution. The following sources of variability within this group of experiments, and between them on the one hand and our study on the other, set limitations to what we can make out of this comparison: 1. The wide differences between drug preparations utilized by various investigators (eg. marihuana; marihuana extract; synthetic THC). 2. The wide range of doses from one experiment to the other. 3. The different methods of administration (being sometimes given orally, and sometimes by smoking). 4. The rather narrow sector of the society to which subjects in most experiments belonged, being either students or graduates. 5. The fact that in some experiments subjects were drug-experienced while in others they were naive. 6. The setting, being always the secure but very arid lab environment.

Clark and Nakashima (1968) gave marijuana extract to a group of educated naive subjects, and used them as their own controls. Those investigators reported no consistent drug effect on the pursuit rotor. But they found that both simple and complex hand and foot reaction times increased under drug effect. They, also, found that the magnitude of increment was positively dose related. Weil and colleagues (1968) gave marihuana to 9 naive university students to smoke. The subjects were also given placebo. The investigators report significant decrements in performance on the pursuit rotor when under marihuana. They also demonstrated a dose-response relationship. Hollister (1971) gave oral doses of synthetic THC to a group of subjects and found that they showed reduced accuracy on a drawing test which was rather difficult and with which the testees were not familiar. No slowing of performance was shown. Hollister, however, did not make it clear whether the task was explicitly speeded or not. Manno and associates (1970; 1971) reported that impairment on pursuit tracking performance would occur under the effect of cannabis smoking whether the smokers were naive or drug-experienced. When the drug was administered on a body weight basis a dose-dependent relationship could be demonstrated. Forney and collaboraters (1971) also demonstrated impaired performance on pursuit meter under drug effect. The Canadian Commission of Inquiry (1972) carried out a number of experiments in which they administered placebo, marihuana and synthetic THC to some 14 university students who were experienced with the drug. The Commission reported no effect of (smoked) cannabis on maximum tapping speed. The Commission's report mentions Bowman's work on Jamaican chronic heavy users pointing out that he found no evidence of significant psychomotor impairment. (1972, p.59). So much for the assessment of psychomotor performance under the effect of cannabis. In summary, most of the work, that came to our knowledge, was confined to one main area, that of visual-motor coordination or accuracy of performance. Apart from the Commission's work on tapping speed and one or two investigations on simple and complex reaction time, speed of performance was seldom amply investigated. Yet this parameter has been repeatedly demonstrated by a whole series of clinicians to be significantly correlated with psychiatric disorder. Babcock (1930; 1933) first demonstrated that performance on error-free psychomotor speed tests-was abnormally impaired in schizophrenics (A. Yates 1973). Shapiro and associates (1955) concluded that psychiatric patients were slowest in tasks which did not involve problem solving and which emphasized speed in the instructions. A few years later Payne and Hewelett (1960) confirmed those findings in a well designed large scale study (A. Yates, 1973). Indeed this part of the clinical literature, was among the sources that inspired us to enquire into the possibility of hashish takers showing any aspect of psychomotor retardation. Not that we imply that the mechanisms operating to bring out psychomotor slowness should be the same in cannabis users as well as in cooperative psychotics. Yet the research worker cannot and should not ignore some analogies. Moreover, the analogy we have adopted here is not in disharmony with Isbell's finding that 2THC can be psychotomimetic. (Nahas, 1973; p.171). As to visual-motor coordination, the general trend of the above cited results was that impairment could be detected. This conclusion is in line with our findings on the Bender Gestalt (Copy). Another conclusion one may draw from the literature pertains to the positive dose-response relationship which could be demonstrated on tasks requiring hand-eye coordination. We investigated this point in a slanted way in our chronics. Two separate comparisons were made between users who took the drug less than 30 times a month and those who took it 30 times or

more. The first comparison was carried out on those chronics who followed this schedule from one to five years. For a one tail test the difference was significant beyond .05 level showing moderates to be better on the Bender Gestalt Copy than heavy takers. We made another totally independent comparison between those who took the drug at the above mentioned frequencies for more than 20 years. The difference, on the same test, at this level of chronicity, was almost nil. Table 2.2 shows the outcome of comparisons.

At face value one might conclude that the obliteration of the difference between the two latter groups was a function of the duration of drug use as such. The implication of such conclusion would then be in agreement with Weil and Colleagues' (1968) who reported that whereas naive subjects' performance on the Digit Symbol Substitution was impaired, chronics started with good baseline and improved slightly. We feel, however, that a more adequate explanation is, still, needed. It should take into account the following two facts: (a) Opium taking was found to be positively correlated with duration of hashish consumption. Figure 2.2 shows this relationship. (b) It was also positively related to heavy cannabis use. (t = 2.16). (Soueif, 1971). We have recently discovered that among users those who took opium over and above cannabis (14 = 243) obtained a much better score on the Bender Gestalt (Copy) than cannabis users who did not take opium. The difference was highly significant. gable 2.3).

This seems to shed light on the question why the dose-response relationship disappeared in our chronics when we moved from the 5-year group to the 20-year group.

It may be that, in the latter group, opium acts as an antidote to cannabis effects. Or it may be, basically, the question of who selects what as the drug or drugs of choice; namely that those who tend to take opium in addition to cannabis have an initially better stand on measures of accuracy of performance. III We shall now consider the literature on the psychometric assessment of acute effects of cannabis on cognition.

Time estimation and short-term memory have been favoured as topics of research among investigators. But with distance estimation we could hardly find anything in the experimental literature apart from a hint concerning the LaGuardia group of researchers. They reported that marihuana produced no significant changes in perception of length of lines on paper (Clark and Nakashima, 1968). Weil and colleagues (1968) reported a tendency among naive subjects (N = 9) to overestimate a period of 5 minutes, when under the effect of smoked marihuana (containing 0.9 per cent THC by weight). Tinklenberg and associates (1972) gave orally social doses of marihuana extract calibrated for THC content (0.35 mg./kg.) to 15 subjects who were all moderate cannabis takers. Placebo was also given within the framework of a double blind design. Time perception was guaged by the technique of time production. The investigators found that the drug induced underproduction on this test compared with placebo. In another experiment by Hollister and Tinklenberg (1973) the same test was used. But this time there were no significant differences between time-production under marihuana and that under placebo. Collating the results obtained by those three groups of researchers together with ours, no consistency could be detected. A number of tests have been utilized to assess various aspects of immediate memory. Relevant among those tests are the Digits Forward and Backward and the Goal Directed Serial Alternation. Melges and others (1970) and Waskow and collaborators (1970), both groups administered Digits Forward and Backward to volunteers under the immediate effect of the drug and of placebo, within the framework of a double blind design. They came out, however, with contradictory results. Melges' group demonstrated significant impairment for both functions. Waskow's group concluded that the drug had no significant effect on either. With the Goal Directed Serial Alternation we were not in a better position. Melges and his group (1970) reported progressive impairment of performance with increased doses of THC. But Tinklenberg and co-workers (1972) and Hollister and Tinklenberg (1973) reported no significant deterioration of function under drug effect. Tinklenberg and associates (1972) added another test labeled Running Memory Span, a variation of the Digits Forward. This test, again, did not reveal any significant effect of the drug. Obviously, good experimental research has been invested in the area of time perception and short term memory. But the results, taken at face value, cannot be meaningfully integrated together, nor can they be reconciled with our corresponding findings on the chronics. However, initial individual differences on related functions can be utilized here. I propose the following as a working hypothesis: The amount of deterioration that is likely to be effected by the drug is a /Unction of the general level of predrug performance: other conditions being equal, the lower the initial level the less deterioration. It is probably the general case behind what Dr Tinklenberg and his colleagues called the initial level of proficiency on the tested variable (1972).

You might well recall that Waskow and co-workers studied a group of criminal offenders whose average I.Q. was 95. They got no significant impairment on Digits Forward nor on Digits Backward (1970). Contrasted with that was Melges and his colleagues who worked on a group of graduate students and got significant results (1970). Relevant here is the fact that the Digits Span test correlated up to 0.51 with the total score on the Wechsler (Wechsler, 1954, p.85). According to the same working hypothesis, conflicting results concerning the Goal Directed Serial Alternation Test could also be resolved (rinklenberg et al., 1972). The same idea may shed light on some of the negative results obtained by Bowman on cannabis takers among Jamaican natives (Commission's Report 1972, p.56). Possibly those natives were initially of low level of proficiency. It is interesting to note that the same idea was mentioned, though en passant, by R.W. Payne in a totally different context. Reviewing the work, done in the sixties, on intellectual deterioration under conditions of mental illness Payne found that he was confronted with two groups of studies: Group A, which ended up with 'patients who had originally above average or average I.Q. and showed significant deterioration, and Group B, concentrating on patients who were initially subnormal and displayed very little deterioration. He drew the following conclusion: 'It is possible to speculate that the dull who had become psychotic deteriorate little intellectually'. (Payne, 1973). Indeed this same paradigm could impose form and meaning on a sizeable amount of data we collected on our takers. The following two sets of derived hypotheses, each comprising two mutually complementary predictions, were formulated: Set A: Prediction 1: We would expect performance on simple psychomotor and cognitive tests to correlate significantly with literacy (up to a certain level). If this was the case, Prediction would then be: The lower the level of literacy, the smaller the size of impairment associated with drug taking. That the two predictions were in fact borne out can be readily seen in Tables 2.4 and 2.5. Table 2.4 shows the outcome of analysis of variance for the data on the same tests mentioned before; this time between 3 levels of literacy irrespective of drug use.

Except for 'Distance over-estimation' F r atios were invariably very highly significant. Table 2.5 presents 3 columns of t values. Each column shows the outcome of comparison between means (or medians) earned by takers and controls within each of 3 levels of the literacy-illiteracy continuum. Set B: This was analogous to set A except for the fact that it hinged upon the parameter of urbanismruralism instead of literacy. Prediction 1: We would expect performance on simple psychomotor and cognitive tests to be significantly related to urbanism. In this case, Prediction 2 would be: The less urbanized cannabis takers were, the smaller the degree of impairment correlated with drug taking.

These two predictions were also confirmed, as can be seen in Tables 2.6 and 2.7.

By and large, our hypotheses were confirmed. Other predictions may also be worked out to amplify our basic formula. For example, we would expect the same pattern of findings to emerge in the area of creative thinking abilities. This means that, contrary to common belief, the highly creative per ns would show more impairment than the mediocre under drug effect. Research in creativity has been making tremendous progress during the last 20 years. To test this prediction the research worker can find available a bit number of good tools for the measurement of factorially purified relevant parameters such as originality, flexibility, ideational fluency and maintaining direction (Soueif and S. Farag 1971; Freeman et al. 1971). Another example still, concerns the difference between age groups. We would expect young cannabis takers (-25 yrs.) to be more seriously affected than older age groups (40+). Relevant data is in the process of analysis. The implications of this theory for any policy making, concerning prevention, education, legislation or direction of future research are quite obvious. The main point would always be

the fact that, citizens who lie within the higher brackets of ability (in the broadest sense of the word) are likely to get the greater damage, when exposed to the effects of cannabis use. The question now is: Where do we go from here? At face value the present state of information concerning the behavioural effects of cannabis consumption suffers from two serious difficulties: gaps and conflicting results. To overcome the consequences of such a situation we may have to choose between two alternatives marking the two opposite poles of a continuum: empiricism or rationalism. An empirically oriented approach would aim at symptomatic treatment of the situation and would suggest simultaneously adopting two lines of action: 1. Filling in the gaps: For example: in spite of the fact that epidemiological reports make it clear that about one third of cannabis takers in some Western countries were females we had almost no information on behavioural sex differences under cannabis. This area, therefore, should be covered. Another area would be speed of psychomotor performance. A third would be speed of intellectual processes. A fourth would be the capacity for distance estimation. A fifth size estimation. A sixth creative thinking ... etc. One could go on counting a number of such relevant areas. 2. Resolving sources of conflict in research findings: One would recommend increasing the number of subjects tested in the various experiments in order to permit individual differences to play their role fully. Or, one would recommend explicitly stating the initial scores of the subjects on specified variables without the investigator being bothered by the question of the size of N. Such an empirically oriented approach, however, will suffer from the shortcomings of any post-hoc framework, no matter how finely its details were worked out. It will always remain remedial, but not preventive; always an after-thought, never a guiding insight. The need for a more rational approach, therefore, seems compelling. In preparing a rational plan of work, field and lab studies accomplished so far, may be viewed as explorations, suggesting an adequate scheme for a well integrated study of the problem. It would be advisable to give optimum chance at the initial stage of developing such a scheme, to all relevant questions to be raised, by simply enumerating all points already investigated, suggested for further research or mentioned by way of speculation. A conference or a circular letter of enquiry for the brainstorming around this inventory of problems should follow. We should then try to obtain a more or less comprehensive list of queries including instructive hints on priorities. If priority is given to a selected group of items (research and applicationwise), a well balanced programme for a broad attack on the problem in its genuine complexities can then be considered. Three main requirements, at least, are necessary in such a programme: (a) it should make use of standardized and calibrated tools for the assessment of drug effects, (b) it should allow for the systematic study of significant individual differences, (c) it should permit transnational and transcultural integration of findings. When we select tools for the objective study of drug effect, they should be representative of a whole universe of tests or scales for the assessment of a certain function (or group of intercorrelated functions). When objective assessment is mentioned, there is always the

hazard of research-workers becoming test-minded rather than function-minded. The trouble with some tests is that they have rather narrow common variance. A sensible rule in factor analysis is never to identify a factor by less than three variables. It may add to the accuracy and meaningfulness of our work to start by explicitly defining what functions to measure, then select and/or devise more than one test variable to use for the measurement of each function. Individual differences, with their multiplicity and diversity, have always been among the main sources of variance in the experimental study of human behaviour. Nevertheless, the systematic handling of this variate is still far from attaining the status of an essential ingredient in the design of experiments on human behaviour. An adequate formula has to be worked out to help future researchers select the right dimensions that would act as good intervening variables to impose optimum regularity on whatever dependent variables are considered for investigation. Sources of inspiration to serve this purpose may be clinical psychiatric chapters dealing with premorbid personality and with indicators of prognostic value. A useful rule is to consider promising intervening variables those dimensions which have been demonstrated (usually in other areas of research) or thought on explicitly stated grounds, to correlate with the variables under study. It would serve our purpose in the long run, to include, among our tools of investigation a minimum of tests that would be 'fair' to a wide variety of cultures. The academic problem of whether there are tests which can really be 'culture-fair' should not disturb us much. What we actually need is to use a number of test variables that can with minimum modification be administered to members of various cultural groups. The research worker should always be able to distinguish between two aspects of human behaviour: action and significance, or the overt and the covert components. The field of attitude research teaches us a good lesson about how intricate the relationship between these two ingredients is. Briefly stated, they should be treated as two separate variates, which sometimes, but not always, are intercorrelated (Dillehay 1973). Speed of psychomotor performance, proficiency of immediate memory, accuracy of fine movements ... etc.; those are demonstrable parameters of overt behaviour which can and should be investigated in their own right. The meaning of the testing situation to an average English= man might differ from what it looks like to an average Egyptian, but that would not negate the fact that relative inaccuracy of fine movements could be shown in both English and Egyptian takers. Such solid facts should be recorded on the widest possible scale without waiting for the problem of the covert component to be solved. It seems quite feasible and indeed should prove rewarding to plan straight away for comparative studies based on the systematic variation of total cultures (or well defined cultural parameters) together with degrees of permissiveness towards the drug. We may need to remind ourselves, all through this kind of work, of the following statement made by Cluckhohn 20 years ago: 'It is true., that for roughly half a century now, anthropologists have concentrated preponderantly upon.. culturally relative behaviors. Nevertheless, there has been more effort at stating and testing universal hypotheses than many psychologists recognize.' (C. Kluckhohn, 1954). True, correction terms will, in all probability, be needed. The safest way to secure such terms is usually to present users' scores in the light of norms for local controls. Meaningful comparisons across national or cultural boundaries can, then, be made in terms of some kind of standard scores. This procedure, however, cannot be a guarantee against the absurdities of unwarranted conclusions. A measure of caution on the part of the investigator is always indispensable.

We must remember that cannabis dependence is the end result of an international activity with many forms of clandestine cooperation to perpetuate the phenomenon. Granting that our scientific efforts in this field are, ultimately, steered towards utilitarian targets, they should be run on an equally broad basis of international collaboration to attain comparable efficiency. True, we are, in a sense, working here on such lines. Our collaboration at the moment, however, takes the form of discussing work already accomplished and some extrapolations therefrom. Something more drastic is needed. We should start designing and carrying out an international project that requires transnational division of labour. How to operate this in actual fact? Probably, the correct approach would be to allow, right from the beginning, for more than one solution to meet this demand. The real challenge is to keep flexible, by drawing up a repertoire of alternative solutions, but never losing direction to the main target. Cannabis consumption is endemic in one part of the world; there, we can find a wealth of data just waiting to be collected and processed. Apart from cultural differences as filtered into personality, those chronics presumably represent the clinical picture of the user we may have to deal with, in the very near future, in other parts of the world where chronicity is just emerging. REFERENCES Babcock, H. and Levy, L. Manual of Directions for the Revised Examination of the Measurement of Efficiency of Mental Rinctioning. Chicago: Stoeling. Clark, L.D. and Nakashima, E.N. Experimental studies of marihuana, Amer. J. Psychiat., 1968, 125/3, 379-384. Commission of inquiry into the non-medical use of drugs: Cannabis, Ottawa, 1972. Dillehay, R.C. On the irrelevance of the classical negative evidence concerning the effect of attitudes on behavior, Amer. Psychologist, 1973, 28/10, 887-891. Forney, R.B. Toxicology of marihuana, PharmacoZ. Rev., 1971, 23, 279-284. Freeman, J., Butcher, H.J. and Christie, T. Creativity: A selective review of research, London: Society for Research into Higher Education, 1971. Hollister, L.E. Marihuana in man: Three years later, Science, 1971, 172, 21-28. Hollister, L.E. and Tinklenberg, J.R. Subchronic oral doses of marihuana extract, Psychopharmacologia, 1973, 29, 247-252 (memeographed). Kluckhohn, C. Culture and behavior, Handbook of social psychology, G. Lindzey ed., Cambridge, Mass.: Addison-Wesley, 1954, 921-976. Manno, J.E., Kiplinger, G.F., Haine, S.E., Bennett, I.F. and Forney, R.B. Comparative effects of smoking marihuana or placebo on human motor and mental performance, Clinical Pharmacology and Therapeutics, 1970, 11/6, 808-815. Manno, J.E., Kiplinger, G.F., Scholz, N. and Forney, R.B. The influence of alcohol and marihuana on motor and mental performance, Clinical Pharmacology and Therapeutics, 1971, 12/2, 202-211.

Melges, F.T., Tinklenberg, J.R., Hollister, L.E. and Gillespie, H.K. Marihuana and temporal disintegration, Science, 1970, 168, 1118-1120. Nahas, G.G. Marihuana, deceptive weed, New York: Raven, 1973. Payne, R.W. Cognitive abnormalities, Handbook of abnormal psychology, H.J. Eysenck ed., London: Pitman Med., 1973, 420-483. Payne, R.W. and Hewlett, J.H.G. Thought disorder in psychotic patients. Experiments in personality, vol. II, Psychodiagnostics and psychodynamics, H.J. Eysenck, ed., London: Routledge and Kegan Paul, 1960, 3-104. Shapiro, MB., Kessell, R. and Maxwell, A.E. Speed and quality of psychomotor performance in psychiatric patients (Memeographed copy, 1955). Soueif, M.I. Hashish consumption in Egypt, with special reference to psychosocial aspects, Bulletin on Narcotics, 1967, 19/2, 1-12. Soueif, M.I. The use of cannabis in Egypt: A behavioral study, Bulletin on Narcotics, 1971, 23/4, 17-28. Soueif, M.I. The social psychology of cannabis consumption: myth, mystery and fact, Bulletin on Narcotics, 1972, 24/2, 1-10. Soueif, M.I. and Farag, S.E. Creative thinking aptitudes in schizophrenics: A factorial study, Sciences de l'Art Scientific Aesthetics, 1971, 8/1, 51-60. Tinklenberg, J.R., Kopell, B.S., Melges, F.T. and Hollister, L.E. Marihuana and Alcohol, Arch. Gen. Psychiat., 1972, 27, 812-815. Waskow, I .E., Olsson, J.E., Salzman, C., Katz, M.M. and Chase, C. Psychological effects of tetrahydrocannabinol, Arch. Gen. Psychiat., 1970, 22, 97-107. Wechsler, D. The measurement of adult intelligence, Baltimore: Williams and Wilkins, 1944. Weil, A.T., Zinberg, N.E. and Nelsen, J.M. Clinical and psychological effects of marihuana in man, Science, 1968, 162, 1234-1242. Yates, A.J. Abnormalities of Psychomotor Functions, Handbook of abnormal psychology, H.J. Eysenck, ed., London: Pitman Med., 1973, 261-283. DISCUSSION Professor Soueif's paper excited considerable interest and discussion, mostly on methodological considerations. Dependence Dr Soueif was asked about his use of the term 'cannabis dependence', and explained that he was adopting a WHO convention in using the term. The user group in his study were

operationally defined by their conviction for a cannabis offence. For him cannabis dependence was synonymous with extended periods of cannabis use: in spite of the constraints against use and the threat to the organism, many people did use cannabis for extended periods, and many reported that they could not stop. ACUTE VERSUS CHRONIC EFFECTS Dr Edwards stressed the distinction between effects of intoxication, and long-term effects of previous use: 'It is important with many sedative-type drugs to ensure that a sufficiently long time period occurs between cessation of use and testing. Do you have any information about this time interval in your study?' Professor Soueif explained that 'we were pretty sure that some of those prisoners didn't stop taking cannabis while they were in prison. So they might have been using cannabis the day before they were tested.' GENERALISABILITY OF RESULTS Mr Hasleton pointed out that 'if one takes groups in prison, then one is not taking a sample from which one can generalise to other samples other than other groups in prison. So if prisoners who have used cannabis differ in various ways from prisoners who have not, it does not necessarily follow that non-prisoners who have and have not used cannabis will differ from one another in the same ways.' Professor Soueif stated that he was not attempting to generalise to other populations. MAGNITUDE OF OBSERVED SCORE DIFFERENCES Several participants were concerned to know the magnitude of the differences between the test scores of users and non-users. Dr Smart remarked "I don't get any feeling from your paper or your remarks of the size of difference you are talking about. In two of the tables the differences that come out significant are really quite tiny. When we're talking about the social policy-implications of the differences, I think that we should know their magnitude.' In response to this and other similar questions about size of differences, Professor Soueif suggested that differences were always a function of the test used, that the means and standard deviations had been published (Bull. Narc. XXIII, 4, 1971), that he did not think it would be appropriate to present detailed data to a large seminar group, and that even small differences were important. Later tests could be developed to find bigger differences, 'the point I am trying to make is that there are significant differences. Are we ready to ignore that?' QUESTION OF EQUIVALENCE OF GROUPS: MATCHING Dr Miller: 'The real significance of these differences, whatever their sizes, depends on how closely you were able to match the experimental and control groups. I wonder if you could give us a list of the specific operational variables on which they were matched and tell us whether or not you ran the same kinds of statistical tests on those matching variables between the two groups.' Professor Soueif: 'What can I say, globally, is that they were similar on a number of variables, and there were also some dissimilarities.' Dr Miller asked if Professor Soueif had tried to correlate the group differences in matching variables with the test score differences, and Professor Soueif said that they were looking

again at the data and might do this. SAMPLE SIZE AND MEANING OF STATISTICAL SIGNIFICANCE Dr Miller suggested that statistical significance, with sample sizes of about 800, must be interpreted with caution, especially bearing in mind the fact that the samples were not exactly equivalent in terms of matching variables. He suggested that two groups of only one hundred subjects might be taken, well matched on a variety of variables, and that the differences in test score between these groups be looked at. In conclusion, it is fair to say that some of the participants felt that insufficient data had been presented to justify full acceptance of the conclusions drawn.

3. Psychological and Cognitive Effects of Cannabis

Jared R. Tinklenberg and Charles F. Darky, Department of Psychiatry, Stanford University School of Medicine. 1. INTRODUCTION Descriptions of the psychological effects of cannabis have been found in Eastern documents and archaeological artifacts dating back to before the time of Christ. Throughout the subsequent centuries writers from various parts of the world have described, at times with remarkable clarity, the subjective and behavioral changes induced by this drug. In recent years, with the application of sophisticated experimental methods, especially the precise quantification of g-tetrahydrocannabinols, the chemicals in cannabis that exert most of the psychoactive effects, our understanding of the drug's psychological and cognitive effects has increased. This paper provides an overview of the present state of knowledge, describes this information in terms of conceptual model, and suggests directions for future cannabis research. 2. BASIC CONSIDERATIONS The effects of cannabis on psychological and cognitive processes are influenced by a number of variables. Since these variables have been discussed in most recent reviews on cannabis, only a brief summary will be presented here. Recent experiments have demonstrated that when reliable quantities of smoked cannabis or tetrahydrocannabinol (THC) are delivered to subjects reproducible dose-dependent and timeaction effects occur with many psychological and cognitive tasks (Casswell and Marks, 1973; Isbell et al., 1967; Kiplinger et al., 1971; Lemberger et al., 1970; Manno et al., 1971; Renault et al., 1971) . As with most psychoactive drugs, larger doses of cannabis produce greater and more prolonged psychological effects. In addition to dose, the route of administration-smoked, oral, or injected--also influences the effects of cannabis (Hollister, 1971). With smoking or injection, the onset of psychological effects is almost immediate'and the total duration of action, depending upon dosage, is usually less than 3 to 4 hours. Drugs effects

from ingested THC are qualitatively similar but differ quantitatively as a result of individual variations in intestinal absorption, the material in which the THC is dissolved, and the phase of the time-action curve (Perez-Reyes et al., 1973). The peak intensity of ingested cannabis is usually only 1/3 or 1/2 of an equivalent smoked or injected dose. The individual's previous pattern of cannabis use is another important variable that influences the psychological and cognitive effects of the drug. With repetitive use, tolerance appears to develop to certain of the behavioral effects, that is, some performance deficits with acute intoxication seen in occasional users not demonstrated in regular users (Meyer et al., 1971; Weil et al., 1968). Although early observations indicated the possibility of increased pharmacological sensitivity to the subjective effects of cannabis with repeated use--so-called 'reverse tolerance'--subsequent studies have provided strong evidence of the development of tolerance to the psychological and cognitive as well as physiological effects of cannabis (Hollister, 1974). The enhanced sensitivity to subjective effects reported by chronic users probably results from learned responses rather than from any direct drug effect (Jones, 1974). The subjective and behavioral effects of cannabis are significantly influenced by 'set'--the attitudes and expectations of the individual user derived from mood, personality, and past drug experiences, and 'setting' the physical and social environment in which drug use takes place. These non-pharmacological variables are especially important at low doses: however, at higher doses pharmacological properties become progressively more dominant. These basic considerations should be remembered in reviewing the psychological and cognitive effects of cannabis described in the following sections. 3. SUBJECTIVE EFFECTS AND EXPERIMENTAL EVIDENCE The subjective and behavioral effects of cannabis depend heavily on the basic factors just discussed and hence vary considerably among individuals. With low dose intoxication, the most commonly reported subjective effects are initial euphoria, increased hilarity, mild restlessness, exhilaration, and later a dreamy, carefree state of relaxation. Although these pleasant effects are frequently reported, some users experience various degrees of anxiety and other dysphoric feelings. Altered time sense is one of the most consistent and marked perceptual effects of cannabis. Even with low doses, users almost invariably report that there is an acceleration of subjective time so that geophysical time seems to pass slowly. This effect, which may be interpreted as a speeding up of the hypothetical 'internal clock' so that 'inner time' is fast compared to geophysical time, has been extensively documented with experimental evidence (Canadian Commission, 1972; Cappell et al., 1972; Clark et al., 1970; Hollister and Gillespie, 1970; Hollister and Tinklenberg, 1973; Karniol and Carlini, 1973; Tinklenberg et al., 1972). Perceptual experience arising from stimulation of sensory receptors also seems altered during cannabis intoxication. Enhanced tactile, olfactory, and gustatory perceptions are often reported. These effects may coalesce to produce the increased hunger, or greater interest in food, often reported during cannabis intoxication. Increased appreciation of the subtle nuances of sound and visual patterns are commonly noted and dose-related increases in subjective reports of imagery have been reported, including increased vividness, clarity, colour, movement, depth, and complexity (Canadian Commission, 1972). Despite these persistent subjective reports of cannabis-induced enhancement of sensory-perceptual processes, objective measures have demonstrated either no change or slight decrements in perceptual

functioning (Caldwell et al., 1969, 1970; Moskowitz et al, 1972; Roth et al., 1973). In other words, there is no firm evidence of changes in peripheral sensory thresholds or actual enhancement of perceptual acuity; instead, any increases in perceptual awareness seem to occur at a subsequent stage of processing. In addition to these subjective effects which entail primarily alterations in the basic sensory modalities, cannabis induces a wide variety of other subjective changes (Tart, 1970). New meanings are found in commonplace events; prosaic ideas seem original. Although many users report that their thinking does not follow logical patterns, they nevertheless feel their reasoning is intuitively correct. Thinking in abstract symbols is reduced while visual imagery is often increased and many users report an increase in spontaneous visual imagery while reading. With higher doses of cannabis, these subjective effects are intensified. The individual may experience rapidly altering emotions, fragmentation of perceptions, an altered sense of self-identity and feelings of enhanced insight. Introspection and social withdrawal are often marked. Synesthesias, the imagery of one sense mode aroused by stimulation of another sense mode, have been reported; for example, the individual may experience colours in response to an auditory stimulus. Dream-like fantasies and vivid hallucinations may occur, usually in a discontinuous or undulating pattern. Laboratory experiments with high doses of THC have produced perceptual phenomena very similar to those induced by LSD including distortions of body image, extreme perceptual aberrations, and a dissolution of personal identity (Isbell et al., 1967; Melges et al., 1970a; Melges et al., 1974; Perez-Reyes et al., 1973). Experienced users usually describe these effects as pleasant, especially when the factors of set and setting are non-threatening and conducive to a pleasant experience. However, the same cannabis effects may also be interpreted as unpleasant, particularly by neophytes who may fail to identify the subjective and behavioral alterations as transient. Severe anxiety and panic may result when inexperienced users interpret the drug state as signs of permanent psychosis or impending death. 4. CANNABIS AND PSYCHOMOTOR TASKS Subjective reports by cannabis users indicate that many feel as if their ability to drive vehicles or perform other complex motor tasks is reduced, but that they are usually able to compensate for any impairment (Hochman and Brill, 1971). Laboratory studies have confirmed that performances on a wide range of psychomotor tasks are impaired during cannabis intoxication. Cannabis has induced deficits in tapping speed, handwriting, simple and complex reaction time, pursuit motor and tracking tasks, continuous performance tests, and driver simulation studies (Clark and Nakashima, 1968; Clark et al., 1970); Dornbush et al., 1971; Galanter et al., 1972; Hollister and Gillespie, 1970; Kiplinger et al., 1971; Manno et al., 1970; Melges et al., 1970b; Mendelson and Meyer, 1972; Meyer et al., 1971; Rafaelson et al., 1973; Roth et al., 1973; Volavka et al., 1971; Weil and Zinberg, 1969; Weil et al., 1968). Previous use of cannabis may also significantly influence performances on psychomotor tasks. At low doses, less experienced users generally show greater impairment on psychomotor tasks than do experienced users (Clark and Nakashima, 1968; Jones, 1971; Jones and Stone, 1970; Mendelson and Meyer, 1972; Meyer et al., 1971; Weil and Zinberg, 1969). How much this phenomenon represents the ability of experienced users to volitionally compensate for drug effects and how much it represents other processes, such as the development of tolerance at a pharmacological level, is difficult to determine. 5. MEMORY - COGNITIVE EFFECTS

One of the most commonly reported effects of cannabis intoxication is difficulty in performing tasks in which information recently stored in memory must be accurately retrieved. In 1932 Beringer described cannabis-induced 'fragmentation of thought processes.. .disturbances of memory.. .interruptions in the stream of thought' (Ames, 1958). Bromberg's classic description of cannabis effects noted that ideas flowed so quickly that they failed to become fixed in memory (1934). Similarly the early investigations of the Chopras and others placed special emphasis on alterations in memory and cognitive processes (Chopra and Chopra, 1939). More recent surveys of cannabis effects have confirmed and elaborated these early clinical observations. Many of Tart's subjects reported that their memory span was shortened so that they forgot what a conversation was about before it had ended (1970). Since concepts often slipped away before they could be grasped, thinking lost clarity. Directed in part by these consistent reports from cannabis users that the drug often interferes with the ability to follow through logical sequences of thought, to engage in conversation, and to recall recent events, researchers have investigated cannabis effects on a variety of memory tasks. The subjective reports and clinical observations were verified in the laboratory; the performance of cannabis subjects was impaired for tasks requiring remembering of simple stimuli for short periods of time such'as digit span (Melges et al., 1970b; Tinklenberg et al., 1970), digit code (Clark and Nakashima, 1968; Clark et al., 1970; Jones and Stone, 1970), and memory scan (Darley et al., 1973a), as well as tasks in which more complex stimuli had to be processed and retained in memory such as Babcock Story Recall (Drew et al., 1972; Miller et al., 1972), picture recognition and Benton Sentence Repetition (Klonoff et al., 1973). Since basic memory processes are fundamental to a wide range of cognitive functions, a considerable amount of research has been focused on gathering data about cannabis induced changes in memory processes. These data now require organization into a conceptual framework. Recently, results from several studies utilizing free-recall memory tasks have been interpreted in terms of a model of human memory which allows the effect of cannabis on memory to be more explicitly identified (Atkinson et al., 1973). The structural components of this model are a very short-term sensory register, a limited-capacity short-term store (STS) or working memory, and a large, permanent Long-term store (LTS). (See Figure 3.1) Stimuli from the environment are received by the sensory receptors and held in the sensory register in an unprocessed form for a period of milliseconds (Sperling, 1960). Each stimulus must then be analyzed so that its internal encoded representation, which is retained in LTS, may be located and activated. The portion of LTS which contains the encoded representations of stimuli is called the conceptual store (CS). From the conceptual store of LTS the encoded stimulus representation is entered into working memory (STS). At this point the individual develops conscious awareness of the stimulus. While the stimulus resides in STS it may be operated upon by control processes such as rehearsal (rote repetition) or the formation of mnemonics or visual images. These rehearsal processes place information regarding the stimulus' occurrence as an environmental event into another portion of LTS, the event-knowledge store (EKS). Thus, the EKS of the long term store acquires and maintains information about the context in which a particular item or event occurred. In contrast, the conceptual store (CS) of LTS contains information about primary characteristics and meaning of stimuli.

Figure 3.1. Information flow through the memory system begins with the processing of environmental input in the sensory register. The stimulus is analysed and its encoded representation in the conceptual store (CS), a partition of long-term store (LTS), is activated. The encoded representation is then sent to short-term store (STS) where it receives additional processing by means of control processes. Control processes cause information to be copied into the event-knowledge store (EKS), a second partition of LTS. From EKS, information about the context in which the stimulus occurred may be retrieved and entered into STS. Depending upon task requirements, a response may be initiated at any stage of processing, utilizing information retrieved from either CS, STS or EKS. The longer an event is actively processed in STS, the more information about that event is transferred to EKS. However, when the control processes of STS do not operate upon an encoded stimulus, the stimulus is lost from STS within about 30 seconds. An encoded stimulus may reside in both STS and the EKS portion of LTS simultaneously; information about the stimulus may pass from the EKS to STS as well as vice versa. In free recall tasks a subject is presented a series of items such as words which he is to recall in any order after the list has been presented. According to the above model, the items (words)

are sequentially received by the sensory receptors, briefly held in the sensory register, encoded in the CS, entered into STS, and transferred to the EKS portion of the LTS by means of control processes. Only a limited number of items can be actively processed in STS, so after the fourth or fifth item has been presented, the arrival of new items causes old items to be dropped from STS. At the time of recall, the retrieval of list items may be from STS, EKS, or both depending on where in the list the item was presented and how soon after list presentation testing occurs. With an immediate free recall test, terminal list items are retrieved mainly from STS and earlier items largely from LTS. When testing is delayed, all items are usually retrieved from LTS (Craik, 1970; Glanzer and Cunitz, 1966). Cannabis intoxication does not impair the delayed free recall of information which had been stored in LTS in a non-cannabis state (Abel, 1971; Darley et al., 1973b; Dornbush, 1974). In other words, cannabis does not appreciably reduce retrieval of items which have been adequately stored in LTS. In fact, information initially presented to subjects during cannabis intoxication is most efficiently recalled when subjects are again under the influence of cannabis ()arley et al., 1974; Eich et al., 1974). These results indicate that cannabis induces asymetrical state-dependency for retrieval of information with free recall procedures, i.e. information presented in the drug state is more efficiently recalled during drug intoxication than during non-intoxication, but information initially stored during non-drug conditions is recalled equally well in either a drug or non-drug state. However, the influence of cannabis on retrieval processes seems to vary according to the type of memory task utilized since retrieval for paired-associate learning is also asymetrically state-dependent for cannabis (Rickles et al., 1973), while performance on a recognition memory task does not show any type of state dependency (Darley et al., 1974). Although cannabis does not impair retrieval on delayed recall tests, cannabis does induce significant deficits on immediate recall testing (Abel, 1971; Darley, et al., 1973b; Pearl et al., 1973). Since retrieval is not impaired by cannabis, this finding indicates that the drug interferes with some phase of memory storage. (Dornbush, 1974). Using the model presented earlier, this phase can be identified. Since terminal list items are recalled equally well by cannabis and placebo subjects, neither storage into nor retrieval from STS is appreciably impaired by the drug. Instead, the cannabis induced immediate recall deficits involve nonterminal list items which can be recalled only if they have been transferred from STS to LTS. Thus, the deleterious effect of cannabis on memory storage must be in the phase where information is transferred from STS to LTS. Since transfer to LTS is controlled by active processes such as rehearsal, a likely explanation for the cannabis effect on memory storage is that drug subjects engage in less rehearsal and other control processes, perhaps due to external and/or internal distractions (Abel, 1971). However, the storage deficit persists when overt fixed-rehearsal procedures are used to equalize the degree to which cannabis and placebo subjects overtly process list items ()arley et al., 1974). Thus, even when cannabis subjects rehearse list items as often as placebo subjects, they store less information about those items.We have suggested that drug subjects may attend to list items only enough to produce the required minimum of rehearsals, while placebo subjects in addition to performing the required overt rehearsals, covertly processed the items for later recall. (Darley and Tinklenberg, 1974). An alternative explanation is that once rehearsal of an item is terminated cannabis causes information which would be transferred to LTS to be lost from STS more quickly. This possibility receives support from a study by Dornbush et al., (1971) which showed that unrehearsed items are lost from STS

more quickly during cannabis intoxication (See Darley and Tinklenberg, 1974, for further details on this issue). Although, as indicated above, certain acute, short-term psychological and cognitive effects of cannabis have been clearly established, little is known about the effects of repetitive, longterm cannabis consumption. Our present paucity of knowledge about this important topic stems directly from the lack of adequate studies. Hopefully, investigations currently being conducted will remedy this situation. 6. FUTURE OF PSYCHOLOGICAL AND COGNITIVE RESEARCH WITH CANNABIS In considering future avenues of research it is often useful to organize what is presently known into a conceptual sketch or model which lends itself to further experimental manipulation. In attempting this with cannabis, we consider five areas of evidence about the effects of cannabis on human psychological and cognitive processes to be especially salient. 1. Cannabis does not markedly alter the earliest (sensory-perceptual) phases of information processing, at least at peripheral levels. 2. However, cannabis often alters the individual's subjective appreciation or interpretation of many sensory-perceptual experiences. 3. Cannabis markedly changes many components of time sense, especially inducing enhanced subjective tempo so that the duration of events seems to be longer than usual. 4. Cannabis does not significantly impair the retrieval of information which has been adequately stored in LTS . 5. However, cannabis does seem to impair the initial storage of information in LTS, possibly by making control processes such as rehearsal less efficient. We propose that cannabis may alter the manner in which encoded information is entered into STS thereby causing the 5 salient effects noted above. To develop our proposal we utilize the memory model presented earlier as well as other recent models which describe how information in LTS is encoded, stored, and retrieved (Atkinson et al., 1973; Melton and Martin, 1972; Norman, 1970; Tulving and Donaldson, 1972). When a stimulus, such as a word, makes contact with sensory receptors it is analyzed initially in terms of its perceptual characteristics. As mentioned previously, this analysis allows the internal encoded representation of the stimulus to be located in the conceptual store (CS) of LTS and activated. The internal representation may have a number of codes which specify the value of the stimulus on a set of dimensions or attributes (Norman and Rumelhart, 1970; Underwood, 1969). For words these attributes may be perceptual, syntactic or semantic properties. In most instances the requirements of the subject's task determine which properties of the stimulus are important and which may be ignored. Therefore, only a subset of the codes which fully describe a stimulus may be placed into STS (working memory) from LTS and thus enter the individual's conscious awareness. For example, if a subject is shown a picture of a ball and required to report what he saw after a short delay, he may enter into STS only the name of the object, i.e. 'ball' and ignore other characteristics such as size, shape and color. The way in which the stimulus is encoded determines to a great extent the subjective perception of the

stimulus and the way in which it is processed by the memory stores. Since items with common characteristics are thought to share some common codes in LTS, the activation of a set of codes for a stimulus is likely to activate codes representing related items. The entry into STS of items associated to the presented stimulus has been termed an 'implicit associative response' (JAR) (Bousfield et al., 1958; Underwood, 1965). For example, if the word 'dog' were presented on a list, the word 'cat' might enter STS (working memory) as an IAR since the two items have some common perceptual and semantic features. We propose that cannabis causes more codes to be entered into STS for each new item presented. For example, the presentation of the word 'apple' to a subject under the influence of cannabis might cause a variety of perceptual and semantic information to enter STS such as 'it's a fruit called an apple which is round, red, shiny, and sweet tasting', whereas for a subject who has not taken cannabis only the name of the stimulus might enter STS (working memory). The more codes activated for a particular stimulus, the more likely it is that codes for related items will also be activated and enter STS. This model may be useful in explaining several different psychological effects of cannabis. First, it is consistent with a limited amount of directly applicable experimental evidence which indicates that during cannabis intoxication there is an increase in intrusion errors, i.e., the subject is likely to erroneously recall information which was not contained in the initial learning task (Abel, 1971; Darley and Tinklenberg, 1974; Dornbush, 1974; Miller et al., 1972). Such intrusions probably occur because IARs are made at the time of stimulus presentation (Ainisfield and Knapp, 1968). The greater number of IARs which result from the increased range of encoding under cannabis would therefore account for such errors of intrusion. Second, the model is consistent with most experimental cannabis work which focuses on memory processes. Cannabis-induced storage deficits arise because each newly-presented stimulus expends more of the limited capacity of STS. During cannabis intoxication the entry into STS of a larger number of codes for each stimulus and related non-list items causes codes from earlier list items to be dropped more quickly from STS than during the non-drug state. Many of the entered codes may be irrelevant for the task requirements; the time spent processing them reduces the time available for processing task-relevant codes. Since the processing applied to an item's relevant codes determines how much information about that item is transferred to LTS, less transfer occurs in the drug state. In addition, since cannabis causes fewer recall items to be represented in STS at any one time, complex control processes which allow list items to be associated with one another become less efficient. Again, the result is less information available in LTS about the occurrence of an item and therefore poorer immediate recall performance for items which must be retrieved from LTS. The model is also consistent with data indicating that cannabis does not impair retrieval of already stored information; if anything, the proposed cannabis effect might enhance retrieval since particular encodings of information previously inaccessible might be located during cannabis intoxication. No objective evidence for enhanced retrieval with cannabis exists at the present time, but appropriate measures may not have been applied. Third, the cannabis induced alterations in time sense, especially the acceleration of subjective tempo would be explicable with this model. The increased flow of codes into STS would induce the greater frequency of inner events which has been theoretically associated with an acceleration of subjective tempo (Ornstein, 1969; Tinklenberg et al., 1972). In other words, the greater number of items impinging on STS (conscious awareness) per unit of geophysical

time would explain the frequently reported effects of cannabis in which geophysical time seems to go more slowly than usual. Fourth, the proposed model would be compatible with the cannabis-induced changes in sensory-perceptual processes where there is often enhancement of subjective awareness of appreciation in the absence of increased peripheral sensory acuity. The increased flow to STS of usually inaccessible codes for a stimulus and related items might subjectively enhance the richness and uniqueness of sensory events. The synesthetic experiences may also be explicable since under marijuana perceptual codes related to different sensory systems (eg. a visual code and an auditory code) may enter STS together. Fifth, since conscious thought processes involve interrelating and manipulating concepts, having available to consciousness a wider range of properties for any particular concept and related concepts might also explain the feelings of increased insight and creativity which occur during cannabis intoxication. Intoxicated subjects may be able to connect usually unrelated concepts or perceive new aspects of objects or events. One prediction which arises from this model is that difficult tasks in which information must be extensively processed in STS should be affected by cannabis while well-learned, automatic tasks in which information may flow directly from LTS to the response apparatus will show little effect. Also, deficits due to an influx of irrelevant memory codes may be most frequent in tasks which place a premium on variability of encoding and interrelating of stimuli. Where task requirements are more limited, so that only a single code for each of a limited number of items must be held in STS, there may be only a sporadic loosening on the restrictions on information flow from LTS. Thus, in free recall learning where optimal performance is achieved only when subjects utilize alternative encodings of items and associate items with one another, there may be a continual flow of irrelevant codes for list items and their non-list associates. On the other hand, in psychomotor tasks, such as simple or choice reaction time, the repetitive nature of the task may allow intoxicated subjects to restrict the flow of irrelevant information into STS during most trials of the task, with performance dropping significantly only on those trials where the restrictions are loosened. This could explain why some investigators have shown that cannabis effects on psychomotor tasks seem to be intermittent, as if subjects were distracted on a portion of the trials (Clark et al., 1970). This model makes many of the same predictions and is conceptually similar to a model proposed by DeLong and Levy (1973). They suggested that cannabis-induced behavioral changes could be understood in terms of various aspects of attention including focusing on relevant aspects of a task and resisting distraction. In the model presented here, the influx of irrelevant material from LTS serves, in a sense, as distracting input which inhibits the processing of task-relevant information. A major contribution of the present model is that it not only explains measurable behavioral effects of cannabis but also reported subjective effects of the drug. In many memory models, the flow of information between memory stores, retrieval of stored information, complex coding operations and many other higher cognitive functions are under the control of an 'executive monitor' or 'executive program'. To meet task requirements the executive monitor must control the flow of memory codes from LTS into STS. Thus, the monitor may be the focus of cannabis effects since such a flow of information seems to be altered by the drug. One direction of cannabis research may be to examine drug effects on

other activities of this monitor. Experiments by Hart (1965) have demonstrated a process he calls 'memory monitoring' which allows individuals to examine their memories to determine if currently irretrievable information is potentially available for recall. Such a process enables the individual to decide whether, after a series of retrieval failures, further similar attempts should be made, retrieval strategies should be altered, or retrieval effort should be abandoned altogether. By utilizing variants of Hart's paradigm, future research may delineate effects of cannabis on the processes which control and direct subjects' behavior and which may be ultimately responsible for the varied psychological and cognitive effects induced by the drug. Future research in the psychological and cognitive effects of cannabis might also usefully include various pharmacological manipulations. One such line of investigation entails the use of other drugs to either inhibit specific effects of cannabis or to interact with the behavioural characteristics of cannabis in other ways. For example, although it has been established that there is little or no cross tolerance with LSD, knowledge about cross tolerance with other drugs is limited (Isbell and Jasinski, 1969). Information about such interactions sometimes provides clues about which receptor sites or physiological systems are responsible for certain behavioural effects of the drug. Similarly information about which agents block drug effects could be exceedingly useful. Except for studies indicating that the cognitive effects of cannabis are not significantly reduced by pre-treatment with the beta-adrenergic blocker, propranolol, little work has been completed in this area ( Drew et al., 1972). Pharmacological differences among various batches of cannabis may also provide useful information. For example, although most of the effects of acute intoxication with cannabis can be explained on the basis of delta-l-THC (delta-9-THC by other nomenclature) and its metabolites, it is possible that other components of cannabis or their metabolites selectively exert certain behavioural effects. Cannabis would thus be directly comparable to alcoholic beverages in which the predominant effects are exerted by ethanol, but various congeners 9ay also induce discernible influences (Katkin et al., 1970). Such a finding might also explain reports suggesting that various cannabis preparations exert different psychological effects. A final area of psychological and cognitive research with cannabis that seems promising involves comparing the psychological and cognitive effects of cannabis with other psychoactive agents and hormones (Drew and Miller, 1974). Cannabis seems to occupy a unique position in the pharmacopoeia; some of its properties are analogous to the sedative hypnotics such as alcohol and secobarbital, but other effects are comparable to low doses of LSD and other psychedelic substances. Comparing the effects of cannabis with other drugs is useful not only at a psychological and cognitive level, but also in attempts to determine the physiological mechanisms that subserve these behavioral effects. Integrating the psychological and cognitive effects of cannabis with the underlying alterations in physiological processes remains one of the ultimate goals of cannabis research. REFERENCES Abel, E.L. Marihuana and memory: acquisition or retrieval? Science, 173: 1038-1040, 1971. Ames, F. A clinical and metabolic study of acute intoxication with cannabis sativa and its role in a model psychosis. Journal of Mental Science, 104: 972-999, 1958. Anisfield, M., and Knapp, M. Association, synonymity and directionality in false recognition. Journal of Experimental Psychology, 77: 171-179, 1968. Atkinson, R.C. and S hiffrin, R.M. Human memory: a proposed system and its control processes. In K.W. Spence and J.T. Spence, eds. The Psychology of Learning and Motivation. Vol. 11, pp.89-195. Academic Press, New York, 1968.

Atkinson, R.C. and Shiffrin, R.M. The control of short-term memory. Scientific American, 224: 82-90, 1971. Atkinson, R.C., Herrmann, D.J. and Wescourt, K.T. Search Processes in Recognition Memory. Technical Report No. 204, Institute for Mathematical Studies in the Social Sciences, Stanford University, June 8, 1973. Bousfield, W .A., Whitmarsh, G.A. and Danick, J.J. Partial response identities in verbal generalization. Psychological Reports, 4: 703-713, 1958. Bromberg, W. Marihuana intoxication, a clinical study of cannabis sativa intoxication. American Journal of Psychiatry, 91: 303-330, 1934. Caldwell, D.F., Myers, S.A., Domino, E.F. and Merriam, P.E. Auditory and visual threshold effects of marihuana in man. Perceptual and Motor Skills, 29: 755-759, 1969. Caldwell, D.F., Myers, S.A. and Domino, E.F. Effects of marihuana on sensory thresholds in man. In D.H. Efron, ed. Psychotomimetic Drugs. pp.299-321. Raven Press, New York, 1970. Canada. Commission of Inquiry into the Non-Medical Use of Drugs. Cannabis. Queen's Printer, Ottawa, 1972. Cappel, H., W ebster, C.E., Herring, B.S. and Ginsberg, R. Alcohol and marihuana: a comparison of effects on a temporally controlled operant in humans. The Journal of Pharmacology and Experimental Therapeutics, 182 (2): 195-203, 1972. Casswell, S. and Marks, D.F. Cannabis and temporal disintegration in experienced and naive subjects. Science, 179: 803-805, 1973. Chopra, R.N. and Chopra, G.S. The present position of hemp-drug addiction in India. Indian Journal of Medical Research Memoirs, 31: 1-119, 1939. Clark, L.D. and Nakashima, E.H. Experimental studies of marihuana. American Journal of Psychiatry, 125: 379-384, 1968. Clark, L.D., Hughes, R. and Nakashima, E.N. Behavioral effects of marihuana: experimental studies. Archives of General Psychiatry, 23: 193-197, 1970. Craik, F.I.M. The fate of primary memory time in free recall. Journal of Verbal Learning and Verbal Behavior, 9: 672-678, 1970. Darley, C.F., Tinklenberg, J.R., Hollister, L.E. and Atkinson, R.C. Marihuana and retrieval from short-term memory. Psychopharmacologia (Berl.), 29: 231-238, 1973a. Darley, C.F., Tinklenberg, J.R., Roth, W.T., Hollister, L.E. and Atkinson, R.C. Influence of marihuana on storage and retrieval processes in memory. Memory and Cognition, 1: 196-200, 1973b. Darley, C.F. and Tinklenberg, J.R. Marihuana and memory. In L.L. Miller and W.G. Drew, eds. Marijuana - Effects of Human Behavior. Academic Press, New York, in press. Darley, C.F., Tinklenberg, J.R., Roth, W.T. and Atkinson, R.C. The nature of storage deficits and state-dependent retrieval under marihuana. Psychopharmacologia (Berl.), in press. DeLong, F.L. and Levy, B.I. Cognitive effects of marijuana, described in terms of a model of attention. Psychological Reports, 33: 907-916, 1973. Dornbush, R.L. Marijuana and memory: effects of smoking on storage. Transactions of the New York Academy of Sciences, 36: 94-101, 1974. Dornbush, R.L., Fink, M. and Freedman, A.M. Marihuana, memory and perception. American Journal of Psychiatry, 128: 194-197, 1971. Drew, W.G. and Miller, L.L. Cannabis: neural mechanisms and behavior - a theoretical review. Pharmacology, 11 (1): 12-33, 1974. Drew, W.G., Kiplinger, G.F., Miller, L.L. and Marx, M. Effects of propranlol on marihuanainduced cognitive dysfunctioning. Clinical Pharmacology and Therapeutics, 13 (4): 526-533, 1972. Eich, J., Weingartner, H. and Stillman, R.C. Availability versus accessibility of information in marijuana state-dependent learning. Paper submitted for publication, 1973.

Galanter, M., Wyatt, R.J., Lamberger, L., Weingartner, H., Vaughn, T.B. and Roth, W.T. Effects on humans of delta-9-tetrahydrocannabinol administered by smoking. Science, 176: 934-936, 1972. Glanzer, M. and Cuntiz, A.R. Two storage mechanisms in free recall. Journal of Verbal Learning and Verbal Behavior, 5: 351-360, 1966. Hart, J.T. Recall, recognition and the memory-monitoring process. Unpublished Ph.D. Thesis, Stanford University, 1965. Hochman, J.L. and Brill, N.Q. Marijuana use and psycho-social adaptation. Paper delivered at American Psychiatric Association meeting in Washington, D.C., May 3, 1971. Hollister, L.E. Marihuana in man: three years later. Science, 172: 21-28, 1971. Hollister, L.E. and Gillespie, H.K. Marihuana, ethanol and dextroamphetamine. Mood and mental function alterations. Archives of General Psychiatry, 23: 199-203, 1970. Hollister, L.E. and Tinklenberg, J.R. Subchronic oral doses of marihuana extract. Psychopharmacologia (Berl.), 29: 247-252, 1973. Hollister, L.E. Department of Medicine and Psychiatry, VA Hospital, Palo Alto and Stanford University. Personal communication, April, 1974. Isbell, H. and Jasinski, D.R. A comparison of LSD-25 with (-)-delta-9transtetrahydrocannabinol (TUC) and attempted cross tolerance between LSD and THC. Psychopharmacologia (Berl.), 14 (2): 115-123, 1969. Isbell, H., Gorodetzsky, C.W., Jasinski, D., Claussen, U., Spulak, F.D. and Korte, F. Effects of (-)delta-9-trans-tetrahydrocannabinol in man. Psychopharmacologia (Berl.), 11: 184-188, 1967. Jones, R.T. Tetrahydrocannibinol and the marijuana induced social 'high' or the effects of the mind on marijuana. Annals of the New York Academy of Sciences, Vol. 191. A.J. Singer, ed. New York Academy of Sciences, New York, 1971. Jones, R.T. Department of Psychiatry, Langley Porter Clinic, University of California. Personal communication, April, 1974. Jones, R.T. and Stone, G.C. Psychological studies of marijuana and alcohol in man. Psychopharmacologia (Berl.), 18: 108-117, 1970. Karniol, I.G. and Carlini, E.A. Comparative studies in man and in laboratory animals on delta-8- and delta-9-trans-tetrahydrocannabinol. Pharmacology, 9: 115-126, 1973. Katkin, E.S., Hayes, W.N., Teger, A.I. and Pruitt, D.E. Effects of alcoholic beverages differing in congener content on psychomotor tasks and risk-taking. Quarterly Journal of Studies on Alcohol, Supplement, 5: 101-114, 1970. Kiplinger, G.F., Manno, J.E., Rodda, B.E. and Forney, R.B. Dose-response analysis of the effects of tetrahydrocannabinol in man. Clinical Pharmacology and Therapeutics, 12: 650657, 1971. Klonoff, H Low, M. and Marcus, A. Neuropsychological effects of marijuana. Canadian Medical Association Journal, 108: 150-157, 1973. Lemberger, C.M., Silberstein, S.D., Axelrod, J. and Kopin, I.J. Marihuana: studies on the disposition and metabolism of delta-9-tetrahydrocannabinol in man. Science, 170: 1320-1322, 1970. Manno, J.E., Kiplinger, G.F., Haine, S.E., Bennett, I.F. and Forney, R.B. Comparative effects of smoking marihuana or placebo on human motor and mental performance. Clinical Pharmacology and Therapeutics, 11: 808-815, 1970. Manno, Kiplinger, T.F., Scholz, N., and Forney, R.B. The influence of alcohol and marihuana on motor and mental performance. Clinical Pharmacology and Therapeutics, 12: 202-211, 1971. Melges, F.T., Tinklenberg, J.R., Hollister, L.E. and Gillespie, H.K. Temporal disintegration and depersonalization during marihuana intoxication. Archives of General Psychiatry, 23:

204-210, 1970a. Melges, F.T., Tinklenberg, J.R., Hollister, L.E. and Gillespie, H.K. Marihuana and temporal disintegration. Science, 168: 1118-1120, 1970b. Melges, F.T., Tinklenberg, J.R., Deardorff, C.M., Davies, N.H.,, Anderson, R.C. and Owen, C.A. Temporal disorganization and delusional-like ideation. Archives of General Psychiatry, in press. Melton, A.W . and Martin, E. Coding Processes in Human Memory. V.H. Winston and Sons, Washington, D.C., 1972. Mendelson, J.H. and Meyer, R.E. Behavioral and biological concomitants of chronic marihuana smoking by heavy and casual users. In United States, National Commission on Marihuana and Drug Abuse, Marihuana: A Signal of Misunderstanding, Appendix, Vol. 1, pp. 68-246. U.S. Government Printing Office, Washington, D.C., 1972. Meyer, R.E., Pillard, R.C., Sapiro, L.M. and Mirin, S.M. Administration of marihuana to heavy and casual marihuana users. American Journal of Psychiatry, 128: 198-204, 1971. Miller, L.L., Drew, W.G. and Kiplinger, G.F. Effects of marijuana on recall of narrative material and Stroop colour-word performance. Nature (Lond.), 237: 172-173, 1972. Moskow itz, H., Sharma, S. and Schapero, M. A comparison of the effects of marijuana and alcohol on visual functions. In M.F. Lewis, ed. Current Research in Marihuana, pp.129-150 . Academic Press, New York, 1972. Norman, D.A., ed. Models of Human Memory. Academic Press, New York, 1970. Norman, D.A. and Rumelhart, D.E. A system for perception and memory. In D.A. Norman, ed. Models of Human Memory, pp.19-64. Academic Press, New York, 1970. Ornstein, R.E. On the Experience of Time. Penguin, Baltimore, 1969. Pearl, J.H., Domino, E.F. and Rennick, P. Short-term effects of marijuana smoking on cognitive behavior in experienced male users. Psychopharmacologia (Berl.), 31: 13-24, 1973. Perez-Reyes, M., Lipton, M.A., Timmons, M.C., Wall, M.E., Brine, D.R. and Davis, K.H. Pharmacology of orally administered delta-9-tetrahydrocannabinol. Clinical Pharmacology and Therapeutics, 14: 48-55, 1973. Rafaelsen, 0.J., Bech, P., Christiansen, J., Christrup, H., Nyboe, J. and Rafaelsen, L. Cannabis and alcohol: effects on simulated car driving. Science, 179: 920- 923, 19 73. Renault, P.F., Schuster, C.R., Heinrich, R. and Fredman, D.X. Marijuana: standardized smoke administration and dose effect curves on heart rate in humans. Science, 174: 589-591, 1971. Rickles, W .H., Jr., Cohen, M.J., Whitaker, C.A. and McIntyre, K.E. Marijuana induced statedependent verbal learning. Psychopharmacologia (Berl.), 30: 349-354, 1973. Roth, W.T., Galanter, M., Weingartner, H., Vaughan, T.B. and Wyatt, R.J. Marijuana and synthetic delta-9-THC: some effects on the auditory evoked response and background EEG in humans. Biological Psychiatry, 6: 221-233, 1973. Sperling, Go The information available in brief visual presentations. Psychology Monographs, whole no. 498, 1960. Tart, C.T. Marihuana intoxication: common experiences. Nature, 226: 701-704, 1970. Tinklenberg, J.R., Melges, F.T., Hollister, L.E. and Gillespie, H.K. Marijuana and immediate memory. Nature, 226: 1171-1172, 1970. Tinklenberg, J.R., Kopell, B.S., Melges, F.T. and Hollister, L.E. Marihuana and alcohol: time production and memory functions. Archives of General Psychiatry, 27: 812-815, 1972. Tulving, E. and Donaldson, W.D., eds. Organization of Memory. Academic Press, New York, 1972. Underwood, B.J. False recognition produced by implicit verbal responses. Journal of Experimental Psychology, 70: 122- 129, 1965. Underwood, B.J. Attributes of memory. Psychological Review, 76: 559-573, 1969.

Volavka, J., Dornbush, R., Feldstein, S., Clare, G., Zaks, A., Fink, M. and Freedman, A.M. Marijuana, EEG, and behavior. Annals of the New York Academy of Sciences, Vol. 191. A.J. Singer, ed. New York Academy of Sciences, New York, 1971. Weil, A.T. and Zinberg, N.E. Acute effects of marihuana on speech. Nature, 222: 434-437, 1969. Weil, A.T., Zinberg, N.E. and Nelson, J.M. Clinical and psychological effects of marijuana in man. Science, 162: 1234-1242, 1968. DISCUSSION As in all discussions at this conference the subject matter of the paper was discussed at length and in depth. In order to organise the material, subject headings have been chosen and relevant contributions placed within these headings rather than sticking strictly to the consecutive order of the contributions. The resulting statement is a summary overview of main themes of the discussion. SUBJECTS AND DOSAGE LEVELS In response to a question by Dr Rubin, Dr Tinklenberg described the choice of subjects for the study. He noted that it had not been possible to use naive subjects because of the possible danger to corrupting them, nor indeed to use women. He therefore had no data on naive users. Experience had shown that individuals over 30 years of age such as school superintendents, police, court judges and others were not suitable because they were more likely to react badly to cannabis. His subjects, therefore, were non-naive graduate students age 1924 who had used 'occasionally' or up to once or twice a week for the last year. Previous experience had shown that it was advisable to avoid medical students, who tended to be neurotic, and psychology or behavioral students who tended to manipulate the study . Stanford University graduate students of civil engineering were found to be particularly suitable since they were reliable, bright, compulsive and obedient. In response to a question by Dr Martens, Dr Tinklenberg stated that the dose used in the study was 20 mg. THC orally, equivalent to a smoking dose of 6-7 mg. per person and described by the subjects as a fairly strong dose. Material was supplied by NIMH and dosage levels could therefore be compared with other studies using NIMH material. THE MODEL PRESENTED It is useful at this point to summarise the main features of the model of sensory processing put forward by Dr Tinklenberg. The model postulates that sensory input is brought into relationship with a conceptual store of previously encoded inputs, and an encoded representation of the input proceeds to short-term memory, where it becomes conscious and open to control processes, such as rehearsal and imagery. Rehearsal may then place the encoded input in the Event Knowledge Store of Long-Term Memory. Dr Tinklenberg's suggestion is that, under cannabis intoxication, a greater number of codes than usual is activated by sensory inputs. Because short-term memory can only handle a limited number of codes at a time, the arrival of a greater number of codes causes previously encoded stimuli to be dropped more quickly than otherwise from short-term memory. Transfer to long-term memory is therefore impaired. The model also offers explanations of increase of intrusion errors, acceleration of subjective time-sense, subjective increase in richness of perception, and formation of new 'creative' associations.

THE CODING PROCESS Dr Mortens asked whether the model merely postulated a greater number of activated codes under intoxication, or whether a different type of code might be activated. The significance of the distinction, he suggested, is that a change in choice of codes would imply a more central effect than would a simple increase in number of codes activated. Dr Tinklenberg replied that existing evidence pointed to a greater number rather than a different choice of codes, and that an experiment was planned to investigate this. Dr Edwards raised the question of a possible analogy between the greater number of codes activated by cannabis intoxication and the kind of overinclusive thinking sometimes observed in schizophrenia. Dr Tinklenberg suggested that alterations in the schizophrenic's processing of incoming information occurred at an earlier stage of processing than was the case with cannabis intoxication. ATTENTION Dr Smart asked about the power of the model in explaining results of studies of attention. Although sensory registration studies (undivided attention, single stimulus) may not show a decrease in performance under cannabis intoxication, studies of divided attention do show deficits. Divided attention studies are more like everyday life. Dr Tinklenberg agreed that such results are not explained parsimoniously by the model, but suggested that whatever was happening was at a central rather than a peripheral level. Professor Paton suggested that the addition of a feedback loop allowing central direction of input might be helpful. Dr Leuw raised the question of motivation as an intervening variable and noted that motivation could overcome the effects of cannabis. Dr Tinklenberg agreed but stated that on certain serial learning tasks there was no significant difference between those under the influence of cannabis and those not under its influence. Dr Hindmarch mentioned task-dependent performance, and stressed that performance could be drastically altered by differences in how one directed the subject. He felt that the effects of motivational and task-dependent factors could be unravelled by careful experimental design. Whilst a number of participants found the model proposed to be interesting and a valuable way of considering a difficult area of cognitive function several participants, notably Drs Hindmarch, Leuw and Smart, were concerned that the model was a little mechanistic and was abstracted memory from other systems. As Dr Smart put it: 'I'm always very happy to see theories develop and models presented. I rarely attempt them myself. This one still bothers me - especially in respect of motivation. There are studies that have shown that deficits under cannabis can be experimentally manipulated. We need to add another box to the model for unknowables.' Acknowledgements: The authors wish to thank Patricia L. Murphy and Peggy Murphy for special assistance in the preparation of this paper as well as Cynthia Demos for technical assistance.

The authors' research discussed in this paper was supported in part by the van Ameringen Foundation.

4. A Selective Review of Studies of Longterm Use of Cannabis on Behaviour, Personality and Cognitive Functioning
C. G. Miles, Formerly Director of Marijuana Studies, Addiction Research Foundation, Toronto, Ontario. In a sense this paper is probably ten years too soon. In researching the literature it is quite apparent that we do not have enough really reliable studies to completely answer the question implied in the title. In general the research used to assess the state of knowledge in the area falls fairly neatly into several categories. First, retrospective studies where users are asked how long they have smoked and various methods of determining change are used. Second, cross-sectional comparisons, where matched groups are compared on the basis of the amount of time they have been smoking. Third, longitudinal studies where smokers and non-smokers are examined at various points in time. Fourth, long-term experimental and observational studies. The emphasis in this paper is on the third and fourth categories. All studies have been hampered by the fact that the drug is illicit in almost every part of the world, and punishments for use and possession have been severe. Consequently, smokers have often been loath to admit usage and submit to extensive investigation of the type required. Only in the last few years have attitudes changed such that governments have been actively promoting investigations into the effects of cannabis, e.g., U.S., Marijuana and Health, Second Annual Report to Congress (1972); Canada, The Commission of Inquiry into the NonMedical Use of Drugs, Cannabis Report (1972); Britain, British Advisory Committee on Drug Dependence, Cannabis (1968), with the significant exceptions of the Report of the Indian Hemp Drugs Commission (1894) and the LaGuardia Report (1944). The later government reports immediately recognized the paucity of information and research efforts started in earnest. However, political pressures on the investigating bodies reflected the concern of a public uncomprehending of the snail-pace of good science which has filtered through to the scientists. In preparing this paper, the author had occasion to contact by telephone and in person a number of researchers who had been pressured into presenting either incomplete summaries or tentative reports. Most of these researchers were cooperative and generous with their impressions, but at the same time they have stressed the fact they have incompletely analyzed data or that their studies are still in progress. In some cases funds for the projects have dried up and whole fleets of ships may be already spoiled for the want of a few pennies' worth of tar. An alternative title for this paper might be, 'The 'amotivational syndrome' re-examined after a further six years of study'. It will be recalled that the amotivational syndrome is a subtle

change in personality manifest in a set of symptoms 'including apathy, loss of effectiveness, reduced drive and ambition, diminished capacity or willingness to carry out complex longterm plans, to endure frustration, to follow routines or to successfully master new material' McGlothlin and West, 1968). To this author studies of long-term cannabis use in which behavioral, personality and cognitive variables have been studied are either implicitly or explicitly searching for evidence which will bear on this hypothesis. There seem to be few experimental studies which have attempted to investigate the effects of cannabis smoking over relatively long periods of time under conditions where the environment, dosage, and the use of other drugs is known or excluded. The advantage of this type of study is that behaviour can be carefully documented with the minimum of error which is inevitable in retrospective and longitudinal studies, which rely heavily on self support and recall material. The first comprehensive study was reported in the LaGuardia Report (1944) and was conducted under the auspices of the New York Academy of Medicine. The subjects in this study were primarily volunteers from New York penitentiaries. In all, 72 subjects were studies in groups of 6 to 10, and their stay in the hospital ward experimental setting varied from 4 to 6 weeks. The drug was administered in the form of cigarettes and oral concentrate. For the psychological tests 54 subjects were used; 36 of them were users of cannabis and 18 were non-users. The subjects were tested while under the influence of the drug and when in an undrugged state. Some were tested with 2-5 cc oral concentrates and others with 3-5 cigarettes. A series of psychomotor functions were tested, including tapping speed, grip strength, static equilibrium, hand steadiness, and simple and complex reaction time. Musical aptitude and auditory acuity tests were also administered. There were almost no effects of the drug on tapping speed and simple reaction time. However, hand steadiness, static equilibrium, and complex reaction were affected adversely by both the higher and the lower doses. Objective time estimation was not affected by the drug, although this effect has been well documented in later acute studies (e.g., Cappell et al., 1972). It is interesting to note that generally the effects of all tests were slightly more pronounced in non-users. Investigations of cognitive functioning were carried out under similar conditions to the above study. Here tests of learning ability, intelligence, speed and accuracy performance tests, and ability to perform routine tasks were administered. A dose-related effect on cognitive functioning as measured by these tests was noted, but again it was present only under the acute effects of the drug. Once again the effects were more noticeable in non-users than users. It is noted, however, that on some of the simpler tasks, some slight improvement occurred with smaller doses. Emotional factors and personality structure were studied, giving various numbers of tests to users and non-users. These included vocational interest, self image, emotional stability, self confidence, drive and ambition, suggestibility, and a clinical evaluation. The acute effects of the smaller doses tended to indicate lower drive, objectivity and aggression, while selfconfidence increased together with a more favourable self image. At higher doses anxiety and paranoia were evident. It is noted that there were drug state differences between users and non-users. In the undrugged state, the users tended to be more introverted and inhibited, while in the drugged state the users changed in the direction of being more outgoing and anxiety prone. In general the study concludes that there was little difference between users and nonusers in intelligence, general behaviour and personality.

There are several problems with this study. First, there is little in the way of statistical evidence presented in the text. Second, the population was drawn from prisons and was under the surveillance of guards throughout the testing. Third, in drug studies the double blind procedure is desirable and was not used. This latter criticism is thought by this author to be of little importance, because in the experience of the writer it is almost impossible to do such studies, since the effects are recognized immediately by the investigator and the subject Niles et al., 1974). Fourth, the doses used are difficult to compare with those such as are routinely reported in the current literature in termsof 2-THC content. However, even with these criticisms it was a commendable effort to study the effects of the drug and drug users experimentally over a longer period of time than had hitherto been attempted. It is interesting to note that no drug-related aggression or psychotic states (Talbot and Teague, 1969) were noted in this study. In what has become a classic experiment, although perhaps less quoted than it deserves, Williams et al. (1946) investigated the effects of daily administration of pyrahexyl compound and marijuana plant material. The two substances were studied in separate experiments. As in the LaGuardia Report, the subjects were all prisoner volunteers, who in this instance were serving sentences for violation of the Marijuana Tax Act. Pyrahexyl compound is a synthetic substance said to have similar effects to that of marijuana, with assumed pharmacological similarities. In general the methods, measuring instruments used, and effects were similar with both substances. Only the marijuana smoking study will be reviewed here. Six subjects were studied in a research ward of a hospital. The period of investigation ran 7 days preliminary observation, 39 days smoking ad libitum, and 7 days post smoking observation. Although the potency of the cannabis was not assayed, it was assessed by these experienced smokers as 'good weed'. The average number of cigarettes smoked per day was a quite remarkable 17, with a range of from 9 to 26. An exception was one subject who limited himself to one cigarette per day. There was an overall trend towards increasing dosage. It is difficult to assess the size of the cigarettes, but an examination of the two figures giving dosage and number of grams smoked suggests that the cigarettes were either experimentally varied in size or the subjects were able to roll their own. The psychological test material was obtained when the subjects were thought to be high during the two weeks after smoking had begun, during the pre-smoking period, and in the post-smoking period. The Rorschach, Stanford-Binet, MacQuarrie Test for Mechanical Ability, Seashore Measure of Musical Talents, and Muller-Lyer Illusion tests were given. Pre-smoking, the Rorschach showed all subjects but one had patterns which were immature and restricted, whereas during smoking the results suggested fewer, but more elaborate patterns. Intellectual functioning was somewhat impaired during the smoking period, but returned to baseline in the post-smoking period. The MacQuarrie Test showed an increase in speed but a loss in accuracy. No improvement in musical ability was reported, although subjects thought they made better discriminations. Judgments in the Muller-Lyer Illusion Test were inconsistent during the smoking period, suggesting lack of motivation and carelessness. It should be remembered that these tests were all done while the subjects were thought to be intoxicated, and after only two weeks of smoking. They can in no way be construed as being indicative of long term effects.

Of greater interest in this study are the general clinical observations, such as an increase in sleeping time, general lassitude, the incapacity of subjects to apply themselves to the achievement of self set goals, or externally set goals, social deterioration and lack of personal hygiene. However, none of the subjects showed anti-social behaviour or psychotic reactions during the study. This presents a picture very much like that of the amotivational syndrome. Also of interest is the fact that no objective evidence of an abstinence syndrome after the abrupt cessation of smoking was observed, although the subjects themselves reported that they were somewhat 'jittery' for a few days. The authors report the general conclusion of a return to baseline testing conditions where tests were administered. Yet a different approach was used by Mendelson and Meyer (1972). figain they used a hospital ward setting, and their subjects were confined for a 31-day period. Subjects were 10 heavy and 10 casual users of cannabis. Heavy users were defined as subjects with 2 to 5 years experience of marijuana smoking, and who smoked daily for at least one year prior to the study. Casual users were defined as having a current smoking pattern of 1 to 4 sessions per month. These two groups were run separately. The subjects typically were multiple-drug users, unmarried, and unemployed. The groups differed primarily in the quantitative patterns of drug use. Both groups were of above average intelligence. The heavy user's tended to have started alcohol and marijuana use at a slightly earlier age, although alcohol use in both groups was said to be infrequent. The groups were matched on total years of marijuana use, but the heavy users tended to smoke a mean of 33 sessions per month, as opposed to casual users, smoking a mean of 7.7 sessions per month. The setting appears to be rather similar to that described in Miles' study Miles et al., 1974), with recreational facilities available, and relatively high security. In both of Mendelson's studies a pre-drug period of 5 days was given; a 21-day period of access to marijuana; followed by a 5-day non-access to marijuana period. The variables of interest to this paper were: 1. Work-contingent operant acquisition and free choice marijuana consumption patterns. 2. Measurements of mood states. 3. Individual and group behaviour observations, clinical and psychological evaluations. Subjects were given an operant manipulandum, which was available to them at all times. By manipulating a bar, they could accrue points which could be used to purchase tobacco during all periods of the study and also marijuana cigarettes during the smoking period. They could also accumulate points which could be converted into money at the end of the study. The operant bar consisted of a 4-digit hand counter, and the operant task was simply to accumulate points on the counter by pressing the bar. Payment for accumulating points was at the rate of one cent for 60 points, but there was a maximum limit on their earnings of ten dollars per day. A pack of cigarettes cost 50 cents, and one marijuana cigarette cost one dollar.

The marijuana used in these studies varied from 1.8 to 2.3 per cent THC, and each cigarette contained approximately 1 gram of material. The counter readings on the operant manipulandum were recorded every two hours. All subjects earned the maximum reinforcement points every day. This contrasts with results obtained in Mendelson's alcohol research, in which subjects showed a cessation of work output when they were consuming alcohol. There was no marijuana-related change in the pattern of work, and subjects often performed high work output while they were actually smoking marijuana. There was a wide range in the amount of consumption among the casual users, ranging from an average of one cigarette a day to an average of 6.2 per day, although there was an overall 46 per cent increase in the use of marijuana during the fourth quarter of the experiment as compared to the first quarter. In the heavy user group, average daily consumption ranged from 3.6 cigarettes to 8.7 cigarettes, and this showed an overall average of 37 per cent increase in the amount smoked in the fourth quarter of the drug period as compared to the first quarter. A 10-point rating scale showed that there was no correlation between 'highness' and smoking day. Thus there appeared to be no tolerance to subjective effects. Over the period of the study, the casual smokers indicated an increase in negative mood states, such as anxiety, hostility, depression, and a decrease in positive mood states, like friendliness, carefreeness, etc. Digit span tests, which were administered regularly, showed a steady performance increment. It should be noted here, however, that subjects were paid five cents for each series of digits reproduced correctly, thus they were financially motivated to increase efficiency, and this motivation was not impaired by the amount of marijuana smoked. Cognitive functioning as measured by the Wechsler Adult Intelligence Scale showed no marijuana-related changes over the course of the study. It is of interest to note that the subjects rarely chose to smoke alone. Less than 6 per cent of the marijuana consumed by either group was smoked in a place where other subjects were not present, although the heavy users tended to join groups more rapidly once smoking had started, and they also tended to be more verbally withdrawn than the casual users during the smoking. Mendelson and Meyer's results suggest no evidence of the amotivational syndrome. However, this may be due to the fact that the task assigned as an earning task was so simple that it would not be impaired by marijuana. Moreover, since there was a ten dollar daily limit on the amount of earnings, it could be that realistic baselines were not established early in the study. That is to say, these results could be interpreted by saying that the task was so well within the grasp of the subjects that the ingestion of marijuana, even in heavy doses over a large number of days, would have no effect in decreasing the efficiency or output of this task. Consequently, these results must be viewed with caution when relating them to the amotivational syndrome. A more recent study in which a double blind procedure was used was conducted by Frank, Epps and Rickles (1973). Here the acute and cumulative effects of daily cannabis smoking were studied. The subjects were 21 healthy college educated males who were hospitalized for 36 days. All were marijuana smokers. The design of the experiment included 4 days of pretest when no smoking occurred, 28 days of continuous daily smoking, and 4 days of post smoking testing. The subjects were randomly distributed to 3 groups, which were given marijuana cigarettes weighing approximately gram. According to which group they were in,

the cigarettes were placebo, 1 per cent THC, or 2 per cent THC. Among the tests of interest in this paper were a test for subjective estimation of 'high' (Katz-Waskow SDEQ), Digit Span Test, Word Association Test, Clomputation Test, MMPI, WAIS, TAT, and Holtzman Inkblot Test. Daily mental status and behavioural observations were also made. These authors noted no bizarre behaviour, nor do they report any cumulative effects, withdrawal, or tolerance effects as measured by the psychological and clinical test instruments. They do note that the placebo subjects showed some rise on physiological indices of anxiety, which they assume was caused by the stress of confinement in hospital. Thus, the marijuana may have had a tranquillizing effect on the other two groups. They note some impairment on mental performance tests, some elevation of moods, but put little emphasis on this point. The data from this and other similar studies are still incompletely analyzed (Frank, 1974, personal communication). Although the experiment was not designed specifically to generate evidence of the amotivational syndrome, some hints might have been obtained on tests like TAT or the MMPI, but no changes were intimated. It should be noted that in a pilot study by Frank (1973) that cigarettes with 4 per cent THC produced some disorganized thinking and paranoid ideation. One subject dropped out of the experiment because of paranoid and psychotic tendencies after only 3 days of smoking. A somewhat different approach was taken by Miles et al. (1974). They designed an experimental framework which they called a 'microeconomy'. The subjects in this experiment were 6 healthy young marijuana smokers. All except one were unemployed at the time they volunteered for the experiment. They signed a contract that they would stay in the experiment for 70 days, whereupon they would receive a bonus of 200 dollars. The essential set-up was a hospital ward where they were given a task for which they were paid in cash-equivalent tokens. The task was the construction of llj inch wooden stools which were pre-cut and held together by friction fit. A sea grass top was then woven into place. Essentially the subjects were offered a job and out of their earnings they had to pay for all entertainment, food, recreation, and clothes; liquor was available at street prices from a store where other goods such as cigarettes and toiletries were available. All other goods were ordered out for. The nursing staff dressed in street clothes and had been trained in interpersonal influence techniques. They were then instructed in a non-involvement role, restricting their interpersonal behaviour with the subjects to a neutral response to subjects' requests for transactions. Behaviour of all subjects was checked and recorded every half hour. Subjects were kept on the ward under high security conditions. On occasional walks, they were always accompanied by a staff member. The design of the experiment was as follows. For the first two weeks no cannabis; during the third week they were allowed to purchase cannabis ad libitum at street prices. During the following 28 days they were required to smoke cannabis and could purchase more at street prices should they so desire. For the following week they returned to a free purchase only situation, and in the final two weeks cannabis was withdrawn. Full medical monitoring was carried out with routine examination, blood and urine testing, etc.

All of the subjects' transactions, production, spending and savings were recorded. The MMPI and Ravens Matrices Tests were given only as screening devices, although in three later and longer studies a battery of personality and intelligence tests was given at regular intervals during the studies. Unfortunately the results from these studies are not ready for reporting. Route of administration was by smoking natural plant material - one gram cigarettes each containing 8.5 mg THC. During the mandatory smoking period the dose was 2 cigarettes, containing a total of 17 mg THC. All smoking was done under supervision. One of the more interesting aspects of the study that should be noted was that while the subjects were all very experienced cannabis users, they were not disinterested in economic gain. After only 8 days in the study they took unanimous action to negotiate a pay raise from $2.00 to $2.25 per unit of production. They threatened to leave the experiment if they were not granted this raise. The second round of negotiations occurred 28 days later; the threat to leave was repeated. Production and savings had been declining, and the amount of time spent in passive entertainment was increasing. The complaint was that the high amounts of marijuana they were required to smoke did not allow them to produce. However, when the price per unit was raised to $2.75 the picture changed dramatically: production rose, less time was spent in entertainment, and less marijuana over and above the required 17 mg THC was smoked. Savings increased, although spending tended to remain stable throughout the experiment. A small but consistent positive correlation between alcohol consumption and marijuana consumption was noted. If the drop in production can be seen as evidence of the amotivational syndrome, it can certainly be overcome by making other rewards greater, in this case, economic incentives. In relation to the amotivational syndrome it should be noted that the efficiency (the amount of time taken to produce each unit of production) did not suffer even when production was low. In fact, a steady improvement in efficiency was noted over the length of the experiment. Moreover, no changes in time spent sleeping were noted, nor was the amount of time spent in toilet and grooming behaviour related to changes in marijuana dosage. In this context, '... the general lassitude and indifference which resulted in carelessness in personal hygiene...' noted by Williams et al. (1946, p.20), was not observed, nor was it noted that the subjects showed evidence of social deterioration such as that associated with '... opiate addiction and chronic alcoholism ...' (Williams et al., 1946, p.20). Since, however, we have no knowledge of the doses of THC in the marijuana administered by Williams et al., this lack of replication may well be explained by the fact that his doses may have been higher. A psychiatric assessment at the beginning and at the end of the study revealed no changes in mental status. Truly long-term studies of a longitudinal sort are rather rare in the literature. There are probably several very good reasons for this: as Robins, Darvish and Murphy ( 1970 ) point out, there has not been enough time to evaluate the long-term effect of marijuana on middleclass white adolescents, which could be interpreted as a social comment on current concerns. Moreover, there are probably very good methodological reasons as to why these studies have not been done. It is extremely difficult and time consuming to contact and recontact at a later date. Robins, Darvish, and Murphy tackled the problem in a rather novel way. They randomly selected and contacted names from old elementary school records. In this sample there were

76 users and 146 non-users of marijuana, all black males of normal or better intelligence. The authors point out that the young black population was using marijuana long before the middle-class white adolescent population. In this sample were 22 subjects who reported only marijuana use, and who started taking the drug before the age of 20. This is an important group, because marijuana use is apparently uncomplicated by other drugs. They used a questionnaire technique designed to find out the effect of marijuana use on education, sex, occupation, criminality, violence, and alcohol use. They found that only 40 per cent of the marijuana users graduated from high school, as compared with 66 per cent of nonusers. With regard to sex, marijuana users tended to show no differences from non-users in terms of age at marriage or length of courtship, but they did tend to exceed non-users in extramarital experience, and the number of illegitimate children they claimed to have fathered. The adolescent users tended to have lower prestige level jobs and much more unemployment. In terms of criminality, although marijuana users proved overall to have arrests as rarely as non-users, the marijuana only group tended to have more adult arrests for offences against property or persons. They also tended to admit to having more violent dispositions. All of the marijuana-only users who started using the drug in adolescence also used alcohol, but drinking preceded the taking of marijuana. When those subjects who used marijuana and alcohol in adolescence were compared to those subjects who used alcohol alone, both groups had a diagnosed rate of alcoholism four times greater than that in the drug free group. Although they conclude that marijuana only users showed better social adjustment than subjects who used other drugs or marijuana plus other drugs, they state, 'Every attempt to rule out a direct harmful effect of marijuana use has failed.' Although these authors state that they had no reason to believe that other drug use was being concealed, the present author's experience with marijuana users suggests that concealment of other drug use is frequently encountered. If this were the case, it would somewhat contaminate the results of this study. A further problem with studies of this nature is also the fact that it is very difficult to establish length and frequency of use and the potency of the drug used. The results reported by Robins, Darvish, and Murphy contrast rather strongly with those of Hochman (1972). Although Hochman's study is characterized as a longitudinal drug research project, which was set up in 1969, the results of the second stage of the study are not yet available, at least to this author's knowledge. Hochman sent 2200 students a 30-page questionnaire and got a 64 per cent return. Of these, 25 per cent volunteered to be tested and interviewed further. They tested and interviewed every fourth one. Their analysis compared three categories of students: experimental smokers; occasional, i.e., up to two years, smokers; and students who used cannabis three or more times per week. The acute effects of marijuana intoxication reported tended to agree with those reported by Tart (1970), and are well known. They report, as does Smart (1971), that there is a significant correlation between the use of marijuana and a tendency to experiment with other drugs. With regard to motivation, they report that a higher percentage of chronic users than non-users plan to go on to graduate school. There were no differences in the feelings of life direction, or in difficulty in choosing a career, although chronic users showed a significant difference in the number of interruptions in college study. Even so, they say that overall grade averages of users were slightly higher than non-users, but not statistically significant. With regard to the

law, the only difference between users and non-users was that users had significantly more traffic offences. Marijuana use did not affect age at marriage, number of times married, number of times divorced or separated, number of children, or marital adjustment, sexual adjustment, or incidence of extramarital sex. However, users tended to have had sexual experiences earlier, have more of them, and, not surprisingly, have a higher incidence of venereal disease. Although marijuana use did not affect the age at which the first job was taken, the total number of full-time jobs, and current employment, three times as many users as non-users quit their job because it was 'dull'. Although Hochman reports that he found no evidence of an 'amotivational syndrome', he also paradoxically reports that twice as many users as non-users could not decide, or did not know, what their plans were for the next five years. He reports also that the groups did not differ with regard to major or minor illnesses or any degree of psychological morbidity, or history of emotional illness, psychiatric or psychological therapy. Chronic users tended to be far more politically active than non-users. It is of interest to note that chronic users report that they reduce or eliminate their use of marijuana when visiting rural or small town environments, indicating that the tranquillizing effect of the drug may be an important element in reducing the pressure of city living. A study by McGlothlin, Arnold, and Rowan (1970) is unique in that it used a sample of subjects who typically first tried marijuana in adolescence, 20 years prior to the study, and subsequently used it on a regular basis for a minimum of 2 years, and who also, during the 1955- 1961 time period, took LSD in a medical setting, half in psychotherapy, and half as volunteers for experiments. The data reported in the paper consist of comparisons between this group and two other samples from the LSD group of approximately the same age, sex ratio, education, and proportion in art-related professions. The majority of subjects who used marijuana heavily as adults began its use in adolescence. Only one of the 17 respondents introduced to marijuana during 1961-1968, typically in the 30 to 40 age range, has ever used it on a regular basis. Most middle-class persons during the time period in this study were never exposed to the drug. Others were deterred by concern over the legal hazards. Another variable of importance is occupation - among the middle class, people in the entertainment field and other art-related professions were much more inclined to find marijuana to their liking than other groups. They were more likely to use it while working, and to report that it either had no effect or that it enhanced their performance. The most consistent correlates of marijuana use in these subjects were a continuing involvement in the various means of altering consciousness, and an unstructured and relatively unstable style of life. The search for altered states of consciousness was indicated by the tendency to use or at least experiment with a wide variety of drugs other than marijuana, but it is also evident in various non-drug use areas. Marijuana users were significantly more prone to non-drug induced regressive states as measured by a hypnotic susceptibility test. They were also much more likely to believe in the validity of astrology, ESP, etc., and were likely to have practised one or more disciplines such as meditation, yoga, or Zen for a period of several years.

In terms of lifestyle, they tended to prefer a high level of stimulation, uncertainty and risk, instead of security and structure. They evidence more than average work and residence instability, they are especially unlikely to maintain a permanent marriage, they are more alienated from a socio-cultural standpoint, but not in a personal sense, i.e., their alienation does not take the form of pessimism, self contempt, and barren interpersonal relationships. As indicated by Robins, Darvish, and Murphy, the user is more likely to have had a history of heavy alcohol use, but strangely, not especially attracted to sedatives and tranquillizers. Other patterns of use include periods of months or years in which marijuana is used virtually to the exclusion of alcohol, or vice versa. Regular users of marijuana in this study cite the relaxing or tranquillizing effect as a major motivation for using marijuana, as well as euphoria. However, the results also tend to support the position that marijuana abusers would be drawn from the population who use, or would use, alcohol to excess. From the behavioural standpoint, there is no simple relationship of substituting one drug for the other. The majority of marijuana users reported that their use of marijuana reduced their total consumption of alcohol. Although the authors made no systematic attempt to explore harmful results of excessive use of alcohol, several respondents reported their use of alcohol had been considerably more disruptive than their use of marijuana. Specific problems associated with excessive use of alcohol were drunken driving, assaultive behaviour, and an inability to work. The degree of dependency was felt to be greater for alcohol. The authors conclude that if a white, middle-class, middle-aged person began using marijuana as an adolescent some twenty years ago and continued on a regular basis for several years, there is a high probability he has a history of fairly heavy use of the intoxicant that is somewhat more socially approved, in other words, alcohol. From the results of this study, it does not seem likely that marijuana use would replace alcohol for many persons, however, the overall consumption of alcohol might be reduced. A study by Sadava (1972) points out the importance of not using personality characteristics exclusively as predictors of cannabis use. He argues that the individual differences approach exclusively is inadequate for explaining cannabis use. He cites, for example, anxiety and alienation as correlates of student drug use, but not exclusive characteristics. Sadava gives great weight to the Becker (1953) hypothesis, where marijuana use seems to be a consequence of social experience, or social learning. His basic conceptual framework is field-theoretical, and he refers to cognitive as well as to behavioural learning. Basically, he views deviant behaviour as consistent with other behaviour in a more general social-psychological framework. In this study, Sadava tested a total of 319 subjects in the freshman week of September, and they were again tested the following March. He was fortunate enough to retain 89 per cent of his original sample. Of the original 319, 110 had used cannabis before college orientation week (the user group); 168 remained as non-users throughout the college year (the non-user group); and 41 of the non-users were initiated to cannabis during the college year (focal initiative group). The data for males and females were consistent, and therefore were combined. His basic measure was the extent to which a subject believed in internal versus external control, i.e., the extent to which a subject sees negative and positive consequences happening

to him as contingent on his own behaviour (internal) or upon luck, or powerful others (external). Thus cannabis use, in a sense, may be a functional substitute for defensive externality. A basic orientation of this research then, was to examine the personal meaning of use or abstinence in terms of gaining positive satisfactions or avoiding negative consequences. Therefore, high positive function scores and low negative function scores should predict cannabis use as a selected deviant behaviour. Other factors, such as religious affiliation, social class, and ethnic background, did not predict marijuana use. In summary, it was hypothesized that the users relative to non-users should have lower expectancies in the area of academic recognition, social love and affection, and high expectancies for independence and internal control orientation. Moreover, they should have positive and fewer negative functions for use. It was also postulated that users should have greater ethical tolerance for use and greater social support. If these variables are good predictors, they should differentiate initiates from non-users prior to their initiation to cannabis use, and the initiate versus non-user distinctions should be increased subsequent to initiation. The author concludes that both personality dispositions and social environmental variables are important in predicting whether a college student will begin to use cannabis in college or not. One interesting point arises from this research, in that the expectations for academic recognition prior to college orientation were somewhat lower for the user group than the nonusers, but this difference disappeared at the time of the second testing in the spring. The postulated lower expectancy for social love and affection was not shown; on the contrary users in both fall and spring testing exhibited a high expectancy. Within a similar theoretical framework, Jessor, Jessor and Finney (1973) studied a student population consisting of junior high, senior high and college students. These authors were concerned to place cannabis use within a more general social psychological framework. In particular, they were using as a background the social learning theory of Rotter (1954) and Rotter, Chances and Phares (1972). The report is a part of a larger, ongoing study of socialization problem behaviour in youth. They point out that the personality-environment interaction has with very few exceptions been ignored by most drug research, and are concerned to get away from the descriptive or epidemiological standpoints. They conceptualize marijuana using behaviour as probably only one of a number of problem behaviours. In this background, four hypotheses are advanced;(i) that marijuana use should co-vary with other kinds of problem behaviours; (ii) variation in marijuana use should be systematically related to variations in the personality and perceived environment; (iii) the onset of marijuana use among non-users should be predictable from initial differences on those same personality and social variables; (iv) irrespective of initial differences, those who begin marijuana use during this period should manifest greater relative change of the personality and social variables in the theoretically expected direction. In general then, marijuana use is seen like any other learned behaviour, as a functional outcome of the interaction of personality, social and behavioural attributes. A theoretical summary profile of the social psychology of the student marijuana user put forward by Jessor, Jessor and Finney (1973) is that first, there should be greater instigation to use, stemming from higher value on independence and lower value on achievement, and lower expectations for achievement. Second, there should be less belief in controls against

use, and greater social criticism and alienation. Third, less personal controls against use, greater tolerance of transgression and less religiosity. Fourth, there should be greater environmental support for use, greater peer-orientation, models and supports; and fifth, greater experience with other behaviours of a 'problem' or 'problem prone' nature. The sample from this study was asked to take part in a 4-year study of personality and social development in youth. It was a large sample and they were given extensive personality testing plus a marijuana-behaviour report, which was a 4-point scale of involvement. The crosssectional analysis gave support to hypotheses(i) and (ii). Users valued achievement less and independence more than non-users, and they also tend to have lower expectations for achievement. In terms of belief structure, users tended towards greater alienation, and were more socially critical. In terms of personal controls, users were more tolerant of deviance and showed relatively less religiosity. In terms of perceiving environmental controls, the users saw less compatibility between peers and parents, and acknowledged that their peers had greater influence on their views than their parents, thus there were more models, pressures, and peer approval for drug use. These findings held for the junior high school, senior high school, and college samples, and in general held across sex and intensity of use. In the longitudinal analysis, which involved retesting after a year, hypotheses (iii) and (iv) were tested. The results showed first, that among non-users there are initial differences among the variables, such that they are predictive of the onset of. marijuana use over time. Second, there was a greater relative change in the problem prone direction for those who share user status than for those who remained non-users. These differences held for both males and females. However, as the authors point out, with regard to hypothesis (iv), it is not possible to determine whether changes in the variables precede or follow the behavioural change. There were certain differences in the college population with regard to hypothesis (iii). The proposed variables did not predict the change from non-user to user status. The authors suggest that this may be in part explained by the high rate of use amongst the college sample: in this case, 70 per cent. They note that the variable of social criticism was much more important at the college level, suggesting a greater idealogical role of marijuana use in college and less of a role in coping with failure, as may be the case with marijuana use in high school. This study (Jessor, Jessor and Finney, 1973) is of particular significance because of its methodological soundness and its attempt to relate a general social-psychological theory to drug using behaviour. Unfortunately, little information is given on the extent of multiple drug use and its relationship to the magnitude of changes. Although in North America it is difficult to find populations who have been using cannabis for long periods of time, it is possible to find communities where large numbers of people have been chronic users of cannabis. Of the more notable work in this area are the investigations directed by Rubin and Comitas (1972) from the Research Institute for the Study of Man. As Rubin (1973) points out, it is important to study the cross-cultural aspects of cannabis smoking, not only because it is possible to find chronic smokers using extremely high dosages, but also to see whether some of the effects which are reported in North America and other parts of the world are culturally conditioned, or whether they are a direct consequence of pharmacological effects. Bearing directly on the amotivational syndrome hypothesis is the work of Schaeffer (1972), reported in Rubin and Comitas (1972). The general air of this research was to find out the relationship of cannabis smoking to agricultural pursuits, to the exploitation of land resources, and to exchange relationships, and its effect on the economy. They were particularly

interested in learning whether acute and cumulative effects of the use of cannabis had any effect on energy expenditure during work in agriculture. Their method was to observe and videotape agricultural work in progress. They were able to observe the number of times per day and they were also able to get estimates of the g-THC content of the material smoked. An important measure in these studies was the number of kilocalories required to complete a task when smoking and when not smoking. They were able to study heavy users who smoked 3 to 8 times daily, smoking as much as one ounce of cannabis; and very light users who use cannabis less than once a week. Jamaican cannabis is usually of very high potency and has been reported as high as 7 per cent THC. In general they report that the use of cannabis is related to effective alterations in the rate and organization of movements, and the expenditure of energy during work. Most of the smokers enacted more movements per minute and they tended to expend more kilocalories per unit of space immediately after cannabis use. However, between 20 to 40 minutes after smoking, the alterations in behaviour appear to return to normal, and between 80 and 140 minutes, feelings of fatigue were often expressed in movement. Dose, frequency of use, and psychological set during use are all important, together with the situational context. The results of the micro-analysis of case studies were summarized as follows: (i) Use of cannabis is related to effective alterations in the rate of organization of movement and expenditure of energy. (ii) Behavioural changes related to light or moderate use are not significant in agricultural pursuits over an extended time period. Behavioural changes related to heavy use are significant in agricultural pursuits over extended time periods. (iii) Alterations related to both moderate and heavy use are appropriate to social cohesiveness during work in group situations. (iv) Total space covered, or amount accomplished, or in number of plants reaped, is usually reduced per unit of time after smoking. The number of movements per minute is often significantly greater after smoking, as is the total number of movements required to complete a given task. Within the same series of studies, Beaubrun and Knight (1973) hospitalized 30 chronic users and 30 matched control non-users for 6 days. All of the subjects were male. Chronic users were defined as those who had smoked daily for not less than 10 years. The object of the study was to make psychiatric and psychological comparisons. On the behavioural and psychological variables, there were no differences between the two groups in the incidence of mental illness, alcoholism, or number of arrests. There were also no differences between the groups on the Eysenck Personality Inventory, or extroversion or neuroticism scores. A mental status examination revealed no significant abnormalities and no difference in the use of other drugs. The work record of each group was very similar, but this may be because the sample was made up of subjects who had occupations where there is little social mobility. Two important differences between the groups appear to be worth mentioning: one is that the family history of mental illness, excluding alcoholism, was significantly greater in the sample of smokers and not surprisingly, the smokers had reported more hallucinatory experiences, although these were when they first used the drug. These reported effects (Beaubrun and Knight, 1973) are consonant with the work of Bowman and Pihl (1973). Bowman and Pihl did two studies which were almost identical, and were designed to study specifically the psychological effects of chronic use of marijuana on

intellectual functioning. The experimental group of subjects were typically smoking daily, and they were smoking material which had 4 to 5 per cent THC content. The users in the sample had typically begun smoking around about 12 years of age, and had been using cannabis for 164 years. The mean consumption per week was 167 grams, which is the equivalent of 23 standard cigarettes per day. Matched controls of non-smokers were used in both studies. Both groups were tested on three sets of tests. The first was concerned with physiological, sensory, and perceptual motor functioning. The second set measured concept formation, and abstracting abilities. The third set was comprised of a number of memory tests. They found no differences between the groups. It is important to note that the subjects were required to be not high at the time of testing. It is worth mentioning that they point out that North American users suggest the drug enhances their sexual pleasure or performance, while users in this study scorned the notion, and gave the impression that they regarded this as a perverted use of the drug. The amotivational syndrome was denied by the subjects here, because the subjects often smoked specifically, as indeed in India, to provide energy and strength in order to accomplish particularly heavy tasks. They underlined Rubin's point (1973) that many findings may be cultural artifacts, and that the effects of the drug may be very much controlled by cultural expectations. They also comment on the possibility of tolerance and cite examples of people who have consumed as much as 5 ounces of cannabis in 5 hours, showing only minimal behavioural effects. Acknowledgement The author wishes to acknowledge the assistance of Janet McDougall and Frances Theodor, research assistants, and Barbara Vary and Evelyn Wollis, for their help in preparing the manuscript. REFERENCES Becker, H.S. Becoming a marihuana user. American Journal of Sociology, 59, 235-242 (1953). Beaubrun, M. and Knight, F. Psychiatric assessment of 30 chronic users of cannabis and 30 marched controls. American Journal of Psychiatry, 130, 309-311 (1973). Bowman, M. and Pihl, R.O. Cannabis: the psychological effects of chronic heavy use. A controlled study of intellectual functioning in chronic heavy users of high potency cannabis. Psychopharmacologia, 29, 159- 170 (1973). Canada, Commission of Inquiry into the Non-Medical Use of Drugs. Interim Report. Ottawa: Queen's Printer, 1970. ('LeDain Report'). Cappell, H., Webster, C.D., Herring, B.S. and Ginsberg,R. Alcohol and marihuana: a comparison of effect on a temporally controlled operant in humans. Journal of Pharmacology and Experimental Therapeutics, 182, 195- 203 (1972). Frank, I.M., Epps, L.D. and Rickles, W. Psychological and physiological effects of chronic marijuana administration in man. Psychopharmacology Bulletin, 9, 28-29 (1973). Great Britain, Advisory Committee on Drug Dependence. Cannabis. London: Her Majesty's Stationery Office, 1968. ('The Wooten Report'). Hochman, J.S. Marijuana and social evolution. Englewood Cliffs, N.J.: Prentice-Hall, 1972. Indian Hemp Drugs Commission, 1893-1894. Report on Indian Hemp. Vol.1-7. Simla: Government Central Printing Office, 1894. Jessor, R., Jessor, S.I. and Finney, J. The social psychology of marijuana use: longitudinal studies of high school and college youth. Journal of Personality and Social Psychology, 23, 1-

15 (1973). Mayor's Committee on Marihuana. The marihuana problem in the City of New York. Lancaster, Penn.: Jacques Cattell Press, 1944. ('The LaGuardia Report'). McGlothlin, W.H., Arnold, D.O. and Rowan, P.K. Marijuana use among adults. Psychiatry, 33, 433-443 (1970). McGlothlin, W.H. and West, L.J. The marijuana problem: an overview. American Journal of Psychiatry, 125, 126-134 (1968). Mendelson, J. and Meyer, R. Behavioral and biological concommitants of chronic marijuana smoking by heavy and casual users. Appendix, VoZ.I, Marihuana: A signal of misunderstanding. The technical papers of the first report of the National Commission on Marihuana and Drug Abuse. Washington, D.C.: U.S. Government Printing Office, 1972. Miles, C.G., Congreve, G.R.S., Gibbins, R.J., Marshman, J.A., Devenyi, P. and Hicks, R.C. An experimental study of the effects of daily cannabis smoking on behaviour patterns. Acta Pharmacologica et Toxicologica, Supplement (In press, 1974). Robins, L.N., Darvish, H.S. and Murphy, G.E. The long term outcome for adolescent drug users: a follow-up study of 76 users and 146 non-users. Proceedings of the American Psychopathological Association, 59, 159-180 (1970). Rotter, J.B. Social learning and clinical psychology. Englewood Cliffs, N.J.: Prentice-Hall, 1954. Rotter, J.B., Chances, J. and Phares, E.J. Applications of a social learning theory of personality. New York: Holt, Rinehart, and Winston, 1972. Rubin, V. The 'Ganja Vision' in Jamaica. Unpublished manuscript. (Research Institute for the Study of Man). Rubin, V. and Comitas, L. Effects of chronic ganja smoking in Jamaica. A report by the Research Institute for the Study of Man, to the Center for Studies of Narcotic and Drug Abuse, National Institute of Mental Health, Contract No. HSM-42-70-97, 1972, Sadava, S.W. Initiation to cannabis use: a longitudinal social psychological study of college freshmen. Canadian Journal of Behavioural Science, 5, 371-384 (1973). Schaeffer, J.H. Cannabis sativa and agricultural work in a Jamaican hill community. In: Rubin, Vera and Comitas, Lambros: Effects of chronic smoking of cannabis in Jamaica. Report by the Research Institute for the Study of Man, to the Center for Studies of Narcotic and Drug Abuse, National Institute of Mental Health, Contract HSM-42-70-97, 1972. Smart, R.G. and Fejer, D. Recent trends in illicit drug use among adolescents. Canada's Mental Health, Suppl. 68 (1971). Talbot, J. and Teague, J. Marijuana psychosis. Acute toxic psychosis associated with the use of cannabis derivatives. Journal of the American Medical Association, 210, 299-302 (1969). Tart, C. Marijuana intoxication: common experiences. Nature, 226, 701-704 (1970). United States Department of Health, Education and Welfare. Marijuana and Health. Washington, D.C.: U.S. Government Printing Office, 1971. (First Annual Report to Congress). Williams, E.G., Himmelsbach, C.D., Wikler, A., Ruble, D.C. and Lloyd, B .J., Jr. Studies of marijuana and pyrahexyl compound. Public Health Reports, 61, 1059-1083 (1946). (Reprint No.2732). DISCUSSION Dr Edwards was concerned to define what is meant by 'amotivation' in comparison with normal behaviour. Dr Connell observed that in England in the late fifties and early sixties, before cannabis use had become widespread, a syndrome had been described entitled the 'lying-abed syndrome' in adolescents. This seemed to

be a feature of adolescent emotional development and was not at that time related to cannabis. If cannabis had been taken the behaviour might well have been attributed to cannabis. Dr Edwards stressed that amotivation means lowered motivation rather than no motivation. MATCHING Professor Paton asked about matching control and experimental groups. 'If one takes a group and matches them with another, one may inadvertantly mask one of the effects one is looking for. If you suspect that amotivation may be an effect of cannabis, and that amotivation may affect income, social class or educational level, then by matching for these factors you delete the effect you are looking for. Or we could find ourselves investigating the possibility of, say, liver damage due to drug use, and matching the control and experimental groups for equal health. Should one leave out of the matching procedure any factor which the drug might affect?' If it is possible to match the experimental to the control group before drug use occurs, then this problem is evaded. This is, however, only possible in experimental laboratory studies and not in longitudinal studies, since the criterion for placement in the control or experimental group is the drug behaviour. In longitudinal studies one can only collect information about individuals in a population before drug use, and then after drug use, try to match some of the users with some of the non-users on the basis of the previously collected information, discarding unmatched individuals. HOSPITALISATION EFFECTS Dr Leuw and Mr Hasleton raised questions about the effects of being hospitalised. Quite apart from effects of being constantly observed in rather artificial circumstances, there is the question of the picture of themselves as cannabis users that subjects might be motivated to present. Dr Miles explained that the staff were trained in techniques of nonverbal communication and influence, and then instructed to monitor their own behaviour and to minimise their influence on the subjects. It was agreed, however, that not all such influences can be completely controlled, and that meanings of the situations, which may have implications for behaviour, cannot be equalised for all subjects. Dr Rubin stressed the need for field studies of users in their natural environment, in order to understand the Gestalt of social behaviour of which cannabis use is a part. Sociological and anthropological techniques of data-collection helped to fill out and place in context the results of experimental and clinical studies. SEX DIFFERENCES Dr Miller pointed out that 'until very recently, nearly all the studies of cannabis use and effects were based on male college students, male psychiatric patients, male prisoners, and male white rats. Typically, if general information about the species is sought, males are used. Almost without exception females have only been studied if some 'specifically female' characteristic is in question. What comparisons were you able to carry out, and what differences did you observe between males and females?' Dr Miles responded that 'we were pretty unsuccessful in comparing males and females, because female behaviour patterns were

so different. For example, they were very generous, giving their money away. And in connection with sex; the men did not care at all, but the women felt extraordinarily deprived and they used to act out. We got a call at 4 o'clock in the morning complaining that the girls were acting out at the windows, naked. These were the main differences; also they didn't smoke so heavily.' CANNABIS AND MOTIVATION Dr Rubin gave some more information on her study of Jamaican agricultural workers' use of cannabis. Cannabis was used in relatively high doses to get energy for arduous tasks, and for weeding, sawing, and cane-loading cannabis use was associated with harder working. This did not always result in a greater throughput of work, or greater efficiency. In spite of the higher expenditure of energy in weeding, there was a drop in efficiency of sowing, and current analysis of videotapes was investigating a possible increase of efficiency of can-loading. Participants were interested to hear that Dr Miles' experimental subjects held a strike and forced renegociation of their system of rewards. Asked about this 'strike', Dr Miles said: 'They threatened to leave. Here you are, you have a staff of 25 people and a hospital floor. What can you do? The strike regulations lasted two days. I put them off one day and they were going to leave the next if they didn't see me.' In general, there was agreement that there was no clear evidence that cannabis in and of itself affected motivation in a particular way.

5. Cannabis Psychosis
N. H. Rathod, St. Christopher's Day Hospital, Horsham, England. INTRODUCTION The relationship between use of cannabis and psychosis has been under discussion for over a century, but opinion still remains divided. The drug has been implicated as the cause of psychiatric disorders it is claimed to cure,and many more besides. Some (e.g. 3, 5, 10, 11, 12, 13, 17, 30, 33, 42) are convinced that not only does it cause psychoses but the syndrome is distinctive. Others (e.g. 18, 37) question these views. And of course, there are others still, (e.g. 2, 6, 7, 44, 48) who give qualified assent. Rates of hospital admission vary between 3 per cent to 27 per cent. The literature on the subject is extensive and varied. Differences in samples, varying authenticity of evidence, cursory attention to clinical details and other factors make valid conclusions difficult. It may be fruitful to view the subject in terms of historical perspective (mid-nineteenth to midtwentieth century), to examine the more recent literature with reference to some of the basic

requirements of clinical research and to propose some ideas for research in the immediate future. HISTORICAL It was in 1828 that O'Shaughnessy(38) observed cataleptic reactions amongst patients receiving liquid extract of cannabis. In his paper published in 1843 he presents detailed observations on one of his three patients suffering from rheumatism developing catalepsy while being treated with cannabis. In the same paper he gives an account of the experimental administration of the drug to animals and humans. The drug produced no harmful effects. Cannabis produced euphoria followed by sleep. In 1844 Jacque-Moreau, according to Oriana Kalant (28), published detailed accounts of effect of experimental as well as therapeutic administration of cannabis. Amongst other effects he observed varied hallucinatory phenomena. He likened these to dreaming whilst awake. He also drew attention to the subject having insight into these experiences and his ability to control the effects, and noted the relationship between dose and effect. He also treated psychotics with benefit and 5 of his 7 manic patients treated with fatty extracts of cannabis recovered. Kalant observes that these two 'contributions, almost simultaneous and independent of each other, constituted the beginning of modern scientific and medical study of cannabis' and, indeed, of experimentally induced psychoses. In 1894, i.e. 50 years later, the Indian Commission on Hemp Drugs (24) published its report. Their enquiries into social aspects of the use of cannabis (insanity) psychosis are to date unsurpassed in scope, meticulousness and detail. They questioned over 400 doctors and 700 other witnesses, made on the spot enquiries in mental hospitals, and also asked for supplementary evidence from sources other than medical. They drew attention to (a) dangers of observer bias, (b) need to check clinical claims and (c) relationship between dose and effect. 222 of the 1344 admissions to mental hospitals in 1892 were attributed to cannabis. The Commission accepted that only in 98 the use of hemp drugs 'may be reasonably regarded as a factor'. When mixed cases, i.e. cases in which the hemp drugs were only one of the 'several possible causes', were omitted the Commission was left with 61 cases. Regarding these 61 cases the Commission commented with commendable caution: 'Even in regard to the remaining 61 cases, it must be borne in mind that it is impossible to say that the use of the hemp drugs was in all, the sole cause of insanity or indeed part of the cause'. The number is reduced by a further 22 because corroborative information was impossible to obtain. Thus a large part of clinical claim was found to be invalid, as, of the 222 reported cases, it was only in 39 cases that cannabis may have been the factor in or cause of insanity. (Pages 239-241). Recently a contribution of Baker and Lucas (4) reveals a parallel situation in the U.K. They scrutinised the records of 79 patients in whose admission cannabis was claimed to be a factor. In 51 or over 60 per cent of cases the role of cannabis was either assessed as irrelevant or the evidence was very inadequate. The strictures of the Indian Hemp Commission on clinical opinion may be found to be as relevant in 1974 as they were in 1894. They say (p.233 'The most striking feature of the medical evidence is perhaps the large number of practitioners of long experience who have seen no evidence of any connection between hemp drugs and disease, and when witnesses who speak of these ill effects from the moderate use are cross examined, it is found that: (a) their opinions are based on popular ideas on the subjects; (b) they have not discriminated between the moderate and excessive use of the drug;

(c) they have accepted the disease as being induced by hemp drugs because the patient confessed to the habit.' No wonder they felt that the claim for hemp insanity was exaggerated. As to classification of disorders attributed to cannabis , mania was the commonest, followed by toxic insanity and these accounted for 81 of the accepted 98 cases, melancholia was seen in 3 and dementia in 4. In 1933 Bromberg (7) from New York published detailed observations on toxic confusional states and psychotic reaction amongst 10 cannabis users. He repeatedly stressed the role of the patient's personality and made a plea for precision in diagnosis. His clinical investigations were thorough and accounts detailed. He divided the clinical picture into three categories Intoxication alone, Reactive states accompanying intoxication, and Toxic psychoses - an admixture of intoxication and basic cyclothymic or schizophrenic reaction. Intoxication (i.e. toxic confusional state) formed the common denominator of all the reactions . In 1939 (11) the Chopras reported on years of their research into varieties of issues related to cannabis. They commented on the epidemiological aspects and estimated the incidence of psychoses amongst chronic users as ranging between 0.5 per cent to 2 per cent amongst the 1,238 subjects they studied. They also differentiated between effects of moderate and heavy use and referred to the popular use of the drug by physicians in indigenous medicine in India for many illnesses. It is interesting that the possible therapeutic value of cannabis, and harmlessness of the substance if used in 'moderation' have received inadequate attention in contemporary scientific literature. The literature is heavily weighted in favour of 'adverse effects of the drug'. In 1944 the first broadly based controlled study of the clinical effects of cannabis came from the Mayor of New York's Commission of Enquiry (36). Referring to the existing literature they were prompted to observe, 'Relationship to varying dosage, to the subject's personality and background and to the environment when the drug was taken is given little or any attention', - remarks so true even today. Their sample was drawn from prison population and included both sexes, negroes and whites. Five of them were novices and 72 had a history of variable abuse of different drugs. DOserelated effects in clinical as well as social settings were studied. Patients were used as their own controls; nine patients developed psychotic reactions; 5 of them were novices. Of these six patients developed psychotic reactions with evidence of confusion which terminated as the drug wore off, and 3 patients developed psychosis after being sent back to prison. Fear, suspiciousness, restlessness were the commonest effects. The role of dose and setting on clinical effect is made obvious by their meticulous attention to details, (e.g. cases of E.C., sex male, novice, H.W. sex male, novice p.47). Allentuk and Bowman (2) have summarised the findings:- 'Characteristic marihuana psychosis does not exist. Marihuana will not produce a psychosis de novo in a well integrated stable person. It may precipitate a psychosis in an unstable, disorganised personality when it is taken in an amount greater than he can tolerate. Under such circumstances the previously mentioned physical and psychic manifestations become quantitatively greater and new events arise,' e.g. assuming grotesque statuesque positions and auditory hallucinations.

In 1946 Freedman and Rockmore (16) reported that none of the 310 chronic users (7 years +) they studied had a history of hospitalisation for psychiatric disturbance. In 1946 Williams, Himmelbach and Wikler (49) reported withdrawal effects on 12 prisoners who were allowed ad lib use of the natural substance or of synthetic compound 'Parahexyl' under controlled experimental setting. The main feature of withdrawal effect was disorientation. In 1949 Fraser (17) described 9 cases of withdrawal syndrome amongst Indian troops fighting in South-east Asia. The patients showed unpredictable violence. The behaviour was waxing and waning and lasted up to 5 weeks. The studies mentioned so far had between them pointed to many ground rules for the study of psychoses in a clinical setting - rules which are by and large ignored. Briefly stated they point to: 1. The need to ensure reliability of the evidence. 2. The need for meticulous attention to clinical details. 3. The need for precision in diagnosis. 4. The dangers of assuming cause and effect relationships too readily. 5. The relationship between amount (or dose) and effects. 6. The value of controlled studies. 7. The value of epidemiological approach. 8. The possibility of withdrawal 'syndrome'. IMPLICATIONS The term 'Cannabis Psychosis' has three important implications: 1. Therapeutic 2. Scientific 3. Social 1. Therapeutic. In a clinical setting diagnosis is based as much on existing knowledge about evidence as on the unique inter-action between the clinicians and the patient. Its primary function is to plan therapy and its implications therefore are limited to the individual patient, e.g. on medical advice he may have to abstain from using cannabis, or reduce its consumption, or ignore its role in the psychiatric disturbance altogether. 2. Scientific. The situation alters materially when the term is used as an aid to scientific knowledge. Its implications are far reaching and crucial to knowledge. Data need to be reliable and uniform; capable of scrutiny by others; and criteria of diagnosis clearly documented. If these points are not met, it is difficult for others to validate them and much worse, faulty conclusions may result. Judging from the literature such approaches are not popular. (e.g.3, 4, 5, 6, 11, 12, 13). Contributors often use evidence collected by others in day to day clinical practice. Meticulousness and uniform standards cannot be guaranteed. The reader is left with the task of drawing valid conclusions when the data does not lend itself to such an exercise. Instead of detailed clinical history and observations authors (e.g. 5, 12) use summary terms such as schizophrenia, depressive mania, aberrational and oneiretic states, and many others; without adequate criteria. This is done, possibly with the misplaced assumption that the reader will comprehend what the author has in mind, but which is not made explicit. It is only in the presence of detailed evidence that the reader is in any position to make

evaluative judgment, e.g. the evidence produced by Campbell and Evans et al (8), enables us to suggest that factors other than smoking cannabis may have contributed to the clinical picture. Head injury and more than occasional use of L.S.D. by some of their subjects may be relevant factors. We also learn that their diagnostic instruments are not adequate to warrant diagnosis of organic brain damage. Kolansky and Moore (33), stated that 'The symptomatology seen in 13 patients imply biochemical or structural change in cerebral cells as a result of chronic cannabis use.' Clinical evidence does not warrant such implications - as no biochemical studies or investigations for organic damage were undertaken. Similarly, it becomes clear that many of their subjects were either showing signs of psychosocial disturbance, and/or were using other hallucinogens before starting regular use of cannabis. These examples illustrate the relevance of detailed evidence. 3. Social Significance. Choice of the least dangerous and at the same time most satisfying intoxicant is not easy for the ordinary person. He looks up to 'experts', e.g. clinicians and scientists, for guidance. These prestige groups have a responsibility in advising society under-estimation of risks is hazardous and over-estimation is likely to cause undue anxiety, and also devalue their authority and authenticity. Furthermore social statements need to be relative, i.e. stating the relative dangers or benefits of various intoxicants in use. In this context it is interesting that the professional literature on cannabis is heavily weighted in favour of its dangers, and reports comparing its effects with others in common use, e.g. L.S.D., amphetamines and alcohol are few, (e.g. 22, 26, 41). CLASSIFICATION OF 'PSYCHOTIC' EFFECTS: Cannabis related psychotic phenomena can be divided into three groups: 1. Dose related acute psychotoxic reactions (e.g.50). They invariably show evidence of organic confusion. This is usually associated with anxiety, suspiciousness, restlessness and maybe violence, and perceptual disturbance. Their duration is measured in hours or days and recovery is complete. They are reported equally in Eastern and Western literature. (e.g. 4, 13, 16). 2. Idiosyncratic reactions: (e.g. 44, 48). These have features similar to acute psychotoxic reactions, except that (A) organic confusion is not always found. (B) Anxiety or panic states dominate the picture; and (C) subjects are usually novices and the amount consumed may be very little. 3. Other psychiatric disorders: Almost all major psychiatric disorders have been ascribed to cannabis. They range from psycho-neuroses, functional psychoses (e.g. Mania/Schizophrenia) to dementia and personality deterioration. Frequency of such diagnoses varies, e.g. schizophrenia was diagnosed in 5 per cent of cases in an Indian study (12) and in between 58 to 60 per cent of cases in a Swedish study (6), and in a Nigerian one (3). On the other hand, Weil (48) claims that marihuana does not seem to trigger true psychotic reaction except in persons with a history of psychoses or hallucinogenic drug experimentation. Duration is equally variable and course unpredictable. They occur predominantly or solely amongst chronic users - and therefore African and Eastern literature is rich in such reports.

The existence of the first two conditions has been observed repeatedly in clinical practice and confirmed by controlled studies. (e.g. 3, 4, 21, 25, 31, 36, 44, 46, 48). However, it is relevant to point out that cannabis is not unique in giving rise to dose related toxic states or idiosyncratic reactions. They occur with many other drugs. The controversy really centres around psychiatric disorders in group (3) i.e. other than toxic states and idiosyncratic reactions. I have chosen to look at the more recent clinical research in this controversial area keeping in mind the historical guide-lines mentioned above. While much of the disagreement may in part be due to defective research techniques, it is worth stressing that clinical research in this field is hampered by practical constraints over which the clinician may have little control, e.g. selection of the sample, availability of reliable witnesses or such witnesses (e.g. parents) not having much knowledge of patients' drug habits. The following discussion is meant only to highlight some of the problems encountered in clinical research in this field. THE TERM Term Cannabis Psychosis presupposes that: 1) The claimed use of cannabis is valid and that it is the only drug involved; 2) There is agreement on criteria for diagnosing psychosis; 3) Cannabis causes the psychosis. Patients' claims should be viewed with some scepticism specially in countries where cannabis is available only through illegal or underground sources. What the patient believes he has bought and used may not be what he has been sold, e.g. Patterson (39) reports that nearly one-third of the samples purchased by users contained no cannabis. In other words, even if the patients' claims are reliable, they need not be valid. The amounts of active ingredients also vary (50). There is also the problem of cannabis being used concurrently or alternatively with L.S.D. and effects of the two drugs being similar. Furthermore, L.S.D. flashbacks are similar to those caused by cannabis and cannabis may precipitate such flashbacks, ( e.g. 29, 30, 32). In India, Datura (Stramonium) is often mixed with cannabis. It too can cause psychotic reactions. By and large little attempt is made to checking the validity of patients' claims. Agreement about Psychosis: Diagnosis in psychiatry is hampered because evidence is often ambiguous. Agreement between independent raters is difficult to achieve because criteria for a particular diagnosis are not uniform and due to individual clinician bias. Diagnosis is also influenced by other factors, e.g. as Hollingshead and Redlich (20) point out, people from lower social classes are more likely to be diagnosed as suffering from psychosis such as schizophrenia than others; and the majority of the cannabis smokers in the Middle East and India belong to the lower strata of society. But an even greater cause for concern about diagnosis is the lack of enough attention to its significance. Discussing the issues involved, Per Dalen (14) observes:- 'The theoretical importance of diagnoses is not heeded. They are not treated as hypotheses to be subject to

test, but rather as conventions or intuitively and implicitly defined concepts of classification.' But even when diagnoses are put forth as hypotheses one needs to guard against personal prejudice because it can affect research design and outcome. In the words of Weil, Zinberg and Nelson (47) 'the researcher who sets out with prior conviction that hemp is psychotomimetic or a mild hallucinogen is likely to confirm his conviction experimentally, but he could probably confirm the opposite hypothesis if his bias were in the opposite direction.' In summary: Reliable and agreed diagnoses, although important, are unlikely to obtain in the absence of uniform criteria, detailed evidence and independent assessment of evidence to exclude personal bias. CAUSALITY Given the history of use of cannabis, cause-effect relationship between the drug and various psychiatric disturbances is too readily assumed. Undoubtedly psychiatric disorders do occur amongst cannabis users, but the implication that cannabis is the or the major cause is more often than not unwarranted. To illustrate: In 1970 I sent a questionnaire to 20 Senior Psychiatrists in India. Twelve replied and estimated that between 1 - 3 per cent of all hospital admissions were due to cannabis psychosis. In 1971 I had an opportunity to see 18 of these people individually in India. Asked, if in the absence of history of cannabis use would they have been able to claim that the psychoses was due to the drug - the answer was 'NO. They could not.' In the same year I visited a well known mental hospital in Northern India. I was shown a ward of 40 males, which housed 10 'typical' cannabis psychosis patients. Their clinical picture and history of cannabis use was similar to twenty or so others in that ward who were not diagnosed as suffering from cannabis psychoses. No explanation was forthcoming to explain this discrepancy. ROLE OF ANTECEDENTS OTHER THAN CANNABIS Little or scant attention is paid to family history, life experiences, and patients' personality (e.g. 3, 6, 5, 30, 33, 44). This is surprising in view of their role in development of psychiatric disorders, including functional psychoses and amotivation syndrome. Cannabis may, in some circumstances, precipitate psychiatric disturbance. In this sense it possibly acts as a nonspecific stress like many others, e.g. trauma, intercurrent infections, personal tragedies, etc. This non-specificity does not receive attention. In order to confirm the role of cannabis as a specific precipitant it is necessary to exclude all other probable factors which may lead to a similar reaction, specially the role of other intoxicants, e.g. use of amphetamines, Lysergic acid diethylamide and other hallucinogens. It is at times surprising to find that history of such use is not given sufficient weight, (e.g. 3, 30). The above observations only go to show the dangers of accepting the immediate and the obvious - as the cause. In the words of Sir Aubrey Lewis (34): 'Interpretations which explain the phenomena too readily lead to a feeling of certainty which closes the mind to further enquiry and makes the believer hostile to any breath of sceptisism.' Similarly in the case of amotivational syndrome varieties of non-drug factors (e.g. Social alienation, Chronic privation and lack of opportunities, etc.) may play a crucial role. (e.g. 43, 50). DOSE AND DURATION Another difficulty arises from insufficient attention paid to dose and duration of use. Eastern authors (such as Chopras and Benabud) assert that moderate use (e.g. UP to 1.3 g per day in

India) is harmless. Observations of Tart (45) on chronic users in the West seem to support such contention. This may imply that in terms of harmful effects what matters is the amount used over a period of time, and not just the use. Relationship between dose and effect has been repeatedly confirmed in controlled studies. (e.g. 20, 25, 31). The role of a dose-frequency and duration may therefore be very significant because of the possibility of cumulative effect (40). Data on these variables in clinical reports is very meagre. Bernhardson and Gunne (6) present data on duration and dosage and find that duration is of little consequence to effects, but they do not discuss the relation between dose and duration combined on one hand and effects on the other. The hypotheses that cannabis causes psychosis will be much strengthened if it could be shown that: a. the frequency of psychoses amongst users is greater than observed in comparable group of non-users; and that this difference cannot be attributed to chance. b. administration of cannabis under controlled conditions will result in predictable recurrence of psychosis amongst those volunteers who have recovered from cannabis psychoses. Such an experiment using double blind technique is permissible as the risk of fatality and sequelae is negligible. Relapsing patients should provide suitable subjects and may not be difficult to find. (3, 6, 13). The procedure will have the benefit of the patient being his own control, (36) and of monitoring dose and the setting if not the mental set. In summary. Evidence in support of causal connections between cannabis use and psychosis is defective on the basis of: 1. Non-critical acceptance of evidence. 2. Short-comings of diagnostic procedures. 3. Scant attention to other probable causes. 4. Concept of single aetiology. 5. Lack of statistical support. 6. Absence of controlled studies. ROLE OF EPIDEMIOLOGY Epidemiological aspects of cannabis psychoses have attracted insufficient attention - a point made by Le Dain Commission in Canada (9). Statistical data based on mental hospital admissions or other special groups can never mirror the actual occurrence or absence of psychoses amongst the users. If anything they must underestimate the phenomena as many who are affected may never come to the notice of the medical profession. In the absence of field studies, incidence of psychoses amongst users is impossible to assess. Chopra (11) estimate of 0.5 to 2 per cent for Indian users, and Benbud's (5) of 0.5 per cent for the Moroccan Kif smokers are difficult to confirm or challenge because of methodological problems. Recent attempts to document the frequency of, adverse effects and their recurrence, in the absence of drug use ('Flashbacks') may be of help. (1, 15, 19, 29, 32). It is, however, not clear from these accounts if the adverse reactions were of such a nature or degree that the users required professional or other assistance or deterred the subjects from further use at least for a period. CONTRIBUTIONS OF RECENT RESEARCH

In the last 6 or 7 years we have seen a resurgence of experimental research into the effects of cannabis. Even though this renewed research has, as Hollister (23) puts it, 'added little not previously known about the clinical syndromes produced by the drug'; it should provide a healthy stimulus to further clinical research. It has (a) made it possible to assess the amount of active ingredient in cannabis; (50) (b) established the relationship between amount of active ingredients and effects; (c) established the basic characteristics of the effects of the drug; thus confirming the claims of users, of folklore and of past observers; (e.g. 21, 25, 51, 31, 32, 41, 47, 52, 53). (d) reminded us of the role of important non-drug factors, such as mental set, the setting (22, 27, 47), and placebo effect (35) in reactions to the use of cannabis. In other words, laboratory studies have provided the field workers, e.g. clinicians and epidemiologists with the basic pre-requisites for field studies. For as Hollister (23) says, 'the clinical effects in regard to social questions about marihuana such as possible deleterious effects from chronic use cannot be answered by laboratory experiments. They must be settled by close observations made on those who experiment on themselves.' Thus he has put the responsibility where it belongs - on to the shoulders of field workers. Some ideas for future research. I would suggest that: 1. We re-examine the usefulness of the term psychosis; 2. We examine the possibility of establishing facilities to study the clinical effects of cannabis. When the term cannabis psychosis is beset with so many difficulties, e.g. definitional one, questions whether it exists, if it exists what is its nature and whether it is a specific or nonspecific entity and so on and so forth one wonders why we retain it at all - why does one have to fit in all adverse reactions into a pre-conceived frame work? I personally cannot see the advantage except that of tradition and the ease with which we can pigeon-hole those affected if these can be called advantages. I would like to suggest that we freeze the term psychosis for the time being and start at the beginning by just documenting the effects (adverse or otherwise) according to their form and degree, and functional consequences on the 'life' of the users and others, in an agreed and uniform manner. It would certainly make communication and understanding between workers much easier, and may help confirm or reject the possibility or a syndrome or syndromes provided of course that there is prior agreement, (a) about the operational definition of adverse effects, and (b) about the criteria to be used. We should be able to establish facilities on national or regional levels to study the adverse effects of cannabis under controlled conditions on volunteers who have suffered from these and have come to the notice of helping professions. I have already referred to this idea. One would expect this facility to carry out extensive clinical, biochemical and psychosocial investigations, and long term follow-up studies on the subjects studied. Unless such or other similar steps are taken sterile arguments as to whether or not cannabis does cause psychosis will continue. REFERENCES 1. Annis, H.M. and Smart, R.G. (1973) Adverse reactions and recurrences from marijuana use, Brit. J. Addict. 68:4, 315-321.

2, Allentuk, S., Bowman, K.M. (1942) The psychiatric aspects of marihuana intoxication, Am. J. Psychiat. 99/21, 3, Asuni, T. (1964) Socio-economic problems of cannabis in Nigeria, Bu21. Narc, 16:2, 1728. 4. Baker, A.A., Lucas, E.G. (1969) Some hospital admissions associated with cannabis. Lancet 1, 148. 5. Benabud, A. (1957) Psychopathological aspects of cannabis situation in Morocco. Statistical data for 1956. Bull. Narc. 9:4, 1-16. 6. Bernhardson, G., Lan-M Gunne (1972) Forty-six cases of psychosis in cannabis users. Int. J. Addict. 7:1, 9-16. 7. Bromberg, W. (1934) Marihuana intoxication, Am. J. Psychiat. Sept. 303-330. 8. Campbell, A.M.G., Evans, M., Thompson, J.L.G. et al. (1971) Cerebral atrophy in young cannabis smokers, Lancet 2: 1219-1224. 9. Interim Report of the Commission of Inquiry into the Non-Medical Use of Drugs, (Ottawa). April 6, 1970. 10. Cassarett, Baselt, R. (1971) A toxicologic view of marihuana. Hawai Med. J. 30:4, 262265. 11. Chopra, R.N., Chopra, G.S., The present position of hemp-drug addiction in India, (1939) Ind. Med. Res. Memoirs, 31: 12. Chopra, R.N., Chopra, G.S., Chopra, I.C. (1942) Cannabis sativa in relation to mental disease and crime in India, Ind. J. Med. Res., 30:1, 155-71. 13. Chopra, I.C., Chopra, R.N. (1957) The use of cannabis drugs in India. Bull. Narc. Jan.March 1-29. 14. per Dalen (1969) Causal explanations in psychiatry - a critique of some current concepts. B.J. Psychiatry, 115/519, 129-139. 15. Fisher, G. (1974) Harmful effects of marihuana use: Experiences and opinions of current and past marihuana users, Brit. J. Addict. 69:1:75-84. 16. Freedman, H.L., Rockmore, M.J., (1946) Marihuana: A factor in personality evaluation and any maladjustment. J. CZin. Exp. Psychopath. 7,765-782; 8,221-236. 17. Fraser, J.D. ( 1949) Withdrawal symptoms in cannabis indica addicts, Lancet 2, 747-748. 18. Grinspoon, L. (1969) Marihuana, Scientific American, 221:6, 17-25. 19. Halikas, J.A., Goodwin, D., Guze, S.B. (1971) Marihuana effects - A survey of regular users, J.A.M.A. 211 715, 692. 20. Hollingshead, A.B., Redlich, F.C. (1958) Social Class and Mental Illness, N.Y. John Wiley and Sons Inc., p.236. 21. Hollister, L.E., Richards, K., Gillespie, H.K. (1968) Comparison of tetrahydrocannabinol and synhexyl in man, Clinic Pharmac. Ther. 9,783-91. 22. Hollister, L.E., Gillespie, H.K. (1970) Marihuana, ethanol and dextroamphetamine. Arch. Gen. Psychiat. 23,199 -203. 23. Hollister, L.E. (1971) Marihuana in man: Three years later. Science, 172: 21-28. 24. Indian Hemp Drugs Commission (1894) Government Central Printing Office, Simla, India. 25. Isbell, H., Gorodetsky, G.W., Jasinski, D., Claussen, U., Spulak, F., Korte, F. (1967) Effects of (-) 2Trans-tetrahydrocannabinol in man, Pharmacologia (Berl.) 11, 184-88, 26, Isbell, H., Jasinski, 0,R. (1969) A comparison of L.S.D.-25 with (-)- 2-transTetrahydrocannabino1 (THC) and attempted cross tolerance between L.S.D. and TCH. Pharmacologica (Berl.) 14/115-123. 27, Jones, R.T. (1971) Marihuana induced high-expectation, setting and previous drug experiences, Pharmac. Rev. 23/1359, 28. Kalant, Oriana J. (1971) Moreau-Hashish and hallucinogens, Int. J. Addict. 6(3) 553-560.

29. Keeler, M., Reifler, C.B., Liptzen, M.B. (1968) Spontaneous recurrences of marihuana effect. Am. J. Psychiat. 125: 384-386. 30. Keup , W. (1970) Psychotic symptoms due to cannabis use (A survey of newly admitted mental patients) Dis. of the Nerv. Syst. 31, 119-126. 31, Klonoff, H. (19 73) Strategy and tactics of marihuana research, Canad. Med. Assn. J. 108. 145-150. 32. Kolansky, H., Moore, W.T. (1972) Toxic effects of chronic marihuana use, J.A.M.A. 222, 35-41. 33, Kiplinger, GoF., Manno, J.E., Rodda, B.E. and Forney, P.B. (1971) Dose response analysis of effects of tetrahydrocannabinol in man. Clinc. Pharmac. and Therapeutic 12/4, 650-657. 34, Lewis, A. (1967) The State of Psychiatry - Essays and Addresses, published Routledge and Kegan Paul, Chapter 13, pages 195-213. 35, Manno, J.E., Kiplinger, G.F., Haine, S.E., Bennett, I.F., Forney, R.B . (1970) Comparative effects of smoking marihuana and placebo on human motor and mental performance, Clinc. Pharma. Therap. 111, 808815, 36, Mayor's Committee on Marihuana (1944) The Marihuana Problem in the City of New York. 370 Murphy, H.B.M. (1963) The cannabis habit. A review of recent psychiatric literature. Bull. Narc. 15:1, 15-23. 38. O'Shaughnessy, W.B. (1843) On the preparations of the Indian Hemp or Ganja, PY,ov. Med. J. Retrospect. Med. Sc. 343-347. 39. Patterson, R.C. (1971) Marihuana and Health - A Report to the Congress for the Secretary of United States, Dept. of Health, Education and Welfare, Washington, D.C. U.S. Govt. Printing Office. 40. Paton, W.D.M. (1973) Cannabis and its problems. Proceedings Royal Soc. Med. 66:718722. 41, Rafaelson, C.J. et al. (1973) Cannabis and alcohol: Effects on simulated car driving. Science 179: 920-923. 42. Spencer, D.J. (1971) Cannabis induced psychoses, Int. J. Addict. 6:2 323-326. 43. Soueif, M.I. (1967) Hashish consumption in Egypt with special reference to psychological aspects. Bull. Narc. 1-:1, 1-12. 44, Talbot, J.A., Teague, J.W. (1969) Marihuana psychosis. J.A.M.A. 210:299-302. 45. Tart, C.T. (1970) Marihuana intoxication. Common experiences. Nature. 226: 701-704. 46. Watt, J.M. (1961) Dagga in South Africa. Bull. Narc. July! Sept 47. Weil, A.T., Zinberg, N.E., Nelson, J.M. (1968) Clinical and psychological effects of marihuana in man. 48. Weil, A.T. (1970) Adverse reaction to marihuana classification and suggested treatment. New azg. Med. J. 282: 997-1000. 49. Williams, E.G., Himmelbach, C.K., Wikler, A., Ruble, D.C. (1946) Studies on Marihuana and Parahexyl Compound. Publ. lath. Report 61. 1059-1083. 50, W.H.O. Techn. Report Series 478 - The Use of Cannabis. Geneva, 1971. 51, Zinberg, N.E., Weil, A.T. (1970) Comparison of marihuana users and non-users. Nature. 226/119-123. 52,, Waskow, I.E., Olsson, J.E., Salzman, C., Katz, M.M. (1970) Psychological effects of tetrahydrocannabinol. Arch. Gen, Psychiat. 22:97-107. DISCUSSION

Discussion on the topic of Cannabis Psychosis ranged widely and can be conveniently brought together under the following headings. THE CONCEPT OF CANNABIS PSYCHOSIS As well as Dr Rathod, Drs Smart, Cameron and Miles, and Mr Hastleton were in favour of dropping the term 'cannabis psychosis' in the present state of knowledge and Professor Soueif and Dr Cameron stressed the need to look at symptoms and signs and define these rather than using syndromes which are not adequately delineated. In general it might be better to talk about adverse or unpleasant reactions. Dr Smart reported that half the high school students who reported adverse reactions or recurrences refused to characterise them as unpleasant and thought that the term psychosis should be restricted to a toxic psychosis on the carb on tetrachloride model. Dr Tinklenberg reported one subject's description of 'cosmic vomiting' and some participants were in favour of talking about 'unusual reactions' or 'atypical reactions'. Dr Edwards drew attention to the possibility that other drugs were also being used, the influence of age and personality and the need to establish the precise pathological mechanism. He highlighted the particular problem of explaining how a psychosis could continue for a long time after withdrawal from a drug. CELL PATHOLOGY Professor Paton, when asked to comment on the question of establishing a pathology, thought that it was too early to relate concepts of cell pathology to clinical findings and that if such a pathology existed it was likely to be found in the field of neuropathology and to affect the finest structures such as dendrites. DEFINITION OF SPECIFIC SIGNS AND SYMPTOMS Dr Rubin noted that 'all research is fraught with difficulties. I think Dr Rathod has made an important contribution in pointing out the kind of problems that seem to plague cannabis research particularly. It seems at times, to me, that cannabis is the current sin to which every thing is being attributed. I think the important contribution is in trying to establish the research methodology that will be able to distinguish between correlation and cause. The Royal Indian Hemp Commission still remains, to my mind, a model of scientific research. The time has come in cannabis research when we need some international agreements on parameters, so that the data can be comparable. Otherwise we are dealing with too many unknowns.' Dr Cameron was concerned about the lack of tools for research, and urged further attention to this aspect. Professor Soueif reported that in his researches, of 120 items collected from classical symptomatology and including behavioural changes, inter-observer reliability had been established in 70 items at a correlation of 0.7 or better. Dr Rathod stressed the need for observers to agree on an operational definition of signs and that a distinction should be made, if possible between direct drug effects and cultural and social effects of expectation and motivation. POPULATIONS TO BE STUDIED

Dr Rubin observing that the fact of illegality influences signs and symptoms stressed the importance of looking at countries where cannabis use is traditional. Dr Cahal stressed 'the need to go to countries where only cannabis is used is becoming a matter of some urgency because more and more countries, as they develop and as communications improve, are beginning to use drugs other than cannabis.' This point was generally agreed. Professor Paton pointed out that since population in Jamaica, India, and South Africa were not characteristic of industrial users, there was a danger of generalisation from researches carried out in them. STRATEGY OF RESEARCH - THE CONTRIBUTION OF EPIDEMIOLOGY Dr Edwards said that he felt that the sort of questions to which the epidemiologist is asked to address himself are often increasingly inappropriate. 'The epidemiologist can only really apply his art when he has some likelihood of describing the distribution through time or through populations of reliably describable entities, syndrome or behaviours. The difficulties of measurement and analyses are such that we should be very wary indeed of mounting new epidemiological studies before we are really certain that we can accurately measure what we want to measure. Research will go forward in this area most effectively by the close definition of hypotheses and close testing of these hypotheses by the most appropriate and economic scientific means available. The most economic means will relatively seldom be the large sample epidemiological enquiry.' PRIORITIES FOR RESEARCH In general, participants agreed with Dr Rathod's suggestion that the priority, at the moment is to develop a reliable check-list of signs to discover which signs, both usual and unusual, occurred across cultures, and which signs were culture-specific. Dr Cameron supported this view saying: 'I don't feel constrained to attribute a diagnosis nor to attribute a cause. I would just like to know if we can see any major differences between groups of heavy users and non-users. I don't get particularly interested in a phenomenon until the phenomenon has been demonstrated. And that is where I think we are, looking at signs and symptoms. It seems particularly difficult to get a population frame from which to draw these two user/non-user groups. We may have to look at more than one population frame in order to get a cross-section. Some population groups have been examined with comparison groups from the same population, but those populations are sometimes very circumscribed and we can't generalise from those.'

6. The Uses and Implications of the Lognormal Distribution of Drug Use

W. D. M. Paton, Department of Pharmacology, University of Oxford. A knowledge of the actual rates of drug use is essential for several purposes; for assessing medical or social risk; for considering the reciprocal problems of the effect of social attitudes on use, and of drug use on social attitudes; and for quantitative studies on supply and demand. The information available comes generally from two sources only; seizures, and surveys by questionnaire. For alcohol, purchases and measurements of alcohol blood levels are also sometimes available. For cannabis, there is a particular need to resolve one way or another the

apparent conflict between the normal 'stereotype' of use - around 5 mg THC absorbed from an experimental reefer (i.e. c. 0.1 mg/kg a few times), against rates of use up to 100 times this, i.e. 10 mg/kg daily or higher, recorded in the WHO Report (No. 478: 1971). The medical and social picture would, of course, be quite different as between a uniform rate of use through a population, and a generally abstinent population with a small group of heavy users. Ledermann (1958) introduced into alcohol studies the log-normal distribution; Lindt and Schmidt (1968) followed this up for alcohol, and Smart and his colleagues (1970) for general use. If one collates the available data from these and other sources, there are signs that a coherent picture of some heuristic use emerges. Since Ledermann's treatment is somewhat cryptic, and includes the assumption of a correlation between the mean and the standard deviation of the distribution, I have used a simpler and more empirical approach, the use of logarithmic probability paper - a procedure very familiar to pharmacologists since the introduction of this mode of analysis by Gaddum in 1933. It is required only that the data are such that rates of use can be assigned to the whole population, without deficit or overlap. The procedure is applied to cumulated data rather than to frequency distribution. When plotted, as % using a drug at a given rate or higher, the probability ordinate scales percentages in such a way that if the distribution is normal, a straight line is produced, whose intercept at 50 per cent gives the median and whose slope gives the standard deviation. Thus, taking Binnie's data on cannabis use (1969), recalculating to include non-user students, we have:

Cumulation here is 'from the right', since for the moment it is the proportion taking a given dose or more that is important. Figures 6.1 and 6.2 show these data plotted. They also include data from Smart and his colleagues (Canadian schoolchildren), from Ledermann (French adults), and from Fisher's study (1972) in the USA Armed Services. First, all the data fit a log-normal distribution well. Second, there appear to be two patterns: one for alcohol, with a lower standard deviation (of the order of 2-5 x in geometrical terms); the other for cannabis, and in fact it seems for most other drug use reported, including other illicit drugs, tranquillizers, and smoking, with a standard deviation of the order of 50 x or higher. The relative constancy of the standard deviations is remarkable, considering the wide range of populations involved, and of ages, rates of use, and drugs.

Graphs of the percentage of various populations using certain drugs at a given rate or higher, plotted with probability scale for percentage ordinate, and logarithm of rate of use for abscissa. Data from Ledermann (1956), Binnie (1969), Smart et al. (1970) and Fisher (1972).

IMPLICATIONS 1. It appears to be worth while to analyse the data in this way, and to make sure that data are obtained in suitable form - some data are not so obtained. At the least, one is then able to express the results of a survey conveniently by means of two quantities, the median rate of use or the mean rate of use, and the standard deviation. Consistency or inconsistency of behaviour, not otherwise easily apparent, can then be recognized. 2. It has been suggested that the fact of a lognormal fit throws light on causes. In general, I think this is not the case. Any variable phenomenon which is the outcome of a large number of small variables tends rapidly to normality, as was shown by Gauss, Laplace and Bessel; and the logarithmic transformation simply implies that it is proportionate rather than absolute variations that are relevant, as is normally the case in drug action. An extra sherry a day to someone only taking a glass a day is a substantial increase in alcohol intake; to a bottle-a-day man, it is trivial, while another bottle a day would be significant. The log-normal fit could mean only that we are dealing with a multifactorial aspect of drug action. It does imply, probably, that there are not two contrasted sub-populations; it does not imply, however, that there may not be as few as 4 or 5 sub-populations, since the sum of several normal curves is liable to be very hard to distinguish from a single one. 3. The most important aspect seems to me that of the magnitude of the standard deviation i.e. of the steepness or flatness of the lines. To take an example, suppose we have a drug which 10 per cent of the population take once a month or more times a month, and a minute fraction 100 times or more a month. But for a standard deviation of 50 x (like tobacco, cannabis and others), 3 per cent will take it 10 or more times a month and only a little under 1 per cent will take it 100 times or more a month. With high standard deviations, the proportionate incidence of high rates of use increases greatly. This seems to me a very important fact about drug use; and, for cannabis, it implies that the WHO collation is, in fact, completely correct. For such a drug, in addition, pharmacological exploration of the effect of high doses becomes essential. 4. As a final aspect, which can not be developed, one may note that one can begin to approach a calculation of total amounts consumed, to construct Lorenz-type curves, to relate patterns of use with estimates of supply, and to make some estimate, given an estimated mean rate of consumption, of the numbers of people using drug at a given rate or higher. In short, there are reasons for hoping that it is possible to do for other drugs what Ledermann did for alcohol. REFERENCES Binnie, H.L. (1969). The attitude to drugs and drug-takers of students at the University and Colleges of Higher Education in an English Midland city. Vaughan Paper No. 4, Dept. of Adult Education, The University, Leicester. Fisher, A.H. (1972). Preliminary findings from the 1971 Department of Defence survey of drug use. Human Resources Research Organization Technical Report 72-8. Gaddum, J.H. (1933). MRC Reports on Biological Standards III. Methods of Biological Assay Depending on a Quantal Response. Special Report Series No. 183. H.M.S.O. Ledermann, S. (1956). Alcool, Alcoolisme, Alcoolisation; Donnes Scientifiques de Caractre Physiologique, Economique et Social. Institut National d'Etudes Dmographique, Travaux et Documents, Cahier No. 29, Presses Universitaires de France. de Lindt, J. and Schmidt, W. (1968). The distribution of alcohol consumption in Ontario. Q.J.

Stud. Alc. 29, 968-973. Smart, R.G., Fejer, D. and Alexander, E. (1970). Drug use among high school students and their parents in Lincoln and Welland Counties. Addiction Research Foundation, Toronto, Canada. WHO Technical Report Series No. 478 (1971). The Use of Cannabis. World Health Organization, Geneva. DISCUSSION DEFINITIONS OF 'HEAVY USE' Referring to Professor Paton's offer of a partial definition of heavy use as 'that point when accumulation in the body occurs', Dr Cameron pointed out that many heavily-used drugs, such as cocaine, accumulated very little, yet were often used frequently. Professor Paton stressed that accumulation was only one aspect of heavy use, and that pharmacologists could not predict or define the meaning of 'heavy use' apart from accumulation, without considering somatic effects. Dr Edwards pointed out that although the alcohol intake per year of two populations might be similar, different populations may often have different distributions of use through time. One can drink a little regularly, or a lot, more seldom. 'This is of more than theoretical importance: it may very much influence the social impact of the drug, and may quite require different social policies'. IS THE GRAPH AFFECTED BY PROPERTIES OF THE DRUG? Professor Paton suggested that there is a consistency of distribution of frequency of use 'which I think is telling us something not about drugs, but about patterns of human behaviour of this kind'. Mr Hasleton asked if there was an implication that one could validly extrapolate the curve to describe patterns of very low and very high use, and Professor Paton said that he would be reluctant to extrapolate without empirical basis. Ledermann had made the assumption that one can't take more than a certain amount of any drug, therefore there has to be a right-hand limit to the curve. Once you say that you are putting a constraint on the rest of the curve. Mr Hasleton and Dr Edwards asked whether the properties of the drug influenced the distribution of frequency of use. Dr Edwards suggested that the shape of the curve was partially determined by the population, partially by the culture, its economies and mores, and partly by the drug. 'Supposing we have drug X, which is non-dependence producing, and that we know its distribution of frequency of use curve. Let us now introduce into drug X a dependence producing potential. Will this change in the characteristic of the drug in no way alter the shape of the curve?' Professor Paton suggested that one could work towards an answer by obtaining distribution curves for commonly used, freely available more 'neutral' drugs, such as nasal drops, aspirin etc. IMPLICATIONS FOR MINIMISATION OF HEAVIER USE Dr Miles congratulated Professor Paton on his paper, and added, 'I've been advocating for a long time micro-economy studies , where the price of various drugs is varied for a captive

audiance. The obvious ones to vary, of course, are alcohol and marijuana. It would be very interesting to see, when people are in a productive situation, when they have access to drugs that they have to pay for, how you could change the curves presented by Professor Paton'. Professor Paton replied that: 'If you wish to diminish the amount of heavier use, there are two ways to do it. You can either reduce the total consumption down or you can try to find a way to move only the heavier users.' Dr Cameron agreed saying that one should stress that there were two possible ways to reduce heavier use. 'First, you can change the mean, or the mode: that is to change the pattern of drinking of the whole population, to shift the whole curve downward. Or you may have another programme which is aimed at changing the standard deviation, aimed primarily at the heavier user and not at the whole population. It needs to be stressed that one need not approach one or the other programme exclusively.' Professor Paton pointed out that it might not be necessary to talk of heavy users as a separate class. A system might be evolved which applies increasing pressure as use goes up. Dr Smart said that there were few well studied situations where drug use was going down 'so one really can't be sure whether it's possible to change the standard deviation or mean or both.'

7. The Psychology of CannabisTechniques for Investigating the Frequency and Patterns of Use of Cannabis in Groups of Drug Users
Ian Hindmarch, Department of Psychology, Leeds University. This paper reports two studies : both investigations of the frequency and patterns of cannabis use in drug using populations. However, inasmuch as the research area might be identical the techniques used to collect information are quite distinct and separate. The first study illustrates the use of the self administered questionnaire to collect frequency and patterns of use statistics and, together with information from personal interviews, shows how a consideration of attitudinal variables provides a paradigm for drug using behaviour within cognitive consistency theory. The second study is an essay in participant observation : frequency of use data was again collected together with complete individual histories of illicit drug use. Before the two research investigations are discussed, a basic consideration of the two methodologies is made, viz, the self administered questionnaire and participant observation. 1. THE SELF ADMINISTERED QUESTIONNAIRE The majority of research investigations conducted in the United Kingdom regarding the patterns and frequency of use of cannabis by 'normal' populations have utilised the selfadministered questionnaire (see Kosviner (1974) in this volume). The self-administered questionnaire has severe limitations in that the behaviour it purports to measure (in this instance patterns of cannabis use) can only be probabilistically inferred from data collected.

The analysis and mathematical treatment of data collected from questionnaire surveys is often done with great finesse (Haselton (1974) in this volume) and usually with a high degree of statistical validity, but the computer testing for statistical significance is not able to discriminate whether the data is valid or not. The behaviour associated with illicit drug use is a complex phenomenon and the accurate reporting of the various aspects of this complicated pattern of behavioural activity will be dependent upon the appropriate response category being contained within the questionnaire. The subjects report of his drug using behaviour is thus limited by the range and suitability of the items contained in the questionnaire. In short, a response box on a questionnaire does not constitute a behaviour analogous to drug use unless the scope of the questionnaire is adequate and the subject is able to make reliable, accurate and valid representation of his behaviour.

Figure 7.1 illustrates how self administered questionnaires relate 'actual' to 'inferred' drug using behaviour and indicates the probabalistic nature of the relationship. However, the efficiency of data collection, with the self-administered questionnaire, in terms of its low 'cost' per unit information make it ideal for administration to large groups. 'Cost' is not simply an economic measure but embodies a consideration of the time taken and effort required to survey a suitable number of individuals from populations where the transient and ever changing nature of their behaviour necessitates immediate information being collected. Whitehead and Smart (1971) and Somekh (1974) (this volume) have shown student respondents in large scale self-administered questionnaire studies to be internally consistent so adding validity to results derived from such surveys. It is suggested by Somekh (loc. cit.) that a more satisfactory approach to the study of drug using behaviour would be to link data from questionnaires with information from other sources such as individual user interviews or 'prevalence' estimates from doctors and clinics. 2. PARTICIPANT OBSERVATION The distinguishing feature of this particular mode for the investigation of the patterns of cannabis use is that a social interaction takes place between interviewer (observer) and client (subject). The validity of the procedure rests on the truism that it is easier to train a single observer to record certain criteria behaviour in a reliable manner than to train a large number

of naive subjects to make consistent responses to a self administered questionnaire. In the illicit drug using situation the participant observer observes directly the behaviour about him and participates to the extent that he has a durable social relationship with the members of the group about him Vindmarch (1972(a)). Direct observation is the oldest method available to the behavioural sciences and is naturally subjective and qualitative but providing some objective rating scale is used it need not be inaccurate since we assume intraobserver reliability to be high.

Figure 7.2 illustrates the relationship between actual and inferred behaviour - using techniques of participant observation - to be, at least, correlative and certainly meaningful. One of the main reasons for using an observer to collect basic data is that such a procedure reduces the constraints that arise when respondents attempt to rate their behaviour on a predetermined set of scales, as in a self administered questionnaire. The unique relationship established between observer and subjects also enables the collection of data not normally available to the researcher. Samples of cannabis and data relating to the weighings of cannabis cited in the second study were only possible because of the intimate and mutual relationship of trust established between the group and the observer. FREQUENCY AND PATTERNS OF CANNABIS USE : ADDITUDINAL CONSIDERATIONS This study reports data collected in 1973 (Einstein, Hughes and Hindmarch (1974) and compares attitudinal profiles collected in 1973 (Hindmarch, Hughes and Einstein (1974)) with profiles collected from a similar population of undergraduates in 1969 (Hindmarch (1970) (1972( b)). A comparison of the 'frequency of cannabis use' statistics collected in 1973 and 1970 is also made. The raison d'etre of the 1973 study was a test of the allegations from a number of reports purporting a link between the consumption of alcohol, smoking of tobacco and the illicit use

of cannabis. The present author (Hindmarch (1970)) - as a result of a participant observation survey of 153 drug users - noted that cannabis users were heavy cigarette smokers. Weitman et al (1972) showed that alcohol and tobacco use correlates with the use of other illicit drugs especially cannabis; and more recently, McKay et al. (1973) concluded that 'drug users' are males who smoke and drink more frequently than the norm. One thousand names were selected at random from the lists of enrolled students and a questionnaire together with a guarantee of anonymity was mailed to each individual. Personal details (age, sex etc) were collected as well as histories and patterns of cannabis, alcohol and tobacco use. The questionnaire also contained full semantic differential ratings for each drug under consideration and a bank of attitudinal statements derived from previous work ( Hindmarch (1972(c)) with undergraduates. To validate the questionnaire and to obtain some idea of the reliability of the information offered, respondents were invited to identify their completed questionnaire with a number/letter combination or nonsense word known only to themselves and then attend for personal interview. Some 10 per cent of the respondents identified themselves in this way, the content of their individual interviews confirmed, that for the most part, the questionnaire posed appropriate questions and that subjects were internally consistent. A full analysis of the patterns of use of alcohol, cannabis and tobacco by the 300 respondents to the questionnaire is to be found in Einstein, Hughes and Hindmarch (1974), but certain of the findings are appropriate to this present symposium. There was no significant effect on alcohol consumption due to the use of cannabis, indeed there was a tendency to refrain from the use of alcohol when using cannabis. However, a high consumption of alcohol was associated with a tendency to try, i.e. experimental use of cannabis . There was also a positive correlation between a heavy use of tobacco and the tendency to 'try' cannabis. Table 7.1 presents the summary data for the overall frequency of use of cannabis, alcohol and tobacco. Table 7.2 is a comparison, using the same criteria of frequency of cannabis use, of data from the 1969 and present studies. It must be emphasised that the present data was obtained via a self administered questionnaire while the 1969 data is from a participant observation study. However, only information from the group of cannabis users is utilised. Essentially of interest is the change in frequency of use of cannabis within a group of cannabis users. It is admitted that a direct comparison of the two groups cannot be made since they are different samples drawn from different populations by different techniques. However, as we will see, evidence from the measured attitudinal profiles of the two samples is consistent with the change in patterns of frequency of use and so a speculative comparison of the two groups is certainly not inappropriate. Table 7.2 suggests that the overall frequency of use of cannabis by cannabis users has declined between 1969 and 1973. In 1969 the greater proportion (72 per centj of cannabis users were using the drug more frequently than twice a week, while of the 1973 sample of users only 16 per cent had that particular frequency of use. On the other hand, 72 per cent of the 1973 sample used the drug less than once a month, and of these 52 per cent had but an experimental exposure to it; while, only 2 per cent of the 1969 sample had such a low frequency of use.

We have already indicated the reservations to be made in comparing these two groups, but it must also be remembered that we are looking at changes in frequency of use of cannabis in groups of drug users and are in no way establishing prevalence or incidence statistics. The comparison of the two groups also appears more reasonable when we link the frequency of use data with information from attitudinal dimensions collected at the same time from both the cannabis users and from a population of non-drug users of similar age. Table 7.3 compares the 1969 and 1973 groups on the scores produced by the evaluative dimension of Osgood's semantic differential. For a further discussion of this technique see Osgood et al. (1957), Hindmarch (1970)(1972(b)) and Hughes, Einstein and Hindmarch (1974).

The three most salient features of Table 7.3 are a. the change in evaluation of cannabis by the non-user group from a positive position (66.7) in 1969 to a negative (36.8) position in 1973.

b. the significant attitudinal separation of user and non-user groups in the 1973 sample and c. in both 1969 and 1973 samples the user groups hold the most positive attitudes. The 10 cm analogue scale represents exceptionally negative attitudes by a score of zero, exceptionally positive attitudes by a score of 190 and neutrality of attitude by scores in the range 50-7. In 1969 both drug users and non-users had a positive evaluation of cannabis, while in 1973 the non-user group were significantly negative in their evaluation of the drug. In discussing these findings and defending this particular technique for measuring attitudes a method of cognitive analysis will be used. Festinger (1957) introduced 'cognition' as 'the knowledge, opinion or belief about the environment about oneself or one's behaviour'; and cognitive organisation thus becomes the individuals internal ordering and processing of his behaviour and a major, if not the sole, predictor of future behavioural and motivational action. The contemporary view of cognition is of a dynamic system which interacts with the environment to modify and control behaviour; and central to the function of the system in making decisions are the processes of commitment, choice and justification.

Moreover, there is a basic tendency in human organisms to maintain a consistent view of themselves and their environment. However, in certain situations, the internal (cognitive ) appraisal of self and an external (behavioural) state may be disparate and conflicting: under these conditions cognitive dissonance is generated (Festinger (1957, 1964); Brehm and Cohen (1962)). Dissonance is the result of two or more inconsistent beliefs being held within the same cognitive framework and is defined 'as a general tension state which motivates behaviour, the terminal response of which results in a reduction in the level of tension' (Zimbardo (1969)). We stated earlier that dissonance was produced within a cognitive system, yet we spoke of the conflict between behavioural and cognitive systems i.e. between external and internal states. However, Doob (1947) argued that an attitude is an internal (implicit) correlate of an objective (overt) behaviour system and Campbell (1963) has also shown attitudes to be internal manifestations of behavioural dispositions. Thus, because overt behaviour finds representation in the implicit cognitive state, cognitive dissonance can occur when external behaviour is at odds and incompatible with internal belief systems. Behaviour is modified by experience and, therefore, in order to maintain consistency the correlated attitude must change appropriately. The intimate relationship between attitudinal and behavioural systems is well illustrated by Collins (1968), Steininger (1973) and Wilson et al. (1973). The scope of the present paper curtails any further elaboration on the notional theories of cognitive consistency, but the generality of such an approach in psychology becomes apparent from the scope of Abelson et al's (1968) discursive review of the topic. The measurement of attitudinal variables is also defended because the author views illicit drug use as a product of drug/society/individual interaction - and the representation of this complex interrelation of variables can be found only in an individual belief system which takes account of the role of environmental factors as well as intrinsic variables of personality, motivation and perception. The appropriateness of attitudinal measures assumes greater importance when one considers the relative failure of 'prone personality' theories (Hill et al. (1960), Gilbert and Lombardi (1967); Halstead and Neal (1968) and Rosenberg (1969) to give consistent views as to the factors which delineate that particular personality structure which is 'at risk'. The lack of evidence attributing the development or escalation of illicit drug use to certain individuals with a 'prone personality' is well illustrated in Mott's (1972) review : she concludes that, individual personality characteristics are 'irrelevant to the study of developing drug misuse because environmental factors are the major determinants of such behaviour'. Furthermore, there is still little agreement between researchers as to which of the many socio-cultural, socio-economic variables are important factors in distinguishing an illicit drug user from his peers (see Kosviner's comparison of U.K. research in this present volume). The cognitive balance model would predict that as an individual became more behaviourally involved with cannabis his attitudinal system would become more positive in its evaluation of the drug - to maintain congruity between internal and external systems. Table 7.3 did show that cannabis users in both 1969 and 1973 samples held more positive attitudes than nonusers. Table 7.3 also shows the tendency for non-user groups to have become more negative in their evaluation of cannabis i.e. the norms of the normal population have 'hardened' and become negative. The causes and origins of such change must needs be speculative but 1969 saw the peak of social activity connected with 'flower power', 'psychedelic music' and 'hippies' and between 1969 and 1973 there has been a great deal of information often conflicting about the botany, chemistry, pharmacology and psychology of cannabis. This information, from a wide variety of scientific and non-scientific sources was disseminated via underground and conventional newspapers and presses. Teratogenic, psychotomimetic, carcinogenic and hallucinogenic effects were attributed to use

of the substance cannabis. Contemporaneous with reports of these negative effects much was written about the relaxing, positive, creative and enhancing effects of the drug. Thus before an individual could take cannabis he would have to resolve the dissonance between the various opinions presented to him, i.e. is cannabis harmful or not? If he decided cannabis was harmful he might not try cannabis, but if he did try it he would certainly not escalate to an intense chronic level of drug use. The behavioural concomitants of these cognitive decisions are to be found in Table 7.2 - i.e. the less frequent (chronic) use of cannabis by drug users in 1973.

Furthermore we would expect that an increased behavioural involvement with the drug would intensify the related attitudinal component. Figure 7.3 shows that as an individual progresses towards a more frequent use of cannabis so his attitudes become more positive. These later figures are consistent with notions of behaviour/ attitude congruity and recent work (Martino and Truss (1973)) which has shown on a variety of measures that attitudes to cannabis were positively related to actual cannabis use. Also of importance (Figure 7.1) is the significant difference (pc-.7.0.001) between non-users and experimenters and between experimenters and casual users in the attitudes they hold towards cannabis. These differences tend to contradict notions of drug use escalation based on a developing pharmacological tolerance since, before an individual can progress from an experimental level of cannabis use to a more persistent level he must

overcome an 'attitudinal barrier'. The 1973 sample (Fig.7.3) analysis also shows the 'non-user' to be separated quite distinctly from the 'experimenter' in terms of his evaluation of cannabis. We have shown earlier that a change in overt behaviour, in this instance frequency of cannabis use, was paralleled by a change in attitudinal valence. An attitudinal valence is regarded as a product of the individual personality but it also embraces the result of environment - organism interaction and other contingent environmental factors relating to past, present and even future behaviour. The individuals attitudinal framework is his internal cognizance of his external behavioural world. It is important for the integrity of his cognitive system, that personal attitudes and beliefs are not inconsistent or dissonant with overt behaviour. The notion that tension or stress is generated when behaviour and belief systems contradict each other or produce disharmony and conflict has been shown to be widely held within psychology (Aronson 1968). It is hypothesised that certain individuals maintain their cognitive integrity by adopting an attitudinal framework which prohibits either an escalation of drug use, or in the case of non users, a commencement of drug use. These individuals could be regarded as internally consistent, in that they maintain their belief system by endogenous as opposed to exogenous factors.On the other hand there are individuals who tend to maintain cognitive consistency by altering their attitudinal framework to fit the change in their own behaviour in the external situation. They therefore place greater reliance on exogenous as opposed to endogenous factors. Such individuals having reached a particular level of cannabis use (in our tables, that of casual use) have to adjust their attitudinal system to take account of the change in overt behaviour, otherwise their attitudes would be dissonant with their behaviour. Once this cognitive conflict has been resolved, i.e. the attitudes restructured, then an increase in the frequency of cannabis use can be tolerated (even to a chronic level) without overt behaviour being inconsistent or dissonant with the new attitudinal framework. If, however, they cannot restructure their cognitive framework, then the only way they can maintain consistency and prevent conflict from arising is to reduce their overt behaviour (cannabis use) to an appropriate level in order to balance their internal beliefs with their drug using behaviour. Two distinct groups of individuals are represented in this present study. The first being internally consistent, do not allow dissonance i.e. conflict between beliefs and actual behaviour to occur, these are the non-users and experimenters. The second group of casual and more frequent users tend to reduce their dissonance by changing their attitudinal framework, and so we observe the increase in positive evaluation of cannabis as concomitant with the increase in reported frequency of cannabis use. Of further interest is the polarised attitudinal separation of non cannabis users from cannabis users of all levels. We have also shown that earlier research (iindmarch (1970)) found the attitudes of both users and non-users of cannabis to be positively biased; and since the 1973 data was collected from a similar population of undergraduates we postulated a 'hardening' of attitudes in the non-using group. Arguably this is for a variety of reasons; but, it is suggested that the vast amount of conflicting information regarding cannabis and its effects produced over the previous four years has been sufficient to produce conflict and dissonance in the attitudinal system of potential users. In order to reduce this dissonance and return their cognitive system to an integral harmonious state, individuals have relied on endogenous norms. In other words, cognitive consistency is maintained by regarding it as 'dangerous', in general, to use a drug about which so much conflicting evidence is presented.

The attitudinal separation of the non-cannabis user from the cannabis user is also an effective 'barrier' to starting cannabis use. Even if a non-user takes cannabis once or twice, say at a party or other social gathering, then he is still unlikely to become a more frequent user because he holds a relatively negative set of attitudes compared with those using cannabis more frequently. Since it is expected that persons need to be 'consistent' in their attitudes they hold then we would expect fewer individuals constantly restructuring their framework to maintain parity with overt behaviour. So we expect fewer chronic and persistent cannabis users than experimenters and this was confirmed in this present sample (Einstein, et al., 1974) and is in accord with results of recent surveys of students (Smart and Whitehead, 1973). FREQUENCY AND PATTERNS OF CANNABIS USE : PSYCHOPHARMACOLOGICAL CONSIDERATIONS In establishing frequency of use statistics via questionnaire methods, the measure used is invariably the number of joints smoked per unit time. Such measures usually preclude any estimate of cannabis use by means other than smoking, and frequently take no account of the relative potency of the cannabis used or the dose administered. The following research shows the relationship of dose, potency and frequency of cannabis use to general parameters of psycho-social behaviour. Any such psychopharmacological study has antecedent, dependent and consequential variables associated with the phenomenon described and measured (Hindmarch (1972(d)). In the present instance antecedent variables concern the chemical and botanical description of cannabis together with measures of dose levels and mode of administration. Situational, emotional and attitudinal variables will also produce an effect at this level. Dependent variables concern the psychological, physiological and pharmacological effects of cannabis administration. Consequential variables, such as the terratogenic, therapeutic and creative effects of cannabis use lie outside the scope of this discussion. One of the most important sets of antecedent variables are those associated with situation, motivation and emotion. Since the present study involves individuals 'at large' in society as members of a drug using 'subculture' these variables are bound to be influencing the observed behaviour, but are impossible to isolate. No attempt at manipulation or control of these situational variables was attempted since Jones (1971) demonstrated how awareness of drug induced states is shaped by companions and environment. Subjects filled out a subjective drug effects questionnaire following solitary smoking of 9 mg of THC and similarly completed a questionnaire following participation in a group of four persons each smoking 9 mg of THC. Subjects in the group situation had a significantly greater rating of the degree of euphoria and the amount of perceptual and cognitive effects than did the same subjects under solitary conditions. The reverse trend was noticed for dysphoria, i.e. a greater feeling of dysphoria was experienced under solitary conditions of cannabis administration. Significantly enough, in view of the comments above on social setting, Jones (op.cit) reported '... subjects frequently commented that it was difficult to make valid judgements because of the laboratory setting.' Furthermore, Miras (1972) and Hindmarch (1972(e)) have commented upon the heightened suggestibility of persons under marijuana intoxication, and have also shown the importance of the group situation in modifying the effects of drug intoxication.

Therefore, the behavioural effects of cannabis use are not only a product of the drug and the immediate physical environment in which the drug is taken, but are also influenced by the particular values of the subcultural group to which the cannabis users belong. Naturally the interaction between these several sources of situational variance is complex but there is no doubt that such variables must be accounted for in any cannabis evaluation performed in man. The placebo effect is well known in man and mention has already been made of the role of other users on the subjective effects of cannabis when taken in a group situation. The 'social high' phenomenon is well documented (Rodin et al. (1970)) and providing the 'set' to experience marijuana effects is sufficiently great, subjects given a placebo do report 'feeling high', 'relaxed', 'a loss of sense of time', and 'feeling more at peace with the world'. (Jones (op.cit)). The most positive attitudes held by cannabis users towards cannabis (Hindmarch (1970)) undoubtedly play a major role in determining the effects experienced by individual users. The 2-THC content of some samples of herbal mixture seized by police drug squads would not be sufficient to induce any psychological effects. However, large quantities of low THC content marijuana are being used daily with most users experiencing a 'high'. Evidence also suggests that totally inert mixtures of 'gum arabic and curry powder' (Hindmarch (1970)) and 'powdered glass and incense' (Fairbairn, Hindmarch et al. (1974)) passed as cannabis can produce typical euphoric reactions when taken by experienced users. Although Weil et al. ( 1968) assert that only practiced users get high and only practiced users can differentiate between placebo and drug effects. Jones' work on the subjective rating of marijuana effects showed that practiced users were no more effective in judging potency of placebo and drug samples than were a naive population. However, Kiplinger et al. (1972) showed that under conditions of constant setting, the effects reported by naive and semi-naive observers were significantly related to the dose level of e-THC administered. Contact with the group of drug users was established using a snowballing technique (Polsky (1971)) and observations were made and data collected from the group using the techniques of participant observation (Hindmarch (1972(a)). Initial contact was established in the Summer of 1969 and maintained until the Autumn of 1972, and it is only because of the long lasting rapport established and mutual confidence gained that the following information was able to be collected. The group comprised 58 (42 a" 169) subjects all of whom had a history of illicit drug use. However, the large group divided into two distinct sub-groups. The first comprising 25 (18e 59) members were persistent and habitual polydrug users (cannabis, hallucinogens, amphetamines, barbiturates and wide experience of prescribed and proprietary medicines, opiates and cocaine); the second group with 33 (240'99) members had a history of polydrug experimentation, but had since September 1970 used only cannabis ( with the exception of a rare and most infrequent use of L.S.D.). The two sub-groups were further distinguished on social behavioural parameters. Eight of polydrug using group (60'29) had a history of psychiatric disorder and two males were then under treatment. Five male subjects had been convicted and fined under the drug abuse legislature and a further male respondent had served a term of imprisonment for drug offences. Only twelve (100' 29) members were in full time employment or completing college education, the remaining thirteen relying on part time jobs or social security benefits to support them. There was also a persistent history of delinquency, minor civil offences (non payment of rent, hire purchase contracts etc) in nine of the group.

By contrast, none of the cannabis using group had ever sought psychiatric help because of their drug use and none had a history of psychiatric disorder. All of the group, bar four married women, were either in full-time employment or completing full-time vocational or college courses. None of the group had been charged by the police for drug offences and the only record of delinquency was a 'shop lifting' offence against one of the female respondents some five years ago. It is the second group of cannabis users that concerns us here. In order to estimate the quantity of cannabis consumed; the group provided, between September 1970 and May 1972, the weight of cannabis (either resin or herbal) consumed during a 24 hour (10a.m. - 10a.m.) period. The cannabis was weighed on a portable 'top-pan' balance prior to use and the weighings represented, therefore, only the cannabis consumed during the 24 hour sample period and not the weight of cannabis possessed by the group. The author or an 'informant' in the group kept a record of the number of people present in the drug using situation which enabled the calculation of the mean weight of cannabis used for each individual during a typical day. The sample days for the weighings were chosen randomly before the collection of data began and all but four were in accord with this principle. As a partial check on the accuracy of the above procedure and to obtain samples for analysis the group was visited 17 times over and above the usual observational sessions. These visits and 3 samples volunteered from the group account for the reefer samples 15 to 28 in the following tables. Five of the reefers collected (15, 22, 25, 26, 27) contained samples of the cannabis weighed during the previous 24 hour sample period. The samples volunteered from the users were collected either 'made up' in cigarette form or were pro-reefers in that the sample supplied was sufficient, when mixed with tobacco to produce a typical reefer. All samples were coded and sent for qualitative and quantitative analysis. Behavioural observations recorded at the time were related to the analytic results via the code number of the sample. The results of the analysis of the cannabinoid content of the reefers together with data from a similar study in London, is reported in full elsewhere (Fairbairn, Hindmarch et al. (1974)). Table 7.4 shows the results of the sample day weighings of cannabis; and Table 7.5 gives the mean daily use of reefers of known cannabinoid content. Table 7.6 gives details of the contents and components of the sample reefers.

Table 7.4 shows the average amount of cannabis consumed per person per day to be 2.8 G. This mean amount is not as important as the wide range of daily dose levels (0.3 G to 12.0 G) and the variety (herbal, resin or mixture) of cannabis consumed. Of interest is the range of modes of administration, and while smoking is predominant eating and cooking are also reported. On sample day 10, 6.4 G of herbal cannabis was infused with water in a coffee percolator for some 20 - 30 minutes, but no samples were collected. However, the group kindly repeated the performance and supplied samples of the infusion and residue for analysis. The tea contained 0.039 mg of THC, a surprisingly low level considering that the preparation was highly valued by the users who claimed the effects of drinking such 'tea' to be 'gentler' and 'higher' than the smoking of an equivalent amount of herbal cannabis. Most users with experience of ingestion of resin or herbal mixtures either in their raw state or when cooked in fudge, stews or biscuits reported the slower onset of drug effects. The subjective effects were rated more like L.S.D. experiences when the drug was ingested in this manner. The 'high' experienced by the herbal tea drinkers is yet another example of the placebo effect and it appears that even habitual users are quite susceptible to such effects. Table 7.5 shows the frequency of use of reefers of known cannabinoid content and indicated the average daily dose of THC per user for the range of samples collected. It is realised that the figures for 'THC consumed' do not represent the dose actually received by the individual. Variations in individual metabolic state, loss of active compound in smoke and 'air stream', number of individuals passing joint, variations in individual smoking habits etc alter the amount of active compound received by the individual. However, even if the individual THC consumption figures are not accurate measures of the dose actually received, they do illustrate the great variability in amount of active compound taken by these users. The net effect of these different sources of variance makes the establishment of any dose response relationships difficult, if not impossible. The following individual effects-relating to the reefers enumerated in Table 7.5 do show, however, a tenuous dose-behaviour relationship. The general pattern of smoking for the majority of the group was usually one reefer during the working day, perhaps at lunchtime, following by a further one or two reefers during the evening. The frequency of use increases somewhat at weekends, and it is usually at weekends that cannabis is ingested or used in cooking. Upon such occasions the individuals present each bring small quantities of cannabis for communal use, and it is not surprising that it is at weekends that the group exhibits noticeable effects of cannabis use.

Reefer No.15: The individual admitted that he could not smoke more than one a day, usually in the evening. On the occasion that he smoked a similar reefer to the one analysed during his lunch break he had to absent himself from his employment because he was to quote 'smashed' and unable to coordinate his movements. He is usually alert during the working day but tends to be more withdrawn at weekends when he occasionally smokes two or three reefers per day. He is essentially isolated within the group and even though he is known and liked by most of the others, he prefers his own company. Reefer No. 22: Two married couples regularly smoke, between them, twenty similar reefers daily. They also use the resin in cooking and have produced a variety of sweetmeats and biscuits containing cannabis. The most characteristic feature of the two female subjects is a persistent hilarity neither are in full-time employment, and they spend most of their time about the flat habited by both couples. The two males appear generally to be somnolent, detached and a little hesitant in their speech when answering questions. Reefer No. 27: Are used habitually by two small sub groups which vary between 3 and 8 members. The daily frequency of use for any one individual is between 3 and 5 reefers. None of the users exhibit any noticeable disorganisation of the cognitive processes and seem

entirely capable of holding full time employment. On occasions some of these users have reported lapses of memory, e.g. forgetting to alight from a bus at the appropriate stop, missing appointments and not posting letters. The most noticeable feature of the above individuals and other members of the group is their apparent 'introversion' and partial detachment from their immediate environment. They all seem more or less withdrawn and somnolent, and most noticeable is the difficulty they experience in placing recent events in their correct temporal order. To a certain extent their perception of time is distorted and minor lapses in short term memory are sometimes obvious. However intentional actions, complex perceptual motor skills, e.g. car driving seem unaffected by their habitual cannabis use. In terms of these observations it would appear that the habitual chronic use of cannabis can make individuals withdrawn, detached, apathetic and lacking intent. However, since all these subjects are coping with normal everyday experiences it would seem that the soporific effects of cannabis are limited to reducing the need to interact with the environment, since there is no impairment of function when interaction is intended, or the behaviour sufficiently motivated. (see also Abel (1970)).

Two interesting observations are made in Table 7.6. The sample (no.15) comprised local grown leaf. The following horticultural details were supplied by the grower ...'the plants were grown from the seeds of Zambian 'bush' germinated in March in pots and transplanted at a height of 9" outdoors in medium clay soil on a well drained embankment with a southern aspect. The plants were cut down in September before flowering by which time they had attained a mean height of 8'.". Perhaps the sample is rich in THC because of its freshness and the apparent lack of CBD is in accord (assuming the seeds to be from Zambia) with Turner and Hadley's (1973) recent results. The relatively high THC content of this sample tends to contradict assertions that potent cannabis cannot be grown in the temperate climate of the U.K., and since this sample contains leaf only it falls outside the present international definition of cannabis as 'the flowering or fruiting tops of the cannabis plant excluding the seeds and leaves when not accompanied by the tops'; as presented in the Single Convention on Narcotic Drugs 1961 (see also Fairbairn et al (1971)). Sample 28, on the other hand,

contained no active cannabinoid whatsoever and comprised a mixture of ground glass and vegetable matter (which turned out to be incense). The subject who used a similar concoction believing it to be Turkish pollen Hashish for which he had paid 18 per ounce, did not experience euphoria but complained of nausea and headache. However, the same substance when used by three other group members ( dose, frequency etc unknown) produced (to quote) 'the usual effects'. Clearly this research is incomplete and few conclusions can be drawn about the generality of the patterns of use reported here. The wide variation of dose, frequency of use, mode of administration and potency are worthy of further consideration. Such variability does imply that the number of joints used per unit time is not a good index by which to classify an individual as a drug user at a particular 'level of use'. This paper has presented two particular approaches to the study of the patterns and frequency of cannabis use. Each approach produces unique data, but it is hoped that future research could be mounted to combine the two techniques and so enable more definitive statements to be made about the psychological effects of cannabis administration in man. REFERENCES Abel, J. (1970): Marihuana and Memory. Nature 227, 1151- 1152 Abelson, R.P., Aronson, E., McGuire, W.J., Newcomb, T.M., Rosenberg, M.J. and Tannenbaum, P.H. (1968): Theories of Cognitive Consistency: A Source Book. Rand McNally: Chicago. Aronson, E. (1968): Dissonance Theory: Progress and Problems, in Theories of Cognitive Consistency p.5-27, ed: Abelson et al. Rand McNally: Chicago. Brehm, J.W. and Cohen, A.R. (1962): Explorations in Cognitive Dissonance. Wiley: New York. Campbell, D.T. (1963): Social Attitudes and Other Acquired Behavioural Dispositions, in Psychology: A Study of a Science ed. S. Koch, vol.6 McGraw Hill: New York. Collins, B.E. ( 1968): Behaviour Theory, in Abelson et al. eds: Theories of Cognitive Consistency ch.13 pp.240-243. Doob, L. (1947): The Behaviour of Attitudes, Psychol. Rev. 54, 135-156. Einstein, R., Hughes, I . and Hindmarch, I. (1974): Patterns of use of alcohol, cannabis and tobacco in a student population. In press, Br. J. Addict. Fairbairn, J.W., Hindmarch, I., Simk, S. and Tylden, E. (1974): Cannabinoid Content of Some English Reefers, Nature 249, no.5454, pp.276-278. Fairbairn, J.W., Liebmann, J.A. and Simic, S.J. (1971): J. Pharm. and Pharmae. 23, 558-559. Festinger, L. (1957): A theory of cognitive dissonance. Stanford Univ. Press: Stanford. Festinger, L. (1964): Conflict, decision and dissonance. Stanford Univ. Press: Stanford. Gilbert, J.G. and Lombardi, D.N. (1967): Personality characteristics of young male narcotic addicts, J. Consulting Psychol. 31, 536. Halstead, H. and Neal, C.D. (1968): Intelligence and Personality in Drug Addicts. Brit. J. Addiction, 63, 237. Hill, H.E., Haertson, C.G. and Glaser, D. (1960): Personality Characteristics of Narcotic Addicts as Measured by the MMPI. J. Gen. Psychol., 62, 36-42. Hindmarch, I., Einstein, R. and Hughes, I. (1974): Attitudes to the use of cannabis, alcohol and tobacco on the campus of a provincial university. In press, Bull. Narc. Hindmarch, I. ( 1970): Patterns of drug use in a provincial university. Brit. J. Addiction 64, 395- 402.

Hindmarch, I. (1972a): Observer participation. Drugs and Society, 1, 11, 6-9. Hindmarch, I. ( 1972b): The patterns of drug use and the attitudes to drug users in school age and student populations, in L'homme et lee Substances Psychoactives, ICAA: Amsterdam, vol.B4, 1-10. Hindmarch, I. (1972c): Patterns of drug use and attitudes to drug users in a school aged population, in Student Drug Surveys ed. Einstein and Allen, pp. 77-95. Baywood: New York. Hindmarch, I. (1972d): The psychopharmacology of cannabis, in Drugs, Sleep and Performance, NATO: Aviemore, p.27.1-27.10. Hindmarch, I. ( 1972e): in Cannabis and its derivatives, ed. Paton, W.D.M. and Crown, J. p.150 OUP: London. Hughes, I., Einstein, R. and Hindmarch, I. (1974): The Semantic Differential: its usefulness in drug user surveys. Papers in Psychology: submitted March 1974. Jones, R.T. ( 1970): Tetrahydrocannabinol and the marijuana induced social 'high', or the effects of the mind on marijuana. Anal. N.Y. Acad. Sci., 191, pp.155-165. Kiplinger, G. (1972): The effects of cannabis derivatives on cognitive functioning, in Cannabis and its Derivatives ed. Paton, W.D.M. and Crown, J. O.U.P: London. McKay, A.J., Hawthorne, V.M. and McCartney, H.N. (1973): Drug taking among medical students of Glasgow university. Brit. Med. J., 1, 540-543. Martino, E.R. and Truss, C.V. (1973): Drug use and attitudes toward social and legal aspects of marijuana in a large metropolitan university. J. Counselling Psychol., 20:2, 120-126. Miras, C.J. (1972): in Cannabis and its Derivatives, ed. Paton, W.D.M. and Crown, J. O.U.P: London. Mott, J. (1972): The psychological basis of drug dependence: the intellectual and personality characteristics of opiate users. Brit. J. Addiction, 67, 89-99. Osgood, C.E., Suci, G.J. and Tannenbaum, P.N. (1957): The measurement of meaning. University of Illinois Press: Urbana. Polsky, N. (1971): Hustlers, Beats and Others, Penguin: London. Rodin, E.A., Domino, E.F. and Porzak, J.P. (1970): The marijuana induced social high. J.A.M.A. 213, 13001302. Rosenberg, C.M. (1969): Young drug addicts: background and personality. J. Nervous Mental Disease, 148, 65. Smart, R.G. and Whitehead, P.C. (1973): The prevention of drug abuse by lowering per capita consumption: distributions of consumption in samples of Canadian adults and British university students. Bull. Narc., XXV, 4, 49-55. Steininger, M. (1973): In defense of measuring attitudes. J. of Psych ol., 85, 131-136. Turner, C.E. and Hadley, K. (1973): Constituents of cannabis sativa L: Absence of cannabidiol in an African variety. J. Pharmaceutical Sciences, 62, 2, 251-255. Weil, A.T., Zinberg, N.E. and Nelsen, J. (1968): Clinical psychological effects of marihuana in man. Science, 162: 1234-42. Weitman, M., Scheble, R., Johnson, K.G. and Abbey, J. (1972): Survey of adolescent drug use: correlations among users of drugs. Amer. J. Public Health., 62, 166-170. Whithead, P. and Smart, R.G. (1971): Validation and relation of self-drug use. mimeograph Toronto: ARF. Wilson, A., Rosenblood, L.K. and Oliver, (1973): Congruity theory and linear models of attitude change. Canad. J. Behav. Sci., 5, 4, 399-409. Zimbardo, P.G. (1969): The cognitive control of motivation. Scott, Fonsman: Illinois. DISCUSSION

A recurring theme of concern to some participants was the comparability of his two university student samples, and the meaning which one could derive from the fact that reported frequencies of use in these two populations of ever-users differed. Since the two samples were not identical, there was some doubt that the lower frequencies of use reported in the later sample could be taken as indicative of any real change in pattern of use over time. This concern over matching, and over whether observed differences might be artifacts of lack of matching of samples, has already been noted in connection with Dr Soueif's paper. Dr Someck congratulated Dr Hindmarch on his paper, and asked 'Do you really believe that the differences between your 1969 and 1974 samples were really not differences in sampling?' Dr Hindmarch: 'There are lots of reasons why there are differences. There are differences in the samples, there are differences in the way that I actually generated the two samples and in the techniques used to collect the information. However the comparison of the two samples in terms of theories of cognitive consistency seems justified since the difference in level of use parallels a difference in attitude. Furthermore, information collected by participant observations suggested a decrease in 'chronic' cannabis use in the area.' AMOUNTS OF THC CONSUMED Dr Miller asked about the ways in which the amounts of cannabis used were determined, and pointed out some differences between amount likely to be absorbed into the body in different conditions of use. In one study carried out for the LeDain Commission, fourteen regular (3-4 times per week) users in their twenties were asked to weight all drugs used and to submit samples for analysis. 'The range of THC consumed ranged from 1 mg. per session up to more than 50 mg in a few instances, and the sessions ranged from one up to twelve hours. The median was about 7 mg THC, and alcohol use was used in about one third of the sessions. There is quite strong evidence from other studies in North America that typical doses are from 2 to 8 mg. Laboratory work indicates that with a tightly controlled smoking paradigm which maximises delivery, retention and absorbtion, about 6 mg put regular (several times a week) users slightly higher than where they prefer to be.' With reference to cross cultural comparisons, Dr Miller said that 'There are pronounced differences in the style of smoking in different countries. In those countries where the total THC consumption per day is very high, for instance Jamaica and India, the style of smoking is one of puffing. In Jamaica, they roll very large spliffs, and they just puff and puff and puff, so much that sometimes you can hardly see the smoker. So he's running through huge quantities of cannabinoids, and then throwing away large roaches, which probably contain a large percentage of the total cigarette weight. They probably absorb a relatively small percentage of total used. But in North America, the smoking style is very anal-retentive, the whole roach is consumed, and retention and absorption is much higher.' Dr Hindmarch said that the smoking style in the UK was comparable to that in North America. Dr Tinklenberg reiterated the importance of looking at styles of smoking, puffing rate, length of butt in determining the amount of THC actually absorbed. 'And I would like to point to research in the area of tobacco, which I think has established quite clearly that consciously or unconsciously the heavy user learns how to modify his consumption of tobacco in order to

maintain a certain level of absorbed nicotine.' Both Dr Tinklenberg and Dr Miller believed that these were issues worth bearing in mind in future research of the kind Dr Hindmarch described, which was well worth while. Dr Rathod agreed and drew a parallel in quoting the style of use in India in which there was no control of dose after ingestion. EDITORIAL NOTE The editors would like to add that the tightness with which the cigarette is packed affects its rate of burning. A more loosely packed joint may burn more quickly, when more will literally go up in smoke between puffs, and less into the lung. Also , more economical use of the joint may occur in a large group, since the ratio of puffing to free burning may be higher then. In a small group, and more particularly with a solitary user, continuous puffing is most unlikely for any length of time unless the joint is extremely weak. Thus the percentage of THC smoked that is actually absorbed may be lower in solitary users than in group using situations. USE OF ATTITUDE MEASURES Dr Edwards suggested that attitudes to cannabis might have to be broken down a little more to be most meaningful. In connecting with some current research into drinking, it is clear that one has an overall evaluation of alcohol, but that really is not all that predictive or meaningful. 'There is an attitude to me drinking, and an attitude to somebody else drinking, and to women drinking, and toward older people drinking, and younger people drinking. And then there are attitudes towards getting drunk as opposed to drinking. And you can invite people not to put one mark on a line to indicate their attitude, but to put marks to show their area of uncertainty. Suddenly you find how asking for only one point on a line has denied your subjects' potential to express what they mean. It may be very important as an indicator of the instability of somebody's position to give them an opportunity to express their area of uncertainty.' ATTITUDES, BEHAVIOUR AND PREDICTION Dr Hawks suggested that it was very important to note that relationships between, say attitudes and behaviour in a sample generated in one manner might be quite different from the relationships between attitudes and behaviour generated in another manner. Dr Hindmarch agreed that such a proviso applied to his study. Dr Edwards expressed his interest in longitudinal studies to discover whether people's stated attitudes to cannabis are predictive of their later behaviour. Mr Dorn reported that the ISDD's work had shown that tte semantic differential did clearly differentiate between non-user with and without drug-using acquaintances in terms of their perception of the 'drug user'. He asked whether these differences in attitude were perhaps the result of the respondent's social location and affiliations. Could one talk of attitudes being predictive of future behaviour, or was the relationship the other way round (meeting a user and thinking - he's not so bad)? Dr Hindmarch replied that he believed behaviour to be largely directed by the environment in childhood, and by the established attitude system in adulthood. Mr Dorn suggested that such a change-over to 'internal' control might never occur: adults might simply have and respond to a more static environment. Research in this area should look at the hypothesis that a change in

social experience and behaviour influences attitudes, and well as the hypothesis that attitudes mould behaviour. COSTS Drs Miles and Tinklenberg asked about costs of cannabis in the UK, and there was some discussion of the difficulties of estimating this. The editors estimate that if a slightly short ounce of 25 gm of resin of 5 per cent THC is bought at a cost of 15 (the current UK price range is about 12-18 per ounce), then the cost per milligram is just over one penny. Since cost goes up for buying in smaller lots than one ounce, we could say that 10 mg THC would cost about 20 pence.

8. Use of Drugs Other than Cannabis and Attitudes to Drugs in U.K. Student Populations
D. SomeIch, Department of Psychology, Bedford College, London. In her presentation to this conference Adele Kosviner catalogues 18 or so studies of student drug use, but for the purposes of this paper, only a few of those she mentions will be reviewed. If one selects those studies published since 1970 which are surveys (as opposed to studies of referred groups such as psychiatric cases) and studies which involve young people who are no longer school-children, only five surveys need be referred to, those of Young and Crutchley, Kosviner et al, Somekh, McKay et al and Fish and Wells. While cannabis use is the main theme of this conference, among students, as in any other population, there are two additional forms of drug-taking, use of socially approved drugs and use of illicit drugs other than cannabis. A consideration of such drug use may serve to throw more light on the patterns of cannabis use by different sub-groups. USE OF SOCIALLY APPROVED DRUGS Use of cigarettes and alcohol by the student population as a whole has been described by Kosviner et al (1973), Somekh (1974), Young and Crutchley (1971) and McKay et al (1973). The first three mentioned studies found that 33 per cent to 48 per cent of students surveyed were smoking cigarettes at the time of the survey. McKay et al reported that 22 per cent of their Glasgow medical students admitted to smoking cigarettes regularly (figures quoted here will refer to McKay et al's second sample, 1971-2). If those who have ever tried drugs are compared with non drug-users across the four studies, 66 per cent to 77 per cent of users were smoking cigarettes compared to 30 to 33 per cent of non-users. From such figures both Young and Crutchley and McKay et al infer that there is a considerable relationship between tendency to use legal drugs and illicit drug use. This view is reinforced by the findings of Kosviner et al (1974) and Somekh (1974), where drug-users were sub-divided according to degree of drug use. In both cases a strong association between numbers of cigarettes smoked and degree of drug use was found. Kosviner et al reported that

only 45 per cent of novice users smoked cigarettes (in Somekh's study a similar category yielded 57 per cent smokers) compared to 74 per cent of the other, more regular users (72 per cent of Somekh's regular users smoked cigarettes). The data available is therefore fairly consistent across studies. A strong association between cigarette smoking and cannabis use, the latter being the principal drug involved, is hardly surprising, as cannabis in this country is most often utilized in the form of cannabis resin, crumbled into tobacco and rolled and smoked as a cigarette. However, there are other possible factors which may influence the relationship between tobacco and cannabis use and these need to be investigated. When comparing areas which show less consistency in the findings, problems can arise because different studies do not yield exactly comparable data. The available material on student alcohol use illustrates this point. McKay et al report that of students who drank 'regularly' 22 per cent had taken drugs once or more, whereas only 5 per cent of those who did not drink 'regularly' had done so. Young and Crutchley actually state that cannabis users are more likely to drink than non-users although their figures are not impressive: 64 per cent of non-users (N = 34) drank once a week or more compared to 74 per cent (N = 39) of users. Kosviner et al (1973) found that only 8 per cent of users reported that they had never been drunk, compared to 31 per cent of non-users. On the other hand, there were only small differences between users and non-users as to the amount drunk on any one occasion or the frequency of drinking in the previous two months. These rather unsatisfactory results are partly clarified by two pieces of evidence. Kosviner et al (1973) were the only workers to ask about respondents being drunk. Their findings are extended by Kosviner et al (1974) who reported that approximately 43 per cent regular users had been drunk more than ten times compared to 24 per cent of novice users (and 13 per cent of nonusers), a strong association between excessive alcohol use and regular drug use. This is supported by Somekh (1974) who reported that 12.5 per cent of regular users drank a maximum of 6 or more drinks at any one time compared to only 4 per cent of novice users and 5 per cent of non-users. These findings lead one to speculate regarding the effects of alcohol in users, but the evidence available is not sufficient to allow determination of whether or not cannabis users' responses to alcohol are different to those of non-users. Secondly, Somekh is the only worker to sub-divide non-users by contact with drugs. When those who had no contact, those who were offered drugs and refused, novice users and heavier users were compared in respect to frequency of drinking and maximum drunk at any one time, it was found that there was little overall differences between the latter three groups. For example, 43 per cent of those who had no contact with drugs drank once or more per week compared to 72 per cent of those who refused drugs, 72 per cent of novice users and 75 per cent of heavier users. Thus it could be postulated that the relationship between drinking and drug use was social rather than related to tendency to use psychotropic substances. This is partly borne out by Kosviner et al (1974)'s finding that significantly more novice users (30 per cent) reported drinking a maximum of six or more drinks at any one time than did regular users (14 per cent), apparently contradicting Somekh's finding, quoted above. However, Somekh has sub-divided heavier users into regular and multi-drug users (heavy users) and found similarly that alcohol use was significantly greater among regular users than among multi-drug (heavy) users. 80 per cent of those classified as regular users (N = 110) drank at least one drink daily as opposed to 60 per cent of heavy users (1 = 90). Thus there is tentative evidence of a 'break-point' in the association beteeen alcohol and cannabis use, with a relative drop-off in alcohol use among the heaviest cannabis users.

To summarize, there is a highly significant association between cigarette smoking and cannabis use among U.K. students but it has yet to be demonstrated whether this relationship has a pharmacological or some other basis. It could be useful to examine the cigarette habit of those who prefer to use cannabis in other ways than smoking it, for example. Those students who do not have any contact with illicit drugs definitely drink less than those who do but among the latter group there seem to be those who use alcohol to excess and those who use illicit drugs to excess, the overlap between these categories being perhaps limited. It may also be that those likely to use illicit drugs are more susceptible to the effects of alcohol. USE OF ILLICIT DRUGS OTHER THAN CANNABIS The majority of students who try other illicit drugs have tried cannabis - in Kosviner et al (1973)'s study less than 1 per cent of the sample had only used drugs other than cannabis and Somekh (1973) reported a figure of less than 2 per cent of his total sample in a like category. Table 8.1 shows a comparison of four recent major U.K. studies. The figures show the frequency and percentage of total sample that have ever tried each of the drugs mentioned. They represent approximate comparisons as there are small differences in categories, e.g. Kosviner et al's figure for amphetamine use includes some other pills and the Somekh figure for heroin use includes physeptone.

It can be seen from Table 8.1 that the data from these recent studies (data collected between October 1970 and late 1972) is very consistent, the only discrepancy being the smaller percentage of students in the Fish and Wells (1974) study who reported cannabis and amphetamine use as compared to the other three studies. These minor differences may simply be a matter of geography, the Fish and Wells study being the only one not carried out in Southern England. From Table 8.1 it seems that while 1 in 3 of students who respond have

tried cannabis at least once, 1 in 10 have tried amphetamines and 1 in 10 tried hallucinogens, but only 1 in 100 has tried heroin. Fish and Wells illustrate the patterns of illicit drug use by reference to step-down rates, that is, the percentage of users of each drug who have also used each of the other drugs. Although this presentation does illustrate well known features e.g. that only a proportion of cannabis users have tried LSD and even fewer heroin, while most heroin users have tried nearly every other drug considered, it does not allow specification of drug use profiles. Further, Fish and Wells data for patterns of use does not specify student drug use, only bulked data being given, despite the fact that this involves combining data from rather disparate sample groups. Kosviner et al (1974) and Somekh (1973) give details of the combinations of drugs used by students. Their categories are more or less the same: thus, 54-56 per cent of users have used cannabis only, 13-14 per cent have used cannabis and amphetamines only, 15 per cent have used cannabis and LSD only, while 12-17 per cent have used cannabis, LSD, opiates and/or other drugs in combination. Furthermore Somekh (1973) has shown that although 40 per cent of those who have used cannabis only have used it more than ten times, 80 per cent of those who have used cannabis and LSD have used cannabis more than ten times and over 90 per cent of hard drug and multi-drug users have used cannabis this often. 22 per cent of drug users had smoked cannabis more than ten times and tried at least one other drug. From these figures it would seem that experimentation with drugs other than cannabis occurs in about 50 per cent of student drug take; as a whole and about 60 per cent of regular users. The common drugs involved are amphetamines and LSD and while regular use of cannabis is associated with experimentation with other drugs, including opiates, there is also a substantial group of regular cannabis users who do not experiment with other drugs. ATTITUDES TO DRUGS Attitudes to cannabis as well as other drugs have been examined in users and non-users. Binnie and Murdoch (1969) inquired into respondents' attitudes to particular drugs from the point of view of risk to mental and physical health, accuracy of public attitude and adequacy of present legal controls. They found agreement between users and non-users in respect to alcohol, tobacco and opiates. User here means a person who has ever experimented with any drug. In respect to cannabis, users' attitudes were consistently more favourable. With regard to LSD and amphetamines, there was agreement on the extent of health risk, in that both users and non-users felt that LSD was dangerous and amphetamines were not, but for both drugs users felt more strongly than non-users that legal controls were too severe. Young and Crutchley found a consistent difference between users and non-users of cannabis with their 106 social science students, fewer non-users agreeing that cannabis should be legalized, and more agreeing that cannabis is addictive or leads to heroin use. These findings could be interpreted either as a cognitive dissonance type response by cannabis users or simply that their information i.e. from experience, was more accurate. ** defined simply in this context as those who have used cannabis more than ten times. Kosviner et al ( 1973) asked users and non-users of cannabis to evaluate the drug and answers tended to suggest that respondents' answers were less categorical if they had had experience of the drug. Greater knowledge of the drug seemed to produce more differentiated answers,

something not evident from Young and Crutchley's data. Cannabis users and non-users differed little in their attitudes to injecting drugs, as Binnie and Murdoch found. Kosviner et al (1974) further analysed the attitudes of sub-groups of cannabis users. Evaluation of cannabis did tend to be more positive, the greater the use. Similarly, heavy users were more in favour of legalization. There were no differences between groups in respect to contact with and attitude to injectable drugs. Somekh (1974) used a 25 question attitude section, each question being answered on a 5 point Likert scale ranging from Strongly Agree to Strongly Disagree. This enabled attitudes to be assessed quantitatively, as the scores for 17 questions were summed to give a 'pro-drug' or drug-favorability score for each subject. When subjects were sub-divided into groups of ascending drug involvement, a clear trend in mean 'pro-drug' score was obtained (Table 8.2).

The biggest differences in means were between groups 2 and 3 (demarcating non-users from possible users) and between groups 4 and 5 (demarcating experiments from regular users). Attitude scores were checked for split half reliability within groups using the Guttman formula and values of rll between 0.75 and 0.85 were obtained, that for the whole sample of 1113 being rll = 0.90. An attempt was made to assess the validity of such a drug involvement score by comparing attitudes to specific drugs with actual use of those drugs. As had been found by other workers, attitudes to cannabis were more favourable, the heavier the use of the drug. Attitudes to LSD showed a similar pattern with non-drug users having a mean score of 4.31 (over 3 questions, maximum LSD-favourable score: 12.00), non- LSD using drug-users 6.15 and LSD users 8.38. Within LSD users (N = 86), 23 who had used hallucinogens once only had a mean score of 7.83 while 21 who had used hallucinogens more than 10 times yielded a mean score of

9.51. Greater involvement with a particular drug therefore gives rise to higher s cores on questions specific to that drug, although it is impossible to say whether this is because the particular drug is viewed more favourably or that this reflects greater overall drug involvement. The fact that multiple drug users in Somekh's survey were the heaviest cannabis users suggests that the two factors go hand in hand and this is borne out by the fact that over 85 per cent of those who had used LSD more than ten times were also to be found among the multiple drug users. It would seem that Somekh's attitude questions have a degree of face validity and that the attitude questionnaire might prove useful in the assessment of the effects of drug education films and the like. THE USE OF UNIVERSITY STUDENTS AS SUBJECTS FOR SURVEY Adele Kosviner, in her paper, has spoken of the study of the student drug taker as approximately to a description of students in general. It seems worthwhile to follow up her point briefly by looking at the appropriateness of using students as a specific survey group. There are two basic disadvantages involved in surveying students, the first being that this results in a 'focussing'phenomenon in that, subsequently, results may be interpreted as if students were the only young people engaged in the behaviour studied. This, despite the fact that (a) the origins of the behaviour lie, in many instances, in a much earlier environment, i.e. school, common to all young people and (b) there is nothing to suggest that drug involvement among non-students is any smaller, rather students are likely, e.g. to be less opiate-involved because of such behaviour being incompatible with life as a student. The other disadvantage is that students are undoubtedly over-exposed to such enquiry methods. This means that students are very often sick and tired of filling in forms and hardly qualify as naive subjects, as many probably have developed their own form-filling set. These disadvantages can be disposed of, to some extent. Students are over-exposed because there are definite advantages in studying them. With students we have a circumscribed group, easier to reach and more approachable than most and the behaviour to be studied can be related to specific aspects of their environment, similar for many students and hence likely to be generalizable within certain limits. The point about focussing is important as it forces us to reconsider the purposes of the investigation - why are we asking? The main reason for wanting to know something about 'how many?' and 'which drugs?' is to confirm the order of magnitude of the situation, little more, so that sensational guesses can be obviated. The essence of the question 'why drugs?' on the other hand may be to provide a reliable (and hopefully valid) description of the student drug taker in relation both to his or her membership of the group 'students' and the group 'drug-takers' and provide corresponding descriptions of students who are not involved with drugs. In doing this, whole focussing on students in a possibly undesirable way, we may hope to provide a paradigm for consideration of other members of more than one group, e.g. delinquent drug-users. In both these examples, membership of one group is not a necessary condition for membership of the other but membership of both groups may produce an interaction which is useful for the investigation of processes underlying membership of either or both groups. Students who live at home illustrate this notion. Several workers have found that students who live at home during term-time are less drug involved than those who live away. For example, Somekh (1974) states that nearly a quarter of students who had no contact with drugs lived at home during term-time as opposed to only 8 per cent of drug users. The danger here is the temptation to use such a finding in a simplistic way, which Young and Crutchley seem to

have done. They support their hypothesis that marijuana use is related to social maturity rather than immaturity by quoting the student living at home as an example of the social isolate who is therefore less likely to use marijuana. In fact the case of the student living at home bears a much more complex relation to student life, as Brothers and Hatch (1971) have shown. Brothers and Hatch found that students living at home during term-time were less satisfied with university life, more highly motivated to achieve success and yet differed only very slightly from hall-based students in respect to contact with staff, participation in student activities and extent of university-based friendships. The point being laboured is that when data from student surveys is interpreted at over and above the most basic demographic level, student status itself becomes a confounding variable. Once this is recognized, however, and the characteristics of student drug use are examined using the premise that student drug use represents the interaction between student status and drug-taking status, a greater understanding of both of these phenomena becomes possible. REFERENCES Binnie, H.L. and Murdock, G. (1969). The attitudes to drugs and drug takers of students at the University and Colleges of Higher Education in an English Midland City : Vaughan Papers No. 14: University of Leicester. Brothers, J. and Hatch, S. (eds) (1971). Residence and Student Life. Tavistock Publications, London. Fish, F. and Wells, B.W.P. (1974). Prevalence of drug misuse among young people in Glasgow, 1970-72. Brit. J. Addict. (in press). Kosviner, A., Hawks, D. and Webb, M.G.T. (1973). Cannabis use amongst British university students: 1 Prevalence rates and differences between students who have tried cannabis and those who have never tried it. Brit. J. Addict. 69 : 35-60. Kosviner, A., Hawks, D. and Webb, M.G.T. (1974). Cannabis Use amongst British University Students: 11 Patterns of Use and Attitudes to Use Amongst Users. (Draft report unpublished). McKay, A.J., Hawthorne, V.M. and McCartney, H.N. (1973). Drug-taking among Medical Students at Glasgow University. B 1 : 540-543. Somekh, D.E. (1973). Prevalence of self reported drug use among London undergraduates (unpublished report). Somekh, D.E. (1975). A survey of self-reported drug use among London undergraduates. 1971-2. (in press). Young, J. and Crutchley, J.B. (1971). Student cultureproject: (Enfield Polytechnic) first annual report submitted to the Social Science Research Council (unpublished). DISCUSSION CANNABIS AND ALCOHOL Dr Hawks said 'it has been demonstrated in one or two studies that there is an association between cannabis use and heavy use of tobacco, suggesting that there is some predisposition towards use of licit and illicit drugs. Whilst the excessive use of alcohol may not be associated with the excessive and concurrent use of cannabis, there is some evidence that people who now use cannabis have in the past been heavy users of alcohol. There is the possibility that alcohol use amongst those people currently smoking cannabis is suppressed because of the potentiating effect of the two drugs.'

Dr Miller spoke of the need for studies to differentiate between-subject at a single point of time and within-subject behaviour over time. If one divides studies according to whether they are describing between-subject correlation at a single time or within-subject correlation over time, then one sees that the within-subject data points towards a decrease in alcohol use with onset of cannabis use. 'As far as I know the statement often made that cannabis merely adds to alcohol use is without substance: not one good study addresses itself to that question. What we need here is prospective research.' Dr Rubin offered some information on use of cannabis and other drugs in Jamaica; there tend to be fewer cigarette smokers among non-cannabis smokers. Cannabis smokers tend to drink less alcohol than non-smokers. Also, cannabis tends to be used quasi-medically instead of prescription drugs - it is an all-purpose drug. Nor is there significant use of any other drug such as heroin in Jamaica. Dr Edwards asked to what extent alcohol and cannabis were substitutable for each other, and suggested that knowing this could tell us quite a lot about the meaning and purpose of the behaviour. 'I'm particularly interested in Kosviner's finding that the heavier cannabis users are the ones who are more often getting drunk. I would be interested to know whether cannabis users were more skilled at getting drunk - getting drunk requires a certain skill just as getting high does. Perhaps the cannabis user, when he does drink, uses drink in a special way. As well as the expensive prospective studies, we can look at the history of drug use of young people with drinking problems, compared with a control group of those without drinking problems. We should look particularly at the motivations, as well as the frequencies, of the behaviour'. Dr Tinklenberg spoke of the cycle of drug use that could sometimes be seen in individuals. There tends to be a period of quite extensive drug use, with a preference for stimulants in adolescence and early twenties. As they get older there may be a levelling off for most of the population. So in considering whether drugs will be addictive or not, we have to specify a stage in the life cycle. CANNABIS AND TOBACCO It was agreed that, with the exception of North America where marijuana is typically smoked alone, cannabis resin is smoked together with tobacco in most parts of the world. Both cannabis and opium were used alone before the introduction of tobacco into the East, but they were now often used with tobacco. Professor Paton pointed out that vomiting, salivation, slowing of the pulse or tachycardia, raised blood pressure, antidiuretic effect, and fainting were possible effects of tobacco. Among the physiological effects attributed to cannabis are tachycardia, diuretic effect and sometimes vomiting and fainting. Thus many cannabis effects might be masked or distorted by tobacco effects if the two are taken together. Dr Miles reported that he had good data showing that cardiovascular tolerance to cannabis develops over a period of 70 days. If the subject is then deprived of cannabis for a long time and then given a small dose of cannabis there is a marked rise in blood pressure.

9. Prevalence, Characteristics and Correlates of Cannabis Use in the U.K. Student Population
Adele Kosviner, Addiction Research Unit, Institute of Psychiatry, London University. Attempting to survey characteristics of students with experience of cannabis is becoming increasingly like attempting to survey characteristics of students generally, if not of life. Experience of cannabis is no longer a small minority activity, and by the size and nature of its appeal it has become 'a phenomenon' - or part of one, depending on one's preconceptions. Anything can gain symbolic meaning but we are in danger of our reaction to cannabis being distorted by the degrees of symbolism with which it is variously attributed. It is no longer just cannabis use, but 'a symptom for which man has no inner cure', 'a symbol of the rejection of bourgeois values', 'an indication of underlying pathology', 'the glue to a bohemian subculture'. Thank goodness, sometimes, for pharmacologists, Meanwhile, its very investigation seems to have become a 'meaningful phenomenon' too, with well mannered but determined manoeuvres for possession going on between chauvinists of various disciplines, and protagonists of differing philosophies. Arguments as to the most relevant (and to whom?) theoretical or conceptual framework within which to consider possibly relevant (and to what?) correlates, is the bane or excitement of all social research. Shame that the search for clarity and meaning almost invariably leads to defensiveness and over-brave assertion of one's own perspective. But perhaps that's what being laudably controversial and inter-disciplinary is all about. It is not the intention of the foregoing to have raised hopes as to any riveting qualities of English student drug surveys. It is merely a reminder (should there be need) that the chosen samples and variables of investigation are, in the main, reflective of these various predilections - from psychiatric referrals to sociology students, from symptoms of confusion and depression to views on the relative merits of violent and non-violent protest. There have been approximately 18 reports of student drug use in the U.K. to date. (This includes three surveys of school children). They range in date of execution from the mid 1960's to the present. In addition to variations in the representativeness of the samples chosen, they range in sample size from just over 100 to just under 8,000; in achieved response rate from under 50 per cent to 100 per cent (but these are, in the main, unspecified); and in ease of access and audience appeal from unpublished reports through Women's Own articles to reports in the drug journals. It is the task of this review to consider what the Nation or Science might have achieved from this somewhat haphazard (if no doubt expensive) counting of heads. I intend firstly to summarize prevalence rates, then to review any common findings as to characteristics and correlates of use, and finally to consider whether there are any well supported indications as to broader contexts in which to view student cannabis use. The 18 reports are summarized in appendix 9.1, with apologies for any unintentional omissions or misrepresentations. PREVALENCE

It can be seen from this listing that prevalence rates of cannabis use amongst school children are estimated at between 3-5 per cent, with the higher estimate being the more recent. The overall rates for 'any' drug use amongst school children (the other major drug of misuse being amphetamines) go up to 9 per cent (ever used). Both figures are generally considered to be under estimates, as non-class attenders (where readers are informed at all) are not represented. One study (Hindmarch, 1972) considers that amphetamines are the drug of choice amongst the earlier age groups (the mean age for first use being 14- 15 years), with cannabis being more popular with older children (the mean age of first experience being 16 years). Considering surveys of student drug use the prevalence rates obtained have varied hugely, from 2 per cent to 50 per cent. This is due to a number of reasons including year and type of survey, type and location of college, and type of student. In the earlier reports (published before 1970), the estimated prevalence range was 2 to 10 per cent ever used, with perhaps 1 to 4 per cent counting as (ill defined) 'regular' users. Since then estimates have ranged from 3 to 50 per cent (ever used) but the source of differences are more distinctly attributable. Excluding studies with very biased samples or exceptionally low response rates, the probable range for use within colleges (rather than broken down by faculty) is, on the evidence available, approximately 10 to 40 per cent. This again, if anything, would be an underestimate: time has passed since the last survey taken and there are indications to suggest both that users are over represented amongst non-respondents and that use is perhaps becoming more widespread. (Kosviner and Hawks unpublished data). Rates appear to be highest amongst more urban, popular, prestigeful colleges, and those oriented towards disciplines in which users are over presented. The faculties which appear to reflect higher rates of use are, again on the basis of available evidence; social science, arts and medicine, with physical science, engineering and business studies reflecting lower rates. However if one is looking for an average figure, then considering only those studies of adequate response rate, covering a variety of subject faculties, which are relatively recent and focused on colleges likely to reflect relatively high rates of use (9, 10a, 13), then the mean prevalence rate found was remarkably consistent at around 33 per cent ever used. CHARACTERISTICS OF USE Students with experience of cannabis seem to be divided almost equally into those who have tried it 10 times or less, and those who have tried it more than 10 times (13,18). Those studies reporting information on current (as opposed to 'ever') use (13, 18) suggest that some 60 to 70 per cent of those who have experienced cannabis see themselves as continuing to use it, and 20 to 25 per cent are using it once a week or more. According to one study (18) , the majority of those who stop using the drug do so after very few experiences of it. The drop off rate was 75 per cent for those who smoked only once, 57 per cent for those who had smoked 2-10 times, and 18 per cent for those who had smoked more than 10 times. Briefly considering other characteristics of student use, it is apparent that the average age for first trying cannabis is in the region of 17 to 19 years and that people who have had experience of cannabis are older at entry to college than their contemporaries who have not. Estimates of the percentage of students who first tried cannabis prior to starting college range from 8 per cent to 65 per cent, with no obvious explanation for the discrepancies. Whilst Dr Somekh will be reporting on the use of other drugs it is perhaps worth mentioning here that cannabis is far and away the most popular drug choice amongst undergraduates, with fewer than 2 per cent of those with any drug experience not having experience of cannabis (10a, 13).

I understand it is not my belief to review details of subjective effects and reasons for cannabis use, but it is perhaps worth mentioning that in this sample, 45 per cent of users said they had experienced some physical ill effects (usually mild), 30 per cent 'nervousness, confusion or panic '; and 25 per cent a 'bad experience' as such. To avoid any appearance of bias it should perhaps be added that 73 per cent reported pronounced feelings of well being. CORRELATES OF USE Before attempting to extract any common correlates of use from the studies listed, another note of warning is appropriate, however familiar. Whether one sees the spread of cannabis use in psychological terms of diffusion of innovation or sociological development of sub-cultures; attitudes, expectations and styles of behaviour can spread or develop in the same way. Therefore attribution of causal or even temporal qualities to correlations between drug use and any other particular variable (including demographic or personality features or even drug effects) can be, as many have pointed out, (e.g. 10, 19, 20) a spurious business. This is not to say, of course, that correlates are not of interest; only that they are to be treated with due respect to their unknown origins. Considering first demographic and background features, a number of studies have investigated association between use and socioeconomic background. Disregarding such niceties as concern for similarity between measuring techniques, three studies have found broad associations between likelihood of having used and higher class origins (4, 9, 11) and three have found no significant differences between users and non-users in this regard (6, 10a, 15). One study (Young, (9)) has placed considerable theoretical importance on finding (uniquely) an association between professional rather than managerial background and likelihood of use amongst social science students. Whilst not at this stage commenting on theory, we are obliged as empiricists at least to note the representativeness and replicability of findings (see Table 9.1, and Appendix 1) in order better to assess any theoretical assertions based on them. A further area of demographic interest has been sex differences. Again findings vary and it seems that where differences are found they are slight, and in terms of a preponderence of males amongst users (4, 11, 13, 15). Other studies have found no sex differences (9, 10a). It may be that sexual differences balance out as cannabis use moves from being primarily of an innovatory nature to a social one. One study (10a) enquired after birth order and found users less likely to be first born children. There are conflicting findings on urban vs. rural background, with one study (6) finding fewer users than non-users came from rural backgrounds, another (10a) finding no such differences. As regards type of schooling one study (6) found no differences in school type, another (10a) found a tendency for users to have had more diverse schooling (i.e. changed school more often), and if anything to be more likely to have been to public schools than non-users. Studies that investigated the students career from school to college (10a, 6) found that users tended to have had a longer time gap between leaving school and starting college. According to three studies (9, 10, 15) users were also more likely to have lived away from home. Considering home backgrounds and family relationships, more studies found a higher incidence of 'distress' of some sort in family history or family relations of users (3, 10a, 11,

15) than no such association (6, 9), but the relevant percentage differences are generally small. In the main this has been investigated by enquiring into whether there had been any sort of break in parental relationship due to death or separation (3, 6, 10a, 15), otherwise through questions directly on family relations (9, 10a, 11). Most studies have included some questions on current attitudinal behavioural and personality correlates, and although in the main these are not directly comparable a few common features are apparent. Those studies that have enquired into political attitudes (4, 9, 10a) found users overrepresented amongst those with 'left wing' and 'fringe' politics. Two studies (9, 10a) found in addition that users were more likely to be active politically (or in favour of it). One of these studies (9) further found that whilst users were consistently more likely to be in favour of non-violent political action (e.g. sit-ins), only social science users were significantly more in favour of violent political action. Turning to religious views, again all studies investigating this concur on finding users very much less likely to hold current beliefs or more especially be currently active in any religion (4, 9, 10a). Considering general social attitudes, values, and behaviour users have been found to be more hedonistic and likely to have opposite sex relationships (7, 9, 10a) and more active socially (10a, 11). They tend to be more likely to shun what might be called the 'puritan ethic', (10a), and express greater dissatisfaction with their course of study and anticipated role in society (9, 10a). One of these studies (9) found users more likely to hold what were termed 'expressive ideals' (subject interest, self understanding, good social life) than 'instrumental ideals' (skills to obtain a higher income and job security). Users were also said to be frustrated in their expressive ideals at college. Other studies (10a, 11) found association between cannabis use and general liberal, permissive 'non-conformist' social attitudes. Briefly anticipating Dr Somekh's report it is relevant to note here that studies which investigated use of alcohol and tobacco (the use of which might be considered partly under a hedonistic umbrella) reported some association with greater use of these drugs and likelihood of cannabis use. There is very little comparability between the studies in terms of any personality correlates measured. One study (3) found SO per cent of 'cases' were 'aware of prolonged depressive feelings' and the same author later considered any dependency problem to be secondary to a range of emotional disturbances including depression. Another study of a specialised sample of psychiatric referrals (17) found that none of the students had attended because of problems associated with their cannabis use and that the only difference in diagnoses between users and non-users was a total lack of manic-depressives amongst users. In our study we found users tended to have higher 'N' and 'E' scores on the EPI, to have had more contact with psychiatric services in the past, and to express less 'purpose in life' as measured by Frankl's test. As mentioned earlier other studies have also cited more social activity amongst users. OVERVIEW AND INDICATIONS OF 'LIFE STYLES' To attempt any synthetic overview of this messy array of correlates runs the risk of that 'overbrave assertion' mentioned earlier. Whenever there is conflicting data, especially from sources of widely differing reliability using widely different measuring techniques, there is a temptation to ignore or belittle that which doesn't fit. To minimize the risk of doing this

without due cause, I shall firstly summarize the existing data (with source specified) under two headings: conflicting and congruent (see Table 9.1). I shall then rely in the main on data from the two recent studies of adequate response rate, clearly specified (and to a degree, heterogeneous) samples, which have attempted to systematically report on variables relevant to consideration of 'life-styles' (9, 10a). However, the findings of other studies will be clearly available for due comparison where there is overlap of relevant data.

Personality. As mentioned earlier there are very few comparable findings. Some indications of higher neurosis and extraversion (10a), less 'purpose in life' and optimism in conventional future (9,10a). Unclear findings on depression (3,17). As can be seen from Table 9.1 it is the 'harder' demographic and background data that reveal most discrepancies in findings. This is perhaps because a greater variety of studies have included these variables in their investigation. There is considerable congruence on the attitudinal and behavioural data, although of course measurement techniques differed. Further isolated findings (i.e. those that have been investigated by one study) are not included in the table but may be mentioned below. It seems certainly to be the case that as a generalization, English students who have had experience of cannabis differ as a group from those with no such experience, in terms of having generally more hedonistic, non-conformist, non-religious and 'left' political values and styles of behaviour. They also appear to be if anything, more emotionally disturbed and frustrated with their lot. Analysis of these variables by frequency or heaviness of use would facilitate differentiation of users according to differing balance of these features in their life styles. Meanwhile approximately one third of students in the 'higher-rate' colleges have had experience of cannabis. Until further multi-factorial analyses have been completed, investigation of cannabis use and its correlates is more profitably regarded as investigation of differing patterns of 'life styles' amongst students generally. Cannabis use may certainly play some role amongst some of these, but there is no reason to assume ad hoc that it is of central, or for that matter, peripheral, importance. It may be of interest not so much as a dependent variable, but as a good predictor of other values and behaviours which in turn, may become the 'interesting phenomenon'. Both studies under consideration (9, 10a) found that use ever was associated consistently with the set of attitudes and behaviours mentioned above. Young termed this a 'bohemian value system'. We: 'hedonism', 'disaffection with societal norms' and 'disaffection with the political status quo'. In addition both studies found indications of frustrations with course of study and anticipated role in society. Young termed this rejecting instrumental values and being frustrated in expressive ideals. We noted less achieved and anticipated 'pay-off' from conformity. It has to be emphasised of course, that these generalisations are rather like 'ideal types' of students, as percentage differences between the groups were not often large. Young has interpreted his findings in terms of a sub-culture theory emanating from Merton and Matza. In brief he argues that all deviant behaviour has purpose in terms of resolving problems, in particular unachieved aspirations. Those students who have rejected 'instrumental values' from 'official morality' (such as discipline, deferred gratification, caution and reliability) might develop an alternative 'bohemian' value system characterised by more 'expressive', 'subterranean' values such as self expressivity, hedonism, autonomy and impulsivity. Such a value system might well incorporate in the pursuit of its goals the use of cannabis or other drugs that enhance its values of pleasure and expressivity. That having had experience of cannabis use and holding these sorts of values are broadly associated is not in dispute. There may be more disagreement when considering the importance of individual differences,

how students may come to hold such values, and what other variables might be relevant to understanding student cannabis use. According to Young, his findings suggest a link between student bohemianism (and therefore cannabis use) and a non-commercial middle class background (i.e. 'professional' value than 'managerial'). He argues that this class has always been critical of 'pecuniary values and instrumentality of the managerial bourgeois class', and that 'it is a short step to conclude that their children....should develop and transform their parents' values' (9). A quick glance at Appendix 1 shows that out of 6 studies only his own found an association between professional rather than managerial class backgrounds (3 finding no class association at all). In considering a number of background and developmental variables we found tentative indications that perhaps users had experienced less 'close or unique association with traditional transmitters of societal morality'. (See Appendix 1 for ingredients). But the findings are again by no means unequivocal - disrupted family background is one variable to be found in the 'conflicting' side of Table 9.1. Young later argues that 'the emergence of marijuana smoking is not a function of family pathology but a bohemian sub-culture which is itself a solution to certain problems of thwarted aspirations faced by students'. I would argue that correlations between variables can be viewed from different angles. Marriages didn't stop breaking up (with varying effect) when subcultures came on the field. Social injustice didn't cease when someone found the meaning of life in a monastery. In an area of so little data, and so much of it conflicting, there is no intellectual need to rule that any one perspective holds the total truth, and that others are irrelevant or misleading. There is a danger of 'the phenomenon' getting treated as a homogeneous thing. In early days of heroin research those searching for a 'typical junkie' got due scoffing. It is boring to be reminded again, whatever one's discipline, that people with experience of cannabis cover a wide spectrum. There isn't going to be an 'explanation' of 'the phenomenon' until it is very much better differentiated within itself. It might be that understanding the development of values is crucial for understanding certain sorts of users, and irrelevant for others. The need for self expression might have to jostle with other 'needs' from curiosity and affiliation to consistency and identity. Individual differences are important. Factors other than values and thwarted aspirations may be important. If those who, for example, Young says, are less likely to be frustrated in their expressive ideals (sociologists and militants) still rank high amongst cannabis smokers, further levels of explanation are required. It might be that associations found in our study between use and current personality characteristics are spurious, or emanate from a particular subgroup, or hold generally. It is not easy to remain coldly academic when social action is involved, as with cannabis. It is not 'right' for 'innocent people' to suffer unnecessarily and unwillingly either 'in prison' or from 'adverse drug effects'. We are all presumably eager to minimize injustice, and have to accept that our conception of it differs and that it probably affects our 'scientific' conviction in varying degrees. To return to the original theme of this paper, we can only try to be aware of the danger of our own tendency to chauvinism, and avoid trying to squeeze the scanty data available into a single model. We can do without another variant on the nature-nurture type of controversy -

in this case the individual blown helplessly about either by his own personal pathology or a world of structural pressures. REFERENCES 1. Linken, A. (1964). In Psychological Aspects of Drug Taking. Proceedings of a one-day conference at University College, Sept., 1964, summarized by Sington, D. Pergamon Press, 1965. 2. Bestic, A. ( 1968). Reported in Turn me on Man, (Ch.8), Tandem Books, London, 1966. 3. Linken, A. (1968). In Adolescent Drug Dependence (Ed.) Wilson, 'The psychosocial aspects of student drug taking'. Pergamon Press, Oxford, 1968. 4. Binnie, H.L. and Murdoch (1969). The attitudes to drugs and drug takers of students at the University and Colleges of Higher Education in an English Midland City. University of Leicester Vaughan Papers, No.14. 5. Linken, A., and Lucas, C.J. (1969). Drug taking among students. Journal of the Medical Women's Federation, October, 1969. 6. Hindmarch, I. (1970). Patterns of drug use in a provincial University. Brit. J. Addict., 1970, 64, pp. 395-402. 7. Weiner, R.S.P. (1970). Drugs and schoolchildren. Longman, 1970. 8. Hindmarch, I. (1972). The patterns of drug abuse amongst school children. Bull. Narcotics, XXIV, No. 3. 9. Young, J. and Crutchley, J.B. et al (1972). Student Drug Use in Drugs in Society, October, 1972. Also 1st and 2nd Annual Reports of Student Culture Project, Aug., 1971 and Nov., 1972. 10. Kosviner, A., Hawks, D.V. and Webb, M.G.T. (1972). 10a. Cannabis use amongst British University Students: I Prevalence rates and differences between students who have tried cannabis and those who have never tried it. Brit. J. Addict., 1973, 69, pp. 35-60. 11. Durham Drugs Liaison Committee (1972). Working Party Preliminary Report. An Enquiry into drug taking carried out among a group of young people in Durham County, July, 1972. 12. Wigfield, W.J. (1972 ). Survey of young people's attitude to drug abuse. Community Medicine, Nov., 1972. 13. Somekh, D. ( 1973). Prevalence of self-reported drug use among London undergraduates. Unpublished report, 1973. 14. Fish, F. and Wells, B. (1973). Reported in Times Educational Supplement. Drugs survey leads to a rethink. April, 1973. Also report in press for Brit. J. Addict. 15. McKay, A.J., Hawthorne, V.M. and McCartney, H.N. (1973). Drug taking among medical students at Glasgow University. Brit. Med. Jour., 1973, 1, p.540. 16. Bestic, A. (1973). Your child and drugs. Woman's Own, Nov., 1973. 17. Duddle, M. (1973). Drug taking amongst emotionally disturbed university students. Brit. J. Addict., 68, 1973, pp.166-169. 18. Kosviner, A. and Hawks, D.V. (1974). Cannabis use amongst British University students: II Patterns of use and attitudes to use amongst users. Draft report, 1974. 19. Blumberg, H.H. Drug taking among children and adolescents. Ch.23, in Recent Developments in Child Psychiatry, M. Rutter and L. Hersov (Eds.). In press, Oxford: Blackwell. 20. Young, J. (1973). Student drug use and middle-class delinquency. In Contemporary Social Problems in Britain. Bailey, R. and Young, S. Lexington Books, 1973.

10. Drug Use in the Netherlands

E. Leuw, Foundation for the Study of Alcohol and Drugs, Amsterdam. INTRODUCTION This survey is intended as a summary of the principal results of the socio-scientific research that has taken place in the last five years in the Netherlands into drug use and some of its backgrounds. This research was in general oriented towards the collection of information on the following points: 1. The nature and extent of drug use, patterns of drug use 2. The sociological and psychological characteristics of drug users 3. The existing knowledge of and attitudes towards drug use. The data coming under the first point give an impression of the distribution of drug use in the Netherlands. More than incidental data on the distribution of the use of drugs exist only on pupils of continued education, and then more specifically of general secondary and higher education. The research in question shows that among the older pupils within this population approximately 20 per cent have some experience with drugs. About half of these are pupils who use drugs experimentally only. Among the drugs used, cannabis occupies by far the most important place. Only about one quarter of the 'users' have ever used something else than cannabis occasionally. In most cases this then relates to the use of pep pills or LSD. Regular use of these drugs occurs only very slightly among the population in question. There is very little information on the distribution of drug use in other populations. Buikhuisen et al. found over 7 per cent drug users among all sixteen to twenty-two year olds in Gieten ( North Drenthe). Here too approx. 75 per cent of the users confined themselves to cannabis. In a student sample it was found that 17 per cent had used cannabis on some occasion. Finally, it was found in a general population of Dutch persons aged 16 years and

over that more than 2.5 per cent had some experience with drugs. As regards the second point, it may be remarked that practically all the relevant research shows that drug use typically occurs in a social context, that it is done together with others. This applies very strongly to cannabis and to a somewhat lesser extent to LSD and opiates. A second general datum is that the use is almost exclusively confined to people below the age of 30. In the relatively older research (up to ca. 1970) it is found that drug use is associated with the male sex, higher socio-economic environment, higher degree of urbanisation and absence of religious affiliation. In the more recent research these relations prove in general to disappear or in any case to decrease. It further emerges clearly from the available research that on many other points too the drug user differs from the current life style. The drug user is often more active and tolerant in respect of all kinds of sex, has a more critical, often leftish attitude towards society, has a greater cultural and intellectual interest, is less achievement-oriented. The (scanty) individual psychological data are consistent with the above differences. Drug users are in general more 'neurotic' and more likely to be in a conflict situation in respect of their parents and education than non-users. However, for all this there are no indications that there is any question of an individual or social pathology among drug users in general. Only a very small minority of drug users cannot maintain themselves physically, mentally or socially. Finally, it should be remarked that so far there are no research data on causality between drug use and the related characteristics of the user. In other words, there is no answer to the question whether a person uses drugs because he is less achievement-oriented or whether he is less achievement-oriented because he uses drugs. The third important point on which recent research has yielded a number of data is the question of the knowledge and attitudes that exist with regard to drugs and drug users. Familiarity with the phenomenon of drugs has increased very strikingly in the last five years. In both adult and juvenile populations there is an increase in the percentage having some familiarity with the phenomenon from approx. 50 to 90 per cent. The great majority of both juveniles and adults are of the opinion that all drugs without exception are dangerous agents to which one becomes addicted and which have an adverse effect on physical and mental health. These ideas also exist with regard to cannabis. This agent is ranked immediately after the opiates and LSD with regard to its harmfulness. It is striking that dangerous agents such as amphetamines, alcohol and barbiturates are regarded as less harmful than cannabis. In the population which is considered to be less opposed, relatively speaking, to the use of cannabis (secondary schoolchildren) 80 to 90 per cent state that they would not even like to experiment with cannabis. As would be expected, drug users have a much more positive attitude to drugs than non-users.

The relatively strong disapproval of drug use is also evident from the opinions that are held about drug users. Drug users are viewed in general as problematic persons and are judged in relatively negative terms. They are regarded as weak, poor achievers and people who are inclined to run away from their problems. Drug users are nearly always assigned negative motivations for their use. These negative judgements of users are more pronounced in adult than in juvenile populations. In the adult population somewhat less than half feel that the use of drugs should be severely punished. Within this population a majority is opposed to the legislation of cannabis; in juvenile populations this is the case only with a minority. Finally, it is striking that in circles involved in drug policy (politicians, people from health care, education and youth work) a much more tolerant attitude exists towards drugs and drug use. In particular a fundamental distinction is made in these circles between cannabis and other drugs and there is an increasing inclination to accept the use of cannabis or in any case to make it a non-punishable offence. REFERENCE This presentation consists of the introduction to a full review of drug use, attitudes of young people and public opinion: Foundation for Alcohol and Drug Research (1972), Drug Use in the Netherlands: a survey of scientific investigations in this field. De Lairessestraat 39, Amsterdam.

11. Social Attitudes of Users

K. Berntsen, Youth Clinic, Copenhagen. In a British TV-play named 'Bartleby', the principal character is a young man who has got a job as a bookkeeper in a small firm. Whenever the manager asks the young man to come in and discuss things etc., he answers 'I prefer not to' or 'I would rather not'. I wonder why this sentence 'I prefer not to' comes into my hear all the times when at home in Copenhagen I started to think of what I should like to point out here concerning the topic of social attitudes and life style of cannabis users. One of the reasons for this 'I prefer not to' may be found especially in one of the topics which I am expected to cover in my paper: 'A discussion of the social attitudes manifested by cannabis users including the extent to which an identifiable life-style can be designated'. My point is that maybe it is a wrong way to set up the problem. We take it as a fact that the cannabis user has special social attitudes and a special life-style. We put the emphasis on the cannabis use. The answer to this is of course that when we meet to a cannabis conference this is the main topic, cannabis is our starting point. I myself followed the line when I wrote my very short draft notes in which I said that marihuana and hashish are still mostly used by young people who are in opposition to the welfare society, against the ECC, who are trying alternative ways of life, etc. If we look at the many researches in the different countries, which during the last 5 to 10 years have tried to find out the numbers of youngsters who have been offered, tried, regularly used cannabis or hashish, the results show that up till now are that they account for an

increasing percentage of the investigated populations. But it is important to bear in mind that the large majority has only been trying or experimenting with marihuana. From my point of view the development in the future concerning this problem can go two ways. The first one, and there is every probability that this will be the result, is that the numbers of individuals who have tried marihuana will increase. If so, that means that more and more so-called 'normal' people will belong to the group so that the differences which are still found between the groups 'ever' and 'never' tried will disappear completely, and the factor 'marihuana' cannot be used as an indicator. It seems perhaps stupid to mention that the other possibility is that the numbers of users decrease in the future. The reason for mentioning it is that this will only happen in my country at least if we continue with our liberal attitude towards cannabis. It must be caused by a rejection of the drug by the children and the youngsters themselves because they find other drugs more interesting for instance alcohol, or make a movement against any drugs at all. To-day there are in Denmark a large number of people, and not only young people, who believe that it is legal to possess and smoke, eat or drink marihuana and hashish. This is rather interesting, especially when one looks upon some legal facts about cannabis, and it shows us at the same time a double-sided moral principle. In a special law about narcotic drugs, the drugs are grouped in two categories. One category must not be allowed in the country because they present special danger due to their intoxicating qualities. Other drugs must be allowed, but only used for medical and scientific purposes. In the first mentioned group one finds heroin, raw opium and cannabis. It is a paradox that a drug like hashish, which is the easiest available and cheapest drug, and - as time goes on - the drug that is considered the least dangerous drug, is from a legal point of view considered so dangerous it is not allowed to be in the country. Bat the confusion becomes greater when we look at the instructions given by the general attorney concerning the work of the police: 'In general there will be no reason to start an investigation concerning liability to punishment for possession when the suspicion solely relates to own consumption of the drug. The possession of hashish or marihuana in small quantities will only be followed by warnings. The consequence of this strategy is, as mentioned before, that many people consider consumption of hashish and marihuana as legal. The fact is that consumption of this and of other drugs is decriminalized which, from my point of view, is a sensible and good thing. We can compare and contrast the situation in Denmark with that in other countries. Whilst Holland adopts a similar policy, some countries have higher penalties. For instance, in Egypt capital punishment is the maximum penalty for smuggling. Going back to the special topic: social attitudes and life-styles of users, I will make the following comments: As mentioned, I think that we will find less and less special extreme kinds of life-styles and social attitudes among the future users of cannabis products. The smoking will not, to the

same extent as previously, be used as a means to oppose against adult authority, in order to demonstrate independency, or to revolt. It has already its own norms, its own ways to be taken, together in groups, in week-ends, as positive elements when listening to music, etc. One also sees that for certain active political youth groups the hashish smoking is decreasing because it makes the participants more passive and disinterested in the political work. But all that I have said up till now does not mean that I think that there are special problems which have relation to the use of marihuana and hashish. And I shall mention two categories of youngsters, whom I know from my daily work with drug addicts and juveniles, with other problems. The first category is that of the drug addicts in treatment. Even when they stop injecting, they may often continue with very heavy use of non-opiates, such as cigarettes, coffee and alcohol, which are all legal, and cannabis. All drugs can be used and can be misused, and it is important to see how heavy users of any drug function. The second category is that of the 'dropout' who takes no part in the educational system or in the normal social life, and prefers to live on a minimum of economic resources. They may smoke cannabis excessively. But cannabis is only one factor among many others, and I believe that the other factors, social and psychological, are of much greater importance. In conclusion, I feel that we should continue to investigate the potential of cannabis and of other drugs for physical and biological harm. In relation to social and psychological research: 1. I prefer to give up talking about the cannabis user as a special sort of person. 2. I prefer to find out how we could influence the attitudes of legislators and of the people who still believe that the drug problem can be solved by legal measures. Just look at the United States. 185 3. I prefer to believe that we will look at all the other more important factors for human wellbeing, such as socio-economic and psychological factors which are so obviously essential. DISCUSSION OF PAPERS BY MS BERNTSEN, MS KOSVINER AND DR LEUW THE CHANGING LEVELS OF USE There was discussion about the rate of growth of cannabis use, and of the very limited evidence available to say whether cannabis use was in fact still increasing. Dr Hawks pointed out that 'in those areas where there is much higher prevalence there may be a plateau effect. They do go up to a point, but whether or not surveys in two points of time show a rise depends very much on the time the comparisons are made.' Mr Hastleton suggested that church attendance is an absolute bar from experimentation with cannabis. One can set up limits for consumption by looking at the distribution of conservatism in the culture. He also suggested that one should be suspicious of research reports of

percentages of ever-users giving up use; admitting to ever-use would seem less dangerous to the respondent from admitting to current use. Dr Miles reported that his longitudinal data indicated a tendency among some heavy users to give up for a while, to try Zen and other non-drug activities, and then to go back to cannabis but to use it as a more moderate rate. Such changes have implications for 'plateau' estimates. Mr Dorn suggested that if one asked teenagers with nicotine experience whether they were current nicotine users, many of those who in fact will go on to become regular users claim to have 'given up' at that time. The implication for surveys of use of marijuana is that asking whether the respondent currently used or had given up is inadequate as a measure of future use. The National Commission on Marijuana's use of data in this way was questionable. 'What you have to ask people is not whether they regard themselves as current users, but whether they have a positive intention of refusing any future offer of cannabis.' NORMALISATION OF CANNABIS USE Dr Leuw saw that the data presented by Kosviner was in line with research done in Holland. In Holland, cannabis use was no longer deviant, and sex, religion and a degree of urbanisation no longer clearly differentiated user from non-user. Nor did lifestyle or attitudinal attributes, and nor was it any longer true that cannabis had a central or core function for social groups. 'In terms of the subculture, cannabis use has been vulgarised. Once straight people began to smoke cannabis, it wasn't something for the sub-culture to be proud of. Heavy use may still be a sub-cultural thing, but it is less of a core element in it'. Ms Kosviner agreed that it is no longer central as an activity, and so it is no longer possible to try to explain it as a unitary phenomenon. Dr Miles said that another change was not so much in the phenomenon of cannabis use, but in the attitude of the scientific community to cannabis research: two years ago, Dr Leuw had been interrupted in the middle of giving a paper at a Conference at Amsterdam by the Chairman, but today the scientific establishment was able to accept such research results. Ms Mott suggested that there was evidence from other UK surveys to suggest that there is a social class factor in cannabis (more likely in middle-class) and that this would not show up clearly in college populations because of their relatively homogenous social class compositions. Dr Rubin pointed out that in Jamaica it is the non-cannabis user who is deviant. Whilst cannabis use may be or have been part of a general constellation of nonconforming behaviour in the USA and UK, this is not the case in other cultures. Mr Hasleton reported that Australian surveys of urban universities also find ever-usership around 30 per cent. In Australia, cannabis is freely available, and he suggested that availability is not a major limiting factor in levels of consumption. Dr Hindmarch, quoting a 1972 study of a school population stressed that the economics of drug use, such as the disappearance of small 'deals' of cannabis, led to some users giving it up. Mr Hasleton and Dr Leuw asked Ms Berntsen about her clinic staff's use of cannabis. Ms Berntsen suggested that the staff coming into work in the drug agencies were now less interested in cannabis than a few years ago. They might smoke as much, but there was less

excitement about it. Mr Berntsen mentioned that as cannabis smoking became more common in places of tertiary education, especially in social science students, so the numbers of young social scientists starting work who smoke cannabis must rise. Although some of her staff might smoke cannabis, and although the clients might also, her staff were forbidden to smoke cannabis with the clients. Dr Edwards suggested that looking at the evidence on alcohol, cigarettes and cannabis use, one could speculate that different variables might best explain different parts of the consumption curve. Sociological explanations are useful for establishing the meanings of experimentation and light use, whilst psychological explanations may have a higher power for relatively heavy use irrespective of time or culture. FUTURE SURVEYS According to Dr Cameron 'What we're ultimately interested in is the extent of problems associated with drug-using behaviour and trying to reduce the seriousness and extent of those problems. Therefore we need incidence figures, figures of heavy use, and figures of problem use, through time with different cohorts'. In other words, longitudinal samples with measures of level and consequence of use will in future be more valuable than cross-sectional surveys at one point in time.

12. Cannabis Permissivism' and Social Response in Australia

Simon Hasleton, Department of Psychology, University of Sydney, N.S.W., Australia. It is worth reminding ourselves that cannabis use is a meaningful activity. By this, I mean only that its non-drug effects are more important than the strictly pharmacologic effect of the active elements. The evidence is that in most cases the dose taken in the naturalistic setting in 'Western' societies is such that the effects are evanescent, subjective, and highly sensitive to set and mood and expectation. (1) Indeed, there is every indication that the user not only requires an extensive coaching in recognising and responding to the effects of the drug, but that he is also capable of 'switching off' the intoxication should he need to do so. (2) Nevertheless, this activity has generated an enormous response. It has been shown that in California alone, in one year, (1968) 51,000 persons were arrested for marihuana offences, at a cost to to the State of some 72 million dollars. (3) The drug has generated in the last few years, two major government sponsored reports, an embarrassing volume of other research, and not least, this series of conferences. This response could not possibly be explained in terms of the pharmacology of cannabis as a toxin, or in terms of its nuisance value as an elective intoxicant. The fact that the recommendations of the various Committees of investigation have been ignored, and that cannabis use continues to be prosecuted with the most draconian penalties and to be classified with the narcotics in many cases, indicates that the response to the use of this substance is not the rational outcome of its denotations. This is not to imply that the response is irrational but rather that it has been generated by connotative aspects other than those which are investigated in ordinary scientific enquiry. In this paper, I wish to report two studies (4) which have explored some aspects of the 'meaning'

of cannabis use, and the social response which this activity has generated. Both studies refer to Australia, a national community which derives some of its mores from Britain, some from Catholic Ireland, some from the United States West Coast, and many from its own recent experiences as the recipient of immigrants from most countries in Europe and many other countries beside. These studies are unique, I think, in that they are based on large stratified random samples of the electorate. The use of such samples permits generalisation to the electorate as the medium of orderly social change, and reflects the belief that cannabis related activity is appropriately discussed in a political or macro-psychological arena. The studies were designed to locate the cannabis issue within the context of other social issues, to examine the structure and complexity of the issues, the support they command, both absolutely and relative to each other, and the social and biographic correlates of that support. Within that framework, cannabis affiliations and the response they generate have been examined in more detail, in order to ascertain whether they comprise a special case, or whether they can be accomodated within the model. METHOD Australian National Polls Pty. Ltd. were commissioned to carry out two separate surveys on samples of the electorate in May and August 1973. A.N.O.P.'s sampling frame is based on a stratified random probability design, in which respondents are selected at random from the Commonwealth Electoral Roll, within electoral divisions which are selected on the basis of a series of social and political indicators. In each capital city, electoral divisions are ranked on a four level SES description and a three level political stability rating. Electorates are then selected in terms of each division's relative population proportion in the 4 X 3 combined levels of these two strata, and following further substratification at the subdivisional level, the random sampling of electors takes place in terms of enrolled electoral populations. Sampling takes place in the whole Commonwealth, with the exception of the Northern Territory and the Australian Capital Territory. In each case the interviewer visited the respondent in his home, the statements being presented verbally and the responses being noted in the presence of the respondent. 1881 subjects were interviewed in one study and 1883 in the other. In order to locate 'marihuana' with other social issues, a bracket of items was selected so as to be representative of those commonly discussed in the context of 'permissiveness' in Australia. The statements presented to the sample are set out below: 1. Do you agree or disagree that people over the age of 16 should be able to buy contraceptives? 2. ... that homosexuality between consenting adults should be legalised? ... 3. .., that censorship of books and films should be tightened up ... 4. ... that a woman should be free to have an abortion in the first three months of pregnancy without risking prosecution ... 5. ... that the community should frown on unmarried couples living together ... 6. ... that the smoking of marihuana should be legalised ... 7. ... that prostitution should be legalised ... 8. ... that divorce on request should be available to couples without children ...

9. ... that hotels should be closed on Sundays ... 10. that there should be more sex education in schools ... Responses were scored as follows, after inversion of items three, five and nine:

Acquiescence is recognised as a problem in studies of this sort. In this case, the three negatively cast items were combined with three positively cast items having a similar factorial composition, and a measure of acquiescence generated. This measure proved to have a very low correlation with the summed scores on the items (0.03) and with the score on the first principal component (0.07), and it was concluded that although acquiescence was present to a statistically significant degree, the findings were not substantially affected. The first analysis of this material was designed to examine whether these items are related in such a manner as to allow discussion of 'permissiveness' as an entity. Product moment correlations were calculated between the items, and a principal component analysis performed for the total sample and a number of subgroups based on age and education. For the total sample, the proportion of the variance extracted by successive components was 30.14, 11.16, 9.45, 8.70, (etc). Factor scores calculated on the first principal component correlate 0.996 with the summed raw scores on the items: the generalized split-half reliability coefficient, (alpha) a measure of consistency, ranged from 0.781 for the tertiary educated to 0.697 for those with less than three years secondary education. Table 12.1 sets out the results of this analysis. (Item mnemonics will be self-explanatory.)

NOTE: Decimal points omitted from component loadings and cases with incomplete data suppressed. It may be concluded that this set of items is related in such a way that it is meaningful to speak of 'permissiveness' as a unitary entity, and furthermore, that the tendency to conceptualise the items as having 'something in common' is higher among the tertiary educated than among those with less formal education. It should be noted that the marihuana issue is embedded in the general factor and is thus not considered as an issue separate from the general questions of personal liberty or licence which are raised by these items. Permissiveness can now be defined as that dimension which best accounts for the common variance in this set of items, in other words the first principal component, and scores on this component may be utilized as a measure of permissiveness. Subsequently, the relative difficulty and scalability of the items was tested entering a Guttmann analysis with a cutting point set above the 'don't know' score of 3. Table 12.2 sets out the results of this analysis. This material raises a number of points for discussion. Among them are the generally high level of approval for permissive issues in Australia, and the way in which marihuana legalization is clearly of a higher order of item difficulty than the other items for all except the best educated and the 18 to 24 age group. Nevertheless, support for this item is fairly substantial. Utilising a recursive multiple regression technique, six variables were identified as predictors of permissiveness. These achieved a multiple R of 0.550 for the whole sample, (F = 135.46, df = 6, 1862), and 0.583 for the 18 to 24 age group (F = 24.12, df = 6, 281). Four of these variables, church attendance, age, education, and Left v. Right voting intention accounted for 96 per cent of the predicted variance in the whole sample. To test for the

presence of interaction effects, the summed raw item scores were now entered into a four way analysis of variance, classifying on the four predictors. The four main effects were significant beyond the 0.001 level, and three of the two-way interactions (age x church, education x church, and church x voting intent) were significant beyond the 0.01 level. The significance of the age by education interaction was obscured by the way the age groups were combined to maintain cell numbers, but emerges with some clarity in the graphical presentation. (5)

We do not know whether these trends are the result of cohort differences which have remained fairly stable over the years, or whether changes occur within individuals over their life span. Material is available to the writer which will throw some light on this, but in the meanwhile it seems probable that both explanations are valid. Of direct relevance, it might be noted that there are two fairly clear patterns here. One is shared by the tertiary educated and those who have more than three years secondary education, but no secondary qualifications. The other is shared by those with secondary qualifications and those with less than three years secondary education.

The explanation of these trends is outside the scope of this paper, but the steep rise in antipermissivism ('Wowserism' in Australia) in the most permissive group, has especial relevance for marihuana when it is recalled that this is the most extreme item in the 'scale'. In all probability, this steep rise is a result of rapid exposure to the competitive world of business and the professions, and the demands of child rearing; one notes a downturn at age 40, when ambition starts to resolve and child rearing responsibilities to decrease. Studies currently in hand will enable a comparison between the present investigation and one conducted some years earlier in another form in Australia, and between the present study and a parallel investigation in the U.K., and these should enable a clarification of these questions and the

underlying issue of the total rate of change in permissiveness, if indeed such change has occurred. 'Permissiveness' concerns a number of aspects of contemporary life. For example, patterns of drug use (including alcohol use) reflect societal norms which might well be discussed under this heading. It can be shown that opinions about marihuana held by the 18 to 24 age group are substantially predictable on the basis of opinions about homosexuality, contraception, censorship, abortion, prostitution and divorce. (R = 0.381, F = 7.77, df = 6,275). When the summed scores on the nine non-marihuana items are added to the biographic indicators marital status, sex, urban residence, political intent, and even in this restricted sample , age, R is increased to 0.454 (F = 11.91). The predictors are cited in the order of their importance. In the latter example the attitudinal predictor accounts for more than half of the predicted variance. There is, however , a significant difference within the groups based on age and education in respect of the relationship between marihuana and general permissiveness. In the whole sample, the loading on the general factor is 0.500, indicating that 25 per cent of the variance in marihuana sentiments is explicable in these terms. In addition, there is a substantial loading (0.683) on a component specific to marihuana indicating that some 46 per cent of the variance is not explicable as 'permissiveness'. This situation is not the case with those educated to the secondary qualified or tertiary level, where 32 per cent of the variance in marihuana sentiments is explicable as permissiveness and there is no component specific to marihuana. On the other hand, the 'separateness' of marihuana sentiments is increased in the numerically larger less well educated group - who are also more 'anti-pot'. Here only 18 per cent of the variance is explicable as permissiveness. Where groups are defined by age, the situation is more complex, but for those over 45, who account for some 45 per cent of the electorate (as represented in this sample), only 16 per cent of the variance is explicable as permissiveness, and 32 per cent as specific. Thus one problem of marihuana control may be explicable as follows: in the least permissive, more anti-pot groups, the older and less well educated, marihuana is conceived as relatively separate from the 'permissiveness' issue. The pro-pot lobby finds few allies among the more permissive of this group (who are not very permissive anyway). The better educated and the younger are not only more permissive and more pro-pot but these two characteristics also share the same meaning. This study has demonstrated that approval of marihuana legalization is a rather complex function of a more general factor, social permissiveness. The raw scores on the items which load on this factor have a platykurtic symetric distribution in the population, unimodal with a mean of 28.05 and a standard deviation of 8.48, indicating that the population is distributed around a mean of indecision on this dimension: the 'silent majority' simply does not exist, but will be referred to further below. This indecision is reflected in a more detailed examination of marihuana affiliations and the social response they generate. Again in the context of this being the most difficult item on a permissiveness scale, a comparison of the various surveys commissioned on this topic since

1970 indicates a steady increase in the proportion of the population approving of legalization. In the general population from 7.4 to 17.3 per cent and in the under 30's, from 10.4 to 26.6 per cent. Currently, 50.7 per cent of the 18 to 20 age group state that they know one or more people who use marihuana and 33.2 per cent of the 18 to 20's believe that marihuana use should be legalized. For the total under 35 group 33.2 per cent know one or more users: promarihuana sentiments cannot be considered statistically deviant in the young population of Australia. It therefore comes as no surprise that support for police action in this area is weak. Respondents were presented with a vignette in which they were asked about penalties in the case of an occasional user apprehended with a small amount of the drug in his possession. 70.1 per cent of the sample would impose a penalty, 26.5 per cent would not impose a penalty, and the balance are unsure. The percentages in favour of penalties are fairly consistent across the States, with Western Australia highest at 80.9 per cent, and New South Wales, the most populous state, lowest at 67.2 per cent. Support for penalties, and for more severe penalties, is clearly related to the discriminators which have been discussed previously: education, age, church attendance, and political affiliation. This is illustrated in Table 12.3.

Examining the congruence of current penalties being imposed in New South Wales courts for possession of a small quantity of marihuana, with current public opinion regarding penalties, responses of the New South Wales sample were compared with the penalties imposed on first offenders for Use or Possession in New South Wales courts of Petty Sessions in 1972. In that year 38.4 per cent were discharged without penalty, or were placed on probation, 11.4 per cent were fined less than $100, 22 per cent were fined between $100 and $199, 24.5 per cent were fined more than $200, and prison sentences were imposed in 2 per cent of the cases. This dichotomy between leniency and severity reflects the balance of public opinion in the State. 43.2 per cent of the N.S.W . sample approve of no penalty or a nominal $10 fine, 23.4 per cent approve a $100 fine, 11.5 per cent approve a $500 find, 10.5 per cent a short period of imprisonment, and 3.3 per cent a long period of imprisonment. There are few 'victims' in marihuana offenses, and the police are reliant to a large extent on information from the public. However, only about 30 per cent of the sample state that they would inform the police or other authorities if they became aware that a group of young adults

occasionally used the drug, as a against 81 per cent who state that they would take similar action if they became aware that 'a parent often severely beats his child'. CONCLUSION Pro-marihuana sentiments are closely related to permissiveness as a factor in social attitudes. In turn, permissiveness is a function of age, education, and disengagement from traditional and conservative attitudes, as indicated by Left voting intentions and 'not going to church'. If the drug is available, and the plant grows freely in Australia, then groups definable in these terms will use it, and approve its use, to the extent that such activity continues to be a facet of permissivism. Perhaps the best way of ensuring that the activity continues to be so regarded is to persist with immoderate legislation and those authoritarian and moralistic statements which can be identified as coming from the opposite end of the dimension. Obversely, the best way of ensuring that such legislation and such statements persist, is to associate the drug with dirt, deviance, and rebellion. But why, finally, do the traditionalists speak with such confidence of a 'silent majority' which manifestly does not exist? There is, I think, a simple reason. The overlap of the over 35's, the church going, the Right voting, and those educated to less than the secondary qualified standard, accounts for some 17 per cent of this sample. This group is the natural recruiting ground for the anti-permissive and is taken by the vociferous minority of the traditionalists to be representative of the general population. The overlap of the under 35's, the better educated, the Left voting, and the non-church goers accounts for a much smaller proportion, some 4 per cent of this sample, so that the vociferous minority of the permissives can call on a smaller cheer squad. Alas, the left liberal tertiary educated are equally prone to speak as if their cheer squad represented the population. The population, meanwhile, is undecided, and probably wishes that neither group would presume to speak so readily on its behalf. This, one suspects, is the most important characteristic of the silent majority. REFERENCES AND NOTES 1. Jones, Reese T., (1971) Marihuana - induced 'high' influence of expectation, setting- and previous drug experience. Pharm. Rev. 23,4. 359-369. See also Howard Becker's classic 'Becoming a marihuana user', Am. J. Sociot. 59, 235-242 (1953). 2. Dr Smart's paper contains a reference to this aspect of cannabis use. The ability to 'come down' at will is part of cannabis folklore. 3. Kaplan, J. Marihuana, the new prohibition Meridian, N.Y. 1970. 4. One of these surveys has formed the basis for a paper in press with the British J. Addiction. Hasleton, S.L. and Simmonds, D.W. 'Is Australia going to pot: some trends relating to marihuana.' 5. A more detailed presentation of these findings is in preparation. DISCUSSION Dr Miles asked about the probable relationship of permissivism to the F scale by Adorno et al. which measures what might be called Fascism or Authoritarianism, and Mrs Mott about the Conservatism Scale, which has

many overlapping items with the F scale. Dr Leuw stated that his group has found a very high correlation between many of the sorts of item used by Mr Hasleton and the F scale. Mr Hasleton replied that the reason for using the item chosen was because they were all issues under current discussion in Australia. Ms Berntsen drew attention to the fact that according to this Australian data, the only issue on which the old were more permissive than the young was on the issue of prostitution! Dr Rubin asked if Mr Hasleton's study, or any other study, had asked why respondents took the position that they did. Mr Hasleton replied that such a question would have to be pursued in a small intensively interviewed sample. Following up this point, Dr Somekh asked how the data presented passed beyond the merely descriptive, and helped to explain the connotative aspects of cannabis that Mr Hasleton referred to at the beginning of his paper. Mr Hasleton replied that the data showed that cannabis was seen as an issue of permissiveness versus non-permissiveness; however, the data did not show why this was the case. Dr Miles suggested that the general population had been given the impression for years that people who smoked pot were politically on the Left. Maybe this is an accurate impression to the extent that cannabis users express leftist attitudes when answering questionnaires. 'To be permissive is one thing, but to be in favour of people who are seen as having revolutionary thoughts is another matter altogether. This may be why your permissiveness scale gives such an apparently strong reading on marijuana.' METHODOLOGY Dr Smart asked about the sampling procedure: how were the elusive deviant younger males found. Mr Hasleton agreed that although attrition in this part of the sample might be high, and that one might be ending up with a rather conservative selection of younger respondents. However, the intention was to get a sample of the population likely to vote (floaters don't vote) rather than a sample representative of the entire population. Dr Edwards suggested 'One of the prime rules of factor analysis is that what comes out is determined largely by what goes in and by the initial choice of questions. The mere fact that you have asked about cannabis at the same time as you have asked about prostitution (as opposed to something like wallpaper), means that you have possibly invited a mental set in the respondent. You talk about attitudes to cannabis being embedded in the permissiveness factor. The word embedded is very appealing, and reifies the notion, but it would be good to be able to look at the raw data in a chi-squared form to see to what extent such a description seemed meaningful. I'm sure that attitudes to cannabis do correlate with these other attitudes, but I'm uncertain that the relationship is as large as your elegant statistics lead us to believe.' Mr Hasleton replied that "The statement that what you get out of factor analysis is what you put into it is, of course, correct. Our task was to test the hypothesis, using component analysis, that a structure of correlations exist to the extent that we could talk about the existence of permissiveness. We were able to support the hypothesis. The question of the effect on the respondent of putting forward these items together is unanswerable by this data, but is answered by studies which find that these permissivism items do come out as a coherent

group from a larger set of items presented to respondents (see a study Eysenck and Wilson, in press B. J. Soc. and din. Psy.)'. POST-CONFERENCE ADDENDUM TO MR HASLETON'S PAPER Since this paper was presented the first results of the U.K. survey have come to hand. The sample was constructed along the same lines as those drawn for the Australian studies, and comprised 1968 respondents. The most important differences between the two communities concern the factorial complexity of permissiveness, and the differing contribution of promarihuana sentiments to permissiveness. In the tralian samples it was clear that the first principal component was an acceptable measure of a permissiveness dimension. The minor cluster of items representing abortion, prostitution, and divorce lay oblique to the major dimension, and the item representing marihuana legalisation was moderately correlated with that dimension in the manner described. In the U.K. sample, the factorial structure is more complex. Three components have eigenvalues greater than unity, accounting for 27.6 per cent, 11.4 per cent, and 10.6 per cent of the total variance respectively. When iterated for communality and rotated to the varimax criterion, three rather 'clean' orthogonal factors emerge. The first, labelled 'permissiveness ' has loadings in excess of 0.40 on contraception, censorship, prostitution, homosexuality, and 'living together'. The second has loadings on sex education and (a new item) drug education, and a low (0.32) loading on contraception, and might be called 'moral education'. The third has a loading of 0.64 on abortion, 0.43 on divorce, and 0.25 on contraception. This is perhaps best thought of as a factor of 'Catholic morality' in which case the low loading on contraception is worthy of note. All the loadings discussed are positive. In the Australian samples, the proportion in favour of cannabis legalisation was 17 per cent. In the U.K., the proportion is very much lower: only 9 per cent of the total sample advocating such a change. This indicates that there has hardly been any change in this respect since 1969. In January of that year, National Opinion Polls Ltd. reported that 8 per cent of the sample believed that the penalties for possessing 'pot' should be reduced, (4 per cent d.k.) and in April 1972 12 per cent believed that the 'smoking of cannabis should be allowed' (5 per cent d.k.). Subsequent analysis may reveal a different trend among the young, but for the total adult population, the early indications are that procannabis sentiments are restricted to a small minority of the population. This is entirely in accord with.the suggestions expressed in the paper: in the U.K. cannabis does not load on the dimension of social permissiveness, and hence has not been a part of whatever changes may have taken place in this respect.

13. Some Suggested Cannabis Research Priorities

R. D. Miller, Former Research Director, Commission of Inquiry into, the Non-medical Use of Drugs, Ottawa.

PAST RESEARCH * * Specific references have not been included in this overview. Thorough bibliographic documentation and further discussion of some of the issues dealt with here in summary form are presented in a detailed review of the literature on 'Cannabis and Its Effects', prepared by the author for the Commission of Inquiry into the Non-Medical Use of Drugs (published in the Commission's Cannabis Report, Ottawa: Information Canada, 1972). The present assessment of research priorities is based primarily on that review, with more recent information from the literature and current research taken into consideration. In the mid-1960s, the cannabis controversy seemed to reach its peak. Many alleged authorities had taken diametrically opposed positions, not only on moral and social policy issues, but on the supposedly hard scientific facts as well. The world literature on cannabis numbered several thousand publications, but few of the papers met modern standards of scientific investigation. The majority of the reports were poorly documented and ambiguous, emotionladen and sometimes incredibly biased, and could, in general, be relied upon for little verified information. Expertise in the area of cannabis was limited by the fact that there was little clearly established scientific information, and preconceived notions often dominated the interpretation of ambiguous data. The lack of modern scientific knowledge of cannabis at that time could be attributed to several factors. To begin with, until recently, governmental restrictions on the medical and scientific use of cannabis have been so strict that the majority of would-be researchers have found it more attractive to work in other areas. Secondly, since the widespread and middleclass use of cannabis in North America, Europe and many other areas is a relatively recent phenomenon, it was not considered a particularly high priority research area. Observations made during centuries of relatively unrestricted cannabis use in regions of the East have rarely been scientifically documented, partly because most of what we consider modern science has been, until recently, a Western phenomenon. In addition, these countries have generally had considerably more pressing public health problems demanding the devotion of limited scientific and medical resources. Thirdly, until the last few years, there was little possibility of properly specifying, standardizing or comparing the cannabis substances studied since the relevant aspects of cannabis chemistry were unknown. Consequently, there was little basis for comparing reports, and generalizations from one study to another were limited. Much of the inconsistent data previously reported may well have resulted from widely varying cannabis doses being studied under different social and experimental conditions. On the other hand, it is worth noting that a number of the earlier reports on cannabis, in retrospect, seem remarkably well done and are deserving of careful critical consideration today. Interestingly, in certain areas, recent formally designed and elaborately instrumented studies have done little more than confirm the observations of some of the well documented, but technologically limited, investigations of the past. In spite of strong disagreement over the years among extremists on many aspects of cannabis, more than a dozen major governmental and international reports by independent commissions and other groups of widely varying backgrounds and orientations, covering three-quarters of a century, have come to remarkably similar conclusions regarding the use and effects of cannabis and its relative potential for personal and social harm. In addition, generally similar conclusions were reached regarding the inappropriateness of a heavy reliance on the criminal law to control the use of the drug. It has frequently been

observed that the impact on government policy of these reports, and of scientific inquiry in general, has been disappointingly limited over the years. In less than a decade, cannabis has emerged as one of the most widely studied drugs in history - perhaps surpassed only by alcohol. Accordingly, much of the past scientific controversy has diminished, although considerable disagreement still exists in some areas. The past five years have witnessed major advances in our knowledge of cannabis, its use and effects: systematic botanical studies are underway; the primary chemical constituents of cannabis have been identified and isolated, and many synthesized; some of the basic factors of cannabinoid absorption, distribution, metabolism and excretion have been elucidated; prominent shortterm physiological, behavioural and psychological effects have been documented; many basic animal toxicology studies have been completed and, in some countries, restrictions on human experimental research have been relaxed accordingly; recent advances in cannabis chemistry and basic pharmacology have stimulated a re-interest in the possible use of natural and synthetic cannabinoids in the medical treatment of a variety of disorders; a number of systematic studies of possible social, psychological and physiological effects of long-term use are underway, and the preliminary findings of several laboratory and field investigations into chronic and sub-chronic use are available; and surveys and various behavioural studies have identified the major dimensions of the extent and patterns of cannabis use in certain countries. In spite of this impressive progress, many important questions are as yet unanswered and must be resolved before a full picture can be presented of the etiology, general characteristics and consequences of the nonmedical use of cannabis in various societies. This is not to say, however, that changes in the legal status or in other aspects of public policy regarding cannabis must necessarily await 'the full picture'. All the answers will never be in. Hopefully, the sophistication of both scientific knowledge and public policy can evolve simultaneously, as a flexible, continuing process.* The following discussion provides some indication of areas where further research is needed. Some suggestions are addressed directly to issues of immediate social concern which urgently need clarification. Others are focussed more on the acquisition of basic scientific information, perhaps of direct and immediate interest primarily to researchers, but which is likely, in the long run, to contribute significantly to our general scientific knowledge of cannabis and, ultimately, to the better understanding of those aspects of its nonmedical use which are of potential concern to society. * A potential role for science with respect to the 'drug problem' is to provide information to better enable individuals and society to make informed and discriminating decisions regarding the availability and use of particular drugs. While scientific enquiry may be able to provide a useful guideline and source of information, science itself is not a policy-making process, but merely a practical system designed to explore and test notions of a certain abstract nature. Even though the aim of science is to maximize objectivity, much of the actual process, including the original definition of the problem to be investigated and the particular research design adopted, on through to the interpretation and application of data, is often a rather subjective venture regardless of the controls maintained in the formal analyses. The practical use of scientific information in the social sphere often entails economic, political, legal, philosophical and moral issues which are not easily amenable to scientific analysis as we know it today. Input from many disciplines is clearly essential. DRUG USING BEHAVIOUR

Some of the major dimensions of the extent and patterns of cannabis use in various countries have been identified in recent years. However, research coverage has been quite uneven, both with respect to geographic area and to the particular focus within the societies studied. Students in North America and Europe have been most frequently investigated groups. The epidemiology of cannabis use among adults and other populations in most parts of the world is much less well understood. In some instances, when the number of users in the general population is relatively small, geographically clustered, or otherwise inaccessible to general survey, anthropological approaches may provide information unobtainable by other means. Much can be gained by ethnographic study of the patterns of cannabis use under natural conditions. Participant observation methods and procedures in which users record their own daily activities, including drug use, can provide much-needed information on the role that cannabis plays in the life of the user as well as its personal and social meanings and functions. We need much more information, from a variety of different cultures, regarding the factors influencing what has been called the 'social career' of cannabis use. This refers to the sequence followed from initial experimental use on, in some cases, to higher levels of regular use through, more rarely, to heavy chronic use. While there is a fair amount known about the variables involved in initial use (e.g., easy availability, curiosity, peer group pressure, etc.), there is a relative paucity of data available on the factors which predispose different individuals to progress through different levels of use, to stabilize use at various plateaux, or to reduce or terminate cannabis use. Further research is nedded regarding the possible roles of such variables as: opportunity and availability; relative cost; other drug use (past and present); legal status and penalties; likelihood of detection; general social and specific peer group acceptability; information and attitudes regarding potential for harm; general pharmacological factors (including tolerance and dependence); individual differences in response; and a variety of other personal psychological and physiological factors under varying degrees of environmental and genetic control. Available evidence suggests that the distribution of levels of use of a drug (i.e., dose and frequency of use) in a population takes the form of a continuous, positively-skewed, unimodal curve, approximating a log-normal function. The most common levels of use occur at the lower end of the scale, with relatively fewer individuals involved in increasingly heavier use. Some researchers have argued, on theoretical grounds, that in order to decrease the total number of users consuming higher (and potentially more dangerous) quantities of a drug, a mean shift towards lower levels of use would have to occur in the entire population of users. On the other hand, it may be possible to reduce the variance of such a distribution (and, therefore, the number of heavy users) to some degree, independently of the mean. Some inventive research methodology is clearly needed here. Many studies have failed to discriminate adequately among various levels of use. Simple dichotomous categories such as 'user' and 'non-user' provide little useful information regarding the full continuum of use patterns and associated personal and social correlates of concern. Social scientists might usefully explore the possibility of gathering and presenting data in a form analogous to the dose-response function in pharmacology. Significant controversy exists regarding the relationships between the use of cannabis and the patterns of use of a variety of other drugs, including alcohol, tobacco, opiate narcotics, amphetamines and LSD. It appears that dynamic and changing personal, social, legal and

economic factors are the primary determinants of patterns of sequential or concurrent multiple drug use, and that the specific pharmacology of the substances involved is often secondary. The relative roles of the various factors predisposing to multiple drug use should be further investigated. We have little adequate information on the effects of cannabis use on levels of alcohol use within a population. It is not clear from the apparently contradictory data available whether, on a large scale, cannabis would tend to replace alcohol as an intoxicant in the user population, or whether the use of these drugs, at various levels, would be additive without significant interaction, or if the use of one might potentiate or increase the consumption of the other. In North America and Europe, cross-sectional surveys indicate that those who use alcohol are much more likely than 'teetotallers' to use cannabis, and that most cannabis users still drink alcohol. However, certain primarily retrospective data suggest that the use of cannabis may reduce or interchange with (but not totally replace) alcohol consumption over time in certain populations. In interpreting or designing studies, it is essential to distinguish cross-sectional between-subject correlations, at a single point in time, from within-subject covariation over time, the latter being the relationship of ultimate interest here. In addition to the alcohol-cannabis controversy, other questions which have received considerable attention in recent years include the role of tobacco smoking in the subsequent use of cannabis and the influence of cannabis use on later opiate narcotic consumption. Neither issue has been really adequately resolved. Systematic retrospective studies would be helpful, but careful prospective investigation of the relationships among the patterns of use of different drugs in various populations over time would be most likely to provide the needed information. There is a general need for retrospective and prospective study of the changing patterns in drug use on a world-wide basis. Not enough is known as to how and why various forms of drug-using behaviour emerge and flow among geographic areas, from one culture to another and among subcultures within societies. Cross-cultural and historical investigation would enable a better understanding of those aspects of the patterns, functions and consequences of cannabis use which transcend cultural change, and those which are relatively amenable to social manipulation or control. It is already clear that substantial drug-culture interaction occurs in many areas of interest and social concern. CHEMICAL AND BOTANICAL ASPECTS While much is known regarding agricultural aspects of hemp fibre production, until recently, little modern botanical research had been conducted on the psychotropic aspects of cannabis. Recent advances in the study of cannabinoid chemistry have enabled considerable progress in this area. Several basic projects are underway in various parts of the world and continued research could be valuable. The relative and absolute levels of the neutral and acid forms of the major cannabinoids, tetrahydrocannabinol (9 THC), cannabidiol (CBD) and cannabinol (CBN) in cannabis samples depends on a variety of genetic, environmental and post- harvest treatment and storage factors. In North America and England, at least, hashish is not simply a concentrated form of the marijuana available. On the average, hashish contains a much higher percentage of the total cannabinoids in the form of CBD and CBN. The pharmacological significance of this difference, with respect to acute and chronic use, has not yet been established.

Variables affecting the shelf-life (i.e., stability and degradation) of different forms of cannabis are not yet adequately understood. Standard storage procedures have been developed for most immediate research purposes, but additional inquiry is needed. Almost all recent human experimental research has been conducted with marijuana or synthetic THC. While there are obvious advantages to the use of uniform substance in experimental work, because of the unusual heterogeneous nature of the various preparations of Cannabis sativa in use around the world, different strains of marijuana and varieties of hashish samples and liquid concentrates should be given chemical and experimental evaluation. Standard supplies of other primary cannabinoids and their metabolites should also be developed and made available to researchers for human and animal studies. Further development of water soluble cannabinoids would greatly facilitate experimental studies of effects relevant to both the medical and non-medical use of cannabis. Accurate information as to the identity, purity and potency of drugs being consumed from illicit sources is essential for meaningful administrative control and regulation, as well as for socially relevant laboratory research and appropriate field studies. Existing data indicate that illicit cannabis around the world varies over a wide range in cannabinoid content. Cannabis is usually smoked with tobacco in most areas (North America being the only significant exception), and, although it is occasionally 'cut' or diluted with inactive materials on the illicit market, there is little evidence that adulteration with other active drugs, unbeknownst to the user, is a significant occurrence. Serious questions have been raised regarding possible contamination with herbicides, pesticides and toxic fungi in certain localities. Appropriate analyses have not been done in these areas. No general systematic 'street drug' data have been collected on an international or national scale. Up-to-date, continuing qualitative and quantitative analyses of randomly selected samples of police seizures, supplemented with data from other sources, such as treatmentoriented laboratories and special field studies, would seem to provide an adequate system for monitoring the purity and potency of illicit cannabis in most countries. Information on the age and original geographic source of the samples would add to the value of such efforts. The uniform collection and analysis of these data on an international scale would seem worthwhile. In the past two years, significant improvements have reportedly been made in chemical analytic techniques for quantifying the major cannabinoids in crude cannabis preparations. Reliable standard methods should be communicated on an international basis, with subsequent follow-up evaluation involving standard test materials submitted to participating analysts. The reliability and validity of the quantitative methods in current use in many laboratories are not adequate, and consequently some recent cannabis studies which have attempted to specify doses can only be interpreted in a general quantitative sense because of questions regarding the chemical assessment of the potency of the samples involved. There is urgent need for simple and convenient techniques for identifying and quantifying the primary cannabinoids and their metabolites in the body. The development of such methods would enable much greater specificity and sophistication in many important areas of cannabis research and should be given high priority. For example, a quantitative chemical test, analogous to the alcohol BreathaZyzer, which could provide evidence of recent cannabis use and a reasonable estimate of the intensity of certain drug effects, would be invaluable to researchers in many fields. Although certain qualitative methods exist which can provide

evidence of use in certain circumstances (e.g., finger swabs, mouth washes and dental scrapings), these techniques are often not appropriate and, in any event, cannot provide information regarding current effect levels. Compared to the relatively simple case of alcohol, cannabis poses significant practical difficulties. Instead of a single primary active molecule, as is the case with ethanol, it appears that 9THC and perhaps several metabolites account for the major effects of cannabis. In addition, other cannabinoids and their metabolites, such as CBD and CBN, are also likely to be present in body tissue and fluid and would likely confound the analysis and perhaps alter the basic pharmacology of THC, as well. After specific analytic methods become available, the relationship between tissue levels of these various cannabinoids and the effects of interest must be established. Available evidence, based primarily on the relatively simple case of isolated THC administration, suggests that blood sample analysis is the most practical approach to estimating effects from levels of active cannabinoids in the body. Although considerable progress is being made towards refining quantitative tests, even those methods which are likely to be developed in the near future may have limited applicability outside of the research laboratory. Several gas-liquid chromatography (GLC)-based methods (usually linked with other tests), which are currently being investigated, may be very sensitive and precise, but are relatively slow and expensive, and they require a high degree of specialised training and delicate laboratory equipment. In addition, venous blood samples are generally needed. Potentially, radioimmunoassay or spin immunoassay techniques could provide very rapid analysis of minute samples (e.g., a finger pin prick of blood), but at present such methods are not adequately specific. Even if simple and efficient blood sample methods were developed, significant practical and legal problems surrounding the acquisition of such samples would likely preclude their use for general control purposes in any way similar to the use of the alcohol Breathalyzer. EFFECTS CHRONIC USE STUDIES Although some important questions regarding the acute effects of cannabis remain unanswered, the possible personal and social consequences of long-term heavy use are of primary immediate concern. Multi-disciplinary longitudinal studies of various populations, including groups of light and heavy users and non-users, would be invaluable and should be initiated. However, further clarification of certain issues is needed and cannot await the completion of such long-term investigations. Cross-sectional studies of persons with varying histories of cannabis use are likely to provide a reasonable short-term pay off. Since the common use of the drug is a relatively recent phenomenon in North America and Europe, it might be rather difficult to locate an adequate number of suitable subjects for extensive epidemiological study in these geographic areas, although certain relevant projects could be conducted. It will be necessary to turn to those nonindustrial countries which have had a longer history of cannabis use for large-scale investigations of this type. Several studies of chronic users in various cultures are currently underway, and some significant reports have recently been published. Considerable additional effort is warranted. Cross-cultural generalizations must be made with caution, however, since many conditions of non-industrial countries may have limited applicability to other social situations.

Research into chronic effects must take into consideration the possible influence of such variables as age, sex, education, socio-economic status, nutritional and hygienic conditions, multiple-drug use and a variety of other ethnic and cultural factors. Appropriate control groups must be compared with experimental subjects for whom reasonable information is available on the frequency and duration of cannabis use and the quantities consumed. Quantitative analysis of the cannabinoid content of typical cannabis samples from the field would enhance such studies. In order to discriminate acute and chronic effects, it is essential to ensure that subjects are, indeed, drug-free at the time of testing. (A simple chemical test for recent use would be invaluable here.) Correlational data of this nature, of course, cannot establish causal factors, but may suggest links and provide clues for further research. If an association were found between cannabis use and other variables, in many instances it would be difficult, if not impossible, to distinguish drug effects from predisposing personal and environmental factors without considerable concentrated follow-up effort. Several sub-chronic studies have recently been conducted involving the daily administration of cannabis to humans under controlled laboratory conditions over periods of weeks and months. Such experiments can provide certain clues regarding the consequences of chronic use, but cannot detect possible effects which might develop only after more prolonged periods of exposure. In some areas of interest, basic information can be obtained from studies of chronic cannabis administration to animals. The significance of such experiments, however, must ultimately be interpreted in the context of human conditions of drug use. Further effort should be made to develop animal tests which parallel pharmacological processes and effects relevant to humans. Because of often substantial inter-species differences, findings involving lower species should be verified in primates. Animal studies have traditionally employed enormous doses, administered under conditions of questionable pertinence to human situations. It is important to explore the acute and chronic effects of cannabis on animals employing dosage ranges and modes of administration which are more comparable to those likely to be used by humans. The assumption is often implied in toxicity studies that one can usefully estimate the effects in humans of long term use of a drug on the basis of information obtained from sub-chronic administration of massive doses to lower species. The predictive validity of such procedures is uncertain, and results must be interpreted with caution. SOME GENERAL EXPERIMENTAL AND PHARMACOLOGICAL CONSIDERATIONS Quantitative information on major cannabinoids in experimental samples administered should be provided in laboratory research reports. Until convenient methods are available for assessing the levels of active cannabinoids in the body, researchers should provide estimates of the actual cannabinoid dose delivered to and absorbed by the subjects under the smoking conditions employed. Specifying doses on a body-weight basis would further aid interpretation of experimental results. In some respects, it would appear advantageous for researchers to adopt more uniform modes of cannabis administration for experimental study. However, the variety of styles and techniques of cannabis consumption around the world requires that different modes of

administration be investigated. The pyrolysis and delivery of cannabinoids smoked in cigarettes, conventional pipes, water pipes, or chillums may not be directly comparable, and possible differences in the acute and chronic effects of these methods of cannabis smoking, as well as ingestion, should be examined. Factors which affect the rate and extent of absorption of cannabinoids in the lung and gastro-intestinal tract have not been adequately delineated. In North America and Europe, there is typically very little cannabis lost during the process of smoking, and users usually consume all of the material in a cannabis cigarette, including the butt. The common deep inhalation and retention procedure is similarly parsimonious. This style of smoking results in a much greater proportion of the THC in the cannabis being delivered to and absorbed by the user, in comparison to the more casual puffing style (with prompt exhalation) commonly seen in countries with a longer history of cannabis use and ready availability - Jamaica or India, for example. These factors, along with probable differences in the efficiency of THC delivery among the various forms of pipes and cigarettes mentioned above, greatly complicate cross-cultural dose comparisons. A clear understanding of the present, and likely future, conditions of cannabis use is essential for optimal laboratory efforts. socially relevant experimental conditions including appropriate drug doses and modes of administration require accurate information from sophisticated socio-pharmacological field studies. In addition, the influence of social attitudes and norms regarding cannabis use on the ultimate social, psychological and physiological effects of the drug needs further study in the laboratory as well as under natural field conditions. Significant advances have been made in our knowledge of the metabolism and distribution of THC in the body, but considerable further research is needed. Since, as noted earlier, CBD and CBN are present in large quantities in some forms of cannabis, the pharmacokinetics of these cannabinoids should be thoroughly investigated, as well. The molecular mechanisms involved in the various central and peripheral effects of cannabis are poorly understood. Structure-activity relationships of the various primary cannabinoids and metabolites should be established. Although it is now clear that L THC is the principal active constituent in cannabis (at least in part via metabolites), it appears that other cannabinoids may alter the pharmacology of THC and its metabolites. For example, CBD and CBN may compete for binding or receptor sites or alter THC metabolism and excretion. They do play a significant role in cannabis tolerance or sensitization and interaction with other drugs. There is evidence that primary cannabinoids or their metabolites may accumulate in the body for prolonged periods with chronic use; possible physiological consequences should be carefully explored. Most laboratory research of this nature has focussed on THC to the relative exclusion of CBD and CBN, even though certain hashish smokers, for example, likely consume a greater amount of these latter cannabinoids than of THC. The possible long-term effects of other, non-cannabinoid, components of marijuana and hashish should also be examined. The vast majority of both human and animal pharmacological studies of cannabis have employed only male subjects. With very few exceptions, when general information about the effects of the drug is sought, males are used; females are rarely studied scientifically except when some specifically female characteristic is under consideration. Recent research has focussed on healthy, young adult males of middle class background (primarily college students), while earlier studies often employed male gaol prisoners as subjects. Other

populations need to be investigated, especially females and older persons of both sexes. Acute and chronic effects of the drug in persons suffering from certain physical disabilities (e.g., diabetes, epilepsy,liver dysfunction, asthma or migraine headaches) should be explored as well. Cannabis effects in persons with varying histories of other chronic drug use should be investigated. Alcohol is perhaps most important in this regard. Tobacco might also be significant. Experimental reports should specify, in detail, the subjects' past and current patterns of medical and non-medical drug use. Unless characteristics of the initial reaction to cannabis, or the adaptation to its effects early in the career of use, are of central concern to the study, the use of cannabis-naive subjects may not be advantageous. In most areas, information on regular users, who have developed a more consistent and stabilized general response, would be of considerably greater immediate social significance. PHYSIOLOGICAL ASPECTS Cannabis produces few prominent short-term physiological effects in normal users. Increased heart rate, a slight reddening of the eye and decreased salivation are the most commonly reported acute responses. Acute and chronic effects in persons suffering from certain prior physical disabilities may be more significant and should be investigated. Minor respiratory, gastrointestinal, cardiovascular and neurological dysfunction has been inconsistently noted in studies of heavy, long-term users of cannabis in non-industrial countries. In most of these reports, likely confounding variables have not been taken into account or controlled. Few cannabis-related disorders have been found in more recent controlled studies, but some significant areas of potential concern exist. Large scale epidemiological studies may be needed to investigate potential problems which might occur relatively infrequently in the user population. It is worth noting here that the major alcohol-and tobacco-related physical disorders which have come to light in recent years (such as liver cirrhosis and lung cancer) are prominent in only a small proportion of regular users of these drugs. Recent evidence suggests that chronic smoking of large quantities of cannabis may have effects on the respiratory system similar to those produced by tobacco. Straightforward inquiry into chronic cannabis use is complicated by the fact that regular cannabis smokers typically also use tobacco, and the two drugs are mixed together for smoking in most countries. In North America, where they are more apt to be used separately, differences in typical patterns and methods of smoking, as well as quantities consumed, make simple comparisons between the substances difficult. The possibility that cannabis (either alone or in combination with tobacco) might produce or complicate respiratory disorders, including cancer, must be investi_ gated further and possible toxic factors identified. Considerable effort must be concentrated on rectifying conflicting data on possible neurotoxic effects of chronic cannabis use. Several recent controlled studies have found no electrophysiological or psychometric evidence of neurological disorders in persons with long histories of heavy use, but some clinical reports have suggested chronic effects (including brain damage) in certain users. Even though major methodological problems exist in these latter reports, the seriousness of the conditions implied requires that further systematic investigation be given high priority.

Reports of the acute and chronic effects of cannabis on sleep are inconsistent. The existence and ultimate neurological and behavioural significance of any cannabis-induced changes in sleep patterns need to be established. Available human data have not indicated that cannabis-induced chromosome abnormalities are a likely occurrence, nor have adverse drug effects on the developing foetus been demonstrated in humans. However, some studies have detected such effects at extreme doses in certain animal species, and there are some preliminary suggestions that cannabis may affect DNA under certain circumstances. Because of the potentially serious consequences of any such effects in humans, further research is required. For example, prospective investigation of young women who are heavy cannabis users, studied through pregnancy with subsequent post-natal follow-up of the mothers and children, would be appropriate. Carefully chosen control groups would be necessary to isolate the influence of likely confounding variables. Although an apparently remote possibility, potential effects on the child of cannabis use by the father should also be considered. Existing research reports of the acute effects of cannabis on muscle strength and physical work output are not conclusive, although some deficit in efficiency is suggested. On the other hand, it has been reported that cannabis is used in certain countries by labourers to reduce fatigue and increase work energy. Several relevant studies are in progress, but the effects of acute and chronic cannabis use on physical labour and athletic performance, under various laboratory and natural conditions, have not been clarified. PSYCHOLOGICAL AND BEHAVIOURAL ASPECTS Cannabis users typically report substantial acute alterations in sensory, perceptual and cognitive processes as a result of taking the drug. Attempts to verify most such subjective effects in the laboratory have net with surprisingly little success, in part because of the insensitivity of standard tests to the more psychedelic aspects of the experience. Since many of the prominent psychological effects reported by users are intensely personal, the laboratory scientist often has little opportunity to make objective measurements, and must rely on subjective, introspective reports communicated verbally through a language system which is frequently inadequate. It would be worthwhile, in order to better understand the growing popularity of the drug, to experimentally investigate, in more sensitive fashion, some of the effects which users claim provide the reinforcement or motivation for continued use at various levels. Of particular interest are those aspects of the response which reinforce chronic heavy use in certain users. There are considerable differences among individuals in reaction to cannabis and in the degree to which various aspects of the response and patterns of use are considered desirable. More thorough investigation is needed of the relative importance of personality, past drug experiences, set and setting of use, and other cultural or genetic factors in determining various cannabis effects and their subjective valence and meaning to the user. Experimental evidence indicates that under some conditions cannabis can produce short-term deficits in certain perceptual, attentional, cognitive and psychomotor abilities. Impairment is most consistently found on complex tasks requiring sustained or divided attention and shortterm memory. Alterations in time perception and increased visual imagery are also highly predictable. The overall significance of the effects found in the laboratory to personal, social, scholastic, and occupational functioning under natural conditions has not yet been established.

Further research, employing both laboratory and field approaches, should be undertaken and the practical predictive value of the experimental tests empirically evaluated. Under certain experimental conditions, cannabis has been shown to have detrimental effects on automobile driving performance. A detailed analysis of such effects, and possible consequences for traffic safety, is needed. Although certain basic parameters can be established experimentally, such studies can provide only a limited basis for predicting the likely effects under natural conditions of drug use and driving. Once appropriate biochemical tests have been developed for the detection and quantification of active cannabinoids in the body, epidemiological studies can be initiated to establish the extent to which cannabis and other drugs are associated with traffic accidents in the general driving population. Studies comparing drug levels in persons involved in accidents, with drug levels in suitable control subjects (who have not been associated with accidents) have been of primary importance in clarifying the traffic hazards of alcohol. Since epidemiological investigations of this nature can yield reliable information only if the incidence of the use of the drug in the driver population is fairly substantial, such studies are most likely to be fruitful in those geographic areas of industrial countries where cannabis use is most prevalent. Acute anxiety or panic reactions to cannabis (generally of little clinical significance) have been reliably reported. Systematic information on contributing factors is scarce, but it appears that such occurrences are infrequent and typically involve higher doses and less experienced users. Only a small proportion of these are ever seen at treatment facilities. However, since such reactions do occasionally occur in laboratory studies of cannabis, valuable information might be obtained if detailed data regarding these incidents were collected from researchers. A prospective study could be undertaken employing standardized report forms outlining potential contributing factors. A considerable body of clinical literature from certain Eastern and other non-industrial countries suggests that the heavy chronic use of cannabis may be associated with a variety of psychological and behavioural disorders. Certain investigators claim that a specific 'cannabis psychosis' exists, while others deny that such a separate clinical entity can be established. Some argue that preexisting psychopathology leads to heavy drug use. Methodological limitations in the reports preclude a comprehensive evaluation of these claims or the identification of causal variables. Modern systematic studies of these populations should be initiated. The view that cannabis may precipitate a significant psychotic reaction in certain predisposed individuals has also been noted in the North American clinical literature. There is no consensus as to the nature of possible predisposing factors or their prevalence in the general population. Prolonged psychological correlates of chronic use of a more subtle nature (including difficulties in concentration and thinking, personality changes and an 'amotivational syndrome') have also been given considerable notice in the clinical literature. Of particular concern are possible adverse effects of heavy cannabis use in adolescents. It is not yet clear what role cannabis might play in the behavioural syndromes described. Recurrences or 'flashbacks' of some cannabis effects in the absence of the drug are occasionally reported, but the concept has not been clearly delineated. We have little knowledge regarding the frequency, intensity and consequences of such experiences. More information is needed.

Additional anecdotal reports of individual adverse reaction cases from poorly defined patient groups would be of little value, but accurate clinical reports, put in a proper population context may provide valuable clues for subsequent systematic study. Clinical investigations, involving appropriate control groups, should focus particular attention on pre-morbid social, psychological and physical conditions, previous and concomitant use of other drugs (including detailed information on the quantities involved and the duration and frequency of use), static and dynamic symptom patterns, and long-term follow-up. It would be of interest, for example, to compare past patterns of drug use in patients presenting similar psychiatric symptoms. Detailed longitudinal study of symptom change in various groups of cannabisusing and non-using patients might prove fruitful. Studies of factors involved in more subtle behavioural syndromes would be considerably more perplexing. It may be extremely difficult to establish cause and effect relationships with present methods of investigation. Given that certain adverse psychological conditions may be associated with heavy cannabis use, it is important to determine the frequency and severity of such reactions in the general population of users. Studies of patient populations alone cannot provide appropriate information for this purpose. There is currently little evidence that serious cannabis-induced disorders occur in a significant proportion of the user population, but adequate data are not available. TOLERANCE AND DEPENDENCE Gross tolerance to the major effects of cannabis does not develop under natural conditions of intermittant or moderate use, but more subtle changes in response to repeated cannabis administration have been noted. 'Reverse tolerance' or 'sensitization' ( where some individuals reportedly experience an increased subjective response after the first few experiences with cannabis) is frequently cited in the literature, but has not been adequately examined under controlled laboratory conditions. Furthermore, there is increasing evidence that, with intermittent or moderate use, some tolerance or behavioural adaptation develops to some of the initial performance-disrupting effects of the drug. The mechanism for such changes in response are uncertain. Available information suggests that learning and other psychological processes may play a more important role in these phenomena than do basic molecular pharmacological factors. There is growing evidence that certain chronic heavy users, who likely maintain a continuously high tissue level of cannabinoids, develop a significant degree of tolerance to the general physiological and psychological effects of cannabis, including those subjective effects which apparently reinforce its use. Such individuals may consume much greater quantities of the drug than are desired or tolerated by the vast majority of users. Evidence of varying degrees of tolerance to certain effects of cannabis in sustained high doses has been observed in some animal species, but not in others. The ultimate epidemiological significance of cannabis tolerance in humans, under various cultural conditions, is as yet undetermined, but the possibility that it might occasion very high levels of use (and, therefore, increase the likelihood of dose-dependent adverse effects) in a small but significant proportion of regular users dictates that the subject be given considerable further study in both laboratory and natural field situations. There is little evidence that even prolonged use of high doses of cannabis produces significant signs of 'classical' physical dependence, although more subtle undesirable physiological and behavioural symptoms may occur on withdrawal in some apparently rare situations. Possible

effects in this regard should be characterized and their influence on sustained patterns of use, under natural conditions, explored. Further effort should be made to define the general concept of behavioural or psychological dependence in an adequate operational fashion, if possible, and the existence and consequences of the phenomenon in various populations of cannabis users investigated. Overreliance on drug dependence concepts may inappropriately focus attention on very extreme, statistically uncommon, patterns of use to the relative exclusion of the wider range of drug using behaviour of basic concern. It would appear that, in practice, drug dependence labels generally provide little verified information about the user which would not be more usefully conveyed by an accurate description of the level (i.e., dose and frequency) of the individual's drug use. Unless adequately defined in each situation in which it is employed, the concept of drug dependence is unlikely to contribute to progress in cannabis research. INTERACTION WITH OTHER DRUGS Interaction between the effects of cannabis and a full spectrum of other drugs commonly in use (both medically and non-medically ) should be explored, and likely physiological, psychological and social components and consequences of such interaction examined. Available evidence suggests that cannabis and alcohol can produce additive detrimental effects on certain psychomotor skills and may enhance common physiological reactions. The characteristics and mechanisms of alcohol-cannabis interaction should be fully investigated, with implications for automobile driving and the operation of complex machinery given special attention. Possible interactions with tobacco, minor tranquilizers, anti-depressants and various popular over-the-counter allergy and cold remedies also seem of likely importance. Although cannabis has exceptionally low lethal toxicity itself, the possibility that it might enhance the toxic effects of an overdose of other drugs, such as alcohol, barbiturates or opiate narcotics, should be thoroughly explored. Cross-tolerance between certain drugs, such as alcohol, and cannabis has been suggested in the literature and should be elucidated. MEDICAL RESEARCH Although the medical use of cannabis is not of primary concern to this paper, related research is likely to advance our general knowledge of cannabinoids and their effects, and may well result in significant medical progress. Any generally accepted use of cannabis, in modern Western medicine, would likely result in significant changes in attitudes regarding its nonmedical use, as well. While cannabinoids have, over the centuries, been reported to produce a wide array of possibly useful medicinal effects, the majority of these claims have either not been adequately investigated using modern scientific methods, or the effects can be duplicated by other more readily available and more convenient drugs. The potential medical uses of natural or synthetic cannabinoids which are currently under investigation include the treatment of anxiety, loss of appetite, insomnia, severe fever, pain, epilepsy, migraine headaches, high blood pressure, glaucoma, sexual unresponsiveness and the secondary symptoms of the common cold and flu. Cannabis is also being studied, alone and in combination with other drugs, as an anesthetic and as a pre-anesthetic agent. In addition, certain cannabinoids have been shown to have significant antibiotic properties. Acknowledgements The valuable assistance of Judith Blackwell, throughout the preparation of this paper, is gratefully acknowledged.

DISCUSSION PATTERNS OF USE Dr Connell started this final discussion by asking whether the types of studies mentioned by Dr Miller would be important in the event that levels of use decreased. Dr Cameron said that in his opinion, given the existing levels of use throughout the world, an increase in use was not necessary to make the studies worthwhile. Dr Miller suggested that cannabis was still filtering to wider social groups in the West, and that there were few signs of any downturn. Cannabis had gone beyond the point where one could expect it to disappear - it was no longer just used by fad groups. Ms Blackwell drew attention to the distinction between ever-use and current use, and said that whilst current use of some drugs may be stabilising, there still are increases in the number of those who have ever used. It was agreed that a monitoring of the changing patterns of use in a range of populations was important. This should go beyond headcounting and investigate the patterns and frequencies and circumstances of use of cannabis and of other non-medical drugs. TOLERANCE AND REINFORCEMENT Dr Tinklenberg pointed out that much of the North American work on tolerance had used very weak preparations of cannabis. 'Recent experimental laboratory data strongly suggests that with more potent preparations tolerance very rapidly develops on a number of different dimensions, such as the cardiovascular, tachychardia, cognitive and so forth (personal communications with Reece-Jones, Frank and Miles). The reason why I bring this up is because Americans often think of the effects of cannabis in terms of what is available in our country - perhaps if only very weak alcohol preparations were available we would have few alcoholic problems. My observation is that when more potent preparations are available, in the Bay Area, cannabis connoisseurs seek those out. When they have the choice of weak local marijuana and some potent imported hashish, they'll pick the more potent preparation.' Dr Miles stressed the need to separate out the physiological aspects from the cognitive aspects. In spite of some physiological tolerance, a group of his subjects did not have significant cognitive tolerance, and were stoned all the time. Dr Miller said that you may get tolerance to certain kinds of effects, but not to those effects that reinforce use, so it may not lead to increased use except possibly by reducing some of the adverse effects. 'Then there's the question of whether all people are reinforced by the same effects. Some like the change between the stoned and the unstoned state, whilst others may find the continuously stoned state reinforcing.' Mr Hasleton suggested that cannabis may achieve secondary reinforcing qualities by being used in pleasurable social, recreational or sexual contexts. 'Its these circumstances in which the stuff is used that provide the reinforcement, it seems to me. An inert substance (like putting a pencil in your ear) under these circumstances, would become a highly likely activity. I think that there is a tendency to see the reinforcement in the pharmacological effects.'

Dr Tinklenberg suggested that there are individuals who, given the choice of putting lowconcentration or high concentration pencils in their ears, would opt for the higher concentration. 'This suggests to me that the pharmacological properties, such as concentration of THC, is important. It's not just the social factors that provide additional reinforcement.' Mr.. Dorn suggested that in Dr Tinklenberg's 'hash is a rarity; its for the connoisseurs. In resin or hash is more common. People over here, are connoisseurs or real heads often try to get grass.' So heavier users do not always seek the centrated preparation. Dr Miller suggested that the English and Canadian experiences provide a control for the American situation. 'In Eastern Canada, imported marijuana is the rarity and people prefer to roll a joint with it than to use the more common resin.' Dr Cameron suggested that with a more potent preparation it is easier to achieve a high dose. 'If you use enough beer, you can become intoxicated. The same is true of a relatively weak preparation of cannabis, you just have to use more. If you are not seeking to become intoxicated, then you may not persevere.' It was agreed that research was needed to isolate the degree of tolerance and reverse tolerance in the whole spread of physiological and cognitive effects. There is no real evidence of any general tolerance or reverse tolerance: one must speak of tolerance to specific effects, isolated by specific objective tests. CANNABIS AS A THERAPEUTIC AGENT Ms Blackwell and Dr Rubin stressed the need to look at the function of the drug for the user. It is known that in some cultures, India and Jamaica for instance, it is used medicinally. The Czechoslavakians have been doing research into folk plants for over 20 years, and selected hemp as the most promising antibiotic out of hundreds of species, and it is now used for a whole range of diseases. Dr Miller mentioned ongoing work in raising seizure thresholds in epileptics, lowering interocular pressure in glaucoma. In Canada there are reports of people taking cannabis to reduce secondary symptoms of flu or colds. Judith Blackwell and Dr Miller had done a small pilot study into possible use of cannabis in reducing discomfort associated with the menstrual cycle. Other possible uses reported by participants are as a pre-operative medication, and as a substitute drug in alcoholism. Professor Paton pointed out that cannabis extract had been given a full clinical trial when O'Shameny introduced it in the 1940s. It might be worthwhile to collect all these reports, since although it was all uncontrolled clinical trial, one should remember that it was by this procedure that many of the currently used major drugs were identified as useful. Dr Edwards suggested that the question was not whether cannabis has medical potential, but whether its balance of main therapeutic effect to side effects was more favourable than that of other drugs. THE USER'S EXPERIENCE Professor Paton said that it would be worthwhile to know what cannabis offered, and to ask how this could be offered by something else. Perhaps cannabis was just plain fun. 'I do think that the question of what cannabis offers is satisfactorily answered by saying that it's

associated with this and that. I think that it's something to do with the drug effect: perhaps the cannabis user himself can tell us.' Dr Miller spoke of the possibility that the immediate powerful reward offered by some drugs may be more strong than the other reinforcers available. Mr Woodcock: 'The fact one form or another of psychoactive drug use is almost universal among human beings suggests that it has a functional purpose about which we really know very little. We should try to find out what it is that people gain from drug use. In much of what is said there is an implied comparison with some kind of human being who is totally healthy and who always works at maximum potential, and this is quite unlike life. If one knew more of what to the individual were not deficits but operational benefits of drug use, one might be able to construct a social policy that could offer this in an alternative way.' Dr Rathod said that there was insufficient talk about the effects as the consumer perceives them. 'How can we talk of what is an abnormal effect, if we don't know the 'normal'? We don't know if the Egyptian smoker experiences the same as the Canadian or English smoker. Unless we have comparable, reliable data then it is very difficult to compare qualitative and quantitative differences.' Professor Soueif stressed the present lack of evidence of reliability of interviews. Dr Miller agreed with the points put by Dr Rathod and Mr Woodcock, saying that he had encompassed the question of the consumer's perspective by referring to factors that reinforce various levels of use. 'I'd like to see why this has become the first new drug to become widely used in several hundred years. There obviously is something important going on here. But I think that if I were talking to a politician or a funder with a limited amount of money then I would have a hard time persuading him to spend a lot of money on asking 'why?' rather than on the effects.' Dr Miller mentioned a scale developed by the Commission on which users can report subjective cannabis effects. Visual imagery is particularly sensitive to dose. While the amount of imagery seems to increase with dose irrespective of culture, the content of the imagery may be culturally determined. Dr Edwards suggested that the degree to which a drug effect is culture-free depends on the dose, with high dose, the effect transcends culture. Participants agreed that how one reacted to one's drug experience was largely culture- specific, and that one's reaction influenced the effect. Dr Rubin took up Mr Woodcock's point, saying that we should ask why human beings, over the millenia, have used drugs. 'Some of the most creative thinking of the adolescent deviant is directed towards the discovery of new psychoactive substances. I think that we ought to use that energy for positive purposes - maybe that's where our major research energies should lie.' EFFECTS ON DEVELOPMENT Dr Hawks said that 'if we are right in assuming that in future there will be more widespread use of cannabis in the UK, then younger age groups will be using the drug. In which case, we have

to ask what the likely effect of the use of such a drug by organisms in state of incomplete physiological and psychological maturation.' Ms Berntsen suggested that this problem was dealt with satisfactorily by other cultures, in which adult patterns of use were not imitated by children. Mr Woodcock suggested that if cannabis ceased to be regarded as 'a drug', then the situation with respect to use by younger age groups might be quite different. 'I don't think that there is a very great tendency for caffeine to be consumed by children. It's not regarded as a drug at all. To see an eight year old child drinking coffee is unusual - yet here is something with no restriction on it at all.' Professor Paton pointed out that children sniff solvents, and Mr Woodcock responded that solvents-sniffing was a secret, non-adult activity. RESEARCH PRIORITIES Dr Hawks asked about the highest priorities out of those discussed in Dr Miller's paper. Dr Miller suggested that in terms of acute effects, effects on driving was a high priority. 'If I were in the position of government I would not spend money on the sort of acute effects experiments that I personally would enjoy doing. I'd look more at the chronic physiological effects. The possibility of respiratory disorder is bothersome. Also, we can no longer attribute all adverse or unusual reactions to bad dope, since we know that they can be produced with pure materials, so we need some work into their causes.' Dr Cameron stressed the need to encourage research into chronic as opposed to acute effects, given a more difficult nature of chronic effects research. Secondly, he stressed the importance of studies of drug interactions. Thirdly, he mentioned the need to relate the patterns of use in a particular area to the nature of the preparation used in that area. These, he suggested, were the priorities. IMPACT ON POLICY Dr Smart pointed out that although the LeDain Commission might be said to have had a liberating effect on court sentencing policy, more people were being convicted. So now there was in fact more criminalisation, but at a lower level: police felt more able to prosecute, and courts to convict. Dr Rubin asked 'I'm not clear whether you feel that public policy should await the outcome of the research that seems necessary.' Dr Miller said that the LeDain Commission was probably a result as much as a cause of change in public attitudes to cannabis. Research is a part of the totality of public decisionmaking, rather than a separate activity that can be fully completed in advance of social change. LIMITS OF SCIENCE 'Most people', said Dr Miller 'have a funny idea about science being very magical, very objective, and don't realise that defining the problem, choosing tools, interpreting results is incredibly subjective. People could be made much more sophisticated and able to evaluate media reports of science without much difficulty.'

14. Conclusions: Asking Better Questions in Cannabis Research

P. H. Connell, Bethlem Royal and Maudsley Hospitals, London. N. Dorn, Evaluation and Research Unit, ISDD, London. INTRODUCTION It was noted in the introduction to this book that the Planning Committee had arranged the proceedings in such a way that participants were not forced into summaries, statements or blue-prints for the future. The Editors, therefore, decided to make an attempt to cull out the proceedings' recurrent themes; to highlight areas where there appeared to be general agreement and to examine and make explicit themes that seemed inherent in the material presented. The interpretations offered (for which we take full responsibility) do not necessarily reflect the views of all the participants and are largely a result of our study of the proceedings, our subjective impressions, and our assessment of the meaning of the Conference viewed in the context of research over the past years. The aims of the Conference were to attempt an overview of the present state of knowledge; to discuss conceptual and methodological problems; to bury dying issues and to point the way to future avenues of worthwhile research. In the event, the Conference seemed to have been successful except in the attempt to bury dying issues. This failure to realise this aim is partly due to the free and undirected discussion but also to other factors which will be discussed later and which involve trends which may bode well for future research. Some participants were concerned that the advent of soluble preparation of cannabis and of delta tetra-hydrocannabinnol and other preparation of active principles might materially and perhaps dramatically alter the whole picture. Although this contribution is partly based upon the discussions, it is in no way intended to be comprehensive and exhaustive. The reader is referred to the summaries of discussion to gain a more detailed overview. METHODOLOGY A major part of the time in discussions was devoted to consideration of methodology. After all, in cannabis research there is as yet no laboratory test which can determine the presence of cannabis in biological fluids outside the very specialised research laboratory (of which there are few) so that whether or not an individual has taken or is in fact taking cannabis cannot often be determined. Nor can the dose which is actually absorbed be determined (taking into account the method of use in which inhalation is wasteful in terms of absorption into the body of active substance) even though the dose of cannabis supplied is known. It is therefore impossible to construct accurate dose-response curves for inhalers and impossible, in the area of field surveys, to know beyond doubt that the person is actually taking cannabis. Thus, however good the methodology of a field study or a clinical study, the pharmacological state of the subject is unknown or relatively unknown.

In the study of individuals from the clinical and psychological viewpoint, it was repeatedly pointed out that there was an urgent need to define symptoms and signs; to create check lists of items which could be validated and could be applicable to different cultures and to employ measures which could be used internationally. The need for standardized interviews was also stressed. From time to time, the question of effects of motivation on drug effect was raised and the need to evaluate this factor in cannabis studies emphasized. These are just a few examples of the general concern of participants about the inadequacy of tools for research and the need to concentrate on this general area in the future. SPECIFIC STUDIES Certain areas for urgent study were proposed and certain priorities were suggested in the discussion of Dr Miller's paper and in the paper itself. There seemed to be some agreement that the following areas should be included and emphasised in future research:1. The development of tests for the determination of the presence of cannabis in body fluids and in particular roadside tests of use in driving research and the evaluation of accidents. Such research would be unlikely to produce results for some years. 2. Research into the effect of cannabis on driving and wider research into driving accidents in relation to other drugs taken and the general 'epidemiology' of driving accidents (including research into the background and meaning of minor damage to cars such as bumps, scratches etc.) 3. Research of populations where cannabis use is the sole drug used before these populations become multiple drug use societies with consequent confusion and blurring of the role of cannabis. 4. Longitudinal surveys in which individuals are assessed regularly over a long period of time rather than cross sectional studies at a given moment in time. S. Cross cultural researches both nationally and internationally. 6. Research into the chronic effects of cannabis use. 7. Research into patterns of use of cannabis and other drugs in geographical areas at different times when the availability of cannabis varies. 8. Research to define dose-response curves whether using inhalation or other methods of use of the drug such as ingestion or even injection of suitable preparation of cannabis or its active principles. 9. Research into what users consider they gain from cannabis use. 10. Research to reassess cannabis in terms of possible use in therapeutics. DIFFERENTIATION OF HYPOTHESES The history of cannabis research is littered with threatening question marks. Does cannabis cause brain damage, madness, heroin addiction, violent crime, degeneration of the body and spirit, etc? are examples. Two characteristics can be identified in these hypotheses. Firstly, their ready appeal to a general audience and secondly, their global, undifferentiated or 'total' quality. These two characteristics are clearly associated. It is easier to think of brain damage than of specific deficit; of madness than of specific clinical signs; of escalation to heroin addiction than use of other drugs, and of overall degeneration than of change of lifestyle.

Furthermore, there is no question of brain damage, madness, addiction, crime and degeneration being other than a bad thing. It seemed, however, that recently there had been a change of style in cannabis research away from the hypothesis of generalised negative effects towards hypotheses of clear, delineated and operationally defined specific effects that were no longer automatically regarded as negative or bad. Cannabis has been the bogy of drug research because of the difficulty of finding scientific evidence (as contrasted with moral emotional and political belief systems) adequate to justify prohibition and because of the difficulty in finding scientific evidence clearly and indisputably proving cannabis to be of no danger to humans. The result would seem to have been a rather panicky attempt to discover something really 'nasty', and has directed attention away from investigation of carefully defined hypotheses and toward the really spectacular, but relatively less likely hypotheses. The end result would seem to have been 'bad' research since one can hardly do good research on hypotheses that are so wide and ill-defined as to be virtually untestable. Thus, data collected in selective ways from particular or peculiar samples, or even subjective impressions, were presented as expert knowledge. Such a reaction to cannabis is understandable when viewed in the context of other drugs. Alcohol, nicotine, barbiturates, amphetamine and opiates, for instance, all have clear dangers such as brain damage, cancer, bronchitis, overdose potential and dependence potential etc. Once a danger is identified the drug is felt to be 'known' and relaxation and action concerning the drug takes place. In the case of cannabis, however, (and this applies to some extent to LSD) there is extreme anxiety because it is difficult to identify the threat which is felt must be there. In the absence of a clearly identifiable threat, a phantasy arises as a focus for anxiety. Thus, paradoxically, extreme and general dangers are hypothesised because it is not possible to locate any moderate or specific danger. This situation seems now to be passing. The scientific community is overcoming its phantasy of cannabis to confront the more complex reality. The greatest improvement in research seems to lie not only in better research design but in cannabis researchers' blossoming willingness to ask more relevant and less ambitious questions. Returning, now, to the intended purpose of this Conference, it is possible to understand why there was little explicit sealing off of dead issues. Global and unsubstantiated hypotheses seem to have given way or been divided up into a greater number of more clearly defined questions. Thus, there is a retreat from the hypotheses of generalised irreversible brain damage to questions such as to what medium and long term cognitive effects of chronic use can be directly attributed to cannabis use, and whether effects, if present, are reversible. The question as to whether tolerance occurs gives way to questions concerning to which specific effects tolerance and reverse tolerance occur. In terms of acute effects questions are now asked as to which kinds of performance in stated conditions are inhibited and which facilitated. These precisely formulated questions seem to lead to answers or potential answers in a way that is impossible if wider questions are asked. So, rather than sealing off dead issues, research would seem to be concerned with subdividing issues to a point where they become more amenable to scientific enquiry. These development have drawbacks. Thus, there is the danger that in building up knowledge in small bits rather than in large chunks, researchers will become exclusively test-minded and select only those areas that can be researched with existing test instruments and procedures or

developments of these. If this tendency takes hold one will be left with a ragbag of atomistic findings devoid of unifying principles. There is thus a need for a series of well-articulated and comprehensive theoretical models that generate mutually exclusive and readily testable hypotheses. These models or systems could be derived from the wider literature in the clinical and social sciences, as well as from the literature on cannabis research itself. A model that ostensibly explains the behaviour of a cannabis user but does not account for the behaviour of the person building the model is a poor model, given that both activities can be regarded as deviant, as purposive and as meaningful in terms of the person's perception of himself to his reference group! It seems to follow, therefore, that in retreating from the vague and almost untestable, illdefined hypotheses of yesterday, the researcher should resist the attraction of a-theoretical empiricism and base his research on hypotheses derived from theoretical working models. Thus, confirmation or disproval of a hypothesis will have implications for a whole model or explanation, rather than just for the acceptance or rejection of that hypothesis, which may be of little interest of itself. THE PERSON AS AN AGENT A theme of the Conference was the interest in the user's subjective reports of the state of acute intoxication and in his accounts of the reasons for his behaviour. When cannabis use was regarded as a kind of disease there were no grounds for taking the perspective of the user seriously since it was assumed that he was not capable of insight into his own experience. More recently, subjective reports of users are coming to be seen to be as valid a source of research hypotheses as the subjective impressions of observers of users. A remaining problem would seem to be that reports of users are always made in relation to and in response to a particular enquirer or social situation, and that, furthermore, subjective experience is itself the result of the interaction of the user with a particular social system and set of expectations and constraints. Stripped of a social context, the user would not exist: embroiled in a social context, be it that of the street or the laboratory, the user constructs his experience in terms of that context. Thus in regarding subjective reports of research data, it is important to acknowledge its social specificity. It would seem imperative, therefore, to draw attention to the importance of relating theoretical models to both individual and social systems. There is an apparent move away from perception of the drug as the agent of drug use, through an appreciation of the power of the person's environment and of the social forces that act upon him, and towards a perception of drug use in which the user himself is re-invested with the power to define the meaning of his behaviour. It is becoming apparent that different persons within a culture take drugs for many reasons and with a variety of intentions, and that the intentions of the user are important in shaping the acute drug effects and may have an influence on some longer-term potential outcomes. CONCLUSION This summary and discussion raises a number of practical and theoretical issues. It seems apposite to close by suggesting, as did one Conference participant, that clinical and psychological cannabis research is becoming more and more a study of 'Life'. The implication is that cannabis research must from now on be judged by the normal criteria of the clinical, behavioural and social sciences. Thus, more imaginative and comprehensive types of

theoretical explanations and higher standards of empirical evidence are called for. But, in the last analysis, success will rest on our ability to ask the right questions.