Evidence for Latent Classes of IQ in Young Children With Autism Spectrum Disorder

Abstract

Autism is currently viewed as a spectrum condition that includes strikingly different severity levels; IQ is consistently described as one of the primary aspects of the heterogeneity in autism. To investigate the possibility of more than one distinct subtype of autism based on IQ both latent class analysis and taxometrics methods were used to classify Mullen IQs in a sample of 456 children with autism spectrum disorder. We found evidence for multiple IQbased subgroups using both methods. Groups differed in level of intellectual functioning and patterns of verbal versus nonverbal ability. Results support the notion of distinct subtypes of autism that differ in severity of intellectual ability, patterns of cognitive strengths and weaknesses, and severity of autism symptoms.

Full text

䉷American Association on Intellectual and Developmental Disabilities 439
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Evidence for Latent Classes of IQ in Young Children
With Autism Spectrum Disorder
Jeffrey Munson, Geraldine Dawson, Lindsey Sterling, Theodore Beauchaine,
Andrew Zhou, and Elizabeth Koehler
University of Washington
Catherine Lord
University of Michigan
Sally Rogers
University of California Davis Medical Center
Marian Sigman
University of California Los Angeles
Annette Estes and Robert Abbott
University of Washington
Abstract
Autism is currently viewed as a spectrum condition that includes strikingly different severity
levels; IQ is consistently described as one of the primary aspects of the heterogeneity in
autism. To investigate the possibility of more than one distinct subtype of autism based
on IQ, both latent class analysis and taxometrics methods were used to classify Mullen
IQs in a sample of 456 children with autism spectrum disorder. We found evidence for
multiple IQ-based subgroups using both methods. Groups differed in level of intellectual
functioning and patterns of verbal versus nonverbal ability. Results support the notion of
distinct subtypes of autism that differ in severity of intellectual ability, patterns of cognitive
strengths and weaknesses, and severity of autism symptoms.
DOI: 10.1352/2008.113:439–452
Autism is characterized by impairments in so-
cial interaction and communication as well as a
restricted repertoire of activities and interests.
Children and adults with autism have specific def-
icits in social and emotional information-process-
ing (Davies, Bishop, Manstead, & Tanta, 1994;
Dawson, Meltzoff, Sterling, & Renaldo, 1998) that
are considered to be common features of individ-
uals with the disorder. Yet autism is also charac-
terized by a wide variability in more specific im-
pairments, range of symptoms, levels of adaptive
and intellectual functioning, and prognosis. These
differences in presentation make conceptualiza-
tion of the disorder difficult, especially given that
diagnosis is based on behavioral observations and
standardized parental interviews administered by
clinicians. Thus, in almost all cases, behavioral
characteristics, rather than laboratory or medical
tests, determine diagnostic assignment.
Because of the significant symptom hetero-
geneity found in autism, it is often conceptualized
as autism spectrum disorder. Variability in IQ is
one of the most salient dimensions of this hetero-
geneity. Both the Diagnostic and Statistical Manual
of Mental Disorders (American Psychiatric Associ-
ation, 1994) and International Classification of Dis-

440 䉷American Association on Intellectual and Developmental Disabilities
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
eases (World Health Organization, 1992) have def-
initions of Asperger syndrome that include cog-
nitive developmental level as one of the key fea-
tures distinguishing it from autism. It is estimated
that 70% of individuals with autism have IQs in
the mentally retarded range (Fombonne, 2003),
yet some individuals have above average intellec-
tual ability (Miller & Ozonoff, 2000). Moreover,
the IQ profiles of individuals with and those with-
out mental retardation tend to differ, with indi-
viduals who have higher IQ typically showing a
higher Verbal IQ, on average (Ghaziuddin &
Mountain-Kimchi, 2004; Gilchrist et al., 2001).
However, there is extreme individual variability in
IQ, making it unlikely that a specific cognitive
profile can be used for differential diagnostic pur-
poses (Filipek et al., 1999; Siegel, Minshew, &
Goldstein, 1996). Nevertheless, researchers have
used various strategies to subtype individuals with
autism. Some have focused on medical conditions
or known biological etiologies contributing to the
disorder. Miles et al. (2005) defined subtypes of
autism based on whether individuals had features
that were stable from birth, suggesting an organic
factor, including all of the syndromes that are cur-
rently acknowledged as causes of autism (e.g., frag-
ile X syndrome). These individuals, comprising
what the authors called the ‘‘complex’’ autism
group, also tend to have more seizures, dysmor-
phic physical features, microcephaly, lower IQs,
and a tendency toward poorer outcome than the
‘‘essential’’ autism group. The essential autism
group is characterized by higher incidence of sib-
ling recurrence and a family history of autism,
higher male to female ratio, higher likelihood of
regression and macrocephaly, and overall higher
IQs. According to Miles et al., assigning individ-
uals to the complex and essential groups allows
for the first stage of characterizing the etiologic
heterogeneity of those with autism spectrum dis-
orders. This separation might be especially useful
for genetic analyses because it provides a more
homogeneous group of individuals (essentials).
However, until the etiological substrates of autism
are identified, it is impossible to know how truly
homogeneous this group is.
Other researchers have focused on defining
autism spectrum subgroups according to behav-
ioral patterns of social interaction. Wing and
Gould (1979) first characterized autism according
to three subtypes: aloof, passive, or active-but-
odd. The aloof subtype, which includes children
who tend to reject contact and avoid gaze, is typ-
ically the most impaired and severely autistic (e.g.,
Castelloe & Dawson, 1993; Sevin et al., 1995).
Levels of IQ tend to correspond to social typol-
ogy, with the aloof group having the lowest IQ,
followed by the passive and then active-but-odd
groups (Borden & Ollendick, 1994). The aloof
group also tends to have the lowest levels of adap-
tive behavior, worse language and communication
skills, and higher ratings of stereotyped behavior/
restricted interests. Intellectual functioning likely
accounts for a large proportion of the variance in
predicting language and communication skills, the
presence of stereotyped behaviors, and other pro-
totypically autistic behaviors, which may partly
contribute to group assignment. Indeed, Volkmer,
Cohen, Bregman, Hooks, and Stevenson (1998)
found that IQ is often a predictor of social sub-
type assignment; however, it may not fully ac-
count for it. Because level of intellectual function-
ing may be among the strongest indicators of sub-
type, investigators have often attempted to divide
the autism spectrum disorder group by choosing
an a priori IQ cutoff in order to designate and
then characterize the resulting high- and low-func-
tioning groups (Allen et al., 2001; Bartek & Rutter,
1976). The lower functioning cognitive subgroup,
defined as having an IQ below 70 or 80, tends to
exhibit more self-injury, stereotypies, and proto-
typical autism behaviors. Yet such cutoff points
are somewhat arbitrary, making it likely that there
is diagnostic overlap between the cognitive sub-
groups generated.
Furthermore, distinctions between verbal and
nonverbal information-processing abilities are of-
ten not explored, but may be important in iden-
tifying subtypes in autism. Tager-Flusberg and Jo-
seph (2003) investigated discrepancies between
verbal and nonverbal IQ in children with autism
and found that children with discrepantly high
nonverbal skills relative to verbal skills had greater
social impairment, independent of absolute level
of verbal ability and overall ability.
Attempts have been made to investigate sub-
groups within a dimensional construct on the ba-
sis of non-unimodal distributions. Meehl (1995)
noted that although bimodality and marked skew-
ness may be suggestive of latent groups, the pres-
ence of bimodality is neither a necessary nor suf-
ficient condition for the existence of latent sub-
groups. For example, when two latent distribu-
tions have a mean difference of 2 SDs and equal
variances, bimodality may not even be apparent.
On the other had, Grayson (1987) noted that even

䉷American Association on Intellectual and Developmental Disabilities 441
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
when bimodality is observed in measured vari-
ables, the underlying structure may still be a con-
tinuous dimension.
Statistical strategies may provide a more em-
pirical basis for characterizing individuals within
possible autism spectrum disorder subgroups. A
review of the literature indicates that most cluster
analytic studies yield two, three, or four subgroups
based on degree of impairment. Sevin et al.
(1995), for example, used cluster analysis to clas-
sify 34 children with autism or pervasive devel-
opment disorder-not otherwise specified (PDD-
NOS) into four groups, described as ranging from
high-functioning to low-functioning (severe) au-
tism, with IQ decreasing with severity, and differ-
ing significantly between groups. Similarly, Eaves,
Ho, and Eaves (1994) used a standard clustering
algorithm and principal components analysis of
variables to assign 166 children into four groups,
ranging from typically autistic and lower function-
ing to a higher functioning group that more close-
ly resembled Asperger syndrome. Again, severity
of autism was related to intellectual impairment
in that the most impaired subtype had the lowest
average IQ. In a longitudinal examination of 138
school-age children with autism, Stevens et al.
(2000) employed hierarchical agglomerative clus-
ter analysis to validate a two group solution, in
which cognitive level was the largest separating
variable. Children who were lower functioning as
defined by nonverbal IQ at preschool age tended
to show poorer outcome at school age, suggesting
that nonverbal IQ is an extremely potent predic-
tor of membership among school-age children.
Often, cluster analytic techniques have been
used to determine which behavioral features of
autism tend to correlate or account for the ma-
jority of variance or which factors cluster togeth-
er. Once a cluster solution is determined and in-
dividuals are assigned to groups, the subtypes are
characterized using various descriptors, including
level of intellectual functioning. In using IQ as a
descriptor only after the groups have been defined,
however, makes it difficult to determine the true
role of intellectual capacity in the formation of
subgroups, and the actual distribution of IQ in
the samples. Few investigators have focused exclu-
sively on cognitive functioning as the empirical
indicator of subgroup classification. Those who
have specifically investigated the role of intellec-
tual capacity in differentiating autism spectrum
disorder subtypes have often found that IQ is the
most significant contributor in discriminating be-
tween groups and the basis of differences between
subtypes (e.g., Miller & Ozonoff, 2000).
Although the goal of cluster analysis is to de-
termine the categories underlying autism spec-
trum disorders, these methods often yield groups
with considerable diagnostic overlap. Unfortu-
nately, under such conditions, cluster analysis of-
ten (a) fails to identify the correct number of clus-
ters in datasets where group membership is known
and (b) performs poorly in sorting individuals into
subgroups (e.g., Krieger & Green, 1999; Tonidan-
del & Overall, 2004). Furthermore, statisticians
have long recognized that clustering algorithms
partition datasets into subgroups, even if the dis-
tributions are known to be continuous (see Beau-
chaine, 2003). Thus, results derived solely from
cluster analysis do not provide strong evidence for
subgroups of autism and do not eliminate the pos-
sibility of a spectrum of autistic-like disorders (Pri-
or et al., 1998). In fact, data from eight cluster
analytic studies suggest that children with PDD-
NOS may fit into one of two overlapping groups
and that the subtypes resemble each other, exist-
ing along a continuum, and differing only by de-
gree of impairment (Myhr, 1998).
In a review of subtyping studies of autism,
Beglinger and Smith (2001) posited their best
guess that symptom heterogeneity can be repre-
sented by three continua (developmental delay,
social impairment, and repetitive behaviors), and
rough divisions can be drawn along these contin-
ua yielding four subgroups. The authors also not-
ed the weaknesses associated with cluster analytic
techniques, including the dependence on the in-
vestigators’ choice of variables and characteristics
of the sample. This conclusion of the presence of
a ‘‘continuum containing subgroups’’ highlights
the continued difficulty researchers in this area
have in determining whether true differences be-
tween subgroups in autism can be reliably distin-
guished.
In part as a result of the limitations of cluster
analysis, additional classification techniques, in-
cluding latent class analysis and taxometrics, have
been developed. Although rarely used to evaluate
whether subgroups of autism exist, these tech-
niques offer several advantages over clustering al-
gorithms (Beauchaine & Marsh, 2006). For ex-
ample, latent class analysis provides objective
measures of fit for comparing alternative sub-
groupings, and taxometric analyses are far less
prone to identify spurious subgroups within con-
tinuous distributions. The lone example of taxo-

442 䉷American Association on Intellectual and Developmental Disabilities
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
metric analysis (based on an adaptation of the re-
gression-mixture model, Golden & Mayer, 1995)
in autism is the Autism and Language Disorders
Nosology project (Rapin, 1996), in which inves-
tigators found evidence for two discrete sub-
groups, or taxa, in a sample of children with PDD
(Fein, et al., 1999) with a nonverbal IQ of about
65, optimally dividing the groups. In the present
paper, we used both latent class analysis and max-
imum covariance (MAXCOV), the most widely
studied taxometric algorithm, to address the ques-
tion of whether subgroups of autism spectrum dis-
order can be identified from the verbal and non-
verbal IQs of probands.
To summarize, although it is unclear whether
distinct subtypes of autism exist, a recurring pat-
tern emerges in which IQ strongly predicts social
functioning, adaptive behavior, severity of symp-
toms, and prognosis (Bolte & Poustka, 2002; Car-
pentieri & Morgan, 1996; Coplan & Jawad, 2005;
Howlin, Goode, Hutton, & Rutter, 2004; Liss et
al., 2001). We used both MAXCOV and latent
class analysis to analyze verbal and nonverbal IQs
obtained from a large sample of preschool-age
children diagnosed with autism spectrum disor-
der, who were evaluated through the National In-
stitute for Child Health and Human Develop-
ment (NICHD) Collaborative Program of Excel-
lence in Autism. Although cluster analysis offers
no proven means of choosing among models with
different numbers of classes and tends to overex-
tract classes when defining subtypes, latent class
analysis and MAXCOV offer an alternative and
more conservative approach in determining
whether there is a bimodal or multimodal distri-
bution of intellectual functioning among individ-
uals with autism. By using young children in this
analysis, we hoped to minimize individual differ-
ence related to experience and treatment.
Method
Participants
Participants were 456 children (370 boys
[81%], 86 girls [19%]) with autism spectrum dis-
order who were between the ages of 24 and 66
months (M⫽43.4, SD ⫽8.7); they were partic-
ipating in studies affiliated with the NICHD Col-
laborative Program of Excellence in Autism. Ex-
clusionary criteria included the presence of a neu-
rological disorder of known etiology, significant
sensory or motor impairment, major physical ab-
normalities, and history of serious head injury
and/or neurological disease. Diagnosis of autism
spectrum disorder was based on administration of
the Autism Diagnostic Observation Schedule-Ge-
neric (ADOS-G) and Autism Diagnostic Inter-
view-Revised (ADI-R). All of the children met cri-
teria for autism (n⫽357, 78%) or autism spec-
trum disorder (n⫽99, 22%) on the ADOS-G.
Nearly all of the children met criteria for a diag-
nosis of autism on the ADI-R (n⫽431, 95%),
with the remaining 25 children within 2 points of
a diagnosis of autism on the ADI-R. Informed
consent was appropriately obtained from each
child’s parent/guardian prior to their participation
in this study.
Measures
Autism Diagnostic Interview–Revised. The
ADI-R (Lord, Rutter, & Le Couteur, 1994) is a
structured, standardized parent interview devel-
oped to assess the presence and severity of symp-
toms of autism in early childhood across all three
main symptom areas (social relatedness; commu-
nication; and repetitive, restrictive behaviors).
This interview has been psychometrically validat-
ed across a wide range of ages and severity levels
in autism. Each site contained one experimenter
who was trained to reliability by one of the au-
thors (C. Lord) on the ADI-R; that person then
trained other raters in her lab to a reliability of
85% or better.
Autism Diagnostic Observation Schedule–Generic
(Lord et al., 2000). The ADOS-G is a semi-struc-
tured standardized interview using developmen-
tally appropriate social and toy-based interactions
in a 30- to 45-min interview to elicit symptoms of
autism in four areas: social interaction, commu-
nication, play, and repetitive behaviors. The
ADOS-G consists of four modules, each directed
at a particular level of language ability. In the pre-
sent study, all participants received Module 1, de-
veloped for preverbal children or those just begin-
ning to speak. The ADOS-G has been psycho-
metrically validated across a wide range of ages
and severity levels in autism (Lord et al., 2000).
Lord trained an experimenter at each site to reli-
ability on the ADOS-G at the University of Chi-
cago; that person then trained other raters in the
lab to a reliability of 85% or better.
Mullen Scales of Early Learning (1997). This in-
strument is a standardized developmental test for
children ages 3 months to 60 months consisting
of five subscales: Gross Motor, Fine Motor, Visual
Reception, Expressive Language, and Receptive

䉷American Association on Intellectual and Developmental Disabilities 443
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
Language. The last four are combined to yield an
overall composite score of intellectual function-
ing. The Mullen Scales of Early Learning dem-
onstrates strong concurrent validity with other
well-known developmental tests of motor, lan-
guage, and cognitive development. We adminis-
tered this test to all participants according to stan-
dard instructions by raters who were trained in
assessing young children with autism and other
developmental disorders. Reinforcers for all par-
ticipants in all groups were used at times to reward
cooperation and attention. A set of four devel-
opmental quotients for each participant was con-
structed dividing the age equivalence score for
each Mullen subscale by the child’s chronological
age (CA) and then multiplying by 100. We cal-
culated an overall verbal score by averaging the
receptive and expressive language scores and a
nonverbal IQ by averaging the visual reception
and fine motor scores. As has been done in other
samples with young children who have autism
(e.g., Lord et al., 2006), we used ratio based scores
throughout this paper because the Mullen sub-
scale T-scores commonly yielded a floor score in
this sample (% with floor T score of 20: Visual
Reception, 59%; Fine Motor, 72%; Receptive
Language, 80%; Expressive Language, 76%).
Vineland Scales of Adaptive Behavior, Interview
Edition. The Vineland (Sparrow, Balla, & Cicchet-
ti, 1984) is a standardized parent interview de-
signed to assess adaptive behavior across four do-
mains: Social, Communication, Daily Living, and
Motor Skills. Standard scores of these four do-
mains were used because floor effects were un-
common given the wide range of normative data
available for these domain scores.
Results
Maximum Covariance Analysis
Taxometric analyses were conducted using
MAXCOV, which is particularly well suited for
identifying overlapping dichotomous distribu-
tions embedded within a range of observed scores.
(A second taxometric algorithm, mean above mi-
nus below a cut was also applied to the data. Be-
cause the results of these analyses were fully con-
sistent with the results obtained from MAXCOV,
the more commonly used taxometric algorithm
[Haslam & Kim, 2002], only the latter are report-
ed.) In contrast to common clustering algorithms
that tend to impose structure upon a dataset re-
gardless of whether it is continuous or discrete in
nature, taxometric procedures begin with the null
hypothesis that the measured traits represent con-
tinua and seek disconfirming evidence of this as-
sumption (Beauchaine, 2003). In MAXCOV, var-
iables are taken three at a time and the covariance
of two is calculated within adjacent intervals of
the third. A smoothed regression function is then
fitted through the resulting covariance values.
Peaked regression functions are suggestive of dis-
crete latent classes, whereas flat regression func-
tions characterize continua (see Beauchaine &
Marsh, 2006; Waller & Meehl, 1998). In the dis-
crete case, the location of the MAXCOV peak
identifies the most efficient cutoff point for divid-
ing the sample, which in turn allows for estima-
tion of the base rates for both groups. With four
variables for analysis (Mullen IQs), 12 nonredun-
dant MAXCOV plots can be generated. When
peaked functions consistently emerge within the
same interval, indicating similar baserates regard-
less of the variable combinations used, more con-
fidence can be placed in the identified classes as
being truly discrete. Given such consistency,
Bayesian-estimated group membership probabili-
ties are calculated by combining information from
each MAXCOV run. Six of the 12 conditional
covariance plots indicate a sharp peak in the func-
tion indicating a low base rate taxon at the high
end of the Mullen IQ distributions. Baserate es-
timates based on the posterior probability of tax-
on membership resulted in a subgroup of high
Mullen scores that comprised 17.8% of the sam-
ple.
Latent Class Analysis
After finding evidence for discontinuity in the
IQ distributions in the sample using the MAX-
COV procedure, we then examined the data using
latent class analysis (Lazarsfeld & Henry, 1968),
which is a maximum likelihood-based method
that provides model-based parameter estimates in
contrast to model-free methods such as cluster
analysis. The software program M-plus (Muthen
& Muthen, 2004) was used to assess whether there
is evidence of multiple, unobserved latent classes
in this sample based on the four subscales of the
Mullen. Models specifying 2, 3, and 4 latent clas-
ses were run. Variances for each class were as-
sumed to be equal in order to minimize the num-
ber of parameters being estimated.
Results indicated that a two-group model fit
the data significantly better than a single group
model (see Table 1). Similarly, three groups fit

444 䉷American Association on Intellectual and Developmental Disabilities
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
Table 1. Latent Class Analysis Fit Indices
No. of
classes
Bayesian
information
criteria Entropy
Loglikli-
hood
Lo-
Mendel-
Rubin
value
One 897.0
Two ⫺1882.3 0.840 1017.7 237.8**
Three ⫺1886.4 0.690 1053.4 70.4*
Four ⫺1892.5 0.773 1090.1 68.8*
Five ⫺1850.9 0.748 1103.0 73.9
*p⬍.05. ** p⬍.01.
Figure 1. Mean Mullen IQs of latent groups.
LCA ⫽latent class analysis. Rec ⫽Receptive Lan-
guage, Exp ⫽Expressive Language, VR ⫽Visual
Reception, and FM ⫽Fine Motor.
better than two, and four groups fit better than
three. No improvement in fit was seen with a five-
group model. Means on the Mullen IQs for the
four groups identified in the four-groups latent
class analysis solution are illustrated in Figure 1.
Rather than simply reflecting overall differences
in intellectual functioning, these groups illustrate
striking differences in both the absolute level of
functioning as well as the relative abilities of the
verbal and nonverbal areas. The lowest group
(59% of the sample) was characterized by extreme-
ly low verbal scores as well as very low nonverbal
scores, with an average difference between verbal
and nonverbal scores of 22 points. The second
group (12.5%) was similar; however, the discrep-
ancy between verbal and nonverbal scores in this
group was even more extreme, with nonverbal
scores averaging 42 points higher than verbal
scores. The third group (21.7%) showed moderate
to mild level of impairment, with scores in the 60
to 70 range, with verbal abilities commensurate
with nonverbal abilities. The final group (7.0%)
reflected a subgroup of children functioning in
the average range, again with verbal and nonver-
bal areas at roughly comparable levels. This fourth
group was the only one in which the pair of Ver-
bal or Nonverbal subtests showed a widely differ-
ent pattern. In this group the Visual Reception
scale was much noticeable higher than the Fine
Motor subscales.
Children in Group 2, which showed the very
large discrepancy between verbal and nonverbal
scores, were, on average, nearly one year younger
than the children in the other groups, F(3, 452)
⫽26.6, p⬍.001. Means and SDs for the age in
months for each group were Group 1, 45.3 (8.5);
Group 2, 34.8 (6.8); Group 3, 43.5 (7.7), Group
4, 42.5 (7.0). Boys and girls were equally likely to
be present in each of the four groups, ␹
2
⫽0.85,
p⫽.839. The number and percentage of boys and
girls present in each group are as follows: boys,
Groups 1 to 4, respectively, 217 (58.6%), 45
(12.2%), 83 (22.4%), and 25 (6.8%); girls, Groups
1 to 4, respectively, 51 (59.3%), 12 (14.0%), 16
(18.6%), and 7 (8.1%).
Comparison Between Latent Class Analysis
and MAXCOV Procedures
Although the four-group model was found to
provide the best fit of the data in the latent class
analyses, we wanted to compare the classification
results of the two class latent class analysis model
with the MAXCOV results to directly compare
the two-group solution of these different meth-
ods. Both methods identified a relatively small
group of higher functioning children, with the
two-group latent class analysis model placing
19.1% of the sample in this high group; and
MAXCOV, 15.1% of the sample in this group.
Overall, there was high agreement between the la-
tent class analysis constrained variance classifica-
tion results and the MAXCOV procedure because
85.6% of individuals were classified similarly by
the two approaches. Thus, 14.4% was classified
differently by these two methods. When examin-
ing the combination of the latent class analysis,
we found that the two-group results and the
MAXCOV results with the latent class analysis
four-group results described above, had a striking
similarity. Children placed in the low group by
the latent class analysis two-group model but in
the high group by MAXCOV show the same ex-
treme difference between their verbal and non-

䉷American Association on Intellectual and Developmental Disabilities 445
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
Figure 2. Mean Mullen scores of latent groups
by combining classification methods. LCA ⫽la-
tent class analysis, RL. ⫽Receptive Language, EL
⫽Expressive Language, VR ⫽Visual Reception,
and FM ⫽Fine Motor.
Figure 3. Convex hull plots and frequency dis-
tributions of Mullen Verbal and Nonverbal IQ by
latent class analysis (LCA) groups. For the convex
hull plot, marginal distributions reflect relative
density plots for each group. Each cross reflects
the bivariate group mean ⫾1SD. The diagonal
line depicts equivalent verbal and nonverbal IQ.
Group 1, Low (dashed line), Group 2, Low/Me-
dium (solid line); Group 3, Medium (dash-dotted
line); Group 4 (dotted line).
verbal scores (Figure 2). Children in the opposite
group (latent class analysis high/MAXCOV low)
show relatively equal verbal and nonverbal scores
in the 60 to 70 range.
Figure 3 illustrates the relationship between
the Mullen Verbal (Mof Receptive and Expressive
Language subscales) and Nonverbal (Mof Visual
Reception and Fine Motor subscales) IQs using a
convex hull plot (Vida & Polar, 2005). Given the
substantial overlap of verbal and nonverbal scores
between the groups, the convex hull for each sub-
group, rather than each individual point, is dis-
played. Relative density plots for each variable are
shown in each margin. This plot reveals the com-
plexities involved in categorizing young children
with autism spectrum disorder in terms of their
intellectual functioning. Clearly the low and low/
medium groups showed strengths in nonverbal
relative to verbal performance. However, the two
higher groups show more commensurate verbal
and nonverbal performance. The relative density
plots nicely reveal the overlapping verbal and
nonverbal distributions; however, information
about the relative sample size is lost. The line
graphs depict the overall verbal and nonverbal IQ
distributions broken down by latent class analysis
group membership.
Notably, the groups depicted in these plots
have been assigned descriptive labels of low, low
verbal/medium nonverbal, medium, and high. Al-
though helpful as a shorthand method to referring
to these latent classes, one should not place undo
emphasis on the meaning of these labels. For ex-

446 䉷American Association on Intellectual and Developmental Disabilities
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
Table 2. Vineland, Autism Diagnostic Interview-Rev. (ADI), and Autism Diagnostic Observation
Schedule-Generic (ADOS) Scores by Latent Class Analysis Group
Measure
Low (1)
Mean SD
Low verbal/nonverbal (2)
Mean SD
Medium (3)
Mean SD
High (4)
Mean SD
Vineland
Socialization 57.6 6.4 60.1 9.0 65.2 8.4 66.8 10.0
Com.
b
54.0 6.8 58.6 5.3 67.3 8.2 77.9 12.8
Daily Living Skills 57.2 7.8 65.3 6.8 64.0 8.3 67.4 8.4
Motor Skills 63.2 14.2 77.8 16.9 72.1 12.9 81.7 16.7
ADOS
Social 11.4 2.1 11.5 2.1 9.2 2.5 8.5 2.7
Com. 6.5 1.4 6.5 1.5 5.7 1.6 5.8 1.6
ADI
Social 20.1 4.7 18.7 3.9 17.1 4.9 15.7 5.0
Com. 11.9 2.3 11.2 2.2 12.5 4.1 13.3 3.7
Repetitive 5.1 1.8 4.3 1.7 5.4 2.2 5.9 2.8
a
Cell values indicate significant pairwise group differences at p⬍.05.
b
Communication.
ample, children placed in the high group in the
latent class analysis will not necessarily be de-
scribed clinically as having high-functioning au-
tism. Some children in both the medium and
high groups have verbal IQs above 80, and some
children in all but the low group have nonverbal
IQs above 80. One can see that the medium
group in Figure 3 overlaps with every other group.
Children whose scores fall into these areas of
overlap have notably lower posterior probabilities
of group membership. Though the mean posterior
probability for group membership was .88, 24%
of the sample had probabilities of less than .80,
and 8%, less than .60.
Group Classification and Relationship to
ADI, ADOS, and Vineland
Finally, we compared the groups identified via
the latent class analysis four-group model on the
Vineland, ADI, and ADOS measures (Table 2)
with ANOVA and Bonferroni post-hoc compari-
sons. Vineland Socialization and Communication
domains both showed a gradation in scores con-
sistent with the findings for the Mullen. Vineland
Communication scores were increasingly higher
across Groups 1 to 4 and Socialization scores were
increasingly higher across Groups 1 to 3, with no
difference between Groups 3 and 4. On the Daily
Living Skills and Motor Skills domains, only
Group 1 had significantly lower scores compared
with the other three groups.
The ADOS social and ADI Social scores
showed similar patterns in which Groups 1 and 2
tended to show greater impairment (i.e., higher
scores) than did Groups 3 and 4. Groups 1 and 2
also showed more impairment than did Group 3
on the ADOS Communication score. In contrast,
on the ADI Communication scores, Groups 3 and
4 had higher scores than did Group 2. This likely
reflects the greater number of items for which ver-
bal children may be rated compared to nonverbal
children. Finally, Groups 3 and 4 had significantly
higher ADI Repetitive and Stereotyped Behaviors
scores than did Group 2.
Clearly, these groups differed in their adaptive
behavior and symptom presentation; however, the
question remains as to whether these differences
simply reflect the overall relationship of verbal
and nonverbal abilities to adaptive behavior and
autism symptoms. To examine this question, we
regressed verbal (Mof Receptive and Expressive
Language subscales) and nonverbal (Mof Visual
Reception and Fine Motor subscales) IQs on each
Vineland, ADOS, and ADI measure. Then, we
added latent class analysis group membership as
a second step in the regression, using three di-
chotomous indicators (Variable 1 coded 1 for
Group 2, Variable 2 coded 1 for Group 3, and
Variable 3 coded 1 for Group 4; Group 1 was
coded zero on all three variables). Table 3 shows
these results in which latent class analysis group
membership added significantly to the prediction

䉷American Association on Intellectual and Developmental Disabilities 447
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
Table 2. Extended
F
Post-hoc Bonferroni comparisons
a
1 vs. 2 1 vs. 3 1 vs. 4 2 vs. 3 2 vs. 4 3 vs. 4
37.72 1 ⬍21⬍31⬍42⬍32⬍4
131.06 1 ⬍21⬍31⬍42⬍32⬍43⬍4
35.01 1 ⬍21⬍31⬍4
27.41 1 ⬍21⬍31⬍4
35.80 1 ⬎31⬎42⬎32⬎4
8.57 1 ⬎32⬎3
16.00 1 ⬎31⬎42⬎4
4.68 2 ⬍32⬍4
5.40 2 ⬍32⬍4
of Vineland Socialization, Communication, Daily
Living Skills, and ADI Social scores above and
beyond that of verbal and nonverbal IQ. This pro-
vides another line of evidence for latent classes of
IQ as opposed to a single or even bivariate (name-
ly, verbal and nonverbal) continuous dimension
of IQ in young children with autism spectrum
disorders.
Discussion
Heterogeneity in autism has long been noted.
Variability in IQ is routinely one of the largest
contributors to this heterogeneity. In the current
study we examined the degree to which variability
in information-processing, both verbal and non-
verbal, may reflect unobserved or latent groups
within a large sample of preschool children with
autism spectrum disorder. Thus, rather than using
a single wide-ranging distribution of IQ, we
sought to determine whether there is evidence
that this wide range of functioning may reflect the
presence of discrete latent groups. By utilizing two
different techniques, latent class analysis and tax-
ometric analysis (Beauchaine, 2003), we avoided
some of the difficulties present in earlier work in
which researchers investigated subgroups in au-
tism based on IQ. First, we employed a taxometric
analysis to assess whether evidence for discrete dis-
tributions of IQ could be found. This approach
did suggest discontinuity in the IQ distribution at
the higher end of the Mullen scale.
With this evidence, we used latent class anal-
yses to explore this question and to assess whether
two or more latent groups based on IQ could be
identified. The latent class analysis revealed a
four-group solution as providing the best fit for
these data. These four groups can be described as
a mixture of the overall level of functioning in the
group, coupled with the presence or absence of a
large discrepancy between verbal and nonverbal
functioning. The largest group, comprised of chil-
dren with low nonverbal and very low verbal
scores, reflected those with the most severe cog-
nitive impairments. A second group also showed
very low verbal abilities but had nonverbal scores
over 40 points higher, on average. The third and
fourth groups reflected commensurate verbal and
nonverbal abilities, with the third group showing
mild to moderate impairments in their cognitive
functioning; and the fourth, highest functioning
group scoring in the low average range.
Groups with identical patterns of Mullen
scores were observed when contrasting the two
group latent class analysis results with those from
the MAXCOV method, which yields, by defini-
tion, two groups when evidence of taxonicity is
present. Children who were classified in a similar
manner with these two methods matched the low-
est and highest functioning groups identified in
the latent class analysis four-class model. Children
classified differently across the two methods com-
prised two distinct IQ-based subgroups that were
different from both a simple low- or high-func-

448 䉷American Association on Intellectual and Developmental Disabilities
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
Table 3. Linear Regression Analyses Predicting Vineland, Autism Diagnostic Interview-Rev. (ADI),
and Autism Diagnostic Observation Schedule-Generic (ADOS) From Mullen IQ and Latent Class
Analysis Group
Measure
Step 1
R
2
Verbal IQ ␤(t) Nonverbal IQ ␤(t)
Vineland
Socialization .69*** 0.75 (19.49***) 0.11 (2.94**)
Com.
a
.31*** 0.46 (8.29***) 0.12 (2.18*)
Daily Living Skills .38*** 0.04 (0.67) 0.59 (10.73***)
Motor Skills .319*** ⫺0.04 (⫺0.76) 0.59 (10.40***)
ADOS
Social .249*** ⫺0.59 (⫺10.13***) 0.14 (2.39*)
Com. .061*** ⫺0.30 (⫺4.56***) 0.07 (1.15)
ADI
Social .166*** ⫺0.40 (⫺6.61***) 0.00 (⫺0.08)
Com. .044*** 0.30 (4.55***) ⫺0.23 (⫺3.56***)
Repetitive .032*** 0.25 (3.82***) ⫺0.20 (⫺3.06**)
a
Communication.
*p⬍.05. **p⬍.01. ***p⬍.001.
tioning group as well as being different from each
other. The verbal versus nonverbal dimension of
intellectual functioning in addition to the abso-
lute level of overall functioning is, therefore, im-
portant to evaluate when considering subgroups
of children with autism.
Though no gender differences were present in
the likelihood of group classification, there was an
intriguing mean difference in age among the
groups. Children in Group 2, who showed an ex-
treme discrepancy between verbal and nonverbal
scores, were, on average, nearly one year younger
than the children in the other groups. In addition
to the low verbal scores and large verbal–nonver-
bal discrepancy, this group showed the lowest lev-
el of repetitive and stereotyped behaviors on the
ADI. One interpretation of these findings is that
Group 2 may reflect a developmental or matura-
tional effect rather than a uniquely different IQ
subgroup. As children in this group age, those
whose language and communication ability im-
proved would look like those in Group 3 (with
more commensurate verbal and nonverbal abili-
ties), whereas those who made slow progress
would tend to resemble members of Group 1
(who had more profound verbal impairments).
Perhaps, as well, levels of repetitive behaviors may
increase to levels similar to the other groups be-
cause these behaviors often follow the social and
communication impairments in young children
with autism. Alternatively, this group may signify
a specific language impairment in these children,
as described by Kjelgaard and Tager-Flusberg
(2001). Whether language deficits relative to non-
verbal IQ in children with autism compared with
children who have specific language impairments
represent the same or different underlying mech-
anisms is not known. However, clarifying areas of
similarity and difference among children with au-
tism spectrum disorder, children with other de-
velopmental disabilities, and typically developing
children continues to be an important line of in-
quiry for understanding the genetics and under-
lying neurobiologic and neurocognitive mecha-
nisms involved in autism.
This age difference in Group 2 also raises an
important measurement issue when working with
children who have moderate to severe cognitive
impairments. Ratio-based IQs were used in this
study to avoid the problem of a floor effect on
low-end scores; differences in mean age among
the groups (which serves as the denominator in
the scores) may come into play. For example,
there may be specific items on the Mullen that
tend to be particularly difficult (e.g., those requir-
ing imitation?) or easy (e.g., those clearly requiring
no spoken instruction) for children with autism in
comparison to children in the normative sample
functioning at roughly the same developmental
level. Simply examining age-equivalence scores

䉷American Association on Intellectual and Developmental Disabilities 449
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
Table 3. Extended
Step 2
⌬R
2
Verbal IQ ␤(t) Nonverbal IQ ␤(t) Group 2 ␤(t) Group 3 ␤(t) Group 4 ␤(t)
.008* 0.91 (14.19***) 0.07 (1.54) 0.05 (1.38) ⫺0.1 (⫺2.23*) ⫺0.11 (⫺2.22*)
.035*** 0.71 (7.86***) 0.18 (2.65**) ⫺0.03 (⫺0.70) ⫺0.17 (⫺2.55*) ⫺0.34 (⫺4.71***)
.027*** 0.23 (2.59**) 0.59 (8.89***) 0.04 (0.75) ⫺0.08 (⫺1.25) ⫺0.25 (⫺3.53***)
.013⫹0.13 (1.36) 0.57 (8.06***) 0.04 (0.84) ⫺0.1 (⫺1.36) ⫺0.17 (⫺2.24*)
.003 ⫺0.56 (⫺5.85***) 0.12 (1.66) 0.01 (0.11) ⫺0.04 (⫺0.63) 0.02 (0.33)
.006 ⫺0.28 (⫺2.56*) 0.06 (0.81) ⫺0.02 (⫺0.34) ⫺0.07 (⫺0.86) 0.04 (0.44)
.019* ⫺0.65 (⫺6.45***) 0.04 (0.50) ⫺0.06 (⫺1.13) 0.16 (2.11*) 0.20 (2.60**)
.008 0.19 (1.72) ⫺0.26 (⫺3.21**) 0.01 (0.20) 0.07 (0.88) 0.15 (1.84)
.013 0.08 (0.70) ⫺0.17 (⫺2.13*) ⫺0.05 (⫺0.94) 0.09 (1.17) 0.16 (1.92)
rather than the ratio-based scores does not address
this issue because functioning at a 20-month-old
level will mean vastly different things for a 2- ver-
sus a 4-year-old child.
The highest functioning group identified by
the latent class analysis (Group 4) had much high-
er mean Mullen scores (verbal IQ ⫽88, nonver-
bal IQ ⫽97) compared to Group 3 (verbal IQ ⫽
63, nonverbal IQ ⫽70). Despite this large differ-
ence in intellectual functioning, these groups did
not differ on the Vineland Socialization, Daily
Living Skills, or Motor Skills domains. The aver-
age standard score across the four Vineland sub-
domains was 73 for Group 4 and 67 for Group 3,
showing a much bigger relative weakness in adap-
tive functioning for Group 4. In addition, Groups
3 and 4 did not differ on any ADOS or ADI
scores. Despite the striking difference in levels of
cognitive functioning between Groups 3 and 4,
this did not translate into systematic differences
in adaptive functioning or level of autism symp-
toms as measured by the ADOS and ADI.
We also found that latent class analysis group
membership accounted for a significant propor-
tion of the variability of Vineland Socialization,
Communication, Daily Living Skills, and ADI so-
cial scores beyond that accounted for by the Mul-
len Verbal and Nonverbal IQs. This provides ad-
ditional evidence of the importance of consider-
ing IQ in young children with autism spectrum
disorder more than simply a single dimension. In-
deed, the results presented here suggest that there
is more than a simple linear relationship between
intellectual functioning and adaptive behavior
and autism symptoms, even when independently
measuring both verbal and nonverbal intellectual
abilities.
This study comprises a very large sample of
young children who have been carefully diag-
nosed as being on the autism spectrum, and, thus,
we believe the findings here should generalize to
the broader population of children with autism
spectrum disorder in this age range. However, this
sample was not collected as a population-based
epidemiologic sample. Replication is clearly need-
ed. One should be particularly cautious in making
assumptions regarding the relative size of each of
the IQ groups described. Of the 1,824 Mullen
subscales (Four Subscales ⫻456 Children), only
11% represent scores in the ‘‘average’’ range (T
scores of 40 or greater, 16th percentile or greater).
In many studies of older children with autism
spectrum disorder, researchers have reported a
much higher proportion in the average range of
intellectual functioning. There are several issues
that may bear on the representativeness of the pre-
sent sample. First, this sample may truly contain
fewer high-functioning children with autism spec-
trum disorder than are present in the population
because families who seek to participate in uni-
versity studies with their preschooler who has au-
tism spectrum disorder may not be representative

450 䉷American Association on Intellectual and Developmental Disabilities
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
of all families with preschoolers who have autism
spectrum disorder. Second, these children were
tested primarily between late 1998 and 2001,
which may represent a cohort effect compared to
more recent samples. Third, by assessing pre-
schoolers and using a developmental instrument
designed for use from infancy through age 5, we
may have produced a sample that included a
greater proportion of more impaired children than
is often found in samples of older children. It is
often a fine line between getting a valid estimate
of a nonverbal child’s level of intellectual func-
tioning and simply being unable to collect valid,
interpretable, quantitative data. Some of these
children will continue to fall further and further
behind their same-age peers in terms of their cog-
nitive development. Many well-developed instru-
ments for older children are limited in their use
with the low end of intellectual functioning be-
cause normative data are simply not available.
Thus, there may be a portion of children in the
present sample, who, by virtue of the difficulties
in obtaining meaningful quantitative data, may
simply never take part in studies of older children
with autism spectrum disorder.
Whether the methods used here or similar sta-
tistical methods will provide evidence for latent
classes based on IQ in older children remains an
open question that necessities further study. How-
ever, careful attention needs to be paid to how
the use of a given measurement instrument will
impact the scores it yields (e.g., How much vari-
ability on the low end of the scale is there?) and
even the resulting sample the study produces.
This is particularly true when verbal information-
processing and language development in children
with autism are studied (Tager-Flusberg, 2000).
The likelihood that a sample of preschoolers with
autism, when assessed 5 or 10 years later, will pre-
sent an even wider range of functioning than they
did at the outset makes the identification and
measurement of the underlying neurocognitive
mechanisms at work in autism that much more
difficult. In this pursuit, we must be aware of how
our measurement tools, presumably impartial and
objective, may also become construction tools
that actively shape and influence the phenome-
non we are studying. Any seasoned child assessor
knows, a la Heisenberg, that observing a child at
a table with an open briefcase full of booklets and
blocks can have a profound influence on what we
end up measuring.
In summary, by examining a large, well-char-
acterized sample of preschool children with au-
tism spectrum disorder, we have provided an im-
portant evidence of the presence of multiple IQ-
based subgroups within autism. Four latent classes
were identified that represent very different levels
of intellectual functioning as well as different pat-
terns of relative verbal versus nonverbal abilities.
We found that group membership was related to
adaptive functioning and social impairment,
above and beyond the direct relationship of verbal
and nonverbal IQ. Cross-sectional samples such
as this must be complemented with longitudinal
data because variability in course represents yet
another area of heterogeneity in autism where
much remains to be learned.
References
Allen, D. A., Steinberg, M., Dunn, M., Fein, D.,
Feinstein, C., Waterhouse, L., & Rapin, I.
(2001). Autistic disorder versus other perva-
sive developmental disorders in young chil-
dren: Same or different? European Child and
Adolescent Psychiatry, 10, 67–78.
American Psychiatric Association. (1994). Diag-
nostic and statistical manual of mental disorders
(4th ed.). Washington, DC: Author.
Bartek, L., & Rutter, M. (1976). Differences be-
tween mentally retarded and normally intel-
ligent autistic children. Journal of Autism and
Childhood Schizophrenia, 6, 109–120.
Beauchaine, T. P. (2003). Taxometrics and devel-
opmental psychopathology. Development and
Psychopathology, 15, 501–527.
Beauchaine, T. P., & Marsh, P. (2006). Taxometric
methods: Enhancing early detection, diagno-
sis, and prevention of psychopathology by
identifying latent vulnerability traits. In D.
Cicchetti & D. Cohen (Eds.), Developmental
psychopathology (2nd ed., pp. 931–967). Ho-
boken, NJ: Wiley.
Beglinger, L. J., & Smith, T. H. (2001). A review
of subtyping in autism and proposed dimen-
sional classification model. Journal of Autism
and Developmental Disorders, 31, 411–422.
Bolte, S., & Poustka, F. (2002). The relation be-
tween general cognitive level and adaptive be-
havior domains in individuals with autism
with and without co-morbid mental retarda-
tion. Child Psychiatry and Human Development,
33, 165–172.
Borden, M. C., & Ollendick, T. H. (1994). An
examination of the validity of social subtypes

䉷American Association on Intellectual and Developmental Disabilities 451
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
in autism. Journal of Autism and Developmental
Disorders, 24, 23–37.
Carpentieri, S., & Morgan, S. B. (1996). Adaptive
and intellectual functioning in autistic and
non–autistic retarded children. Journal of Au-
tism and Developmental Disorders, 26, 611–620.
Castelloe, P., & Dawson, G. (1993). Subclassifi-
cation of children with autism and pervasive
developmental disorder: A questionnaire
based on Wing’s subgrouping scheme. Journal
of Autism and Developmental Disorders, 23,
229–241.
Coplan, J., & Jawad, A. F. (2005). Modeling clin-
ical outcome of children with autistic spec-
trum disorders. Pediatrics, 116, 117–122.
Davies, S., Bishop, D., Manstead, A. S., & Tan-
tam, D. (1994). Face perception in children
with autism and Asperger’s syndrome. Journal
of Child Psychology and Psychiatry, 35, 1033–
1057.
Dawson, G., Meltzoff, A. N., Osterling, J., & Rin-
aldi, J. (1998). Neuropsychological correlates
of early symptoms of autism. Journal of Au-
tism and Developmental Disorders, 69, 1276–
1285.
Eaves, L. C., Ho, H. H., & Eaves, D. M. (1994).
Subtypes of autism by cluster analysis. Journal
of Autism and Developmental Disorders, 24, 3–
21.
Fein, D., Stevens, M., Dunn, M., Waterhouse, L.,
Allen, D., Rapin, I., & Feinstein, C. (1999).
Subtypes of pervasive developmental disor-
der: Clinical characteristics. Child Neuropsy-
chology, 5, 1–23.
Filipek, P. A., Accardo, P. J., Baranek, G. T.,
Cook, E. H., Dawson, G., Gordon, B., Gravel,
J. S., Johnson, C. P., Kallen, R. J., Levy, S. E.
et al. (1999). The screening and diagnosis of
autistic spectrum disorders. Journal of Autism
and Developmental Disorders, 29, 439–484.
Fombonne, E. (2003). Epidemiological surveys of
autism and other pervasive developmental
disorders: An update. Journal of Autism and
Developmental Disorders, 33, 365–382.
Ghaziuddin, M., & Mountain-Kimchi, K. (2004).
Defining the intellectual profile of Asperger
syndrome: Comparison with high-function-
ing autism. Journal of Autism and Developmen-
tal Disorders, 34, 279–284.
Gilchrist, A., Green, J., Cox, A., Burton, D., Rut-
ter, M., & Couteur, A. (2001). Development
and current functioning in adolescents with
Asperger syndrome: A comparative study.
Journal of Child Psychology and Psychiatry, 42,
227–240.
Golden, R. R., & Mayer, M. J. (1995). Peaked in-
dicators: A source of pseudotaxonicity of a
latent trait. In D. Lubinski & R. Dawis (Eds.),
Assessing individual differences in human behav-
ior: New concepts, methods, and findings (pp. 93–
115). Palo Alto, CA: Davies-Black.
Grayson, D. A. (1987). Can categorical and di-
mensional views of psychiatric illness be dis-
tinguished? British Journal of Psychiatry, 151,
355–361.
Haslam, N., & Kim, H. C. (2002). Categories and
continua: A review of taxometric research. Ge-
netic, Social, and General Psychology Mono-
graphs, 128, 271–320.
Howlin, P., Goode, S., Hutton, J., & Rutter, M.
(2004). Adult outcome for children with au-
tism. Journal of Child Psychology and Psychiatry,
45, 212–229.
Kjelgaard, M. M., & Tager-Flusberg, H. (2001). An
investigation of language impairment in au-
tism: Implications for genetic subgroups. Lan-
guage and Cognitive Processes, 16, 287–308.
Krieger, A. M., & Green, P. E. (1999). A caution-
ary note on using internal cross validation to
select the number of clusters. Psychometrika,
64, 341–353.
Lazarsfeld, P. F., & Henry, N. W. (1968). Latent
structure analysis. Boston: Houghton Mifflin.
Liss, M., Harel, B., Fein, D., Allen, D., Dunn, M.,
Feinstein, C., Morris, R., Waterhouse, L., &
Rapin, I. (2001). Predictors and correlates of
adaptive functioning in children with devel-
opmental disorders. Journal of Autism and De-
velopmental Disorders, 31, 219–230.
Lord, C., Risi, S., DiLavore, P. S., Shulman, C.,
Thurm, A., & Pickles, A. (2006). Autism from
2 to 9 years of age, Archives of General Psychi-
atry, 63, 694–701.
Lord, C., Risi, S., Lambrecht, L., Cook, E. H.,
Leventhal, B. L., DiLavore, P. C., Pickles, A.,
& Rutter, M. (2000). The Autism Diagnostic
Observation Schedule–Generic: A standard
measure of social and communication deficits
associated with the spectrum of autism. Jour-
nal of Autism and Developmental Disorders, 30,
205–223.
Lord, C., Rutter, M., & Le Couteur, A. (1994).
Autism Diagnostic Interview–Revised: A re-
vised version of the Autism Diagnostic Inter-
view for caregivers of individuals with possi-
ble pervasive developmental disorders. Journal

452 䉷American Association on Intellectual and Developmental Disabilities
VOLUME
113,
NUMBER
6: 439–452 円
NOVEMBER
2008
AMERICAN JOURNAL ON MENTAL RETARDATION
Latent classes of IQ in autism J. Munson et al.
of Autism and Developmental Disorders, 24,
659–685.
Meehl, P. E. (1995). Bootstraps taxometrics: Solv-
ing the classification problem in psychopa-
thology. American Psychologist, 50, 266–275.
Miles, J. H., Takahashi, T. N., Bagby, S., Sahota,
P. K., Vaslow, D. F., Wang, C. H., Hillman,
R. E., & Farmer, J. E. (2005). Essential versus
complex autism: Definition of fundamental
prognostic subtypes. American Journal of Med-
ical Genetics, 135A, 171–180.
Miller, J. N., & Ozonoff, S. (2000). The external
validity of Asperger disorder: Lack of evi-
dence from the domain of neuropsychology.
Journal of Abnormal Psychology, 109, 227–238.
Mullen, E. M. (1997). Mullen Scales of Early Learn-
ing. Los Angeles: Western Psychological Ser-
vices.
Muthen, L. K., & Muthen, B. O. (2004). Mplus
user’s guide (3rd ed.). Los Angeles: Author.
Myhr, G. (1998). Autism and other pervasive de-
velopmental disorders: Exploring the dimen-
sional view. Journal of Abnormal Psychology, 43,
589–595.
Prior, M., Eisenmajer, R., Leekam, S., Wing, L.,
Gould, J., Ong, B., & Dowe, D. (1998). Are
there three subgroups within the autism spec-
trum? A cluster analysis of a group of children
with autism spectrum disorders. Journal of
Child Psychology and Psychiatry, 39, 893–902.
Rapin, I. (1996). Preschool children with inade-
quate communication: Developmental lan-
guage disorders, autism, low IQ. Clinics in De-
velopmental Medicine (No. 139). London:
MacKeith Press.
Sevin, J. A., Matson, J. L., Coe, D., Love, S. R.,
Matese, M. J., & Benavidez, D. A. (1995). Em-
pirically derived subtypes of pervasive devel-
opmental disorders: A cluster analytic study.
Journal of Autism and Developmental Disorders,
25, 561–578.
Siegel, D. J., Minshew, N. J., & Goldstein, G.
(1996). Wechsler IQ profiles in diagnosis of
high–functioning autism. Journal of Autism
and Developmental Disorders, 26, 389–406.
Sparrow, S. S., Balla, S., & Cicchetti, D. V. (1984).
The Vineland Adaptive Behavior Scales (Survey
Form). Circle Pines, MN: American Guidance
Service.
Stevens, M. C., Fein, D. A., Dunn, M., Allen, D.,
Waterhouse, L. H., Feinstein, C., & Rapin, I.
(2000). Subgroups of children with autism by
cluster analysis: A longitudinal examination.
Journal of the American Academy of Child and
Adolescent Psychiatry, 39, 346–352.
Tager-Flusberg, H. (2000). The challenge of study-
ing language development in children with
autism. In L. Menn & N. Bernstein Ratner
(Eds.), Methods for studying language production
(pp. 313–332). Mahwah, NJ: Erlbaum.
Tager-Flusberg, H., & Joseph, R. M. (2003). Iden-
tifying neurocognitive phenotypes in autism.
Philosophical Transactions of the Royal Society of
London, B, 358, 303–314.
Tonidandel, S., & Overall, J. E. (2004). Determin-
ing the number of clusters by sampling with
replacement. Psychological Methods, 9, 238–
249.
Vidmar, G., & Pohar, M. (2005). Augmented con-
vex hull plots: Rationale, implementation in
R and biomedical applications. Computer
Methods and Programs in Biomedicine, 78, 69–
74.
Volkmer, F. R., Cohen, D. J., Bregman, J. D.,
Hooks, M. Y., & Stevenson, J. M. (1998). An
examination of social typologies in autism.
Journal of the American Academy of Child and
Adolescent Psychiatry, 28, 82–86.
Waller, N. G., & Meehl, P. E. (1998). Multivariate
taxometric procedures: Distinguishing types from
continua. Newbury Park, CA: Sage.
Wing, L., & Gould, J. (1979). Severe impairments
of social interaction and associated abnor-
malities in children: Epidemiology and clas-
sification. Journal of Autism and Childhood
Schizophrenia, 9, 11–29.
World Health Organization. (1992). ICD-10 clas-
sification of mental and behavioral disorders:
Clinical descriptions and diagnostic guidelines.
Geneva: Author.
Received 2/4/07, accepted 4/12/08.
Editor-in-charge: Frances Conners
Correspondence regarding this article should be
sent to Jeffrey Munson, University of Washing-
ton, Psychiatry and Behavioral Sciences, Seattle,
WA 98195. E-mail: jeffmun@u.washington.edu