Sunteți pe pagina 1din 7

Applied Neuropsychology Copyright 2001 by

2001, Vol. 8, No. 3, 148–154 Lawrence Erlbaum Associates, Inc.

A Comparison of WAIS–R Profiles in Adults With High-Functioning


Autism or Differing Subtypes of Learning Disability A COMPARISON OFGOLDSTEIN
WAIS–R PROFILES
ET AL.

Gerald Goldstein
VA Pittsburgh Healthcare System and School of Medicine, University of Pittsburgh,
Pittsburgh, Pennsylvania, USA
Sue R. Beers, Don J. Siegel, and Nancy J. Minshew
School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

To examine cognitive differences among adults with differing developmental disorders, a com-
parison of Wechsler Adult Intelligence Scale–Revised profiles was made with samples of 35 in-
dividuals with high-functioning autism (HFA) and 102 individuals with adult learning
disability (LD). All participants had Verbal and Performance IQ scores of 70 or higher. The LD
group was divided into 3 subtypes based on relative achievement levels in mechanical reading
and arithmetic. The group with HFA had a profile characterized by a high score on Block De-
sign with a low Comprehension score. The HFA group most resembled the LD subtype that had
superior achievement in reading relative to arithmetic, with the exception of their poor perfor-
mance on measures of social perception and judgment. Results are discussed in terms of the
substantial differences in cognitive structure between these 2 neurodevelopmental disorders
and are considered in the context of the learning deficits reported for Asperger’s Disorder and
nonverbal learning disability.

Key words: autism, learning disability subtypes, WAIS–R

Clinical and neuropsychological prototypes in the tory abilities (Francis, Espy, Rourke, & Fletcher,
form of Wechsler Adult Intelligence Scale (WAIS) 1991). Although the intelligence test may be supple-
profile configurations of the scaled scores (Wechsler, mented with additional psychoeducational and
1944, 1955, 1981, 1998) have been described from neuropsychological tests, it provides unique informa-
the time of Rapaport, Gill, and Schafer (1945) to more tion about complex abilities in a number of areas,
recent times (Reitan & Wolfson, 1993). Numerous such as working memory or complex spatial prob-
group comparisons have been reported, demonstrating lem-solving skills.
often substantially different intellectual profiles in nu- In this research, we provide a comparison between
merous developmental and psychiatric disorders. Ad- high-functioning autism (HFA) and adult learning dis-
ministration of an intelligence test is often advised or ability (LD). Such a comparison is of interest because
required for diagnosis and description of the client be- both are neurodevelopmental disorders that emerge
ing assessed, not only to rule out mental retardation early in life but persist into adulthood (McCue &
but also to provide information about the profile of Goldstein, 1991). Shea and Mesibov (1985) believed
cognitive abilities that may contribute to the academic that there may be an unclear boundary between HFA
deficit or to provide data about potentially compensa- and severe LD and remarked on the resemblance be-
tween the core deficits of autism and those of the social
and nonverbal developmental disorders. An uneven
This research was supported by National Institute of Neurologi- profile of skills involving aspects of language, cogni-
cal Disorders Grant NS33355 and the Medical Research Service,
tion, and social adjustment is common to each disorder.
Department of Veterans Affairs.
Requests for reprints should be sent to Gerald Goldstein, VA HFA and LD are distinguished from each other, how-
Pittsburgh Healthcare System, 7180 Highland Drive (151R), Pitts- ever, by differences in the nature of deficient language
burgh, PA 15206, USA. E-mail: ggold@nb.net and communication and academic and social skills.

148
A COMPARISON OF WAIS–R PROFILES

Extensive neuropsychological research has demon- except for Digit Symbol, approaching the average
strated substantial differences in the cognitive func- range. However, the use of a heterogeneous LD sample
tioning of children and adults with HFA and may have obscured the possible existence of differing
individuals with LD. Individuals with autism have been profiles among LD subtypes.
characterized alternatively as having executive dys- A comparison between HFA and LD in general may
function (Griffith, Pennington, Wehner, & Rogers, be misleading because although HFA is a reasonably
1999; Ozonoff, 1995) or impaired complex information well-defined condition, LD is not a unitary disorder.
processing (Minshew, Goldstein, & Siegel, 1997) as Various subtyping systems have been developed based
the core dysfunction. Their weakness lies in conceptu- on characteristic behavioral and psychometric patterns
alizing and organizing information. However, they typ- (Fletcher & Satz, 1985; Rourke, 1985, 1991). Rourke
ically have intact mastery of the mechanical aspects of (1982) developed an empirically based system that clas-
language, including phonetics, spelling, and simple cal- sifies individuals with LD on the basis of performance on
culation, and often remarkably good spatial skills reading and arithmetic tests. Utilizing the Wide Range
(Minshew, Goldstein, Muenz, & Payton, 1992). Indi- Achievement Test–Revised (WRAT–R; Jastak &
viduals with LD may have fully adequate nonverbal ab- Wilkinson, 1984), it encompasses individuals uniformly
stract reasoning abilities and spatial skills but are deficient on the Reading and Arithmetic subtests
typically significantly impaired in phonetics and re- (Global group), individuals who are poor in Reading but
lated mechanical aspects of language (Ozols & Rourke, better in Arithmetic (A > R group), and individuals who
1991). Individuals with autism typically have substan- are at least average in Reading, but poor in Arithmetic (R
tial difficulties with the pragmatic and metaphoric as- > A group). The Global group is characterized by defi-
pects of language (Brook & Bowler, 1992) and may cient reading and mathematical achievement relative to
often show a pattern marked by fully intact oral reading general intellectual level. The A > R group typically does
but poor reading comprehension (Frith & Snowling, not exceed average-level arithmetic ability, but arithme-
1983; Minshew, Goldstein, Taylor, & Siegel, 1994; tic performance is substantially better than reading.
Whitehouse & Harris, 1984). In a direct demonstration Thus, individuals in this group are often severely dys-
of such differences in the adult population, Rumsey and lexic. The R > A group is characterized by average-level
Hamburger (1990) found no differences between reading skill but is deficient in arithmetic. Rourke’s
groups of average-IQ adults with either autism or dys- (1985) research showed that these subtypes are associ-
lexia on measures of perceptual organization, word flu- ated with differing neuropsychological characteristics.
ency, tactile perception, and motor speed and Summarizing this work briefly, the R > A group is char-
coordination. The performance of these groups on mea- acterized by relatively good verbal abilities and rela-
sures of simple language and visual–spatial skills was tively poor spatial, fine rapid psychomotor, and higher
similar. The group with autism performed more poorly conceptual skills. The A > R group has the reverse pat-
on tests of verbal and nonverbal reasoning, comprehen- tern. Good spatial and relatively poor linguistic abilities
sion, and recall of complex verbal and visual material, also characterized the group that was disabled in both
but they outperformed the dyslexia sample with respect reading and arithmetic.
to auditory recall of simple information and on tests of This investigation compared WAIS–R performance
basic academic skills. Thus, the literature suggests the of adults with HFA and those with the three Rourke LD
presence of important cognitive differences between subtypes to investigate differences in the WAIS–R
individuals with HFA and with LD. subtest profile of each group. A major concern was that
Studies using Wechsler Adult Intelligence the consideration of LD subtypes may substantially re-
Scale–Revised (WAIS–R; Wechsler, 1981) subtest duce the confounding that may result from a compari-
profiles also suggest the presence of such a difference. son with a heterogeneous LD sample. We hypothesized
A study and literature review of WAIS–R performance that adults with HFA and LD would have distinct
in adults with autism noted the high prevalence of a pat- WAIS–R profiles. In regard to subtest patterns, an ele-
tern of relatively elevated scores on Block Design, with vated Block Design and Digit Span with depressed
diminished performance on Comprehension (Siegel, Comprehension was anticipated in the autism group.
Minshew, & Goldstein, 1996). In the case of a hetero- Many studies have demonstrated that individuals with
geneous sample of adults with LD, McCue and HFA have intact mechanical reading abilities
Goldstein (1991) described a WAIS–R profile charac- (Minshew et al., 1994). The WRAT–R Reading subtest
terized by relatively low scores on Information, Vocab- involves only mechanical reading and not comprehen-
ulary, and Arithmetic, with Performance IQ subtests, sion, particularly in adults (Coneley & Kramer, 1989).

149
GOLDSTEIN ET AL.

We therefore predicted that the HFA profile would and were also excluded, as were cases of autism with a
most greatly resemble the R > A LD subtype profile, known etiology, such as Fragile-X syndrome. These
not because remarkably poor Arithmetic scores are an- participants constituted essentially the same sample as
ticipated but because of the excellent mechanical oral was used in the Siegel et al. (1996) study.
reading abilities often shown by individuals with HFA. The LD group consisted of 102 participants (34%
Indeed, children with autism have been characterized women). All were referrals to a neuropsychological
as having hyperlexia, or advanced ability to read and clinic from the Office of Vocational Rehabilitation or
pronounce written words (Frith & Snowling, 1983). other community agency. The sample included only in-
dividuals meeting Diagnostic and Statistical Manual of
Mental Disorders (3rd ed., rev. [DSM–III–R]; Ameri-
can Psychiatric Association, 1987) criteria for specific
Method developmental disorder. These criteria require marked
impairment of academic skills not explained on the ba-
Participants sis of mental retardation, inadequate education, or sen-
sory defects. Appropriate clinical and psychometric
To meet accepted diagnostic criteria for HFA, and to assessment documenting a history of academic diffi-
exclude individuals with mental retardation in both culties and significantly impaired performance on
groups, all participants had Verbal, Performance, and psychoeducational tests established the presence of
Full Scale WAIS–R IQ scores of 70 or above. The these criteria. Individuals with acquired academic defi-
group with HFA consisted of 35 participants (12% cits (e.g., acquired alexia) were excluded. This sample
women). The diagnosis of autism was established was utilized in several other studies (Goldstein, Katz,
through expert clinical evaluation in accordance with Slomka, & Kelly, 1993; Katz, Goldstein, Rudisin, &
accepted clinical descriptions of HFA (Minshew, 1996; Bailey, 1993) where more detail concerning the estab-
Minshew & Payton, 1988; Rapin, 1991; Rutter & lishment of the diagnoses is provided. To summarize,
Schopler, 1987) and two structured diagnostic instru- all participants met both DSM–III–R and State of Penn-
ments, the Autism Diagnostic Interview (ADI; sylvania Office of Vocational Rehabilitation require-
LeCouteur et al., 1989; Lord, Rutter, & LeCouteur, ments for an LD diagnosis. Demographic and academic
1994) and the Autism Diagnostic Observation Sched- data for the autism group and the three LD subtypes are
ules (ADOS; Lord et al., 1989). A review and rescoring presented in Table 1.
of the videotaped ADI and ADOS for all participants
with autism by a developer of these instruments docu-
mented diagnostic accuracy. Eligibility for inclusion
was dependent on consistency of diagnosis across all Procedure
assessments, with cases of disagreement excluded. In-
dividuals with impairments in social interaction and re- To classify participants in the group with an LD di-
stricted patterns of behavior, but with no clinically agnosis into the Rourke subtypes, we used standard
significant delay in language, cognition, and adaptive scores from the Reading and Arithmetic subtests of the
behavior, were considered to have Asperger’s Disorder WRAT–R. A presence or absence of a 15-point dis-
Table 1. Demographic and Psychometric Data for the High-Functioning Autism and Learning Disability Groups

Autism A>R R>A Global

Variable M SD M SD M SD M SD F

Age 26.62 9.23 32.17 10.99 19.13 2.66 23.75 8.16 7.80**
Years of Education 12.06 2.65 10.67 3.18 12.13 1.26 11.31 2.09 1.93
WAIS–R Verbal IQ 94.91 19.57 82.94 9.17 96.94 13.45 86.69 10.31 5.86**
WAIS–R Performance IQ 89.44 11.92 88.89 13.63 95.50 16.87 88.19 14.86 1.14
WAIS–R Full Scale IQ 91.88 16.05 84.83 10.62 96.13 15.01 86.28 10.31 3.98*
Reading Achievement 88.97 17.85 62.17 14.55 100.19 16.49 77.69 14.09 21.20**
Arithmetic Achievement 82.88 22.92 84.50 12.09 77.81 11.03 78.93 11.50 1.10

Note: WAIS–R = Wechsler Adult Intelligence Scale–Revised.


*
p < .01. **p < .001.

150
A COMPARISON OF WAIS–R PROFILES

crepancy between Reading and Arithmetic determined subtypes of the LD sample resulted in a statistically sig-
subtype membership. WAIS–R profiles were com- nificant main effect, F(33, 363) = 3.68, p < .001. The
pared with multivariate analysis of variance subtest profiles for the four groups are presented graph-
(MANOVA) followed by individual comparisons. ically in Figure 1. Univariate F tests indicated signifi-
Scheffe’s test (p < .05) was used for making multiple cant differences for all subtests except Similarities,
comparisons. Block Design, Object Assembly, and Digit Symbol.
Because autism is a rare disorder, it took some time Further analyses were accomplished by multiple com-
to obtain a sufficient sample, and data collection began parisons utilizing Scheffe’s test (p < .05). In the cases of
before the recently published Wechsler Adult Intelli- Information, Digit Span, and Arithmetic, the HFA
gence Scale–Third Edition (WAIS–III; Wechsler, group performed significantly better than the A > R and
1998) appeared. However, because of the stability of the Global group, but not the R > A group. For Compre-
structure and subtest composition of the different edi- hension, Picture Completion, and Picture Arrange-
tions of the Wechsler scales, and because the study in- ment, the R > A group performed significantly better
volved an evaluation of cognitive profile rather than than the HFA group. Although an overall significant
assessment of individuals relative to the most contem- difference was found for Vocabulary, there were no
porary norms, it was felt that WAIS–R data remain per- significant between-group differences. These data are
tinent. The user of the WAIS–III, which has an reported in Table 2.
extended set of subtests, would, of course, have to use
the WAIS–R subtest set to make comparisons with this
study. Discussion
WRAT–R data were not available for the HFA par-
ticipants. Because we wished to provide information This study compared WAIS–R subtest mean pro-
concerning their level and pattern of academic achieve- files between groups of adults with HFA and individu-
ment, we reported data from the Reading Decoding and als with three different academic profiles in an effort to
Mathematics Computation subtests of the Kaufman determine cognitive profile characteristics that discrim-
Test of Educational Achievement (Kaufman & inated among these groups. The adults with LD were
Kaufman, 1985). We provided descriptive data using divided into subtypes based on performance levels in
those tests for the HFA sample only for reference pur- mechanical reading and arithmetical computation. In
poses with regard to academic status, without any im- general, the HFA group performed better than the LD
plication concerning the presence or absence of LD in subtypes on the WAIS–R subtests. Examination of the
any individual case. The academic performance of the profiles suggests that the HFA group was most like the
HFA group was found to be commensurate with gen- R > A LD subtype with regard to both verbal and per-
eral intellectual level and not suggestive of substantial formance abilities. The HFA group, however, did more
difficulties in mechanical reading or calculation.

Results

Demographic and academic data are presented in


Table 1. There was not a statistically significant differ-
ence among groups in education, but there was in age.
The A > R group was significantly older than the R > A
and Global groups but was not older than the autism
group. The autism group was significantly older than
the R > A group. Correlations between age and the
WAIS–R scales were significant (p < .05) for the Infor-
mation, Vocabulary, Picture Arrangement, and Digit
Symbol subtests. However, analyses of covariance us-
ing age as the covariate did not substantially alter the
findings for these four subtests with regard to statistical
significance. A MANOVA of the 11 WAIS–R subtests Figure 1. WAIS–R subtest profiles for the autism and learn-
providing comparisons among the HFA and the three ing disability group

151
GOLDSTEIN ET AL.

Table 2. WAIS-R Subtest Scores for the High-Functioning Autism and Learning Disability Groups

Autism A>R R>A Global

Subtest M SD M SD M SD M SD F Scheffea

Information 9.03 4.26 6.28 1.74 7.38 2.39 6.68 2.28 5.97*** 1 > 2, 4
Comprehension 7.35 3.70 8.28 3.18 10.44 3.29 8.57 2.84 3.51* 3 > 1, 2, 4
Arithmetic 8.68 3.96 6.39 1.91 6.88 1.50 6.56 1.97 5.77* 1 > 2, 4
Similarities 8.53 3.52 7.72 3.01 10.38 2.87 8.72 2.98 2.16
Digit Span 9.00 3.58 5.94 1.95 8.06 1.91 6.76 1.75 9.38*** 1 > 2, 4
Vocabulary 8.41 3.60 6.39 2.30 8.56 2.56 6.97 2.17 4.11**
Digit Symbol 6.85 2.65 6.89 2.63 7.31 2.06 7.25 2.35 0.29
Picture Completion 7.41 2.08 8.06 3.35 9.94 2.69 8.43 2.58 3.56* 3>1
Block Design 9.21 3.21 8.11 2.45 9.00 2.53 8.26 2.60 1.19
Picture Arrangement 8.24 2.55 7.39 2.52 10.63 2.75 8.38 2.56 4.91** 3 > 1, 2, 4
Object Assembly 8.74 3.45 8.72 2.63 8.56 3.46 8.29 2.77 0.21
a
1 = Autism; 2 = A > R; 3 = R > A; 4 = Global.
*p < .05. **p < .01. *** p < .001.

poorly than the R > A subtype on subtests involving so- mental Arithmetic Disorder. That is, the members of
cial judgment, including Comprehension, Picture this group typically did not have dyslexia and appeared
Completion, and Picture Arrangement. Otherwise, to have relatively good social cognition, but they had a
there were no significant differences, or the HFA group rather specific difficulty with working with numbers.
outperformed the A > R group and the Global group but Klin, Volkmar, Sparrow, Cicchetti, and Rourke
not the R > A group. (1995) reported that the cognitive profile in
The results of this study suggest that individuals Asperger’s Disorder, but not HFA, is similar to non-
with HFA have a very different cognitive profile from verbal LD (Rourke, 1989). Findings from this study
LD subtypes. Individuals with HFA have relatively would therefore parallel their conclusions in that al-
strong points in the areas of semantic memory, atten- though we did not study individuals with Asperger’s
tion, and spatial–constructional abilities. They are rela- Disorder, the profile of the HFA participants con-
tively weak with regard to tasks requiring social tained many differences from what is observed in
judgment and perception. The strengths of the LD par- nonverbal LD. Particularly notable in the HFA profile
ticipants depend to some extent on the subtype. The R > was the absence of deficits in arithmetic and in spa-
A subtype had relatively good social judgment, verbal tial–constructional abilities. Perhaps the profile for the
reasoning, and spatial–constructional abilities. The A > HFA group can be best characterized as being similar
R subtype had generally low subtest scores but rela- to the R > A group with the exception of tasks requir-
tively good spatial–constructional abilities. Rourke ing social judgment, on which the HFA group demon-
(1989) described this subtype as being low functioning strated relatively impaired performance both in verbal
in the sense that although arithmetic may be better than and performance modalities. Thus, the difference is
reading, it is typically at a below-average level. The better conceptualized in terms of social versus
Global subtype has almost precisely the same WAIS–R nonsocial task requirements, rather than verbal versus
subtest profile as does the A > R subtype, but it is performance abilities.
slightly elevated. Although this study was conducted with an adult
These results indicate that the cognitive profile of in- sample, there are implications for children in a number
dividuals with HFA is most like what is seen in individ- of respects. With regard to autism, the course of this
uals with LD who are not significantly impaired in disorder is lifelong and there is no indication of a remis-
reading but who have academic difficulties in other ar- sion during adolescence or adulthood. However, there
eas. Examination of the R > A WAIS–R profile indi- is a developmental pattern characterized by differing
cates that this group did most poorly on those subtests courses for different cognitive abilities. The situation
that involve numbers, with other test scores ranging appears to be that individuals with autism do not deteri-
around the average level. They may be best character- orate but, rather, do not keep up with their normal age
ized as having the DSM–III–R diagnosis of Develop- peers for certain abilities. For basic language skills such

152
A COMPARISON OF WAIS–R PROFILES

as phonetics, they start out at normal levels and retain References


normal function into adulthood. However, for compre-
hension skills, they start out at a level comparable to American Psychiatric Association. (1987). Diagnostic and statisti-
age peers but lose ground as they grow older. Still other cal manual of mental disorders (3rd ed., rev.). Washington,
DC: Author.
abilities, particularly those involving conceptual rea- Brook, S. L., & Bowler, D. M. (1992). Autism by another name? Se-
soning, start out abnormally and remain that way. This mantic and pragmatic impairments in children. Journal of Au-
material is described in detail in a cross-sectional tism and Developmental Disorders, 11, 293–302.
age-difference study by Goldstein, Minshew, and Cohen, J. (1957). Factor analytically based rationale for Wechsler
Siegel (1994). In the case of LD, it is now no longer be- Adult Intelligence Scale. Journal of Consulting Psychology,
21, 451–457.
lieved that it is typically outgrown, and remarkable
Coneley, J. C., & Kramer, J. J. (1989). Wide Range Achievement
similarities have been found in academic profile and Test–Revised. In O. K. Buros (Ed.), Tenth mental measure-
academic performance in children and adults (McCue ments yearbook: Vol. 2 (pp. 897–905). Lincoln: University of
& Goldstein, 1991). Nebraska Press.
Lincoln, Courchesne, Kilman, Elmasian, and Allen Fletcher, J. M., & Satz, P. (1985). Cluster analysis and the search for
(1988) performed a factor analysis of the Wechsler learning disability subtypes. In B. Rourke (Ed.),
Neuropsychology of learning disabilities (pp. 40–64). New
scales based on a sample of individuals with autism. The York: Guilford.
first factor received high loadings from all of the Verbal Francis, D. J., Espy, K. A., Rourke, B. P., & Fletcher, J. M. (1991).
Scale subtests, the second from Block Design, Object Validity of intelligence scores in the definition of learning dis-
Assembly, and Digit Symbol/Coding, and the third from ability: A critical analysis. In B. Rourke (Ed.),
Picture Completion and Picture Arrangement. Thus, it Neuropsychological validation of learning disability subtypes
(pp. 15–44). New York: Guilford.
was quite discrepant from the conventional Verbal Com-
Frith, U., & Snowling, M. (1983). Reading for meaning and reading
prehension, Perceptual Organization, and Freedom for sound in autistic and dyslexic children. British Journal of
from Distractibility factors (Cohen, 1957). The factor Developmental Psychology, 1, 329–342.
structure obtained by Lincoln et al. (1988) would there- Goldstein, G., Katz, L., Slomka, G., & Kelly, M. A. (1993). Rela-
fore be consistent with our finding of a large discrepancy tionships among academic, neuropsychological, and intellec-
in the HFA group between Block Design and both Pic- tual status in subtypes of adults with learning disability. Ar-
chives of Clinical Neuropsychology, 8, 41–53.
ture Arrangement and Picture Completion, suggesting Goldstein, G., Minshew, N. J., & Siegel, D. J. (1994). Age differ-
that the structure of intellectual function in HFA is quite ences in academic achievement in high-functioning autistic in-
different from what is the case for the general population dividuals. Journal of Clinical and Experimental
and for LD. Neuropsychology, 16, 671–680.
For users of the WAIS–III, it has been reported that Griffith, E. M., Pennington, B. F., Wehner, E. A., & Rogers, S.
the factor structure of the WAIS–III is quite similar to (1999). Executive functions in young children with autism.
Child Development, 70, 817–832.
that of the WAIS–R, containing Verbal Comprehen- Jastak, J. F., & Wilkinson, G. S. (1984). The Wide Range Achieve-
sion, Perceptual Organization, and Freedom from ment Test–Revised. Wilmington, DE: Jastak Associates.
Distractibility (now called Working Memory) factors, Katz, L., Goldstein, G., Rudisin, S., & Bailey, D. (1993). A
with a Processing Speed factor added. Furthermore, the neuropsychological approach to the Bannatyne
correlations between the two editions are .94, .86, and recategorization of the Wechsler intelligence scales in adults
with learning disabilities. Journal of Learning Disabilities, 26,
.93 for Verbal, Performance, and Full Scale IQ, respec-
65–72.
tively (Tulsky, Zhu, & Prifitera, 2000). Thus, the find- Kaufman, A. S., & Kaufman, N. L. (1985). Kaufman Test of Educa-
ings reported here are likely to be applicable to the tional Achievement. Circle Pines, MN: American Guidance
WAIS–III. Clearly, however, further research in this Service.
area should use this newly developed, state-of-the-art Klin, A., Volkmar, F. R., Sparrow, S. S., Cicchetti, D. V., & Rourke,
instrument. Limitations of this study are insufficient B. P. (1995). Validity and neuropsychological characterization
of Asperger syndrome: Convergence with nonverbal learning
sample size to perform factor analyses and more ad- disabilities syndrome. Journal of Child Psychology and Psy-
vanced analyses of profile similarities and differences chiatry, 7, 1127–1140.
as well as use of the WAIS–R rather than the LeCouteur, A., Rutter, M., Lord, C., Rios, P., Robertson, S.,
WAIS–III, necessitated by the low prevalence of HFA. Holdgrafer, M., & McClennan, J. (1989). Autism diagnostic in-
As larger samples of individuals with HFA become terview: A standardized investigator-based instrument. Jour-
nal of Autism and Developmental Disorders, 19, 363–387.
available, further exploratory as well as confirmatory
Lincoln, A. J., Courchesne, E., Kilman, B. A., Elmasian, R., & Al-
factor analytic studies might yield valuable new infor- len, M. (1988). A study of intellectual abilities in high-func-
mation regarding ability structures in these different tioning people with autism. Journal of Autism and Develop-
developmental disorders. mental Disorders, 18, 505–524.

153
GOLDSTEIN ET AL.

Lord, C., Rutter, M., Goode, S., Heemsbergen, J., Jordan, H., Reitan, R. M., & Wolfson, D. (1993). The Halstead–Reitan
Mawhood, L., & Schopler, E. (1989). Autism diagnostic obser- Neuropsychological Test Battery: Theory and clinical inter-
vation schedule: A standardized observation of communicative pretation (2nd ed.). Tucson, AZ: Neuropsychology Press.
and social behavior. Journal of Autism and Developmental Dis- Rourke, B. P. (1982). Central processing deficiencies in children:
orders, 19, 185–212. Toward a developmental neuropsychological model. Journal
Lord, C., Rutter, M., & LeCouteur, A. (1994). Brief report: Autism of Clinical Neuropsychology, 4, 1–18.
diagnostic interview–revised: A revised version of a diagnostic Rourke, B. P. (Ed.). (1985). Neuropsychology of learning disabili-
interview for caregivers of individuals with possible pervasive ties: Essentials of subtype analysis. New York: Guilford.
developmental disorders. Journal of Autism and Developmen- Rourke, B. P. (Ed.). (1989). Nonverbal learning disabilities: The
tal Disorders, 24, 659–685. syndrome and the model. New York: Guilford.
McCue, M., & Goldstein, G. (1991). Neuropsychological aspects of Rourke, B. P. (Ed.). (1991). Neuropsychological validation of learn-
learning disability in adults. In B. Rourke (Ed.), ing disability subtypes. New York: Guilford.
Neuropsychological validation of learning disability subtypes Rumsey, J. M., & Hamburger, S. D. (1990). Neuropsychological di-
(pp. 311–329). New York: Guilford. vergence of high-level autism and severe dyslexia. Journal of
Minshew, N. J. (1996). Autism. In R. D. Adams & M. Victor (Eds.), Autism and Developmental Disorders, 20, 155–168.
Principles of child neurology (pp. 1713–1729). New York: Rutter, M., & Schopler, E. (1987). Autism and pervasive develop-
McGraw-Hill. mental disorders: Concepts and diagnostic issues. Journal of
Minshew, N. J., Goldstein, G., Muenz, L. R., & Payton, J. B. (1992). Autism and Developmental Disorders, 2, 159–186.
Neuropsychological functioning in non-mentally retarded au- Shea, V., & Mesibov, G. (1985). Brief report: The relationship of
tistic individuals. Journal of Clinical and Experimental learning disabilities and higher-level autism. Journal of Autism
Neuropsychology, 14, 749–761. and Developmental Disorders, 15, 425–435.
Minshew, N. J., Goldstein G., & Siegel, D. J. (1997). Siegel, D. J., Minshew, N. J., & Goldstein, G. (1996). Wechsler IQ
Neuropsychologic functioning in autism: Profile of a complex profiles in diagnosis of high-functioning autism. Journal of Au-
information processing disorder. Journal of the International tism and Developmental Disorders, 26, 389–406.
Neuropsychological Society, 3, 303–316. Tulsky, D. S., Zhu, J., & Prifitera, A. (2000). Assessment of adult in-
Minshew, N. J., Goldstein, G., Taylor, H. G., & Siegel, D. (1994). telligence with the WAIS–III. In G. Goldstein & M. Hersen
Academic achievement in high functioning autistic individu- (Eds.), Handbook of psychological assessment, (3rd ed., pp.
als. Journal of Clinical and Experimental Neuropsychology, 97–129). Amsterdam: Pergamon.
16, 261–270. Wechsler, D. (1944). The measurement of adult intelligence (2nd
Minshew, N. J., & Payton, J. B. (1988). New perspectives in autism. ed.). Baltimore: Williams & Wilkins.
Part I: The clinical spectrum of autism. Current Problems in Wechsler D. (1955). WAIS manual. New York: Psychological Cor-
Pediatrics, 18, 561–610. poration.
Ozols, E. J., & Rourke, B. P. (1991). Classification of young learn- Wechsler, D. (1981). WAIS–R manual. San Antonio, TX: Psycho-
ing disabled children according to patterns of academic logical Corporation.
achievement: Validity studies. In B. Rourke (Ed.), Wechsler, D. (1998). WAIS–III manual. San Antonio, TX: Psycho-
Neuropsychological validation of learning disability subtypes logical Corporation.
(pp. 97–123). New York: Guilford. Whitehouse, D., & Harris, J. C. (1984). Hyperlexia in infantile au-
Ozonoff, S. (1995). Executive functions in autism. New York: Ple- tism. Journal of Autism and Developmental Disorders, 26,
num. 423–438.
Rapaport, D., Gill, M., & Schafer, R. (1945). Diagnostic psychologi-
cal testing (Vol. 1). Chicago: Year Book Publishers.
Rapin, I. (1991). Autistic children: Diagnosis and clinical features. Original submission March 31, 2000
Pediatrics, 87(Suppl. 2), 751–760. Accepted February 8, 2001

154