Psychological Medicine Volume 33(5), July 2003, pp 897-905

Autobiographical memory and theory of mind: evidence of a relationship in schizophrenia

CORCORAN, RHIANNON1; FRITH, CHRISTOPHER D.
From the Department of Psychology, University of Manchester; and Wellcome Department of
Cognitive Neurology, Institute of Neurology, University College London
1
Address for correspondence: Dr Rhiannon Corcoran, Department of Psychology, University of
Manchester, Oxford Road, Manchester M13 9PL.
ABSTRACT
Background. It has been proposed that inferences about the mental states of others are drawn after a
referral to autobiographical memory. This study explored the relationship between autobiographical
memory retrieval and performance on tests of theory of mind in people with schizophrenia.
Method. Fifty-nine people with a DSM-IV diagnosis of schizophrenia and 44 healthy participants
matched for age, sex and estimated IQ were given the Autobiographical Memory Interview (AMI),
two measures of theory of mind (ToM) and a logical memory test.
Results. There was clear evidence that the people with schizophrenia were under-performing on all
tasks. Within the schizophrenia sample, robust relationships existed between the total scores achieved
on the AMI and the ToM tasks. Furthermore, there was evidence that the participants with
schizophrenia had a tendency to recollect odd or negative events when prompted by the standard
questions of the AMI.
Conclusions. The results of this study indicate that when people with schizophrenia attempt to think
about the beliefs and intentions of others they use analogical reasoning. Whether this approach is also
adopted by other clinical and normal adult groups is a question for future research.
INTRODUCTION
Autobiographical memory in schizophrenia
In 1990, McKenna et al. argued that the memory impairment of schizophrenia was similar in extent
and nature to that seen in the amnesic syndrome. This position was reinforced in a later study by the
same authors (Tamlyn et al. 1992). It is now widely acknowledged that people with schizophrenia do
have poor memories (Saykin et al. 1991; McKenna et al. 1995) and that they also often show
impoverished executive skills (Shallice et al. 1991; Frith, 1992). Given that schizophrenia is widely
understood as reflecting a disintegration of self-awareness (Bleuler, 1950; Frith, 1992) and that
delusions typically refer to the functioning of the self in the social world, it is surprising that there has
been little exploration of autobiographical memory (AM) in schizophrenia.
In the first of three studies Baddeley et al. (1995) used a small group cognitive neuropsychological
approach to explore the profiles of schizophrenic patients with and without delusions. Few differences
were found between the deluded and the non-deluded patients with the exception of their performance
on the AMI (Kopelman et al. 1990). Interestingly, the patients who were not currently deluded
performed worse on the autobiographical incidents section of the task compared to those with active
delusions. The deluded patients had a normal capacity to recollect events from the past but they
tended to recall 'odd' events. Thus, both the qualitative and the quantitative aspects of
autobiographical retrieval were implicated as being of significance to schizophrenia and the nature of
the dysfunction appeared to be related to the presence of symptoms.
In a study exploring the characteristics of remote memory in schizophrenia, Feinstein et al. (1998)
found that the temporal gradient of AM in a small sample of people with schizophrenia was different

to that seen in normal samples. A U-shaped function was found in schizophrenia indicating poor
recollection of early adult life, a time when most people with the condition first present with clear
symptoms. This pattern of dysfunction was reported as unique in the neuropsychiatric literature.
Kaney et al. (1999) explored the nature of autobiographical recollection in people with persecutory
delusions using Williams & Broadbent's (1986) method. These authors found that, like people with
depression, deluded participants had a tendency to produce 'over-generalized' memories of a
categorical nature instead of recollecting a specific event that is situated in time and place. Kaney and
colleagues interpreted this over-general memory effect as serving a defensive function so that past
events of a negative nature do not get recalled. In support of this, the authors referred to the other
biases that arise when depressed or deluded people attempt to recall episodic information (Kaney et
al. 1992; Bentall et al. 1995; Williams et al. 1997). The authors also noted the apparent importance of
histories of abuse to the recollection of these kinds of memories (Kuyken & Brewin, 1995).
Autobiographical memory, social cognition and schizophrenia
Theoretical links have been proposed between the retrieval of events from AM and the ability to infer
other people's mental states. Corcoran (2000, 2001) has suggested that the mentalizing process may
begin with a referral to AM in order to retrieve any analogous event from the past that might help to
disambiguate the present scenario. The proposal is similar to the simulation model of theory of mind
(ToM) proposed by Harris (1992). Briefly, Corcoran has proposed that the ToM problems seen in
people with schizophrenia can be explained by the presence of dysfunctions of AM or of social
conditional reasoning. In particular, it was argued that people with negative features, who often have
extensive cognitive deficits (Frith, 1992) and who also may have prolonged histories of social
isolation (Foerster et al. 1991) will have impoverished AM stores. Drawing on Baddeley et al.'s
(1995) finding, it was also suggested, that a bias in the retrieval of negative or odd events from AM
might characterize people with persecutory delusions. The study reported here explored the strength
and nature of AM in people with DSM-IV schizophrenia and the relationship between AM
impairment and ToM ability.
METHOD
Subjects
Fifty-nine people with a DSM-IV diagnosis of schizophrenia and 44 normal subjects contributed data
to this study. The majority of the participants with schizophrenia were men who lived independently
or semi-independently in the community. All, with the exception of three, took antipsychotic
medication, with most receiving depot or atypical neuroleptics. The normal volunteers were mostly
recruited after the clinical sample was collected using various methods and comprised largely of men
working in semi-skilled or unskilled employment. The demographic and medical details of the
samples can be seen in Table 1.
The current symptomatology of the schizophrenic sample was assessed prior to testing using the
Present State Examination (PSE version 9; Wing et al. 1974). According to their responses and
behaviour during this interview, the patients were classified according to the primary sign or symptom
type displayed or reported. Ten participants had negative signs of poverty and social withdrawal or
blunted affect with or without active positive symptoms. In ten there was evidence of formal thought
disorder and delusions. Sixteen had delusions of persecution, reference or misidentification (paranoid
delusions) with no behavioural signs. Eight reported delusions of control or influence (passivity
delusions) in the absence of paranoid delusions and behavioural signs. The remaining 15 were free of
signs and symptoms on the day of testing and reported being 'well' for at least 2 weeks (see Table 1).

The tests
Each subject performed the following five tasks.
1 The Hinting Task (Corcoran et al. 1995)
Ten short vignettes were read out to, and left in front of, the participants. The vignettes end with one
of the characters dropping a hint to the other character. The participant is asked what the character
really means by their utterance. An appropriate inference scores 2 points. If no inference is drawn, a
second more obvious hint is added and the participant is asked to attempt to infer the intention a
second time. A correct response at this stage is given a score of 1 and an incorrect response is scored 0
and the next item is introduced. The task has good face validity for this participant group and has
proven sensitive to 'theory of mind' difficulties in a number of studies to date (Corcoran et al. 1995;
Pickup, 1997; Swarbrick, 2000).
2 The theory of mind stories (after Frith & Corcoran, 1996)
Four short stories with accompanying cartoon pictures were read out to the participants. These
explored first and second order false belief and deception. A first order false belief story is one in
which a character has a false belief about the state of the world while a second order false belief story
is one in which a character has a false belief about another's mental state. The deception stories
involved bluff (first order) or double-bluff (second order). After each story, two questions were
posed. One assessed the inference of mental state while the other was a reality control question. If an
incorrect answer was given to this reality question, the answer given to the ToM question was
automatically marked incorrect. While this might appear to be a rather strict treatment of the data it
was considered preferable to excluding cases and reducing the sample size. In fact, it was extremely
unusual for cases to fail the reality question while passing the ToM question.
3 The autobiographical incidents section of the AMI (Kopelman et al. 1990)
Here, participants are asked to recollect three events from different stages of life: childhood, early
adult and recent life (within the past year). Each episode recalled is scored such that a score of 3 is
given to an event specific in time and place, 2 is given for a specific event for which detail about time
and place is not given or a personal but non-specific event, 1 is given for a vague personal memory
and 0 for a recollection from semantic memory or when nothing can be recalled. Each subsection is
scored from a total of 9 with a grand total of 27. For the purposes of the present study each episode
was also coded using a simple dichotomous scale. A score of 0 was used to refer to an event which
was in no way unusual and 1 was used to represent an event that was 'odd' or negative. The events
reported by the participants that had been allocated a code of 1 by the main investigator were
subsequently presented to two nave raters who were asked to state whether the event being described
was normal, negative or odd. There was full agreement for all events described. A sample of the 'odd'
or negative responses are given in the Appendix.
4 The Quick Test (Ammons & Ammons, 1962)
This is a measure of word-to-picture matching that provides an estimate of the level of general
functioning based upon the Wechsler Adult Intelligence Scale (Wechsler, 1955). The participant is
asked to state for each word which of four pictures the word goes best with. As such it is a measure
that assesses word - knowledge as well as more 'fluid' contextual analysis. As the test measures the
kinds of general skills that might be used when answering ToM questions, it is a well-suited measure
of general functioning for this study.
5 The Story Recall subtest from the AMIPB (Coughlan & Hollows, 1985)
This was used to assess recall of a short prose passage immediately after hearing it. Since people with
schizophrenia have poor episodic memories and socio-cognitive tasks require the recollection of
information presented in prose passages, it is important to explore the potential influence of this
variable.

RESULTS
Demographic characteristics
The demographic characteristics of the samples can be seen in Table 1 where the medical details of
the participants with schizophrenia are also given. The samples are matched for sex ([chi] 2 = 0.47, df
= 1, NS) and age (schizophrenia mean = 40.5 10.0; control mean = 40.0 13.7).
Performance on the cognitive tasks
The two groups do not differ in terms of estimated intelligence level (schizophrenia mean = 100.7
10.6; control mean = 102.7 5.8; t(equal variances not assumed) = 1.19, df = 93.4, NS). The IQ data
can be seen in Table 1.
Table 2 gives the details of the performance of the schizophrenic and normal sample performance on
the other four tasks. Highly significant differences were found on the performance of all of these tasks
with the schizophrenic group performing worse than the normal sample. For the AMI this held for the
total score and the scores on each of the subsections (Hinting task, t = 5.535, df = 68.8, P < 0.001;
ToM stories, t = 7.055, df = 68.7, P < 0.001; AMI total, t = 6.429, df = 66.4, P < 0.001; AMI
childhood, t = 4.303, df = 72.7, P < 0.001; AMI early adult, t = 6.026, df = 63.5, P < 0.001; AMI
recent life, t = 5.981, df = 63.7, P < 0.001; Immediate logical memory t = 12.8, df = 101, P < 0.001.
All figures are for two-tailed tests).
Comparing the data on AM to the cut-off scores published in the AMI test manual, it is clear that the
greatest deficit exists in those participants with prominent negative features with 60% of these
patients having probably or definitely abnormal autobiographical retrieval. This compares to 40% of
the thought disorder group, none of the patients who report delusions only and 13% of those currently
in remission.
The relationship between the tests
Given that the normal sample were performing at or close to ceiling on the ToM tasks and the AMI,
reliable indices of covariation could not be obtained for the normal group.
The Hinting Task and the ToM stories
The Pearson correlation between hinting task performance and ToM stories performance in the
schizophrenia group stands at 0.63 (P = 0.01 level, two-tailed test). Thus, approximately 40% of the
variance on each task can be accounted for by the reliance on the same skills in their performance.
When the impact of immediate logical memory performance and estimated IQ on this relationship are
accounted for using a partial correlation the figure falls to 0.41 with a P value = 0.001.
The hinting task and the AMI
Fig. 1 shows the relationship that exists between the hinting task and the AMI total score in the
schizophrenia sample. The Pearson correlation stands at 0.66 and is highly significant. Thus,
approximately 44% of the variance on the Hinting task in this sample can be accounted for by the
ability to recollect events from the past as assessed by the AMI. When the impact of immediate
logical memory and estimated IQ are accounted for using a partial correlation, the figure falls to 0.48
but remains highly significant (P < 0.001).
The relationship between the ToM stories and the AMI
The Pearson correlation between the ToM stories and the AMI total score stands at a modest 0.42,
which is significant at P = 0.01 level two-tailed. Thus, approximately 17% of the variance on ToM
stories performance can be accounted for by the ability to recollect events from the past as measured
by the AMI. The partial correlation falls to 0.29 with a P value of < 0.05 after accounting for
immediate logical memory and general ability.

The nature of the recollections

Table 3 shows data describing the number and percentage of the schizophrenic sample as a whole, and
when divided according to their primary presenting sign/symptom, who reported at least one
markedly 'odd' recollection. At least one odd recollection was reported by 42% of the sample. This
compares with the normal sample who reported no 'odd' recollections ([chi] 2 = 12.39, df = 1, Fisher's
exact P < 0.001). The 'odd' recollections came almost exclusively from the childhood section of the
AMI. According to these data however, the recollection of odd events from the past is not a particular
feature of those with paranoid delusions or delusions generally since a Kruskal-Wallis test indicated
that no significant differences existed between the subgroups in the number of odd or negative events
reported ([chi]2 = 2.76, df = 4, P = 0.59).
The hypothesis that, those participants with schizophrenia who report some odd or negative events
from their past lives will perform more poorly on ToM tasks compared with those who do not report
these kinds of events, was explored using t tests. There was no evidence that this was the case
(t(hinting task) = -0.019, df = 57, NS; t(ToM stories) = -0.79, df = 57, NS).
DISCUSSION
The study has added significantly to our knowledge of ToM and AM in schizophrenia. First, there is
clear evidence of impoverished performance on two distinct tests of theory of mind in this sample of
people with schizophrenia. Secondly, diminished retrieval from AM is characteristic of people with
schizophrenia. Thirdly, there is evidence that people with schizophrenia have a tendency to retrieve
more markedly odd events when asked to recollect events from their past lives than do normal
participants. However, this tendency is not associated with any particular symptom type. Finally, there
is evidence of a relationship in the schizophrenia group between the strength of autobiographical
retrieval and performance on tests of mental state inference with a strikingly strong relationship
apparent between the comprehension of hints and the ability to recollect events from the past. These
four distinct areas of evidence have implications for the studies reviewed earlier.
Inferring mental states
Problems inferring intentions behind veiled speech acts and understanding situations of false belief
and deception have a tendency to arise together in schizophrenia. This supports the argument that
these two tasks do assess the same underlying skill, namely the ability to think about the mental states
of others.
A closer inspection of the ToM stories data indicted that the sample found the second order deception
question problematic with 86.5% of the sample failing it. Given that this study shows a strong
relationship between these two types of ToM task, perhaps it would be sufficient to focus on the
appreciation of higher order deception or double bluff situations in future assessments of ToM in
schizophrenia.
Impoverished autobiographical memory
This study, like the few that have preceded it, has found that poor AM is a feature of people with
schizophrenia. Unlike Feinstein et al. (1998) however, there was no evidence that early adult life was
differentially compromised in this sample. In fact, each section of life defined by the AMI seems to be
equally affected. Furthermore, this study has found that AM impairment is most profoundly affected
in those with negative behavioural signs. Patients in remission and those with active delusions
perform at about the same level on the AMI. This finding is contrary to Baddeley et al.'s (1995)
finding. The results are, in part, consistent with those of Kaney et al. (1999) who found evidence of
over-generalized memory in schizophrenia. The scoring system of the AMI is sensitive to the ability
to place a recollection in time and to give adequate personal details about it. Thus, poor scores on the
measure can be regarded as indicative of over-generalized memories. However, Kaney et al. sampled
people with persecutory delusions. In this study the most impoverished memory was seen in those
with negative signs. It is not clear whether the group of people studied by Kaney et al. also had
negative signs.

Retrieval of odd events

Again contrary to Baddeley et al. (1995) there was no evidence that markedly 'odd' events tended to
be reported by those with current delusions nor were they a feature of those with exclusively paranoid
type delusions. Corcoran's (2000) suggestion that poor mentalizing in paranoia may be a consequence
of a biased retrieval process is not supported by this data. It was noted that the odd or negative
recollections apparent in the schizophrenia sample tended to come from childhood and it was, in some
cases, likely that these were delusional memories or confabulations (Nathaniel-James & Frith, 1997).
A more in-depth exploration of the incidence of these kinds of 'recollections' in schizophrenia would
be informative, as would a large-scale study of the incidence of reports of potentially traumatic events
in childhood.
Relationship between autobiographical memory and ToM
The relationship between AM and ToM was found to be particularly strong when ToM was measured
using the hinting task. There seem to be two related explanations as to why the relationship is most
powerful for this ToM task. First, the task has 'ecological validity'. The hinting examples involve, for
the most part, everyday scenarios for which clear analogues probably exist in autobiographical
memory. Secondly, both inferring the meaning behind hints and autobiographical retrieval require
reconstructive skills. While context is to an extent preserved in AM, precise contextual details would
often not be available. In the scenes provided in the hinting task only limited contextual detail is
provided and so context must be constructed to some extent. However, in the ToM stories all of the
contextual detail that is required to answer the questions is provided. So, while the ToM stories can be
answered using deductive skills as the problems are well defined, the ill-defined problems of the
hinting task can only be answered using inductive reasoning. However, there was a modest correlation
between ToM stories and AMI and perhaps with more stories a stronger relationship would be shown.
One of the most problematic aspects of empirical work exploring social cognition in adult samples is
the fact that normally-functioning adults are, by definition, experts. Therefore, normal control groups
will perform close to ceiling, causing difficulties analysing the data. Given the limited range of scores
on both ToM and AM in this normal sample, analyses aimed at exploring the extent of the relationship
between the measures within this group was not performed. We cannot therefore say whether the
relationship that exists in schizophrenia also exists in normally-functioning adults. This is an
important issue because cognitive accounts of psychotic symptoms often make the assumption that
the way psychotic patients process information is similar to how normal adults process the
information. If people with psychosis are processing information differently then it follows that we
need to re-think the kinds of models we use to explore the cognitive basis of psychosis.
The above issue has important implications for the work described here. If people with schizophrenia
try to understand the mental-states of others using skills that are different from those invoked by the
'default' brain, then it is not surprising that their ToM judgements can be poor. Furthermore, if the
skills that they rely upon to achieve these judgements are themselves weak, as the memory skills
focused on in this study are, their ToM judgements will be even more unreliable. This issue raises a
clear limitation of this study that is the lack of a psychiatric control group. A particularly appropriate
psychiatric control group would have been people with unipolar depression who have overgeneralized memories (Williams & Broadbent, 1986). As this group of patients is characterized by
weak autobiographical retrieval it may have provided adequate data to enable an exploration of the
relationship between ToM and AM in a sample without schizophrenia. Data from this particular group
is being gathered in an ongoing study.
Conclusion
The study has highlighted a number of important issues. In particular, a robust relationship between
the abilities to recollect events from the past and to understand the mental states of others has been
demonstrated in the schizophrenia sample. The schizophrenia data therefore support the model of
mentalizing proposed by Corcoran (2000) suggesting that mentalizing is accomplished using
inductive or analogical reasoning since to solve these social inference problems we rely upon

information gathered in the past. This proposal has similarities to Harris's (1992) simulation theory of
mentalizing.
Several areas for future work are highlighted. First, studies using more challenging tasks could
perhaps achieve a better range of scores enabling reliable tests of the relationship in a normallyfunctioning adult sample. If designing such tests proves to be too troublesome, it may be possible to
explore the relationship in young children who are at critical ages for ToM understanding. Another
approach would be to gather data from amnesics or to test the relationship in a sample of people with
high-functioning autism. Finally, studies exploring the relationship between mentalizing tasks and
tasks that require inductive reasoning would be enlightening.
This work was funded by a project grant awarded by the Wellcome Trust (ref 046366/Z). We are
grateful to all of the Healthcare trusts who supported this work and to the participants who supplied
the data.
REFERENCES
Ammons, R. B. & Ammons, C. H. (1962). The Quick Test. Psychological Test Specialists: Missoula,
MT.
Baddeley, A., Thornton, A., Chua, S. E. & McKenna, P. (1995). Schizophrenic delusions and the
construction of autobiographical memory. In Remembering Our Past: Studies in Autobiographical
Memory (ed. D. C. Rubin), pp. 384-428. Cambridge University Press: New York.
Bentall, R. P., Kaney, S. & Bowen-Jones, K. (1995). Persecutory delusions and recall of threat-related,
depression-related and neutral words. Cognitive Therapy and Research 19, 331-343.
Bleuler, E. (1950). Dementia Praecox (Translated by J. Zinkin). International Universities Press: New
York.
Corcoran, R. (2000). Theory of Mind in other clinical samples: is a selective theory of mind deficit
exclusive to autism? In Understanding Other Minds. Perspectives from Autism and Developmental
Cognitive Neuroscience, 2nd edn (ed. S. Baron-Cohen, H. Tager-Flusberg and D. J. Cohen), pp. 391421. Oxford University Press: Oxford.
Corcoran, R. (2001). Theory of mind in schizophrenia. In Social Cognition and Schizophrenia (ed. D.
Penn and P. W. Corrigan), pp. 149-174. APA: Washington, DC.
Corcoran, R., Mercer, G. & Frith, C. D. (1995). Schizophrenia, symptomatology and social inference:
investigating theory of mind in people with schizophrenia. Schizophrenia Research 17, 5-13.
Coughlan, A. K. & Hollows, S. E. (1985). Adult Memory and Information Processing Battery.
Coughlan: Leeds. [Context Link]
Feinstein, A., Goldberg, T. E., Nowlin, B. & Weinberger, D. R. (1998). Types and characteristics of
remote memory impairment in schizophrenia. Schizophrenia Research 30, 155-163.
Foerster, A., Lewis, S., Owen, M. & Murray, R. (1991). Pre-morbid adjustment and personality in
psychosis. Effects of sex and diagnosis. British Journal of Psychiatry 158, 171-176.
Frith, C. D. (1992). The Cognitive Neuropsychology of Schizophrenia. Lawrence Erlbaum Associates:
Hove.
Frith, C. D. & Corcoran, R. (1996). Exploring theory of mind in people with schizophrenia.
Psychological Medicine 26, 521-530.
Harris, P. L. (1992). From simulation to folk psychology: the case for development. Mind and
Language 7, 120-144. [Context Link]
Kaney, S., Wolfden, M., Dewey, M. E. & Bentall, R. P. (1992). Persecutorydelusions and the recall of
threatening and non-threatening propositions. British Journal of Clinical Psychology 31, 85-87.

Kaney, S., Bowen-Jones, K. & Bentall, R. P. (1999). Persecutory delusions and autobiographical
memory. British Journal of Clinical Psychology 38, 97-102.
Kopelman, M. D., Wilson, B. A. & Baddeley, A. D. (1990). The Autobiographical Memory Interview.
Thames Valley Test Company: Bury St Edmunds. [Context Link]
Kuyken, W. & Brewin, C. R. (1995). Autobiographical memory functioning in depression and reports
of early abuse. Journal of Abnormal Psychology 104, 585-591.
McKenna, P. J., Tamlyn, D., Lund, C. E., Mortimer, A. M., Hammond, S. & Baddeley, A. D. (1990).
Amnesic syndrome in schizophrenia. Psychological Medicine 20, 967-972.
McKenna, P. J., Clare, L. & Baddeley, A. D. (1995). Schizophrenia. In Handbook of Memory
Disorders (ed. A. D. Baddeley, B. A. Wilson and F. N. Watts), pp. 271-292. John Wiley: Chichester.
[Context Link]
Nathaniel-James, D. A. & Frith, C. D. (1996). Confabulation in schizophrenia: evidence of a new
form. Psychological Medicine 26, 391-399. Ovid Full Text
Pickup, G. (1997). The Representation of Mental States in Schizophrenia. Ph.D. thesis. University of
London. [Context Link]
Saykin, A. J., Gur, R. C., Gur, R. E., Mozley, P. D., Mozley, L. H., Resnick, S. M., Kester, D. B. &
Stafniak, P. (1991). Neuropsychological impairment in schizophrenia: selective impairment in
memory and learning. Archives of General Psychiatry 48, 618-624.
Shallice, T., Burgess, P. W. & Frith, C. D. (1991). Can the neuropsychological case study approach be
applied to schizophrenia? Psychological Medicine 21, 661-673.
Swarbrick, R. (2000). Social Cognitive Processes in Persecutory Delusions. Ph.D. thesis. University
of Manchester. [Context Link]
Tamlyn, D., McKenna, P. J., Mortimer, A. M., Lund, C., Hammond, S. & Baddeley, A. D. (1992).
Memory impairment in schizophrenia: its extent, affiliations and neuropsychological character.
Psychological Medicine 22, 101-115.
Wechsler, D. (1955). Manual of the Wechsler Adult Intelligence Scale. Psychological Corporation:
New York. [Context Link]
Williams, J. M. G. & Broadbent, K. (1986). Autobiographical memory in attempted suicide patients.
Journal of Abnormal Psychology 95, 144-149.
Williams, J. M. G., Watts, F. N., MacLeod, C. & Mathews, A. (1997). Cognitive Psychology and
Emotional Disorders, 2nd edn. Routledge: London.
Wing, J. K., Cooper, J. E. & Sartorius, N. (1974). Measurement and Classification of Psychiatric
symptoms. An Instruction Manual for PSE and the Catego Program. Cambridge University Press:
Cambridge.
APPENDIX
A sample of recollections given by the participants with schizophrenia that were coded as 'odd' or
negative in quality. (The sessions were not tape-recorded but reported episodes were summarized in
writing by the experimenter at the time of telling.)
AMI Question Try to remember something that happened to you before you ever went to school.
A man with negative signs replied that when he was four he was out with his mother and aunt and saw
a women hit by a car when she was running for a bus. The woman died.
A young woman with paranoid delusions remembered walking up the stairs eating coal. She reported
eating all sorts of things she should not have.

A woman with passivity features answered: 'I used to try to write on hats.'
A young man with passivity features remembered that at the age of two and a half his dad took them
to the children's home and he cried when his dad left.
A young man in remission had a first memory of his ginger cat being kicked off the balcony by one of
his dad's enemies.
A man in remission recollected a first memory at the age of four of crying lots and playing with the
other children at the children's home.
A man in remission reported that his first memory was of being at nursery and being told that his
sister had died.
AMI Question Try to remember something that happened to you while you were at primary school.
A man with formal thought disorder remembered that his teacher had 'smashed me in the face' when
he was nine and broken his nose. This happened because he had not answered his name when the
register was called. Instead he had answered the name of the person whose body he was in. He had
been taken away into a quiet room, hit in the face and lost consciousness.
A man with paranoid delusions recalled that at the age of five he was run over by a car that went right
over him. He got up and ran away. He added that the wheels had gone right over him but he was 'like
rubber' and only got a few scratches.
AMI Question Try to remember something that happened to you while you were at secondary school.
A man with prominent negative signs recalled being assaulted by one of his teachers in the classroom
because he had not done his homework. He was told to wait behind after class when the teacher
shouted at him, pinned him against the blackboard, ripped his shirt and picked him up by the neck.
A man with formal thought disorder answered that at the age of 11 he had found a room in the house
of which he said there was 'perfection in that room'. He added: 'They say I still live there'.
A man with formal thought disorder reported that the headmaster used to molest boys in the showers.
The boys used to give themselves 'chicken scratches' and he used to make his nose bleed to get off
lessons.
A man with paranoid delusions recalled being 'done for GBH'. A bunch of skinheads had told him
they were going to beat him up. He went out with a knife and his opponent had a 'heavy duty chain'.
He got done for GBH and the other boy got done for ABH.
One man in remission recalled his auntie's friend 'feeling him all over' when he was 13.
One man in remission reported that aged 12 he cried in his class because he could not do his maths.
The teacher was unpleasant and called him a baby. After this the children picked on him and he had to
change school.