Adaptive Behavior in Autism and Pervasive Developmental Disorder Not Otherwise Specified : Microanalysis of Scores on the Vineland Adaptive Behavior Scales

483929.qxd 3/18/04 21:37 Page 223
Journal of Autism and Developmental Disorders, Vol. 34, No. 2, April 2004 (© 2004)
Adaptive Behavior in Autism and Pervasive Developmental
Disorder-Not Otherwise Speci?ed: Microanalysis of Scores
on the Vineland Adaptive Behavior Scales
Rhea Paul,1,2,4 Stephanie Miles,1 Domenic Cicchetti,1 Sara Sparrow,1 Ami Klin,1
Fred Volkmar,1 Megan Co?in,3 and Shelley Booker3
The purpose of this study is to provide a microanalysis of differences in adaptive functioning
seen between well-matched groups of school-aged children with autism and those diagnosed as
having Pervasive Developmental Disorder-Not Otherwise Specified, all of whom functioned in
the mild to moderate range of intellectual impairment. Findings indicate that the major area of
difference between children with autism and those with Pervasive Developmental Disorder-Not
Otherwise Specified, was expressive communication; specifically, the use of elaborations in syn-
tax and morphology and in pragmatic use of language to convey and to seek information in dis-
course. Linear discriminant function analysis revealed that scores on just three of these expressive
communication item sets correctly identified subjects in the two diagnostic categories with 80%
overall accuracy. Implications of these findings for both diagnosis and intervention with chil-
dren with Autism Spectrum Disorders will be discussed.
KEY WORDS: Autism; Pervasive Developmental Disorders; adaptive behavior; communication; expressive
language; socialization.
Ever since the initial delineation of the autistic syn-
the American Psychiatric Association (1994) for the
drome (Kanner, 1943), children have been recognized
diagnosis of autism. The diagnosis of PDD-NOS, how-
who demonstrate social and communicative disabilities
ever, is much less specified; DSM-IV criteria merely
but who—because of late onset, atypical presentation,
state that, “a severe and pervasive impairment in the
or subthreshold symptoms—do not fully meet diag-
development of social interaction associated with im-
nostic criteria for autism. Diagnostic and Statistical
pairment in either verbal or nonverbal communication
Manual of Mental Diseases Fourth Edition (DSM-IV;
skills or . . . stereotyped behaviors, interests and activ-
American Psychiatric Association [APA], 1994) spec-
ities” (APA, 1994, p. 84) are present, and that the di-
ifies a condition referred to as Pervasive Developmen-
agnostic criteria for autism, schizophrenia, schizotypal
tal Disorder-Not Otherwise Specified (PDD-NOS) to
personality, or avoidant personality are not met. Al-
cover such conditions. Specific diagnostic criteria,
though the diagnosis of autism is made with a good deal
which have been validated by a large-scale field trial
of guidance from the literature, with a wide range of
(Volkmar et at., 1994), are provided in the DSM-IV of
carefully studied instruments (Volkmar & Lord, 1998),
conferring a diagnosis of PDD-NOS forces clinicians
1 Yale Child Study Center, New Haven, Connecticut.
to rely primarily on subjective judgment as to whether
2 Southern Connecticut State University, New Haven, Connecticut.
the social and communicative deficits observed rise to
3 ASHA Mentored Research in Higher Education Visiting Fellow,
the explicitly defined level of the diagnostic category.
New Haven, Connecticut.
4
In addition, it is as yet unclear whether the distinction
Correspondence should be addressed to Rhea Paul, Yale Child
between autism and PDD-NOS has any implications
Study Center, P. O. Box 207900, New Haven, CT 06520-7900;
e-mail: rhea.paul@yale.edu
for treatment of children with these disorders. This
223
0162-3257/04/0400-0223/0 © 2004 Plenum Publishing Corporation

---

483929.qxd 3/18/04 21:37 Page 224
224
Paul et al.
study will examine specific aspects of adaptive behav-
assessment of children with PDD. Children received a
ior in these two subject groups to find out whether any
comprehensive evaluation, which included administra-
of these behaviors help to differentiate these diagnoses.
tion of the survey form of the Vineland, tests of com-
Although cognitive level is a significant predictor
municative ability, a psychiatric examination, and a test
of outcome in autism, as it is in other conditions
of intellectual functioning.
(DeMyer, Hingtgen, & Jackson, 1981; Prior &
Subjects for this study had been assigned to diag-
Ozonoff, 1998), adaptive skills are another aspect of
nostic groups on the basis the clinical diagnoses they
development that contributes strongly to prognosis
received following their assessment. Diagnostic char-
(Gillham, Carter, Volkmar, & Sparrow, 2000). Adap-
acterization included the Autism Diagnostic Interview-
tive skills are those involved with using whatever ca-
Revised (ADI-R; Lord, Rutter, & LeCouteur, 1994) and
pacities the individual possesses to function within the
the Autism Diagnostic Observation Schedule-Generic
everyday environment. These skills are particularly
(ADOS-G; Lord et al., 2000). To be included in the
important in individuals with autism and related con-
autistic or PDD-NOS group, each subject had to meet
ditions because it is these, rather than cognitive level,
either DSM-IIIR (APA, 1987) or DSM-IV (APA, 1994)
that contribute most to the individual’s ability to func-
criteria for the disorders. Clinical diagnoses were con-
tion successfully and independently in the world (Liss
firmed independently by two experienced clinicians
et al., 2001). Literature attesting to the adaptive deficits
(A. Klin and F. Volkmar) who were blind to Vineland
in autism dates back at least to Volkmar et al. (1987).
interview data and results. Interrater reliability between
Several later studies confirmed that the Vineland Adap-
these clinicians for diagnostic assignment was high, with
tive Behavior Scales (Sparrow, Balla, & Cicchetti,
kappa values ranging from .80 to .95 in related research
1984), a well-standardized, semistructured caregiver
projects (Klin, Lang, Cicchetti, & Volkmar, 2000).
report instrument for assessing adaptive behavior,
Twenty subjects were selected who had received
could be used to document delays in adaptive devel-
a diagnosis of autism and whose ages and full-scale IQs
opment in individuals with autism (Carter et al., 1998;
(FSIQs) fell within the ranges specified at the outset of
Liss et al., 2001; Loveland & Kelley, 1991; Rodrigue,
the study (ages 4–12 years; FSIQ 50–100). These sub-
Morgan, & Gefken, 1991; Schatz & Hamdan-Allen,
jects represented all the eligible candidates with autism
1995). Gillham, Carter, Volkmar, and Sparrow (2000)
in the clinical database whose Vineland data were avail-
reported that autism could be differentiated from both
able for item-level analysis. A group of 20 children who
PDD-NOS and nonautistic developmental disorder
were diagnosed as having PDD-NOS was selected to
(DD) by means of scores on the Socialization and
match the group with autism in terms of age, IQ range,
Daily Living scales of the Vineland Adaptive Behav-
and Vineland data availability. Mean age in the group
ior Scales (Sparrow et al., 1984). These authors sug-
with autism was 6.5 years (SD: 1.8); in the group with
gested that the broad domains of the Vineland
PDD-NOS, it was 6.6 years (SD: 1.2). Mean FSIQ in
employed in their study might miss subtle differences
the group with autism was 71.2 (SD: 13.4; range:
between performance in autism and PDD-NOS. Look-
50–99; mean mental age based on this average IQ was
ing more closely at the specific items that make up the
4.6). In the group with PDD-NOS, mean IQ was 76.3
scales on the Vineland might yield a better ability to
(SD: 13.0; range: 56–98; mean mental age based on
discriminate PDD-NOS from autism in several areas
this average IQ was 5.0). Average age and IQ scores
of adaptive skills. The purpose of this study is to
between the two groups were not significantly differ-
provide a microanalysis of differences in adaptive
ent. Eighteen of the 20 subjects in the group with PDD-
functioning seen between well-matched groups of
NOS were male; 19 of the 20 in the group with autism
school-aged children with autism and those diagnosed
were male.
as having PDD-NOS.
Measures
The Vineland Adaptive Behavior Scales Survey
METHOD
form (Sparrow et al., 1984), a nationally standardized
semistructured caretaker interview instrument that as-
Subjects
sesses day-to-day adaptive functioning, was administered
The sample comprised 40 subjects between
to primary caregivers by research assistants extensively
the ages of 4 and 11 years who were assessed at a
trained in Vineland interview and scoring procedures.
university-based clinic that specializes in diagnosis and
The Vineland consists of four domains: Communication,

---

483929.qxd 3/18/04 21:37 Page 225
Adaptive Behavior in Autism and PDD-NOS
225
Table I. Vineland Adaptive Behavior Scales Domains
The average percentage score for each subject for each
and Subdomains
set was then calculated. Age equivalents for each set
were derived from the Vineland standardization data,
Domains
Subdomains
to determine which sets could be expected to fall within
Communication
Expressive: What an individual says
the range of expected performance for these subjects.
Receptive: What an individual understands
Written: What an individual reads and writes
Daily Living
Personal: How an individual eats, dresses,
Data Analysis
and practices personal hygiene
Domestic: What household tasks an
Analysis of variance was used to test for differences
individual performs
in raw scores between groups on each domain score
Community: How an individual uses time,
(Communication, Daily Living, Socialization) and on
money, the telephone, and job skills
each subdomain score (Communication-Expressive,
Socialization
Interpersonal: How an individual interacts
Communication-Receptive, Communication-Written;
with others
Play and Leisure: How an individuals plays
Daily
Living-Personal,
Daily
Living-Domestic,
and uses leisure time
Daily Living-Community; Socialization-Interpersonal,
Coping: How an individual shows
Socialization-Play/Leisure,
Socialization-Coping).
responsibility and sensitivity to others
These results appear in Tables II and III. Multivariate
analysis of variance, using FSIQ as a covariate, was used
to test for differences in raw scores between the autism
Daily Living, Socialization, and Motor. For the purpose
and PDD-NOS groups on each item set. Although all sta-
of this study, only data on the first three domains were
tistics were performed on raw scores, data in Table III
used because the Motor domain is only administered
are presented as percentage scores to enable comparison
to children below 6 years, so that data were not avail-
across items. Linear discriminant function analysis was
able for all subjects in the cohort. Each domain contains
also performed, to examine the ability of item set scores
several subdomains, which are listed and described in
to correctly classify individuals within the two diag-
Table I.
nostic groups.
Within each subdomain, the Vineland is divided
into sets of items that probe a particular area of devel-
opment. These sets each contain two to seven individ-
RESULTS
ual items. Each item within the set is scored as a 0
(never), 1 (sometimes; partially), or 2 (usually), ac-
Differences between Groups
cording to criteria detailed in the Vineland manual. For
Vineland Domain and Subdomain Scores
the comparisons used in this study, each subject’s score
for each set of items was derived by summing the num-
As in previous studies (e.g., Klin, Volkmar, &
ber of points assigned for each item within the set. In
Sparrow, 1992; Gillham et al., 2000), univariate analy-
addition to these set total scores, the percentage of pos-
sis of variance (using the SPPS computer program) in
sible points earned within each set by each individual
this study revealed significant differences in raw scores
was also computed to compare performance across sets.
between the group with autism and the group with
Table II. Mean (SD) Vineland Adaptive Behavior Scales (Sparrow, Balla, & Cicchetti, 1984) Domain
Raw Scores and Standard Scores (in italics) for Two Groups
Vineland Adaptive Behavior Scales domain scores
Age (years)
IQ
Communicationa
Socializationa
Daily living
Autism
6.5
71.8
136.9 (34.9)
93.7 (22.5)
131.7 (32.2)
(n = 20)
(1.8)
(13.4)
57.8 (14.0)
56.8 (9.3)
50.3 (17.0)
PDD-NOS
6.6
76.3
171.4 (33.4)
113.7 (24.2)
149.0 (40.5)
(n = 20)
(1.2)
(13.0)
65.3 (13.6)
60.8 (8.3)
53.4 (16.8)
Note: PDD-NOS, Pervasive Developmental Disorder-Not Otherwise Specified.
aSigni?cant difference between groups at p < .01.

---

483929.qxd 3/18/04 21:37 Page 226
226
Paul et al.
Table III. Percentage Scores of Subjects in Two Groups for Item Sets in Which Signi?cant Differences Were Found
Percentage
Percentage
Item set age
Difference between
score score
equivalent
autism
and
Vineland item set
autism
PDD-NOS
(months)a
Multivariate F
PDD-NOS groups
Communication Domain
Expressive Subdomainb
Talking in sentences
60.6
88.1
25.5
36.2
p < .007
Asking questions
31.2
69.6
26.0
15.6
p < .0001
Abstract concepts
53.8
82.5
29.3
8.4
p < .006
Relating experiences
15.5
49.5
31.6
10.1
p < .003
Using prepositions
29.0
65.0
39.2
14.1
p < .001
Using function words
37.0
70.5
33.6
12.6
p < .001
Plurals/Tenses
28.8
63.8
36.8
10.7
p < .002
Gives information about self
37.8
68.2
50.4
8.7
p < .005
Socialization Domain
Interpersonal Subdomainb
Identifying others
46.3
73.8
24.5
14.4
p < .001
Play/Leisure Subdomainb
Coping Subdomain
Manners in conversation
7.5
29.2
65.6
8.0
p < .008
Daily Living Domain
Community Subdomain
Phone use
22.5
45.0
36.3
14.9
p < .001
Note: PDD-NOS, Pervasive Developmental Disorder-Not Otherwise Specified.
a Age-Equivalents derived from Vineland normative data.
b Significant difference between groups was found on total subdomain score. Boldface indicates items achieved at <4.5 years (average MA of
subjects in sample) in typical development, based on Vineland standardization sample (Sparrow, Balla, & Cicchetti, 1984).
PDD-NOS on the Vineland domains of Communication
subdomains: Communication; Receptive, Written;
and Socialization. Significant differences ( p < .01)
Socialization; Play/Leisure (although a significant dif-
were also found on the Expressive subdomain within
ference was found on this set as a whole); and Daily
the Communication Domain and on both the Interper-
Living; Personal, Domestic.
sonal and Play/Leisure subdomains within the Social-
Seventy percent of the differences found were on
ization domain. No significant differences were found
item sets with age-equivalents below the mental age of
on the Daily Living domain, nor on any of its subdo-
the subjects; that is, on sets that were clearly within the
mains. These results appear in Tables II and III.
expected range of achievement for these individuals.
Vineland Item Set Scores
Discriminant Function Analysis
Table III presents the percentage scores for the
Work by Fletcher, Rice, & Ray (1978) indicates
item sets on which significant differences were found
that a minimum of five subjects per group per predic-
between raw scores of the two groups, using multi-
tor variable is required for valid application of linear
variate analysis of variance with FSIQ as a covariate
discriminant function analysis. Thus, our sample size
(using the SPSS computer program multivariate analy-
would allow only four predictors to be entered into this
sis package). It should be noted that in all cases, scores
analysis. Examining the significant differences between
for subjects with PDD-NOS were higher than for those
groups indicated that the Expressive Communication
with autism. Significant differences between the autism
subdomain was the area in which the largest number
and PDD-NOS groups were seen on 64% of the item
of differences was found; seven item sets with age-
sets in Communication-Expressive, 7% of the sets
equivalents below 4.5 showed significant differences
in Socialization-Interpersonal, 9% of the sets in
between groups. Of these, one appeared to be a general
Socialization-Coping, and 20% of the sets in Daily
index of multiword communication (“Talking In
Living-Community. No significant differences were
Sentences”), three appeared to assess grammatical
found between any sets in the following remaining
ability (“Function Words,” “Plurals/Tense Use,” and

---

483929.qxd 3/18/04 21:37 Page 227
Adaptive Behavior in Autism and PDD-NOS
227
“Preposition Use”), two indexed pragmatic use of
present data indicate that, relative to PDD-NOS,
communication (“Asking Questions” and “Relating
pragmatics in autism is more severely affected. These
Experiences”), and one measured conceptual ability
data also indicate that syntax is an area of significant
(“Abstract Concepts”). For the discriminant function
difficulty for children with autism.
analysis, then, the one variable from each of these four
Only three other significant differences were found
categories with the highest p value in the multiple
outside the Expressive Communication subdomain; one
analysis of variance was chosen to enter into the analy-
in each of the Daily Living-Community, Socialization-
sis. This resulted in the use of Asking Questions, Using
Interpersonal, and Socialization-Coping subdomains.
Prepositions, Talking in Sentences, and Abstract
The first was the “Phone Use,” set in Daily Living-
Concepts as predictors. Using these four variables, the
Community. This set is clearly related to skill in ex-
discriminant analysis correctly predicted group mem-
pressive language. The second difference, in the
bership for 80% of children with autism and 75% of
“Identifying Others” set on the Socialization-
children with PDD-NOS, for an overall classification
Interpersonal subdomain, indicates that children with
accuracy of 77.5%. Further analysis indicated that re-
autism are less likely than those with PDD-NOS to en-
moving Abstract Concepts from the analysis resulted
gage others through the use of their names or to name
in improving sensitivity for the diagnosis of autism to
others without prompting. This, again, appears to be
85%, with no change in specificity for the diagnosis of
strongly related to their pragmatic use of expressive
PDD-NOS. Thus, including only the Asking Questions,
communication. The third difference, “Manners In Con-
Using Prepositions, and Talking in Sentences item sets
versation,” on the Socialization-Coping subdomain, is
in the analysis was sufficient to yield 80% accuracy of
the only difference with an age equivalent above 4.5.
classification overall. Applying criteria proposed by
It is noteworthy that even this relatively sophisticated
Cicchetti, Volkmar, Klin, and Schowalter (1995), this
set of behaviors, which relies primarily on verbal ex-
level of classification would be considered “good.”
pression, reveals significant differences between the
Additional analyses indicated that no other combina-
two groups. This finding again emphasizes the primacy
tion of item sets exceeded this level of classification
of the role of expressive communication in differenti-
accuracy.
ating between autism and PDD-NOS in this mildly to
moderately impaired sample of school-aged children.
In summary, the major differences in adaptive
DISCUSSION
function found in this study are in the area of expres-
sive communication; specifically, use of elaborations
This study finds a relatively small number of dif-
in syntax and morphology, beyond the most basic ex-
ferences in adaptive function between children with
pression of sounds and words, and use of language to
autism and PDD-NOS, and all of these seem to reflect
initiate interaction and to seek and convey information
a deficit on the part of the children with autism in ver-
in discourse. Other areas in which differences are found
bal expression. In the Expressive Communication sub-
seem to be the direct result of this deficit in verbal ex-
domain, children with PDD-NOS scored higher than
pression. Moreover, a small number of item sets that
those with autism in areas that involve more elaborated
index the major deficits in expressive language
language usage and more socially oriented criteria for
correctly classify 80% of the subjects into their diag-
success, including: using grammatical combinations of
nostic groups.
words; using function words (e.g., a, the, in, on) and
grammatical markers (e.g., plurals, -ed endings); using
language to communicate personal experiences and
IMPLICATIONS
commonly known information (e.g., age, birthday);
using language to impart new information, such as to
This study indicates that children with PDD-NOS,
communicate simple messages; using language to gain
when compared with those with autism, appear not to
new information by asking questions; talking about ob-
be more mildly impaired in all areas of adaptive func-
jects and events removed from the “here and now”; and
tion. Instead, these groups differ only in very specific
talking about less concrete, more conceptual topics.
areas, primarily the use of expressive language for
These items can be construed to reside in two basic
communication—namely, the areas of syntax and
areas of language function: a) syntax/morphology and
pragmatics—and the areas of adaptive function on
b) pragmatics. Pragmatics is an aspect of communica-
which these skills have a direct effect, such as phone
tion well known to be impaired in all ASDs, but the
use, manners in conversation, and using language to

---

483929.qxd 3/18/04 21:37 Page 228
228
Paul et al.
identify and initiate interaction with others. The dis-
Gillham, J., Carter, A., Volkmar, F., & Sparrow, S. (2000). Toward a
criminant function analysis indicates that focus on
developmental operational definition of autism. Journal of
Autism and Developmental Disorders, 30, 269–278.
assessing the syntactic and pragmatic expression of
Kanner, L. (1943). Autistic disorders of affective contact. Nervous
mildly to moderately impaired school-aged children
Child, 2, 217–250.
with autistic spectrum disorders will be useful in help-
Klin, A., Lang, J., Cicchetti, D. V., & Volkmar, F. R. (2000). Brief
Report: Interrater reliability of clinical diagnosis and DSM-IV
ing to establish categorical diagnoses for individuals
criteria for autistic disorder: Results of the DSM-IV autism
with these disabilities. Moreover, these findings high-
field trial. Journal of Autism and Developmental Disorders, 30,
light the importance of intensive intervention in the area
163–167.
Klin, A., Volkmar, F., & Sparrow, S. (1992). Autistic social dys-
of expressive language, particularly for children with
function: Some limitations of the theory of mind hypothesis.
autism, as this is an area of weakness over and above
Journal of Child Psychology and Psychiatry, 33, 861–876.
other social and adaptive deficits seen in this group.
Liss, M., Harel, B., Fein, D., Allen, D., Dunn, M., Feinstein, C.,
Morris, R., Waterhouse, L., & Rapin, I. (2001). Predictors and
Finally, the findings indicate that although social-
correlates of adaptive functioning in children with develop-
pragmatic skills are greatly in need of intervention in
mental disorders. Journal of Autism and Developmental Disor-
the group with autism, grammatical production should
ders, 31, 219–230.
Lord, C., Risi, S., Lambrecht, L., Cook, E. H., Jr., Leventhal, B. L.,
not be ignored in intervention programs for these
DiLavore, P. C., Pickles, A., & Rutter, M. (2000). The Autism
children. Although it has been thought for some time
Diagnostic Observation Schedule–Generic: A standard measure
that grammar is relatively “spared” in autistic commu-
of social and communication deficits associated with the spec-
trum of autism. Journal of Autism and Developmental Disor-
nication (Tager-Flusberg, 1995), these results indicate
ders, 30, 205–223.
that syntactic deficits do contribute to the severity of
Lord, C., Rutter, M., & LeCouteur, A. (1994). Autism Diagnostic
the disability in expressive communication and should
Interview-Revised: A revised version of a diagnostic interview
for caregivers of individuals with possible pervasive develop-
be addressed as part of a comprehensive program for
mental disorders. Journal of Autism and Developmental Disor-
improving social communication in autism.
ders, 24, 659–685.
Loveland, K., & Kelley, M. (1991). Development of adaptive
behavior in preschoolers with autism or Down syndrome.
American Journal of Mental Retardation, 96, 13–20.
ACKNOWLEDGMENTS
Prior, M., & Ozonoff, S. (1998). Psychological factors in autism. In
F. Volkmar (Ed.), Autism and Pervasive Developmental Disor-
This research was funded by grants from the
ders (pp. 64–108). Cambridge: Cambridge University Press.
Rodrigue, J., Morgan, S., & Gefken, G. (1991). A comparisative eval-
National Institute of Child Health and Development
uation of adaptaive behavior in children and adolescents with
(DC030080), the National Alliance for Autism
autism Down syndrome and normal development. Journal of
Research, and the National Institutes of Health
Autism and Developmental Disorders, 21, 187–198.
Schatz, J., & Hamdan-Allen, G. (1995). Effects of age and IQ on
STAART Program (U54MH66494).
adaptive behavior domains for children with autism. Journal of
Autism and Developmental Disorders, 25, 51–60.
Sparrow, S., Balla, D., & Cicchetti, D. V. (1984). The Vineland Adap-
tive Behavior Scales (Survey Form). Circle Pines, MN: American
REFERENCES
Guidance Service.
Tager-Flusberg, H. (1995). Dissociations in form and function in the
American Psychiatric Association (1987). Diagnostic and Statistical
acquisition of language in autistic children. In H. Tager-Flusberg
Manual of Mental Diseases-IIIR. Washington, DC: Author.
(Ed.), Constraints on language acquisition: Studies of atypical
American Psychiatric Association (1994). Diagnostic and Statistical
children (pp. 175–194). Hillsdale, NJ: Erlbaum.
Manual of Mental Diseases. (4th ed.). Washington, DC: Author.
Volkmar, F., & Lord, C. (1998). Diagnosis and definition of autism
Carter, A., Volkmar, F., Sparrow, S., Wang, J., Lord, C., Dawson, G.,
and other pervasive developmental disorders. In F. Volkmar
Fombonne, E., Loveland, K., Mesibov, G., & Schopler, E.
(Ed.), Autism and Pervasive Developmental Disorders
(1998). The Vineland Adaptive Behavior Scales: Supplementary
(pp. 1–31). Cambridge: Cambridge University Press.
norms for individuals with autism. Journal of Autism and
Volkmar, F., Klin, A., Siegel, B., Szatmari, P., Lord, C., Campbell, M.,
Developmental Disorders, 28, 287–302.
Freeman, B., Cicchetti, D. V., Rutter, M., Kline, W., Buitelar, J.,
Cicchetti, D., Volkmar, F., Klin, A., & Schowalter, D. (1995). Diag-
Hattab, Y., Fombonne, E., Fuentes, J., Werry, J., Stone, W.,
nosing autism using ICD-10 criteria: A comparison of neural
Kerbeshian, J., Hoshino, Y., Bregman, J., Loveland, K.,
networks and standard multivariate procedures. Child
Szymanski, L., & Towbin, K. (1994). Field trial for autistic
Neuropsychology, 1, 26–37.
disorder in DSM-IV. American Journal of Psychiatry, 151,
DeMyer, M., Hingtgen, J., & Jackson, R. (1981). Infantile autism
1361–1367.
reviewed: A decade of research. Schiophrenia Bulletin, 7,
Volkmar, F., Sparrow, S., Goudreau, D., Cicchetti, D., Paul, R., &
388–451.
Cohen, D. (1987). Social deficits in autism: An operational
Fletcher, J., Rice, W., & Ray, R. (1978). Linear discriminant func-
approach using the Vineland Adaptive Behavior Scales. Journal
tion analysis in neurophychological research: Some uses and
of the American Academy of Child and Adolescent Psychiatry,
abuses. Cortex, 14, 564–577.
26, 156–161.

---