J Autism Dev Disord (2014) 44:915–925

DOI 10.1007/s10803-013-1945-1

ORIGINAL PAPER

National Survey of Sensory Features in Children with ASD:

Factor Structure of the Sensory Experience Questionnaire (3.0)
Karla Ausderau • John Sideris • Melissa Furlong •
Lauren M. Little • John Bulluck • Grace T. Baranek

Published online: 6 October 2013

Ó Springer Science+Business Media New York 2013

Abstract This national online survey study characterized The SEQ-3.0 demonstrates an empirically valid factor
sensory features in 1,307 children with autism spectrum structure specific to ASD that considers sensory response
disorder (ASD) ages 2–12 years using the Sensory Expe- patterns, modalities, and social context.
riences Questionnaire Version 3.0 (SEQ-3.0). Using the
SEQ-3.0, a confirmatory factor analytic model with four Keywords Autism spectrum disorders
Factor
substantive factors of hypothesized sensory response pat- analysis
Sensory processing
terns (i.e., hyporesponsiveness; hyperresponsiveness; sen-
sory interests, repetitions and seeking behaviors; enhanced
perception), five method factors of sensory modalities (i.e., The rate of autism spectrum disorders (ASD) is increasing
auditory, visual, tactile, gustatory/olfactory, vestibular/ with recent prevalence estimates reaching 1 in 88 in the
proprioceptive), and one of social context were tested with United States (Baio 2012). Although sensory features have
good model fit. Child and family characteristics associated been listed as commonly associated features [Diagnostic
with the sensory response patterns were explored. The and Statistical Manual-IV-TR; American Psychiatric
effect of sensory response patterns on autism severity was Association (APA) 2000] the new ASD diagnostic criteria
tested, controlling for key child and family characteristics. proposed in the DSM-5 includes sensory features as core
features and specifically describes hypo- and hyper-reac-
tivity to sensory input, unusual interests in sensory aspects
K. Ausderau
L. M. Little
J. Bulluck
G. T. Baranek
Division of Occupational Science, Department of Allied Health of the environment, and restricted and repetitive interests in
Sciences, University of North Carolina at Chapel Hill, sensory based activities (DSM-5; APA 2012). Assessments
Chapel Hill, NC, USA are needed that comprehensively characterize sensory
features in children with ASD and facilitate accurate
K. Ausderau (&)
Department of Kinesiology, Occupational Therapy Program, diagnostic practices. Few instruments measure sensory
University of Wisconsin at Madison, 3195 Medical Sciences features specific to ASD, thereby limiting large scale
Center, 1300 University Avenue, Madison, WI 53706, USA studies of prevalence and heterogeneity of sensory
e-mail: kausderau@wisc.edu
response patterns among children with ASD.
J. Sideris Sensory features are highly prevalent in ASD (Baranek
Frank Porter Graham Child Development Institute, University of et al. 2006; Ben-Sasson et al. 2009; O’Donnell et al. 2012),
North Carolina at Chapel Hill, Chapel Hill, NC, USA with studies consistently showing that children with ASD
present with higher rates of sensory features than typically
M. Furlong
UNC Gillings School of Global Public Health, Department of developing children (Dunn et al. 2002; Ermer and Dunn
Epidemiology, University of North Carolina at Chapel Hill, 1998; Schoen et al. 2009; Watling et al. 2001) as well as
Chapel Hill, NC, USA those with other developmental disabilities (DD) (Baranek
et al. 2006, 2013; Rogers et al. 2003).
L. M. Little
Department of Occupational Therapy, University of Kansas The literature suggests that sensory features among
Medical Center, Kansas City, MO, USA children with ASD constellate into four distinct behavioral

123
916 J Autism Dev Disord (2014) 44:915–925

categories, or sensory response patterns—hyporesponsive- sensory features across both social and nonsocial contexts
ness (HYPO), hyperresponsiveness (HYPER), sensory (Baranek et al. 2006; Hilton et al. 2007; Liss et al. 2006);
interests, repetitions and seeking behaviors (SIRS) and however, findings vary depending upon assessment meth-
enhanced perception (EP). HYPO is considered a lack of or ods and specific sensory patterns studied.
delayed response to sensory stimuli (e.g., a lack of ori- Research is needed to develop refined sensory process-
enting to loud sounds, slow to react to pain) (e.g., Ben- ing instruments based on current conceptual models, and
Sasson et al. 2009; Watson et al. 2011). HYPER is defined validate their utility for the ASD population. Although a
by an exaggerated or avoidant response to sensory stimuli number of caregiver report measures are designed to assess
(e.g., discomfort to grooming activities, covering ears in sensory features in children, most instruments have the
response to sounds) (e.g., Baranek et al. 2006; Schoen et al. following limitations: (a) were not developed to measure
2009). SIRS is characterized by a fascination with or unique sensory patterns among children with ASD (Sen-
craving of sensory stimulation which is intense and may be sory Profile; Dunn 1999; Sensory Processing Measure;
repetitive in nature (e.g., fascination with flickering lights Parham et al. 2007), and (b) were designed to assess only
or rubbing textures) (e.g., Ben-Sasson et al. 2007; Liss one sensory pattern, such as hyperresponsiveness (Sensory
et al. 2006). Although previous research has focused much Sensitivity Questionnaire-Revised; Talay-Ongan and
on the prevalence of HYPO, HYPER, and specific aspects Wood 2000) and/or (c) may conflate co-occurring sensory
of SIRS among children with ASD (Ben-Sasson et al. patterns within the same subscales [e.g., the Short Sensory
2009), EP has emerged as a fourth pattern of sensory Profile (McIntosh et al. 1999); ‘‘Underresponsive/Seeks
response possibly unique to individuals with ASD (Happé Sensation’’ subscale includes HYPO, HYPER, and SIRS
and Frith 2006; Mottron et al. 2006, 2009). EP is charac- items; ‘‘Auditory Filtering’’ subscale includes HYPER and
terized by superior acuity in the awareness of specific HYPO items]. Moreover, most were standardized on typ-
sensory stimuli and focuses on specific perceptual elements ical children or were validated with relatively small sam-
(e.g., recognizing perfect pitch, superior ability to recog- ples of children with ASD (Dunn 1999; Glennon et al.
nize minor changes in visual appearance) (Mottron et al. 2007; McIntosh et al. 1999; Talay-Ongan and Wood 2000).
2006). Recent neuropsychological theories suggest that EP The Sensory Experiences Questionnaire Version 3.0
is likely characterized by strengths in locally oriented (SEQ-3.0; Baranek 2009) is a 105 item parent report tool
visual and auditory perception and enhanced low-level designed specifically to measure behavioral responses to
discrimination (Mottron et al. 2006, 2009), and may be naturally occurring sensory stimuli in the context of
related to hypersensitivity (e.g., lower threshold detection) everyday situations in children with ASD, ages 2–12 years.
and hyper-systemizing cognitive styles (Baron-Cohen et al. Previous research on earlier versions of the Sensory
2009). Experiences Questionnaire (Version 1.0, Version 2.1) has
While these four constructs are uniquely defined, demonstrated the tool’s reliability and validity (Baranek
research suggests certain patterns may co-occur within et al. 2006; Little et al. 2011). However, emerging empir-
individuals creating significant phenotypic heterogeneity ical research on the sensory response patterns among
across the population (Baranek et al. 2006; Ben-Sasson children with ASD has resulted in the need for the
et al. 2007, 2009; Hilton et al. 2007; Lane et al. 2010; Liss expansion of items and a confirmation of the proposed
et al. 2006). HYPO and HYPER have been found to co- four-factor model designed to capture these features uti-
occur in children with ASD (Baranek et al. 2006; Ben- lizing a sufficiently large sample which may provide
Sasson et al. 2007), SIRS has been associated with both researchers and clinicians with better ability to characterize
HYPO and HYPER among children with ASD (Boyd et al. the heterogeneity of sensory features specific to children
2010; Gabriels et al. 2008), and HYPER and EP have also with ASD.
been associated (Mottron et al. 2006). Although patterns of The main purpose of this study is to provide empirical
sensory response do co-occur, some research suggests that validation for the Sensory Experiences Questionnaire
HYPO better discriminates between children with ASD Version 3.0 (SEQ-3.0) with children with ASD ages
when compared with children with DD or typical devel- 2–12 years. The following research questions were
opment (Baranek et al. 2013; Ben-Sasson et al. 2009; addressed:
Rogers and Ozonoff 2005). While others have suggested
1) Is there an empirical validation for a four factor model
HYPER, may differentiate children with ASD (Joosten and
of sensory features in children with ASD?
Bundy 2010). Further, due to the core deficits of ASD, the
2) To what extent do sensory response patterns predict
impact of social context on sensory features is an important
autism severity while controlling for associated child
consideration in formulating conceptual models appropri-
and family characteristics?
ate to this population. Studies have shown the presence of

123
J Autism Dev Disord (2014) 44:915–925 917

Table 1 Child and family characteristics Exclusion Criteria

Primary Diagnosis (allowed to select more than one) (n = 1,307) (%)
Autism/Autistic Disorder 63.0
Exclusionary criteria for the study included the following:
co-morbid conditions of ASD such as fragile X syndrome
Asperger’s Disorder 22.1
and tuberous sclerosis, genetic disorder or syndrome
PDD-NOS 24.4
associated with a developmental disability (e.g., Down
Gender (n = 1,307)
syndrome), severe physical impairment (e.g., cerebral
% Male 82.3
palsy), significant visual or hearing impairment (e.g.,
SRS/SRS-P Total Score (SD) (n = 1,241) 107.1 (27.3)
blindness or deafness), traumatic brain injury or brain
Chronological Age (SD) (n = 1,307) 7.7 (2.7) years
malformation, psychotic disorder such as bipolar disorder
IQ Proxy (SD) (n = 1,124) 81.4 (28.8)
or schizophrenia, and seizure activity within the last
Maternal College Education (n = 1,299) (%)
12 months. The final sample included 1,307 children, as
Partial High School or Lower 0.8
100 participants were excluded on the basis of these
High School or GED 15.6
criteria.
Associates Degree/Partial College 29.1
Bachelor or Master Degree 50.4 Study Instruments
Advanced Degree such as doctorate 4.2
Annual Household Income (n = 1,209) (%) The Sensory Experiences Questionnaire Version 3.0 (SEQ-
Less than $20,000 8.3 3.0; Baranek 2009) is a caregiver report instrument
$20,000 to $39,999 17.2 designed to characterize sensory features in children ages
$40,000 to $59,999 19.3 2–12 years with ASD and/or DD in social and non-social
$60,000 to $79,999 16.9 contexts. The SEQ framework considers whether sensory
$80,000 to $99,999 14.3 experiences occur in a predominantly social context (e.g.,
$100,000 or more 24.0 experiencing contact with people), or a nonsocial context
Race/Ethnicity (allowed to select more than one) (n = 1,307) (%) (e.g., experiencing loud sounds or textured objects). Defi-
African-American 5.0 cits in social cognition and communication may impact a
American Indian/Alaskan Native 3.5 child’s ability to understand other people’s intentions (e.g.,
Asian 3.9 Baron-Cohen 2002) as well as create difficulties in
Native Hawaiian/Pacific Islander 0.3 expressing sensory preferences and needs appropriately
Other 3.4 (e.g., Wetherby 2006), which adds variability in responses
White 93.0 to particular types of sensory experiences.
Hispanic or Latino Origin 9.9 The SEQ-3.0 has been revised from its earlier version to
better meet research needs and an evolving conceptual
model. In particular, recent research on enhanced percep-
Methods tion among individuals with ASD, and potential contribu-
tions to hyperresponsiveness (e.g., Baron-Cohen et al.
Participants 2009; Mottron et al. 2006), the SEQ-3.0 has been expanded
to include items that tap this construct. Additional revisions
Caregivers (n = 1,407) of children with ASD (i.e., diag- included refining items on the basis of previous psycho-
nosis of Autistic Disorder, Asperger’s Disorder, or Perva- metric studies, expanding the total number of items to
sive Developmental Disorder-Not Otherwise Specified provide more balance across the sensory response patterns,
[PDD-NOS]) (DSM-IV-TR, APA 2000) ages 2–12 years sensory modalities, and contexts (social and nonsocial),
were recruited from a national registry augmented by and adding control items. The items based on sensory
national and local autism advocacy organizations to par- features included in the SEQ-3.0 were developed from a
ticipate in this survey (see Study Procedures for further review of existing literature on sensory features in children
details of demographic and diagnostic information). with ASD, including empirical studies, parental report
Primary caregivers [mothers (95.1 %), fathers (3.3 %), studies, expert clinical reports, conceptual models of sen-
other adults (1.6 %) such as grandparents that lived in the sory processing, and neuropsychological theories of ASD
home] completed an SEQ 3.0 for each study participant. describing core and associated features. Through a con-
All caregivers provided informed consent as approved by sensus process, a team of experts grouped the items into the
the Institutional Review Board of the University of North sensory response patterns, modality categories, and con-
Carolina at Chapel Hill. Table 1 summarizes demographic texts (social or non-social), consistent with the conceptual
data. framework of the SEQ.

123
918 J Autism Dev Disord (2014) 44:915–925

The SEQ-3.0 has 105 items that measure the frequency functioning’’. An IQ proxy was then calculated based on
of sensory behaviors across four sensory response patterns this developmental estimate using the following formula:
(i.e., HYPO, HYPER, SIRS, EP), five modality categories [(PEDA/CA)*100]. This was done to obtain a standardized
(i.e., auditory, visual, tactile, gustatory/olfactory, vestibu- metric for all participants that would not be correlated with
lar/proprioceptive), and two contexts (i.e., social and non- CA to use as a covariate in the analyses. A subset of 316
social). The first 97 items measure the frequency using a participants had reported previous IQ testing scores which
5-point Likert scale ranging from 1 (never/almost never) to showed a positive correlation with the calculated IQ proxy
5 (always/almost always) with a higher score indicating (r = .67).
more sensory symptoms. The measure also includes eight
items which address broader issues related to the children’s
Study Procedures
sensory behaviors and allow caregivers to elaborate with a
qualitative response. The questionnaire takes approxi-
The IAN Research Database at the Kennedy Krieger
mately 15–20 min to complete.
Institute and Johns Hopkins Medicine-Baltimore (spon-
Autism symptom severity was assessed using the Social
sored by Autism Speaks), the University of North Carolina
Responsiveness Scale (SRS; Constantino and Gruber 2005)
at Chapel Hill Research Registry, and other online autism
for children ages 4–18 years, and Social Responsiveness
organizations were used to recruit a demographically
Scale- Preschool Version (SRS-P; Pine et al. 2006) for
diverse sample of families in the United States to partici-
children ages 35–48 months. The SRS and SRS-P consist
pate in a longitudinal study on the sensory features of
of 65 items and takes 15–20 min to complete, and both
children with ASD. Approximately 50 % of the sample
measures were designed as quantitative trait assessments of
was recruited through the Interactive Autism Network
children’s autism symptoms in social settings with higher
(IAN), a web-based autism registry, which recently
scores indicating more autistic symptoms (Constantino and
authenticated the parent-report ASD diagnosis in their
Gruber 2005). Total scores on the SRS and SRS-P, versus
registry (Daniels et al. 2012). The remainder of participants
T-scores, were used in analysis to allow for a direct com-
was recruited through online autism advocacy and parent
parison of the measures, as the SRS-P is in prepublication.
support groups.
A Background Information Questionnaire (BIQ;
Recruitment occurred in phases and was done solely
unpublished questionnaire) was developed specifically for
through online recruitment material. Participants were
this study to obtain information about the child and family
screened for inclusion/exclusion before initiation of the
in four domains: family characteristics, child characteris-
survey. Once determined eligible, they were sent an elec-
tics, child’s functioning level, and services the child
tronic invitation to participate, followed by up to three
receives. The BIQ is a parent-report instrument that takes
electronic contacts to complete the survey if needed. All
approximately 10–15 min to complete. The following
questionnaires were converted to an electronic format
demographic and developmental variables were derived
using Qualtrics software (Qualtrics Labs 2011) and the
from the BIQ to provide descriptive data regarding key
online versions of the surveys were approved by the pub-
child and family characteristics: Chronological age (CA)
lisher or author, as appropriate, before administration. The
was calculated from the child’s birth date. Autism diagnosis
surveys took 45–60 min to complete, and families were
included Autism/Autistic Disorder, Asperger’s Disorder,
offered a $5.00 gift card upon completion.
and PDD-NOS. Diagnosis was obtained per caregiver
report for all participants; a subset included previously
authenticated diagnoses through a registry (see below in Data Analysis
Study Procedures). Household income was reported in
increments of $20,000, with categories ranging from To address research question 1, the proposed structure of
\$20,000 to[$100,000, and the analysis used the floor for the SEQ-3.0 was tested with a confirmatory factor analysis
each income category (e.g., $20,001–$40,000 was recoded (CFA). Factor analysis offers substantial benefits in
as $20,000). Maternal College Education indicated whe- assessing the relationships between item-level data and
ther or not the child’s mother completed at least a bache- underlying latent variables (factors). In these models,
lor’s degree. Race was categorized as either white or non- responses to items are viewed as arising from underlying,
white due to the small number of participants in non-white unobserved latent variables. That is, the latent variables are
categories (see Table 1). Parents’ Estimated Develop- the source of item covariance. In this case, the unobserved
mental Age (PEDA) was measured in 6 month increments latent variables are the four sensory patterns of the SEQ-
between \12 months to 3 years and 12 months increments 3.0. In CFA, a specific latent structure is proposed to
from 3 to 19 years. Specifically, caregivers were asked explain covariance among manifest variables (item-level
to estimate their child’s ‘‘overall level of cognitive data).

123
J Autism Dev Disord (2014) 44:915–925 919

Four distinct factors (HYPO, HYPER, SIRS, EP) were result from the superficial similarity of items in a given
tested, which characterize the sensory response patterns in sensory modality (e.g., visual or auditory items). To
our conceptual model. Given the literature describing co- account for the covariance, each sensory modality was
occurrence of these sensory features in children with ASD, included as a measurement factor that would act as latent
correlations among the four constructs were expected measures of the error covariance (see Kenny and Kashy
(Baranek et al. 2006; Ben-Sasson et al. 2007, 2009; Hilton 1992 for comprehensive discussion). This covariance
et al. 2007; Lane et al. 2010; Liss et al. 2006). The five should be unrelated to the constructs of interest in the
sensory modality categories (i.e., auditory, visual, tactile, model (e.g. sensory response pattern) and so the correla-
gustatory/olfactory, vestibular/proprioceptive) presented tions of these measurement factors to each other and to the
additional sources of item covariance beyond the sensory sensory response pattern factors were fixed to zero. There
response patterns. In any scale, some items may be related were several items on the scale regarding behaviors that
to one another for reasons outside the factors of interest occur in a social context. We believed this provided
(HYPO, HYPER, SIRS, EP); measurement error associated another source of error variance similar to that from the
with one item is correlated with that of another item. In sensory modalities. A sixth measurement factor was pro-
these data, it was anticipated that correlated errors would posed to account for shared variance among items

Fig. 1 Factor analytic model

Factor Analytic Model

Method Factors Content Factors

Context

HYPO

Social

Modalities HYPER

Auditory

SIRS

Visual

EP

Tactile

Gustatory
Olfactory

Vestibular
Proprioceptive

123
920 J Autism Dev Disord (2014) 44:915–925

occurring in the social context. Like all of the measurement Results

factors, the social factor was fixed to be uncorrelated with
all other factors in the model. Factor Analysis
In sum, the model had a total of 10 latent variables: four
sensory response pattern variables, the social context Model fit was good using standard fit measures. See
measurement variable, and the five sensory modality Table 2 for indices. Chi square is sensitive to both sample
measurement variables, with 97 measured variables (items) size and the number of parameters in the model (Schu-
from the SEQ-3.0. See Fig. 1 for the factor analytic model. macker and Lomax 2004), and in this case both were large.
The model was estimated in Mplus (Muthén and Muthén Jöreskog (1969) suggested the use of a normed Chi square
2010) under maximum likelihood estimation. Factor means (Chi square/degrees of freedom) provides some protection
were fixed to zero and factor variances to one, with all from this sensitivity. In these data, the normed Chi square
hypothesized loadings and between factor correlations of 4.35 indicates reasonable model fit (Bollen 1989). Both
estimated freely. The 97 measured variables (items) were the RMSEA and the SRMR fall within common guidelines
allowed to have 2–3 cross loadings on the latent variables for good model fit (Kline 2010). Post hoc assessment
depending on the item content (e.g. sensory pattern, sen- suggested some small modifications to the model, but
sory modality, social context). However, any one measured model fit did not significantly change when they were
variable was only allowed to load on one sensory pattern included and so the initial model was retained.
variable, one sensory modality variable, and the social The factor loadings for the latent sensory factors were
variable if appropriate. generally strong and all significant (p \ .001); all were
The SEQ-3.0 is intended for use across a relatively wide greater than .20 and the vast majority were .40 or greater.
population, so measurement invariance was tested for the Factor variances were set to one. This was considered
factor loading across gender and age. The sample was split strong support for the existence of our distinct hypothe-
into two age groups, preschool or younger (\5 years old) sized constructs. As expected, between factor correlations
and school aged ([5 to \13 years old) for the age model. were positive and significant, (p \ .001). With the excep-
Invariance indicates that the model parameters are the same tion of EP and HYPO (r = .22), the correlations were
across the two age groups. Testing typically proceeds in a moderate to large ranging from .44 to .74. See Table 3 for
series of tests each requiring more stringent levels of sensory factor correlations.
invariance (Bollen 1989; Horn and McArdle 1992). Invariance testing confirmed the SEQ-3.0 can be used
To address research question 2, first we computed a across the different gender and age groups. Model fit for
correlation matrix to look at associations among sensory each of the separate groups, boys versus girls and preschool
response patterns (HYPO, HYPER, SIRS, EP), autism versus school-age was essentially unchanged from the
severity (SRS), and various child (i.e., CA, PEDA, IQ overall model. Further, comparison of the configural
Proxy, gender, race) and family (income, maternal educa- models to the strong invariance models where factor
tion) characteristics. Next, a mixed model regression in loadings and item intercepts are constrained to be equal,
SAS 9.2 (SAS Institute 2010) was used to predict autism
severity (SRS) from the four sensory response patterns,
controlling for key child and family characteristics found to Table 2 Model fit indices
be associated with the sensory patterns and/or autism Statistic Values
severity. The presence of siblings (69 cases) from 33
families who had more than one child with ASD in the Chi Square 16, 724.18 (3984)**
sample gave rise to non-independence between these RMSEA .051 (.050–0.052)
observations. Mixed model regression, which allows for SRMR .07
nesting of observations within child and within family,
manages this non-independence through the inclusion of
random effects which provide separate parameter estimates
Table 3 Between factor correlations for sensory patterns
for each observation, in addition to the fixed effects esti-
mates provided in general linear models (see Bryk and HYPO HYPER SIRS EP
Raudenbush 1992; Burchinal et al. 2006 for complete HYPO 1.00
discussions). Finally, graphic depictions were used to HYPER 0.49 1.00
demonstrate how mean sensory response pattern factor SIRS 0.64 0.44 1.00
scores differed by ASD diagnostic categories (i.e., Autism/ EP 0.22 0.74 0.51 1.00
Autistic Disorder, Asperger’s Disorder, PDD-NOS).

123
J Autism Dev Disord (2014) 44:915–925 921

indicated no significant difference either between boys and positively associated with all sensory response patterns
girls, v2 (92) = 95.89, ns, or between younger and older (.33–.57) as well as to maternal education, income, IQ
children, v2 (92) = 105.71, ns. proxy, and gender. Moderate to high inter-correlations
were detected among maturational variables, particularly
Correlation Matrix between PEDA and CA (r = .39) as well as PEDA and IQ
(r = .70).
Table 4 presents the inter-correlations between the four
sensory response patterns (HYPO, HYPER, SIRS, EP), Mixed Model Regression Analysis
autism severity (SRS), and child (i.e., CA, PEDA, IQ
Proxy, gender) and family characteristics (i.e., income, Based on previous literature and the pattern of inter-cor-
maternal education). Although statistically significant, relations in the above analyses, the following child and
most correlations between sensory response patterns and family characteristics were selected as the key covariates
child and family characteristics were small to medium for inclusion in the mixed model regression analysis: CA,
(r = -.34– .18) (Cohen 1988). Income and maternal edu- IQ Proxy, gender, and income. In the mixed model
cation were moderately correlated at r = .32, but of these regression analysis (Table 5), all sensory response patterns
two variables, income had a slightly stronger negative were found to significantly predict autism severity. As
correlation with the sensory response patterns, likely due to expected, both CA and IQ Proxy were negatively related
the restricted binary nature of the maternal education var- with SRS, such that increases in either were associated with
iable. Small, but statistically significant, correlations were lower autism severity. Boys tended to be about 2 points
found between sensory response patterns and both CA and lower on the SRS than girls (mean difference = 2.03,
PEDA (-.21–.11). Autism severity was significantly p = .002). A number of significant two-way interactions

Table 4 Correlations among sensory response patterns, autism severity, and child and family characteristics
Income College (yes) Race (white) Gender (Male) CA PEDA IQ Proxy SRS Total

Income 1.00
1,209
College (yes) 0.32*** 1.00
1,203 1,299
Race (white) 0.04 0.06* 1.00
1,209 1,299 1,307
Gender (Male) 0.02 0.01 -0.01 1.00
1,209 1,299 1,307 1,307
CAa 0.10*** 0.07* 0.03 0.05 1.00
1,209 1,299 1,307 1,307 1,307
PEDAb 0.05 0.03 0.02 0.00 0.39*** 1.00
1,209 1,299 1,307 1,307 1,307 1,307
IQ Proxy 0.15*** 0.13*** 0.10*** -0.09** 0.04 0.70*** 1.00
1,044 1,117 1,124 1,124 1,124 1,124 1,124
SRSc Total -0.19*** -0.14*** 0.00 0.15*** 0.00 -0.06* -0.24*** 1.00
1,157 1,243 1,251 1,251 1,251 1,251 1,080 1,251
HYPO -0.27*** -0.14*** -0.06* 0.03 -0.17*** -0.13*** -0.24*** 0.57***
1,209 1,299 1,307 1,307 1,307 1,307 1,124 1,251
HYPER -0.18*** -0.08** 0.03 -0.01 0.11*** 0.11*** 0.11*** 0.50***
1,209 1,299 1,307 1,307 1,307 1,307 1,124 1,251
SIRS -0.28*** -0.14*** -0.08** 0.02 -0.21*** -0.16*** -0.24*** 0.50***
1,209 1,299 1,307 1,307 1,307 1,307 1,124 1,251
EP -0.15*** -0.07* -0.01 -0.02 0.14*** 0.11*** 0.15*** 0.33***
1,209 1,299 1,307 1,307 1,307 1,307 1,124 1,251
* \ .05 ** \ .01 *** \ .001
a b c
Chronological Age, Parents’ Estimated Developmental Age, Social Responsiveness Scale/Preschool

123
922 J Autism Dev Disord (2014) 44:915–925

Table 5 Mixed model predicting SRS/SRS-P score from sensory

patterns (n = 1,156)
SRS Total
Parameter (SE)

Seek 2.03 (0.50)***

Hypo 1.88 (0.52)***
Hyper 5.63 (0.59)***
EP -2.98 (0.65)***
Seek*Hypo -0.84 (0.30)**
Hypo*Hyper -1.91 (0.33)***
Hyper*EP 0.65 (0.28)**
Male -2.03 (0.61)**
CA -0.35 (0.10)**
Income -0.01 (.01)
IQ Proxy -0.06 (.009)***

were present (see Fig. 2). Three and four way interactions
were tested, but were not significant. The test of random
effects for the intercept was significant, z = 2.37,
p = .009, indicating significant between-subject variance.

Sensory Response Pattern Scores by ASD Diagnosis

Figure 3 depicts the mean factor scores for each of the

sensory response patterns by ASD diagnosis (Autism/
Autistic Disorder, Asperger’s Disorder, PDD-NOS). Each
diagnostic group showed a unique representation of the
four sensory response patterns (HYPO, HYPER, SIRS, EP)
as follows: Children with a PDD-NOS diagnosis, on
average, had scores below the mean on all four sensory
response patterns, indicating fewer sensory features. Chil-
dren with Autism/Autistic Disorder and Asperger’s Disor-
der presented with a split profile: HYPO and SIRS were
above the mean for children with Autism/Autistic Disorder,
while HYPER and EP were above the mean for children
Fig. 2 Two way interactions predicting SRS/SRS-P score
with Asperger’s Disorder.

Discussion

This study aimed to empirically validate the factor struc-

ture of the SEQ, and test the effect of sensory response
patterns on autism severity while accounting for key child
and family characteristics in a large national sample of
children with ASD ages 2–12 years. A complex factor
Asperger’s PDD-NOS
analytic model was confirmed for the specified factor
structure, including four sensory response patterns (HYPO, Autism

HYPER, SIRS, EP), and one social context and five

modality constructs as measurement variables. The model
fit and factor loadings provided validation for the SEQ 3.0
conceptual model with four distinct sensory patterns
among children with ASD. Inclusion of social context and Fig. 3 Mean factor scores by ASD diagnosis

123
J Autism Dev Disord (2014) 44:915–925 923

modality constructs as part of the measurement structure of into the co-occurrence of sensory response patterns and
the model allowed for a more precise estimation of the specific characteristics associated with various types or
individual sensory features that contributed to each sensory functioning levels of children with ASD. For instance,
pattern, by removing the error variance or statistical noise children with Asperger’s Disorder are often described as
associated with the modalities and social latent variables. having increased IQ as well as speech and language skills
The limited number of necessary instrument modifica- (APA 2000; Green et al. 2006). In this study, the distri-
tions provided further evidence of a sound theoretical bution of sensory patterns among children with Asperger’s
model. All measured variables representing the 97 Likert Disorder was driven by higher means of HYPER and EP,
scale items from the SEQ-3.0 were included in the model, which was also related to higher IQ Proxy. Although the
and all items made significant contributions to at least one DSM-5 no longer differentiates on the basis of these
of the latent variables. Within each sensory pattern con- diagnostic subcategories, the current findings have impli-
struct, the high factor loadings provided strong evidence cations for parsing out the heterogeneity in this population
for the underlying latent concepts. For example, in the on the basis of sensory response patterns, and uncovering
SIRS construct, items such as ‘‘seems fascinated with sensory phenotypes that may differentially predict devel-
specific textures’’, ‘‘fascination with specific visual opmental outcomes.
effects’’, and ‘‘movement such as jumping up and down or A unique contribution of this study was the inclusion of
spinning in circles’’ are all examples of the types of items enhanced perception in the four factor model. The valida-
that were driving the SIRS construct with high factor tion of this construct expands current conceptualizations in
loadings between .57 and .68. The item analysis of the 97 the literature, and provides support for addressing
measured variables had no strong indications of missing enhanced perception in the comprehensive clinical assess-
pathways between SEQ-3.0 items and sensory factors, ment of sensory features and diagnostic practices for
further supporting distinct sensory response patterns. children with ASD. This may be particularly important as
Despite clear distinctions between sensory patterns, various combinations of sensory features are differentially
findings from the present study include the estimation of associated with adaptive and maladaptive outcomes (Boyd
between sensory factor correlations to characterize how et al. 2010; Green et al. 2012; Lane et al. 2010; Watson
sensory patterns are related to one another. The moderate et al. 2011), and moreover, enhanced perception reflects
correlations between sensory patterns are consistent with some areas of strength in sensory-perceptual functions
the literature that sensory patterns often co-exist in children (Baron-Cohen et al. 2009; Mottron et al. 2006) as opposed
with ASD (Baranek et al. 2006). However, stronger asso- to deficits commonly attributed to other sensory response
ciations emerged specifically between HYPER and EP as patterns. Intriguing patterns of concomitant strengths and
well as between HYPO and SIRS, which adds to our deficits are found to be characteristic of individuals with
understanding of these features in this sample of children autism across various developmental domains (Wallace
with ASD. These findings also suggest a potential co- et al. 2009) and these patterns deserve further study in
occurrence of these patterns in ASD. Furthermore, the issues related to sensory processing as well.
mixed model regression analysis and the mean distribution
of sensory patterns by ASD diagnosis provided further Limitations and Future Directions
support for this co-occurrence.
While sensory features have long been described as While the study procedures allowed for testing of a large
common symptoms in ASD, there is little information heterogeneous sample of children with ASD, online
about how they impact autism severity. The current find- recruitment and survey administration presented with some
ings suggested that sensory patterns contributed to autism limitations. Specifically, the findings rely on parent-report
severity even when controlling for key child (i.e., CA, IQ and we were not able to validate sensory features or other
Proxy, gender) and family (i.e., income) characteristics. child characteristics (e.g., ASD diagnostic category, IQ)
Furthermore, novel findings from this study demonstrated through observational methods in this study for such an
that each of the four sensory response patterns contributed extensive sample. Online methods provide limited access
to autism severity through two-way interactions rather than to people without computers and internet access, resulting
through simple additive effects. For example, as the effect in less racial and socio-economic diversity. A more strat-
of HYPER on autism severity score increases, the effect of ified sample in the future could explore whether or not
HYPO decreases. By examining sensory response pattern these demographic variables may differentially affect par-
mean scores by diagnosis, clear combinations of sensory ents’ perceptions of their children’s sensory features. Other
response patterns emerged for each ASD diagnosis (i.e., studies have shown modest correlations between parent
Autism/Autistic Disorder, Asperger’s Disorder, PDD- reported sensory features and clinician observed lab mea-
NOS). Such differential distributions may provide insight sures (Miller et al. 1999) and may be an underestimate

123
924 J Autism Dev Disord (2014) 44:915–925

compared to self-reports (Parush et al. 2006). Nonetheless, Baranek, G. T., Watson, L. R., Boyd, B. A., Poe, M. D., David, F. J.,
the factor structure confirmed by this method was consis- & McGuire, L. (2013). Hyporesponsiveness to social and
nonsocial sensory stimuli in children with autism, children with
tent with previous work using multi-trait, multi-method developmental delays, and typically developing children. Devel-
procedures for children with ASD in this age group using a opment and Psychopathology, 25(02), 307–320.
similar conceptual model (e.g., Watson et al. 2011). Future Baron-Cohen, S. (2002). The extreme male brain theory of autism.
studies and instruments are needed to test the extent to Trends in Cognitive Sciences, 6(6), 248–254.
Baron-Cohen, S., Ashwin, E., Ashwin, C., Tavassoli, T., &
which this conceptual model is applicable to individuals Chakrabarti, B. (2009). Talent in autism: Hyper-systemizing,
with ASD older than 12 years. hyper-attention to detail and sensory hypersensitivity. Philo-
Currently, sensory features in children with ASD are sophical Transactions of the Royal Society, 364, 1377–1383.
often the target of a wide range of interventions. The Ben-Sasson, A., Cermak, S. A., Orsmond, G. I., Tager-Flusberg, H.,
Carter, A. S., Kadlec, M. B., et al. (2007). Extreme sensory
development of psychometrically sound assessment tools modulation behaviors in toddlers with autism spectrum disor-
such as the SEQ-3.0 should lead to increased diagnostic ders. American Journal of Occupational Therapy, 61(5),
accuracy, which may aid in designing more targeted 584–592.
treatment interventions for sensory features in ASD. Ben-Sasson, A., Hen, L., Fluss, R., Cermak, S. A., Engel-Yeger, B., &
Gal, E. (2009). A meta-analysis of sensory modulation symp-
Additionally, future research could further validate these toms in individuals with autism spectrum disorders. Journal of
sensory response patterns, as well as their inter-correlations Autism and Developmental Disorders, 39(1), 1–11. doi:10.1007/
with specific behavioral and physiological measures to s10803-008-0593-3.
deepen our understanding of neurobiological mechanisms Bollen, K. A. (1989). Structural equations with latent variables. New
York, NY: Wiley.
that may give rise to such features. Understanding the vast Boyd, B. A., Baranek, G. T., Sideris, J., Poe, M. D., Watson, L. R.,
heterogeneity in ASD, including how sensory features Patten, E., et al. (2010). Sensory features and repetitive
contribute to this heterogeneity, is an ongoing quest for behaviors in children with autism and developmental delays.
researchers and clinicians in this field; the SEQ-3.0 may be Autism Research, 3(2), 78–87. doi:10.1002/aur.124.
Bryk, A., & Raudenbush, S. (1992). Hierarchical linear models for
one useful tool toward this goal. social and behavioral research: Applications and data analysis
methods. Newbury Park, CA: Sage.
Acknowledgments Thank you to the families that participated in Burchinal, M. R., Nelson, L., & Poe, M. (2006). Growth curve
the study as well as the research team at the Sensory Experiences analysis: An introduction to various methods for analyzing
Project. Thank you also to the Interactive Autism Network Research longitudinal data. Monographs of the Society for Research in
Database at the Kennedy Krieger Institute and John Hopkins Medi- Child Development, 71(3), 65–87. doi:10.1111/j.1540-5834.
cine-Baltimore (sponsored by Autism Speaks), the University of 2006.00405.x.
North Carolina at Chapel Hill Research Registry, and the multiple Cohen, J. (1988). Statistical Power Analysis for the Social Sciences.
other autism organizations who assisted in recruitment. This study (pp 79–80). New York: Psychology Press.
was supported by the National Institute of Child Health and Human Constantino, J. N., & Gruber, C. P. (2005). Social responsiveness
Development/National Institutes of Health; ARRA Supplement A10- scale manual. Los Angeles, CA: Western Psychological
0589 (R01-HD042168). Recruitment was partially supported by the Services.
Intellectual and Developmental Disabilities Research Center; Corp, Stata. (2011). Stata statistical software (12th ed.). College
P30HD03110. Station, TX: Stata Corp, LP.
Daniels, A. M., Rosenberg, R. E., Anderson, C., Law, J. K., Marvin,
A. R., & Law, P. A. (2012). Verification of parent-report of child
autism spectrum disorder diagnosis to a web-based autism
References registry. Journal of Autism and Developmental Disorders, 42(2),
257–265. doi:10.1007/s10803-011-1236-7.
American Psychiatric Association. (2000). Diagnostic and statistical Dunn, W. (1999). The sensory profile: Examiner’s manual. San
manual of mental disorders (4th ed., text rev.). Washington, DC: Antonio, TX: The Psychological Corporation.
Washington, DC: American Psychiatric Publishing, Inc. Dunn, W., Myles, B. S., & Orr, S. (2002). Sensory processing issues
American Psychiatric Association. (2012). Proposed Autism spectrum associated with Asperger syndrome: A preliminary investigation.
disorder. Retrieved from http://www.dsm5.org/proposedrevisions/ The American Journal of Occupational Therapy, 56(1), 97–102.
pages/proposedrevision.aspx?rid=94. doi:10.5014/ajot.56.1.97.
Baio, J. (2012). Prevalence of autism spectrum disorders–autism and Ermer, J., & Dunn, W. (1998). The sensory profile: A discriminant
developmental disabilities monitoring network, 14 sites, United analysis of children with and without disabilities. The American
States, 2008. Centers for disease control and prevention Journal of Occupational Therapy, 52(4), 283–290. doi:10.5014/
surveillance summaries, 61(3), 1–19, Retrieved from http:// ajot.52.4.283.
www.cdc.gov/mmwr/pdf/ss/ss6103.pdf. Gabriels, R. L., Agnew, J. A., Miller, L. J., Gralla, J., Pan, Z.,
Baranek, G.T. (2009). Sensory experiences questionnaire version 3.0. Goldson, E., et al. (2008). Is there a relationship between
Unpublished manuscript. restricted, repetitive, stereotyped behaviors and interests and
Baranek, G. T., David, F. J., Poe, M. D., Stone, W. L., & Watson, L. abnormal sensory response in children with autism spectrum
R. (2006). Sensory experiences questionnaire: Discriminating disorders? Research in Autism Spectrum Disorders, 2(4),
sensory features in young children with autism, developmental 660–670. doi:10.1016/j.rasd.2008.02.002.
delays, and typical development. Journal of Child Psychology Glennon, T. J., Miller-Kuhaneck, H., Henry, D. A., Parham, L. D., &
and Psychiatry, 47(6), 591–601. doi:10.1111/j.1469-7610.2005. Ecker, C. (2007). Sensory processing measure manual. Los
01546.x. Angeles: Western Psychological Services.

123
J Autism Dev Disord (2014) 44:915–925 925

Green, V. A., Pituch, K. A., Itchon, J., Choi, A., O’Reilly, M., & Muthén, L. K., & Muthén, B. O. (2010). Mplus user’s guide (6th ed.).
Sigafoos, J. (2006). Internet survey of treatments used by parents Los Angeles: Muthén & Muthén.
of children with autism. Research in Developmental Disabilities, O’Donnell, S., Deitz, J., Kartin, D., Nalty, T., & Dawson, G. (2012).
27, 70–84. Sensory processing, problem behavior, adaptive behavior, and
Green, S. A., Ben-Sasson, A., Soto, T. W., & Carter, A. S. (2012). cognition in preschool children with autism spectrum disorders.
Anxiety and sensory over-responsivity in toddlers with autism AJOT: American Journal of Occupational Therapy, 66,
spectrum disorders: Bidirectional effects across time. Journal of 586–594.
Autism and Developmental Disorders, 42(6), 1112–1119. Parham, L.D., Ecker, C., Miller-Kuhananeck, H., Henry, D.A., &
Happé, F., & Frith, U. (2006). The weak coherence account: Detail- Glennon, T. (2007). In Los Angeles: Western Psychological
focused cognitive style in autism spectrum disorders. Journal of Services (Ed.), Sensory processing measure (SPM) manual. Los
Autism and Developmental Disorders, 36(1), 5–25. doi:10.1007/ Angeles: Western Psychological Services.
s10803-005-0039-0. Parush, S., Doryon, Y. D., & Katz, N. (2006). A comparison of self-
Hilton, C., Graver, K., & LaVesser, P. (2007). Relationship between report and informant report of tactile defensiveness amongst
social competence and sensory processing in children with high children in Israel. Occupational Therapy International, 3(4),
functioning autism spectrum disorders. Research in Autism 274–283. doi:10.1002/oti.41.
Spectrum Disorders, 1(2), 164–173. doi:10.1016/j.rasd.2006.10. Pine, E., Luby, J., Abbacchi, A., & Constantino, J. N. (2006).
002. Quantitative assessment of autistic symptomatology in pre-
Horn, J. L., & McArdle, J. J. (1992). A practical and theoretical guide schoolers. Autism, 10(4), 344–352. doi:10.1177/136236130
to measurement invariance in aging research. Experimental 6064434.
Aging Research, 18(3), 117–144. doi:10.1080/03610739208 Qualtrics Labs, Inc. (2011). Version 21269 of the Qualtrics Research
253916. Suite. Provo, UT: Qualtrics Labs, Inc./http://www.qualtrics.com.
Joosten, A. V., & Bundy, A. C. (2010). Sensory processing and Rogers, S. J., & Ozonoff, S. (2005). Annotation: What do we know
stereotypical and repetitive behavior in children with autism and about sensory dysfunction in autism? A critical review of the
intellectual disability. Australian Occupational Therapy Journal, empirical evidence. Journal of Child Psychology and Psychiatry,
57(6), 366–372. 46(12), 1255–1268. doi:10.1111/j.1469-7610.2005.01431.x.
Jöreskog, K. G. (1969). A general approach to confirmatory Rogers, S. J., Hepburn, S., & Wehner, E. (2003). Parent reports of
maximum likelihood factor analysis. Psychometrika, 34(2), sensory symptoms in toddlers with autism and those with other
183–202. doi:10.1007/BF02289343. developmental disorders. Journal of Autism and Developmental
Kenny, D. A., & Kashy, D. A. (1992). Analysis of the multitrait- Disorders, 33(6), 631–642. doi:10.1023/B:JADD.0000006000.
multimethod matrix by confirmatory factor analysis. Psycholog- 38991.a7.
ical Bulletin, 112(1), 165–172. doi:10.1037/0033-2909.112.1. Schoen, S. A., Miller, L. J., Brett-Green, B. A., & Nielsen, D. M.
165. (2009). Physiological and behavioral differences in sensory
Kline, R. B. (2010). Principles and practice of structural equation processing: A comparison of children with autism spectrum
modeling. New York, NY: Guilford Press. disorder and sensory modulation disorder. Frontiers in Integra-
Lane, A. E., Young, R. L., Baker, A. E. Z., & Angley, M. T. (2010). tive Neuroscience, 3, 29–40. doi:10.3389/neuro.07.029.2009.
Sensory processing subtypes in autism: Association with adap- Schumacker, R.E., & Lomax, R.G. (2004). A beginner’s guide to
tive behavior. Journal of Autism and Developmental Disorders, structural equation modeling. Lawrence Erlbaum.
40(1), 112–122. doi:10.1007/s10803-009-0840-2. Talay-Ongan, A., & Wood, K. (2000). Unusual sensory sensitivities
Liss, M., Saulnier, C., Fein, D., & Kinsbourne, M. (2006). Sensory in autism: A possible crossroads. International Journal of
and attention abnormalities in autistic spectrum disorders. Disability, Development and Education, 47(2), 201–212.
Autism, 10(2), 155–172. doi:10.1177/1362361306062021. doi:10.1080/713671112.
Little, L. M., Freuler, A. C., Houser, M. B., Guckian, L., Carbine, K., Wallace, G. L., Happe, F., & Giedd, J. N. (2009). A case study of a
David, F. J., et al. (2011). Psychometric validation of the sensory multiply talented savant with an autism spectrum disorder:
experiences questionnaire. The American Journal of Occupa- Neuropsychological functioning and brain morphometry. Philo-
tional Therapy, 65(2), 207–210. doi:10.5014/ajot.2011.000844. sophsical Transactions of the Royal Society: Biological Sci-
McIntosh, D. N., Miller, L. J., & Shyu, V. (1999). Development and ences, 364(1522), 1425–1432.
validation of the short sensory profile. In W. Dunn (Ed.), Sensory Watling, R. L., Deitz, J., & White, O. (2001). Comparison of sensory
profile manual (pp. 59–73). San Antonio, TX: Psychological profile scores of young children with and without autism
Corporation. spectrum disorders. The American Journal of Occupational
Miller, L. J., McIntosh, D. N., McGrath, J., Shyu, V., Lampe, M., Therapy, 55(4), 416–423. doi:10.5014/ajot.55.4.416.
Taylor, A. K., et al. (1999). Electrodermal responses to sensory Watson, L. R., Patten, E., Baranek, G. T., Poe, M., Boyd, B. A.,
stimuli in individuals with fragile X syndrome: A preliminary Freuler, A., et al. (2011). Differential associations between
report. American Journal of Medical Genetics, 83, 268–279. sensory response patterns and language, social, and communi-
doi:10.1002/(SICI)1096-8628(19990402)83:4\268:AID-AJMG7[ cation measures in children with autism or other developmental
3.0.CO;2-K. disabilities. Journal of Speech, Language, and Hearing
Mottron, L., Dawson, M., Soulieres, I., Hubert, B., & Burack, J. Research, 54(6), 1562–1576. doi:10.1044/1092-4388(2011/10-
(2006). Enhanced perceptual functioning in autism: An update, 0029.
and eight principles of autistic perception. Journal of Autism and Wetherby, A. (2006). Understanding and measuring social commu-
Developmental Disorders, 36(1), 27–43. doi:10.1007/s10803- nication in children with autism spectrum disorders. Social &
005-0040-7. communication development in autism spectrum disorders: Early
Mottron, L., Dawson, M., & Soulières, I. (2009). Enhanced perception identification, diagnosis, & intervention (pp. 3–34). New York,
in savant syndrome: Patterns, structure and creativity. Philo- NY US: Guilford Press.
sophical Transactions of the Royal Society B: Biological
Sciences, 364(1522), 1385–1391. doi:10.1098/rstb.2008.0333.

123