Conversational gestures in autism spectrum disorders: asynchrony but not decreased frequency.
ABSTRACT Conversational or "co-speech" gestures play an important role in communication, facilitating turntaking, providing visuospatial information, clarifying subtleties of emphasis, and other pragmatic cues. Consistent with other pragmatic language deficits, individuals with autism spectrum disorders (ASD) are said to produce fewer conversational gestures, as specified in many diagnostic measures. Surprisingly, while research shows fewer deictic gestures in young children with ASD, there is a little empirical evidence addressing other forms of gesture. The discrepancy between clinical and empirical observations may reflect impairments unrelated to frequency, such as gesture quality or integration with speech. Adolescents with high-functioning ASD (n = 15), matched on age, gender, and IQ to 15 typically developing (TD) adolescents, completed a narrative task to assess the spontaneous production of speech and gesture. Naïve observers rated the stories for communicative quality. Overall, the ASD group's stories were rated as less clear and engaging. Although utterance and gesture rates were comparable, the ASD group's gestures were less closely synchronized with the co-occurring speech, relative to control participants. This gesture-speech synchrony specifically impacted communicative quality across participants. Furthermore, while story ratings were associated with gesture count in TD adolescents, no such relationship was observed in adolescents with ASD, suggesting that gestures do not amplify communication in this population. Quality ratings were, however, correlated with ASD symptom severity scores, such that participants with fewer ASD symptoms were rated as telling higher quality stories. Implications of these findings are discussed in terms of communication and neuropsychological functioning in ASD.
- [Show abstract] [Hide abstract]
ABSTRACT: We focus on the analysis, quantification and visualization of atypicality in affective facial expressions of children with High Functioning Autism (HFA). We examine facial Motion Capture data from typically developing (TD) children and children with HFA, using various statistical methods, including Functional Data Analysis, in order to quantify atypical expression characteristics and uncover patterns of expression evolution in the two populations. Our results show that children with HFA display higher asynchrony of motion between facial regions, more rough facial and head motion, and a larger range of facial region motion. Overall, subjects with HFA consistently display a wider variability in the expressive facial gestures that they employ. Our analysis demonstrates the utility of computational approaches for understanding behavioral data and brings new insights into the autism domain regarding the atypicality that is often associated with facial expressions of subjects with HFA.Multimedia and Expo (ICME), 2013 IEEE International Conference on; 01/2013
- [Show abstract] [Hide abstract]
ABSTRACT: This study examined the relationship between child language skills and parent and child gestures of 58 youths with and without an autism spectrum disorder (ASD) diagnosis. Frequencies and rates of total gesture use as well as five categories of gestures (deictic, conventional, beat, iconic, and metaphoric) were reliably coded during the collaborative Tower of Hanoi task. Children with ASD had lower Peabody Picture Vocabulary Test scores and gestured less and at lower rates compared to typically developing children. Gesture use was unrelated to vocabulary for typically developing children, but positively associated with vocabulary for those with ASD. Demographic correlates of gesturing differed by group. Gesture may be a point of communication intervention for families with children with ASD.Journal of Autism and Developmental Disorders 02/2014; 44(8). · 3.06 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: The present study concerns individual differences in gesture production. We used correlational and multiple regression analyses to examine the relationship between individuals' cognitive abilities and empathy levels and their gesture frequency and saliency. We chose predictor variables according to experimental evidence of the functions of gesture in speech production and communication. We examined 3 types of gestures: representational gestures, conduit gestures, and palm-revealing gestures. Higher frequency of representational gestures was related to poorer visual and spatial working memory, spatial transformation ability, and conceptualization ability; higher frequency of conduit gestures was related to poorer visual working memory, conceptualization ability, and higher levels of empathy; and higher frequency of palm-revealing gestures was related to higher levels of empathy. The saliency of all gestures was positively related to level of empathy. These results demonstrate that cognitive abilities and empathy levels are related to individual differences in gesture frequency and saliency. (PsycINFO Database Record (c) 2013 APA, all rights reserved).Journal of Experimental Psychology General 08/2013; · 5.50 Impact Factor
Conversational Gestures in Autism Spectrum Disorders: Asynchrony
but not Decreased Frequency
Ashley de Marchena and Inge-Marie Eigsti
Conversational or ‘‘co-speech’’ gestures play an important role in communication, facilitating turntaking, providing
visuospatial information, clarifying subtleties of emphasis, and other pragmatic cues. Consistent with other pragmatic
language deficits, individuals with autism spectrum disorders (ASD) are said to produce fewer conversational gestures, as
specified in many diagnostic measures. Surprisingly, while research shows fewer deictic gestures in young children with
ASD, there is a little empirical evidence addressing other forms of gesture. The discrepancy between clinical and empirical
observations may reflect impairments unrelated to frequency, such as gesture quality or integration with speech.
Adolescents with high-functioning ASD (n515), matched on age, gender, and IQ to 15 typically developing (TD)
adolescents, completed a narrative task to assess the spontaneous production of speech and gesture. Naı ¨ve observers rated
the stories for communicative quality. Overall, the ASD group’s stories were rated as less clear and engaging. Although
utterance and gesture rates were comparable, the ASD group’s gestures were less closely synchronized with the
co-occurring speech, relative to control participants. This gesture–speech synchrony specifically impacted communicative
quality across participants. Furthermore, while story ratings were associated with gesture count in TD adolescents, no such
relationship was observed in adolescents with ASD, suggesting that gestures do not amplify communication in this
population. Quality ratings were, however, correlated with ASD symptom severity scores, such that participants with fewer
ASD symptoms were rated as telling higher quality stories. Implications of these findings are discussed in terms of
communication and neuropsychological functioning in ASD.
Keywords: conversational gestures; autism; synchrony; communication; narratives
Individuals with autism spectrum disorders (ASD) exhibit
striking impairments in socialization and communica-
tion. In many language-disordered populations, affected
individuals compensate for decreased verbal output via
nonverbal communication; thus, increased rates of
gesturing are observed in adults with acquired aphasia
[Buck & Duffy, 1980] and children with expressive
language delay [Thal & Tobias, 1992] and Down syn-
drome [Stefanini, Caselli, Nusbaum, & Small, 2007].
These data suggest a speech-specific origin of commu-
nicative impairments in these populations: communica-
tive intent appears relatively intact despite decreased
verbal output. In contrast, communicative impairments
in ASD are characterized as more global. Clinicians have
long reported abnormalities in both verbal and nonverbal
forms of communication [e.g., Kanner, 1943]. For
example, the original account of Asperger syndrome
described the ‘‘large,’’ ‘‘clumsy,’’ and ‘‘inappropriate’’
nature of gestures [Asperger, 1944, from Wing, 1981].
Early clinicians described the atypical composition of
gestures, their poor integration with speech, and a
general decrease in gesture quantity. A precise character-
ization of nonverbal communication impairments in
ASD, including conversational gestures, is essential to our
broader understanding of communication in ASD. In the
current study, we characterize communicative gestures
produced by high-functioning adolescents with ASD, link
this characterization to broader ASD symptomatology,
and suggest that gesture is primarily affected in ASD via
its synchrony with speech, as opposed to simple measures
of gesture frequency.
Gesture is studied within the fields of neuropsychology
and psycholinguistics. Within neuropsychology, gesture
(or praxis) is conceptualized as a motor rather than a
communicative process, and is defined as the ability to
perform purposeful motor movements with or without
objects [Heilman & Gonzalez Rothi, 2003]. In contrast,
psycholinguists define gestures as spontaneous commu-
[McNeill, 1992]. Co-speech gestures, in this latter sense,
are the focus of this study. Specifically, we focus on iconic
gestures, termed ‘‘descriptive gestures’’ on the Autism
Diagnostic Observation Schedule [ADOS; Lord, Rutter,
DiLavore, & Risi, 2002]. Iconic gestures depict physical
Autism Research 3: 311–322, 2010311
Received March 10, 2010; accepted for publication July 19, 2010
Address for correspondence and reprints: Inge-Marie Eigsti, University of Connecticut, 406 Babbidge Road, Unit 1020, Storrs, CT 06269.
Grant sponsor: University of Connecticut Research Foundation; Grant number: 458938.
Published online 9 December 2010 in Wiley Online Library (wileyonlinelibrary.com)
& 2010 International Society for Autism Research, Wiley Periodicals, Inc.
From the Department of Psychology, University of Connecticut, Connecticut (A.d.M., I.-M.E.)
properties of referents, often giving information that
complements co-occurring speech (for example, a throw-
ing motion complements, ‘‘he threw the coconut’’).
The presence of communicative gesture deficits in ASD
is widely asserted in the clinical literature. Impairments in
gesture are codified on gold-standard ASD diagnostic
measures and screeners such as the ADOS [Lord et al.,
2002], the Autism Diagnostic Interview [ADI; Lord, Rutter,
& LeCouteur, 1994], and the M-CHAT [Robins, Fein,
Barton, & Green, 2001]; on these measures, the absence or
infrequency of gesture is rated as symptomatic. Scoring
criteria for these diagnostic measures suggest that indivi-
duals with ASD use all gestures less than their typically
developing (TD) peers; however, this assertion has not
been demonstrated in the empirical literature. Protode-
clarative pointing (i.e., pointing to share attention), but
not instrumental pointing (i.e., pointing to request), is
found to be reliably reduced in frequency in ASD [Bono,
Daley, & Sigman, 2004; Camaioni, Perucchini, Muratori, &
Milone, 1997; Loveland & Landry, 1986; Mundy, Sigman,
Ungerer, & Sherman, 1986; Mundy, Sigman, & Kasari,
1990], a finding which highlights that the social aspects of
these gestures likely contribute to their delayed produc-
tion [Baron-Cohen, 1989; Klin, Jones, Schultz, Volkmar, &
Cohen, 2002]. However, the literature on non-pointing
gestures in ASD is sparser and less conclusive; specifically,
reductions in gesture frequency are not well replicated. It
may be that children with ASD simply appear to gesture
less due to overall reductions in communicative acts. For
example, early studies found reduced rates of gesture in
children with ASD [Bartak, Rutter, & Cox, 1975], and more
recent studies have observed gesture delays [Charman,
Drew, & Baird, 2003; Luyster, Lopez, & Lord, 2007].
However, others have failed to find group differences in
gesture frequency after controlling for overall amount of
speech [Attwood, Frith, & Hermelin, 1988; Capps, Kehres,
& Sigman, 1998]. In fact, some have found iconic gestures
to be a relative communicative strength for children with
ASD, perhaps because enacting experiences is more
accessible than verbalizing them [Capps et al., 1998],
and consistent with gesture use in other speech-disordered
populations (described above). Children with ASD also
show a reduced variety of gestures [Colgan et al., 2006;
Wetherby & Prutting, 1984], which may strengthen the
impression of fewer gestures.
In addition to reductions in overall communicative-
ness, clinicians may be sensitive to gesture quality, a less-
studied aspect of nonverbal communication in ASD.
Studies of adults with ASD show few differences in the
frequency or duration of nonverbal communication
directed at conversational partners [Garcı ´a-Pe ´rez, Lee, &
Hobson, 2007; Tantam, Holmes, & Cordess, 1993];
however, ASD groups differ in more subjective ratings,
such as ‘‘conversational flow’’ [Garcı ´a-Pe ´rez et al.,
2007]. Qualitative differences in gestures, such as ‘‘odd’’
greeting waves, have been noted [Hobson & Lee, 1998].
In fact, unusual quality, in addition to quantity, of
gestures has long been central to clinical accounts of ASD
[e.g., Asperger, 1944; Wing, 1981]. Qualitative aspects of
conversational gestures (e.g., ‘‘exaggerated’’ gestures), as
well as the integration of distinct forms of verbal and
nonverbal communication, including gesture, are coded
on ADOS modules intended for individuals with fluent
speech. One study of adults with ASD examined the co-
occurrence of speech and nonverbal communication
during conversation [Tantam et al., 1993]; this study
measured the co-occurrence of speech and gesture within
half-second intervals. Co-occurrence of these behaviors
was found in both the ASD group and controls, though
co-occurrence was reduced in the ASD group; unfortu-
nately this observation was not addressed statistically. In
addition, since the behaviors were coded in half-second
intervals (a fairly low resolution for speech and gesture),
the degree of asynchrony could not be determined. The
authors suggested that gesture and speech might be less
synchronous in ASD, but did not quantify this asynchrony.
In TD children and adults, gesture accompanies and
amplifies information conveyed in speech [McNeill,
1992]. Conversational partners are sensitive to gestural
content, and will incorporate information provided only
in gesture into speech [Cassell, McNeill, & McCullough,
1998]. The main motion (stroke phase) of a gesture
reliably occurs in synchrony, and is exquisitely well-
timed, with its semantically related speech [Chui, 2005;
Levelt, Richardson, & La Heij, 1985; McNeill, 1992; Nobe,
2000]. Although gesture can communicate information
that is not found in speech, its communicative power is
reduced in the absence of speech [Krauss, Morrel-
Samuels, & Colasante, 1991], suggesting that it is the
integration of gestures with speech that provides commu-
nicative power. In typical adults, gesture–speech compre-
hension is dependent on timing [Habets, Kita, Shao,
O¨zyurek, & Hagoort, 2010]; thus, when gestures are
poorly synchronized with speech, their communicative
power may be diminished. The integration of gesture and
speech—the coordinated timing of these behaviors—is
the focus of this study. Impairments in gesture–speech
coordination by individuals with ASD may account for
discrepancies in the clinical and empirical literatures.
Specifically, gestures produced by individuals with ASD
may be poorly integrated with speech, thereby reducing
communicative power. Thus, while gestures may be equally
present in ASD (consistent with the empirical literature),
their effectiveness in communication may be reduced (con-
sistent with the clinical literature).
Gesture–speech synchrony requires the efficient co-
ordination of distinct behaviors. There is a growing
literature demonstrating impairments in behavioral tim-
ing in ASD. Specifically, electrophysiological studies have
shown delayed responses to social stimuli by children
312 de Marchena and Eigsti/Asynchrony of conversational gestures in ASD
with ASD, as compared to typical peers [McPartland,
Dawson, Webb, Panagiotides, & Carver, 2004; Webb,
Dawson, Bernier, & Panagiotides, 2006]. A recent electro-
myography study of facial mimicry showed that children
with ASD differed from typical peers only in terms of
their latency to mimic, but not in the amount and
appropriateness of mimicry [Oberman, Winkielman, &
Ramachandran, 2009]. These findings suggest that defi-
cits in interpersonal synchrony may reflect inefficient
timing, rather than execution. Timing impairments may
also be present in non-social cognitive processes [Sears,
Finn, & Steinmetz, 1994].
Abilities in ASD appear to be more normalized in
conditions of high structure and explicit attention,
relative to spontaneous production. This phenomenon
applies to gesture: when children with ASD are specifi-
cally instructed to gesture, their gestures appear more
typical [Bartak et al., 1975]. Research on how gesture, as a
communicative tool, is limited in ASD, should thus begin
with spontaneous, rather than prompted, communica-
tion. The current study used a narrative task from the
ADOS [Lord et al., 2002] to elicit speech and gesture. This
task prompts participants to tell a story based on six
picture cards, allowing for a high degree of spontaneity;
participants were not instructed to gesture. This task thus
provides a spontaneous communication sample that is
structured enough to allow comparisons between groups
and individuals. We use this task to assess verbal and
gestural communication, and to measure the degree of
synchrony between these two modalities. These data
provide quantitative information on gesture production,
its integration with speech, and the degree to which
gestures are related to autistic symptomatology.
Participants were 20 adolescents with ASD and 16
adolescents with TD matched on chronological age,
gender, and full-scale IQ [Stanford-Binet, Fifth edition;
Roid, 2003]. Table I provides demographic information
by group. Diagnoses, based on DSM-IV [APA, 2000]
criteria, were confirmed in the ASD group and ruled out
in the TD group using the ADOS [Lord et al., 2002], the
Social Communication Questionnaire [Rutter, Bailey, &
Lord, 2003], and clinical judgment. Nine participants met
criteria for Autistic Disorder, five for PDD-NOS, and one
for Asperger’s Disorder.
Participants with ASD were not excluded for comorbid
learning or psychiatric disorders. Likewise, participants
Table I.Demographic Information for Participants with ASD and TD Control Participants
ASD M (SD)
TD M (SD)
Chronological age (years)
0.93 15.0 (1.5)
Verbal 1.630.21 0.06
Full-scale IQ 0.0030.95
Communication (C)2.7 (1.0)
SCQ (total score)b
ASD, autism spectrum disorder; TD, typically developing; ADOS, Autism Diagnostic Observation Schedule; SR, social reciprocity; SCQ, Social
aOn the ADOS, 7 is the cutoff for a diagnosis on the autism spectrum, 10 is the cutoff for autism. All ASD participants in the final sample, except one,
were above the cutoff for an ASD diagnosis on the ADOS; this participant had a high SCQ score (20) and was judged to carry an ASD diagnosis by clinicians
on the study.
bWhen used as a screening instrument, a cutoff score of 15 is recommended as an indication of a possible ASD [Rutter et al., 2003].
de Marchena and Eigsti/Asynchrony of conversational gestures in ASD313
with TD were not excluded for non-ASD learning or
psychiatric disorders, to avoid obtaining a ‘‘hypernormal’’
comparison group. Participants with TD were excluded,
however, if they had any first-degree relatives with ASD or
any history of neurological problems.
Five participants from the ASD group were excluded for
the following reasons: ASD diagnosis not confirmed
(n51), IQ below 80 (n53), and failed digital recording
(n51); these individuals are not included in subsequent
analyses. One participant with TD was excluded because
he had a mild form of cerebral palsy. The final sample
consisted of 15 adolescents with ASD and 15 adolescents
with TD, all of whom had IQ scores within the average
This study was approved by the University of Connecti-
cut Institutional Review Board. Written consent or assent
was obtained from parents and participants.
Autism Diagnostic Observation Schedule [ADOS;
Lord et al., 2002].
The ADOS is a semistructured
assessment for the diagnosis of pervasive developmental
disorders. Participants completed either Module 3 or
Module 4, depending on their maturity level. Six
participants from each group completed Module 3, and
nine participants from each group completed Module 4.
Modules 3 and 4 provide comparable scores, so scores for
the two modules were collapsed; data are presented in
Table I. The cartoons task from the ADOS is an optional
task designed to assess gestural communication and the
integration of verbal and nonverbal communication.
This task (hereafter, the narrative task) served as the
primary experimental measure of gesture and speech
Rutter et al., 2003].
questionnaire for the screening of ASD symptoms
in children, based on the ADI-R [Lord et al., 1994], a
major tool used for diagnosing ASD. Parents of 27
questionnaire; three parents (one from the TD sample,
and two from the ASD sample) were unable to complete
and return the measure.
The SCQ is a 40-item parent
Stanford-Binet Intelligence Scale: Fifth Edition
The Stanford-Binet is a factor-analytic
providing a reliable estimate of verbal and nonverbal
room at the University of Connecticut, in their homes, or
at school. Measures included in this study were collected
as part of a larger battery lasting approximately 4hr over
one or two sessions.
All participants were tested in a quiet
Stimuli for the narrative task were six (7 by 8.5in) cards
containing black-and-white line drawings depicting the
story of two monkeys.1Participants were instructed to
examine the cards and then they would be asked to tell
the story to the experimenter. Participants viewed the
cards one at a time; the cards were then removed and the
participant was asked to stand up and tell the story.
Participants who began speaking with their hands in
their pockets were asked to remove them. No further
instructions were given, and participants were never
explicitly instructed to gesture. Narrations were recorded
on digital video for transcription, gesture coding, and
assess the frequency and types of gestures used spon-
taneously during storytelling, naı ¨ve listeners’ perceptions of
story quality, and the temporal synchrony between gestures
and speech. All coders and story raters were trained research
assistants (RAs) naı ¨ve to study hypotheses and participant
The narrative task was used to
recordings using CLAN [Computerized Language Analysis
Software; MacWhinney, 2000]. All words and pauses were
transcribed, andspeech was
utterances based on a combination of linguistic and
Speech was transcribed from digital
Identifying and coding gesture categories.
categorized according to McNeill’s  taxonomy.
Gesture coding quantified the number and types of
gestures produced during narrations.2Movements that
were determined to be gestures were classified as either:
(1) iconic gestures (‘‘descriptive’’ gestures on the ADOS,
which depict the physical properties of objects or
actions), (2) metaphoric gestures (i.e., gestures with
abstract referents), (3) deictic gestures (i.e., pointing),
(4) beat gestures (i.e., gestures with minimal semantic
content that are timed with speech prosody), or (5)
1The six cards depict the following story: (1) A monkey (Monkey A) is
up in a tree picking coconuts. (2) A second monkey (Monkey B) is leaning
against a tree and a coconut lands in front of him; he appears surprised.
(3) Monkey B picks up the coconut. (4) Monkey A looks down to where
the coconut was and appears confused. We also see Monkey B running off
with the coconut. (5) Monkey A drops a second coconut from the tree,
and we see Monkey B running back to retrieve it. (6) Monkey B leans
over to take the new coconut, as Monkey A throws a coconut and hits
Monkey B in the head.
2One challenge of gesture coding is differentiating simple movements
from gestures. Adolescents with ASD might make movements that are
more erratic and difficult to interpret. If coding is relatively more difficult
or unreliable for gestures produced by the ASD group, this introduces a
potential confound. To address this challenge, prior to beginning gesture
coding, videos were watched in their entirety with a focus on the
participant’s movements and mannerisms. Then, once coders began
gesture coding, it was easier to differentiate gestures from non-commu-
nicative movements, as the coder already had a sense of the participant’s
314 de Marchena and Eigsti/Asynchrony of conversational gestures in ASD
emblems (‘‘conventional’’ gestures on the ADOS, such as
an ‘‘ok’’ sign).
Ratings of story quality.
two 7-point scales. These RAs had not participated in
prior speech or gesture coding. Each story was rated
according to two questions: (1) How well were you able to
follow this story? and (2) How engaged were you during
this story? To determine the relative contributions of
verbal and nonverbal communication to story ratings,
raters were instructed to listen to half of the stories
without the accompanying video (radio condition) and
watch the other half with both audio and video (TV
condition). Condition and group were counterbalanced
RAs (n510) rated each story on
determining the start and endpoint of a gesture and its
semantically related speech. Coders must be confident
about what speech co-occurs with a given gesture. It is
more difficult to determine the semantically related
speech for gestures with ambiguous meanings. This
might be especially confounding if gesture clarity is
reliably lower for one group. To address this issue, only
iconic gestures were included, as these gestures were
produced frequently by participants across groups, and
because the intended meaning of these gestures is the
most transparent. Gesture–speech synchrony was coded
for all participants who produced at least one iconic
gesture during their narration. For each iconic gesture, the
co-occurring speech was determined in two stages. First,
based on the form and motion of the gesture, the gesture
meaning was determined (e.g., throwing); coders also rated
their confidence in the gesture’s meaning. There was no
difference in confidence between groups, F(1,24)50.81,
P50.38, with ratings of M (SD) of 3.66 (0.68) and
3.83 (0.25) out of five for the TD and ASD groups,
respectively, suggesting that coders could confidently
identify gesture meanings. Second, the coder selected one
or two words from the speech transcription that contained
most of the information present in the gesture meaning
(e.g., ‘‘threw’’). In one case, the exact meaning of the gesture
was not present in the (TD) adolescent’s speech (the
utterance, ‘‘picked it up,’’ and the gesture, glossed as ‘‘a
coconut’’, contained distinct information); in this case, the
word (‘‘picked’’) that complemented the gesture’s meaning
After selecting the gesture–speech pair, the start and
endpoints of the speech and gesture were coded in
Noldus Observer [Noldus, Trienes, Hendriksen, Jansen, &
Jansen, 2000]. The audio portion of the video was
replayed at half-speed until the beginning of the first
phoneme and the end of the last phoneme in the target
word could be identified. During speech coding, the
video display was hidden; hence, coders would not be
biased by participants’ movements.
Once speech was coded, coders muted the audio signal
while coding gestures, so that speech would not bias
gesture coding. For gesture coding, videos were first
watched in real time to determine the stroke phase of the
gesture. In most accounts, the stroke carries most of the
gesture’s information [McNeill, 1992]. The stroke was
defined as the middle of three gesture phases (i.e.,
preparation, stroke, and retraction), and was the phase
that included the major motion of the gesture. Once the
stroke was identified, the picture was advanced frame-by-
frame to code its onset and offset.
As an additional measure of gesture quality, coders
rated their own confidence about the onset and offset of
each gesture, from 1 (uncertain) to 5 (very confident).
There was a significant group difference in coder
confidence, F(1,24)55.61, P50.03, with the TD group
gesture endpoints rated more confidently than the ASD
group, with group means (SD) of 4.58 (0.67) and 3.93
(0.73), respectively. The groups shared a similar range of
confidence ratings (from 3–5 in each group), and for both
groups, mean confidence was greater than 3, the
midpoint of the scale (anchored with the word ‘‘con-
fident’’). Thus, while RAs felt more confident coding the
TD group, both groups were seen as reliably ‘‘codable,’’
an impression consistent with the high inter-rater
reliability for this analysis (reported below).
inter-rater reliability. Because these ratings were entirely
subjective, and raters were given only limited scoring
instructions, there was a concern that raters would be
unreliable. In fact, inter-rater reliability was high.
Chronbach’s alpha (collapsed across diagnosis) was 0.93
when raters were asked how well they could follow the
story and 0.91 when asked how engaged they were.
Collapsed across question, Chronbach’s alpha was 0.87
for the TD group and 0.91 for the ASD group. Thus,
although the ratings were subjective in nature, raters
generally agreed on the communicative quality of the
stories, for both diagnostic groups.
Three aspects of speech coding were assessed for
reliability: utterance count, co-gesture speech, and the
timing of speech onset. To determine the reliability of
speech transcription, two coders independently tran-
scribed nine narratives (four by adolescents with ASD).
The intraclass correlation coefficient (ICC) for utterance
count per narrative was 0.95. Co-gesture speech was
independently coded by two coders for ten narrations
(four by adolescents with ASD). Percent agreement for the
words that best captured the meaning of the co-occurring
iconic gesture was 0.90. ICC for the timing of speech
onset was 0.99.
Three aspects of gesture coding were analyzed for
reliability by two independent coders: gesture count,
gesture type (iconic vs. other), and gesture timing.
Coding nine participants (seven with ASD), percent
agreement was 0.90 and ICC was 0.98 for the number
of gestures per narration. Kappa for gesture type (iconic
vs. other) was 0.86. ICC was 0.99 for gesture onsets.
Story quality ratings were analyzed for
de Marchena and Eigsti/Asynchrony of conversational gestures in ASD315
The narrative task provided information about how
adolescents with ASD gesture spontaneously during
storytelling. To investigate overall rates of verbal and
ANOVA was conducted with diagnostic group as the
independent variable, and utterance count, gesture
count, and gesture rate (gestures per utterance) as
dependent variables. Group differences were nonsignifi-
cant, F(3,26)50.36, P50.78, Z2
on this task, participants with ASD spoke and gestured
with the same frequency as participants with TD. Data are
presented in Table II. Interestingly, gesture rate varied
dramatically within groups, from as many as two gestures
per utterance (both groups) to as few as two gestures in
total (ASD group) or no gestures at all (TD group).
One possibility is that individuals with ASD may gesture
with similar frequency as their typical peers, but use
different types of gesture. We tested this hypothesis by
looking at the distribution of gesture types used during
narrations in the two groups. All gestures were categorized
as iconics, deictics, beats, and ‘‘other’’ (a category which
included metaphoric gestures and emblems, both of
which were very infrequent). While deictics, iconics, and
others were normally distributed, beat gestures violated
this assumption; as such, non-parametric Kolmogorov–
Smirnov (KS) tests were used to probe for group differences
in this category of gestures. Analyses indicated no
significant group differences in the frequency of different
categories of gesture, all P’s40.13. While beat gestures
appeared to be produced more frequently in the TD group
(4.7 vs. 1.7 per narration), this difference was not
significant, KS (28)50.91, P50.38.
Several subsequent analyses focused on iconic gestures.
Participants who produced sufficient iconic gestures
included 14 adolescents with ASD and 12 adolescents
with TD. The number of iconic gestures produced during
narratives was similar in across groups, t(25)5?1.46,
P50.16, Cohen’s d50.58; see Table II. A multivariate
ANOVA revealed no group differences in the length of
gesture stroke phases or co-occurring speech, F(2,23)5
0.64, P50.54, Z2
suggests that gestures produced by the ASD and TD
groups did not differ in raw quantity or duration.3
In contrast to the similar rates of communicative
events for the ASD and TD groups, sharp differences
a one-way multivariate
p50.04, suggesting that
p50.05; see Table III. This further
were observed in narrative quality. Naı ¨ve listeners rated
these stories while watching and listening to them (TV
condition) or only listening to them (radio condition).
Stories produced by the ASD group were rated across
conditions as significantly harder to follow, F(1,28)5
12.71, P50.001, Z2
5.02, P50.03, Z2
Correlational analyses supported the validity of the story
quality ratings. For the ASD group only, ratings of ‘‘how
engaged were you during this story?’’ were significantly
scores, r(15)5?0.57, P50.03; adolescents whose stories
were more engaging exhibited fewer social symptoms.
There was also a trend, in the expected direction, for
these ratings to be correlated with ADOS communication
although this effect failed to reach significance.
Correlational analyses within groups also indicated
that, for the TD group, gesture count was positively and
significantly correlated with ratings of communicative
quality in the TV condition, r(15)50.69, P50.005.
In contrast, for the ASD group, gesture count was
P50.83. The strength of these correlations differed
significantly between ASD and TD groups, z5?2.48,
p50.31, and less engaging, F(1,28)5
p50.15; data are presented in Table IV.
Table II.Utterance and Gesture Count, by Group
ASD M (SD)
TD M (SD)
Number of utterances 10.5 (3.3)
0.98 (0.76) 0.72 (0.65) 1.01 0.32
0.12 0.73 o0.01
Number of gestures (total)0.30 0.590.01
Number of iconic gestures2.52 0.12 0.08
ASD, autism spectrum disorder; TD, typically developing.
aGesture rate computed as number of gestures (total) divided by number
Speech by Group
Length of Gesture Stroke Phase and Co-occurring
ASD M (SD)
TD M (SD)
Gesture length (ms) 690 (510)
Speech length (ms)1.290.270.05
ASD, autism spectrum disorder; TD, typically developing.
3In addition, coding examined whether participant consistently linked
story characters with a particular hand or location. Results indicated that
gestures were predominantly produced by the right hand, M(SD) of 80%
(34%) and 67% (44%), for the ASD and TD groups, respectively, and
produced in a similar location (center of body) across groups. Both
analyses were non-significant, with P40.39. While more fine-grained
analyses may provide more detail, these preliminary results suggest no
group differences in the ability to mark characters via hand or location.
316de Marchena and Eigsti/Asynchrony of conversational gestures in ASD
P50.007. This finding is consistent with the hypothesis
that gestures augment communicative quality for TD
participants, but not for participants with ASD.
In addition to gesture and speech production, and
narrative quality, we investigated the temporal dynamics
of gesture–speech pairs for iconic gestures. The onset times
of each gesture and its co-occurring speech were marked;
the absolute value of the difference in onset times was taken
as a measure of gesture–speech synchrony. Gesture–speech
pairs that begin simultaneously have a value of zero; larger
values indicate less synchrony. One adolescent with ASD,
whose mean asynchrony was more than four standard
deviations above the ASD group mean, was excluded from
this analysis as an outlier. Gesture–speech onsets differed by
490 (SD5250) ms in the ASD group and by 240 (SD5200)
ms in the TD group, a difference that was significant with a
large effect size, t(23)5?2.72, P50.01, Cohen’s d5?1.13.
Data (excluding the outlier participant) are shown in Figure
1. Adolescents with ASD produced gestures that were
strikingly less synchronized with speech, compared to TD
In the gesture literature, gestures typically anticipate the
related item in speech by 100ms [Morrel-Samuels & Krauss,
1992]. In the current data set, gestures sometimes preceded
and sometimes followed the related speech, and the ASD
and TD groups did not differ in this dimension; 58.3% of
gestures produced by the ASD group and 57.1% of gestures
produced by the TD group started prior to the co-occurring
speech, a difference that was not significant, w2(1,n526)5
0.95, P51.0. In the TD group, gesture strokes preceded as
much as 460ms in advance of the speech, or lagged by as
much as 280ms, and in the ASD group, preceded and lagged
by as much as 2,770 and 930ms, respectively. Thus, the
relative onsets of gesture and speech did not vary by group;
rather, it was the absolute synchrony of gesture and speech
onset that distinguished them.
A linear regression analysis was conducted to investigate
the contribution of gesture–speech synchrony to commu-
nicative quality. Communicative quality is multiply
determined; it is influenced by (at least) language skills,
the ability to consider a listener’s perspective, and
experience with the narrative form. To control for some
of these additional factors, a regression was conducted
with communicative quality ratings as the dependent
variable. Full-scale IQ was entered at the first step, and
gesture–speech asynchrony values at the second step;
diagnostic status and the interaction of diagnosis with
asynchrony were entered in the final step, to test whether
the effect of asynchrony on story quality was more or less
pronounced for the ASD group. Results indicated that IQ
scores accounted for a nearly significant 13% of the
variance in communicative quality ratings, F(1,24)5
3.49, P50.07. Controlling for full-scale IQ, gesture–
speech asynchrony accounted for an additional signifi-
cant 20% of the variance in communicative quality,
F(1,23)56.65, P50.02. Finally, testing for group effects,
the addition of diagnostic status contributed to an addi-
tional significant 26% of the variance, F(2,21)56.43,
P50.007. Unsurprisingly, given the significant group
differences in asynchrony, adding diagnostic status to
the regression model meant that asynchrony no longer
accounted for specific significant variance, b5?0.188,
t5?1.16, P50.26; diagnostic status was a significant
contributor to the model, b50.672, t53.13, P50.005.
The interaction of asynchrony with diagnostic status
did not account for significant independent variance,
gestures produced stories that were harder to understand,
specific negative impact on communication across
groups; furthermore, this relationship was consistent
Averaged Across Ten Different Raters
Listener Ratings of Story Quality (on a 1–7 scale),
ASD M (SD)
TD M (SD)
‘‘How well were you able to follow this story?’’ (15somewhat; 75very well)
3.3 (1.5)5.0 (1.0)
‘‘How engaged were you during this story?’’ (15somewhat engaged;
3.6 (1.1)4.6 (1.3)
ASD, autism spectrum disorder; TD, typically developing.
** p = 0.01
Gesture-speech offset (ms)
by diagnostic group (??P50.01).
Synchrony of iconic gestures and co-occurring speech,
de Marchena and Eigsti/Asynchrony of conversational gestures in ASD317
Although clinicians, and the diagnostic literature, have
often described lifelong gestural impairments in ASD, to
date there has been little empirical work to quantify how
gesture is used in combination with fluent speech in
older (i.e., beyond preschool) individuals with ASD. The
data presented here contribute to our understanding of
how adolescents with ASD spontaneously use gesture and
speech to communicate, and how these distinct commu-
nicative modalities are integrated.
Using a structured narrative task in which participants
were not instructed or prompted to gesture, we found
that adolescents with ASD and TD spontaneously
produced the same number of utterances and gestures
in their narratives. While inconsistent with some prior
work [e.g., Bartak et al., 1975], this finding is consistent
with studies showing comparable gesture rates for ASD
and other populations [e.g., Attwood et al., 1988; Capps
et al., 1998; Tantam et al., 1993], and suggests that raw
quantities of gestures cannot account for group differ-
ences. In addition, we found the iconic gestures produced
by the two samples to be equally interpretable by trained
coders, suggesting that the gestures produced by teens
with ASD were communicative. However, despite simila-
rities in gesture and speech frequency between groups,
results indicated that naı ¨ve listeners rated the stories
produced by adolescents with ASD as significantly harder
to follow, and less engaging, than stories produced by
adolescents with TD. This finding is consistent with the
literature on narratives in ASD, which suggests that
narratives are less coherent in this population [Baron-
Cohen, Leslie, & Frith, 1986; Diehl, Bennetto, & Young,
2006]. Although differences in communicative quality
were apparent, as suggested by listener ratings, these
differences could not be reduced simply to differences in
the frequency of communicative acts. Interestingly, for
the adolescents with TD, communicative quality was
related to increased gesture rates, such that participants
who produced more gestures were rated as telling stories
that were more engaging and easier to follow. However,
in the ASD group, there was no relationship between
gesture production and listener ratings, suggesting that
gestures, while present, did not contribute to story
quality. Furthermore, higher quality ratings given by
naı ¨ve listeners were associated with fewer autism symp-
toms, suggesting the relevance of this task to real-world
Our most robust finding was that gesture and its
semantically related speech were less synchronized in
adolescents with ASD than adolescents with TD. This
effect was highly reliable despite our relatively small
sample size and relatively small number of iconic gestures
produced, and points to the strength of integration
impairments in this population. Integration of verbal
and nonverbal communication is clinically relevant in
ASD (e.g., on diagnostic tools). Empirically, children and
adolescents with ASD exhibit difficulties integrating
audio and visual information from linguistic stimuli
[Bebko, Weiss, Demark, & Gomez, 2006; Smith &
Bennetto, 2007]. In fact, one of the only studies to look
at gesture comprehension in ASD [Silverman, Bennetto,
Campana, & Tanenhaus, 2010] found that adolescents
with ASD were slower to select a picture matching a
spoken sentence when that sentence was accompanied
by an informative gesture; TD participants were faster
when the gesture was present. This suggests that
adolescents with ASD could not integrate information
from speech and gesture, and in fact may have found this
information distracting. Given the parallels between
language comprehension and production, it stands to
reason that integration impairments in communicative
production should be expected in ASD. However, the
specific nature of integration impairments has not been
well described to date.
Our findings provide evidence that the integration of
verbal and nonverbal communication is impaired in ASD,
even given comparable baseline rates of communication.
Further, the degree of gesture–speech coordination was
associated with communicative quality ratings by naı ¨ve
listeners. These findings are consistent with recent work
on interpersonal synchrony, which has shown differences
in timing but not in quantity of facial mimicry [Oberman
et al., 2009], and on ERP studies of social responsiveness
[McPartland et al., 2004; Webb et al., 2006], as reviewed
in the Introduction; this work suggests differences in
interpersonal synchrony in ASD, with negative social and
Our study differs from this prior work in its focus on
synchrony of behaviors within an individual. There is no
a priori reason to believe that a single modality would be
delayed in the ASD sample, as some gesture theorists
propose that gesture and speech arise out of a single
semantic origin [e.g., ‘‘growth points,’’ McNeill, 2005].
Other theorists have proposed that gesture and speech
arise out of separate, but parallel processes that are highly
coordinated during production [Chu & Kita, 2008;
Kita, 2000; Kita & O¨zyu ¨rek, 2003]. If either of these
proposals is true, instead of a specific impairment in a
single modality (gesture vs. speech), one might expect
decreased overall synchrony, consistent with current
findings. In addition to group differences in gesture–
speech synchrony, results indicated that the individual
degree of gesture–speech synchrony played a significant
role in listener ratings of story quality, over and above
the contribution of IQ, further supporting the hypo-
thesis that gesture–speech integration contributes to
The study of gesture–speech
our understanding of communicative strengths and
318de Marchena and Eigsti/Asynchrony of conversational gestures in ASD
weaknesses in ASD. It also provides support for specific
theories of brain function in this population. Although
our data cannot distinguish between these theories, they
do lend support to several current proposals regarding
brain function in ASD. The integration of speech with
gestures, which are often produced bilaterally, should
regions and across hemispheres [McNeill, 2005]. Further,
if gesture and speech represent separate but parallel
processes [Chu & Kita, 2008; Kita, 2000; Kita & O¨zyu ¨rek,
2003], then their integration should require connectivity
between distinct neural areas. Connectivity between
distant brain regions is reduced in ASD [Just, Cherkassky,
Keller, & Minshew, 2004; Kana, Keller, Cherkassky,
Minshew, & Just, 2006]. Our finding of decreased
synchrony between speech and gesture provides support
for theories of reduced connectivity.
The impaired coordination of speech and gesture
reported here may also suggest the involvement of
Broca’s area, which is involved in the retrieval of
semantic information [Gough, Nobre, & Devlin, 2005]
and action production [Nishitani, Schurmann, Amunts,
& Hari, 2005]. In fact, Broca’s area appears to be involved
in integrating information acquired from both speech
and gesture and generating semantic representations
[Skipper, Goldin-Meadow, Nusbaum, & Small, 2007].
Functional neuroimaging studies have suggested that
Broca’s area functions atypically during language tasks in
ASD, although some studies report increased activation
[Knaus, Silver, Lindgren, Hadjikhani, & Tager-Flusberg,
2008] and others report decreases [Harris et al., 2006].
The iconic gestures evaluated in this study are rich
in semantic information; differences in Broca’s area
function could contribute to decreased integration of
semantic information from speech and gestural modali-
ties [McNeill, 1992, 2005].
Although it has yet to be investigated, gesture–speech
synchrony likely involves the cerebellum. Atypicalities of
the cerebellum have been found in individuals with ASD
in multiple studies [Chugani, Sundram, Beham, Lee, &
Moore, 1999; Courchesne, Yeung-Courchesne, Press,
Hesselink, & Jernigan, 1988; Fatemi et al., 2002; Otsuka,
Harada, Hisaoka, & Nishitani, 1999; Ritvo et al., 1986].
The cerebellum has been shown to be involved in tool
use [Higuchi, Imamizu, & Kawato, 2007] and tool-use
gestures [Choi et al., 2001]. Further, the cerebellum is
involved in the timing and integration of behaviors.
Eyeblink conditioning, mediated by the cerebellum
[McCormick & Thompson, 1984], requires rapid and
precise timing, and is inefficient in ASD [cf. a study of
eyeblink conditioning in ASD which demonstrated intact
learning but poorly coordinated timing of eyeblinks;
Sears et al., 1994]. The cerebellum also controls the
timing of behaviors that have both a cognitive and a
motor component [Glickstein, 2006], and that require
close synchrony [Katz & Steinmetz, 2002], such as speech
production [Ackermann, Mathiak, & Ivry, 2004]. Gesture
is not only timed closely with speech, but is also part of a
larger suite of behaviors that are rhythmically coordi-
nated [Loehr, 2007]. Although our data are purely
behavioral, the finding that gesture and speech are
asynchronous in ASD is consistent with atypical cerebel-
lar development in this population [Allen, 2005].
There are several limitations to this work, many of
which motivate future studies. This study did not include
a measure of gesture form. Gestures may be poorly
formed in ASD, which may negatively impact their
communicative power. Ongoing research in our labora-
tory will include a standardized praxis assessment to
disentangle the role of fine motor control in gesture
production. Another limitation of this study is that we
used subjective ratings of story quality as a measure of
communicative skill. While there are certain limitations
to using a relatively short seven-point scale as an
outcome measure, we believe that it is important to
obtain holistic ratings of how listeners interpret commu-
nicative exchanges, rather than simply reducing them to
their component parts. Finally, the gesture production
tasks used in this study involve communication pro-
duced in a ‘‘monologue’’ format. That is, although an
experimenter was present and responded nonverbally
throughout, participants told their stories uninterrupted.
Beattie and Aboudan  found that typical adults
produce more gestures during conversations than mono-
logues. The distinction between conversation and mono-
logue will be particularly important to address in
adolescents with ASD, who struggle with conversational
skills. Although we found that adolescents with ASD
produced the same quantity of gestures as adolescents
with TD, it is certainly possible that in conversation these
adolescents might produce fewer gestures, consistent
with clinical impressions [though cf. Tantam et al., 1993,
which failed to find differences in the frequency of
nonverbal communication during conversations].
A major advantage of this study is the use of
spontaneous communication samples to study gesture–
speech synchrony. Individuals with ASD often perform
more similarly to controls when given explicit instruc-
tions, relative to their spontaneous behaviors [Charlop,
Schreibman, & Thibodeau, 1985]. For example, the
timing of spontaneous but not explicitly instructed facial
mimicry is delayed in ASD samples relative to controls
[Oberman et al., 2009], suggesting that the timing of
behaviors in ASD may be affected by explicit instruction,
and providing further support for the study of sponta-
neous over prompted behaviors. To truly understand the
constraints on how individuals with ASD communicate,
we must investigate how communicative behaviors, such
as speech and gesture, occur spontaneously. Future work
on the integration of communicative modalities should
de Marchena and Eigsti/Asynchrony of conversational gestures in ASD319
address how these distinct systems come together during
We thank all the undergraduate and graduate members of
the Developmental Cognitive Neuroscience Laboratory at
the University of Connecticut, and Ashley Lepack in
particular for assistance with synchrony coding. Many
thanks to the staff at The Learning Clinic in Brooklyn,
Connecticut for facilitating the data collection process.
Finally, thank you to all the adolescents who participated
in this study and their gracious families. Research was
supported by grant ] 458938 from the University of
Connecticut Research Foundation to I.M.E.
Ackermann, H., Mathiak, K., & Ivry, R.B. (2004). Temporal
organization of ‘‘internal speech’’ as a basis for cerebellar
modulation of cognitive functions. Neuroscience Reviews, 3,
Allen, G. (2005). The cerebellum in autism. Clinical Neuropsy-
chiatry, 2, 321–337.
APA. (2000). Diagnostic and statistical manual of mental
disorders, text revision. Arlington, VA: American Psychiatric
Asperger, H. (1944). Die ‘‘autstichen Psychopathen’’ im Kinder-
salter. Archiv fu ¨r Psychiatrie und Nervenkrankheiten, 117,
Attwood, H.H., Frith, U., & Hermelin, B. (1988). The under-
standing and use of interpersonal gestures by autistic and
Down’s syndrome children. Journal of Autism and Develop-
mental Disorders, 18, 241–257.
Baron-Cohen, S. (1989). Perceptual role-taking and protodeclara-
tive pointing in autism. British Journal of Developmental
Psychology, 7, 113–127.
Baron-Cohen, S., Leslie, A.M., & Frith, U. (1986). Mechanical,
stories in autistic children. British Journal of Developmental
Psychology, 4, 113–125.
Bartak, L., Rutter, M., & Cox, A. (1975). Comparative study of
infantile autism and specific developmental receptive lan-
guage disorders. I. The children. British Journal of Psychiatry,
Beattie, G., & Aboudan, R. (1994). Gestures, pauses, and speech:
An experimental investigation of the effects of changing
social context on their precise temporal relationships.
Semiotica, 99, 239–272.
Bebko, J., Weiss, J., Demark, J., & Gomez, P. (2006). Discrimina-
tion of temporal synchrony in intermodal events by children
with autism and children with developmental disabilities
without autism. Journal of Child Psychology and Psychiatry,
Bono, M.A., Daley, T., & Sigman, M. (2004). Relations among
joint attention, amount of intervention and language gain in
autism. Journal of Autism and Developmental Disorders, 34,
Buck, R., & Duffy, R.J. (1980). Nonverbal communication of
affect in brain-damaged patients. Cortex, 16, 351–362.
Camaioni, L., Perucchini, P., Muratori, F., & Milone, A. (1997).
Brief report: A longitudinal examination of the communica-
tive gestures deficit in young children with autism. Journal of
Autism and Developmental Disorders, 27, 715–725.
Capps, L., Kehres, J., & Sigman, M. (1998). Conversational
abilities among children with autism and children with
developmental delays. Autism, 2, 325–344.
Cassell, J., McNeill, D., & McCullough, K.-E. (1998). Speech-
gesture mismatches: Evidence for one underlying representa-
tion of linguistic and nonlinguistic information. Pragmatics
and Cognition, 6, 1–24.
Charlop, M.H., Schreibman, L., & Thibodeau, M.G. (1985).
Increasing spontaneous verbal responding in autistic children
using a time delay procedure. Journal of Applied Behavior
Analysis, 18, 155–166.
Charman, T., Drew, A., & Baird, C. (2003). Measuring early
language development in preschool children with autism
spectrum disorder using the MacArthur Communicative
Development Inventory (Infant Form). Journal of Child
Language, 30, 213–236.
Choi, S., Na, D.L., Kang, E., Lee, K., Lee, S., & Na, D. (2001).
Functional magnetic resonance imaging during pantomim-
ing tool-use gestures. Experimental Brain Research, 139,
Chu, M., & Kita, S. (2008). Spontaneous gestures during mental
rotation tasks: Insights into the microdevelopment of the
motor strategy. Journal of Experimental Psychology: General,
Chugani, D.C., Sundram, B.S., Beham, M., Lee, M., & Moore, G.J.
(1999). Evidence of altered energy metabolism in autistic
children. Progress in Neuropsychopharmacology and Biolo-
gical Psychiatry, 23, 635–641.
Chui, K. (2005). Temporal patterning of speech and iconic
gestures in conversational discourse. Journal of Pragmatics,
Colgan, S.E., Lanter, E., McComish, C., Watson, L.R., Crais, E.R.,
& Baranek, G.T. (2006). Analysis of social interaction gestures
in infants with autism. Child Neuropsychology, 12, 307–319.
Courchesne, E., Yeung-Courchesne, R., Press, G.A., Hesselink, J.R.,
& Jernigan, T.L. (1988). Hypoplasia of cerebellar lobules VI and
VII in autism. New England Journal of Medicine, 318,
Diehl, J.J., Bennetto, L., & Young, E.C. (2006). Story recall and
narrative competence in high-functioning children with
autism spectrum disorders. Journal of Abnormal Child
Psychology, 34, 83–98.
Fatemi, S.H., Halt, A.R., Realmuto, G., Earle, J., Kist, D.A., et al.
(2002). Purkinje cell size is reduced in cerebellum of patients
with autism. Cellular and Molecular Neurobiology, 22,
Garcı ´a-Pe ´rez, R.M., Lee, A., & Hobson, R.P. (2007). On inter-
subjective engagement in autism: A controlled study of
nonverbal aspects of conversation. Journal of Autism and
Developmental Disorders, 37, 1310–1322.
Glickstein, M. (2006). Thinking about the cerebellum. Brain,
320 de Marchena and Eigsti/Asynchrony of conversational gestures in ASD
Gough, P.M., Nobre, A.C., & Devlin, J.T. (2005). Dissociating
linguistic processes in the left inferior frontal cortex with
transcranial magnetic stimulation. Journal of Neuroscience,
Habets, B., Kita, S., Shao, Z., O¨zyurek, A., & Hagoort, P. (2010).
The role of synchrony and ambiguity in speech-gesture
integration during comprehension. Journal of Cognitive
Neuroscience, Early Access, 1–10.
Harris, G.J., Chabris, C.F., Clark, J., Urban, T., Aharon, I., et al.
(2006). Brain activation during semantic processing in autism
spectrum disorders via functional magnetic resonance ima-
ging. Brain and Cognition, 61, 54–68.
Heilman, K.M., & Gonzalez Rothi, L.J. (2003). Apraxia. In:
Heilman KMH, Valenstein E, editors. Clinical neuropsychology.
New York: Oxford University Press, pp 215–235.
Higuchi, S., Imamizu, H., & Kawato, M. (2007). Cerebellar
activity evoked by common tool-use execution and imagery
tasks: An fMRI study. Cortex, 43, 350–358.
Hobson, R.P., & Lee, A. (1998). Hello and goodbye: A study of
social engagement in autism. Journal of Autism and Devel-
opmental Disorders, 28, 117–127.
Just, M., Cherkassky, V., Keller, T., & Minshew, N. (2004). Cortical
activation and synchronization during sentence comprehen-
sion in high-functioning autism: Evidence of underconnec-
tivity. Brain, 127, 1811.
Kana, R.K., Keller, T.A., Cherkassky, V.L., Minshew, N.J., &
Just, M.A. (2006). Sentence comprehension in autism:
Thinking in pictures with decreased functional connectivity.
Brain, 129, 2484–2493.
Kanner, L. (1943). Autistic disturbances of affective contact.
Nervous Child, 2, 217–250.
Katz, D.B., & Steinmetz, J.E. (2002). Psychological functions
of the cerebellum. Behavioral and Cognitive Neuroscience
Reviews, 1, 229–241.
Kita, S. (2000). How representational gestures help speaking.
In: McNeill D, editor. Language and gesture. Cambridge:
Cambridge University Press, pp 162–185.
Kita, S., & O¨zyu ¨rek, A. (2003). What does cross-linguistic
variation in semantic coordination of speech and gesture
reveal? Evidence for an interface representation of spatial
thinking and speaking. Journal of Memory and Language, 48,
Klin, A., Jones, W., Schultz, R., Volkmar, F., & Cohen, D. (2002).
Defining and quantifying the social phenotype in autism.
American Journal of Psychiatry, 159, 895–908.
Knaus, T.A., Silver, A.M., Lindgren, K.A., Hadjikhani, N., & Tager-
Flusberg, H. (2008). fMRI activation during a language task in
adolescents with ASD. Journal of the International Neuro-
psychological Society, 14, 967–979.
Krauss, R.M., Morrel-Samuels, P., & Colasante, C. (1991). Do
conversational hand gestures communicate? Journal of
Personality and Social Psychology, 61, 743–754.
Levelt, W., Richardson, G., & La Heij, W. (1985). Pointing and
voicing in deictic expressions. Journal of Memory and
Language, 24, 133–164.
Loehr, D. (2007). Aspects of rhythm in gesture and speech.
Gesture, 7, 179–214.
Lord, C., Rutter, M., & LeCouteur, A. (1994). Autism Diagnostic
Interview-Revised: A revised version of a diagnostic interview
for caregivers of individuals with possible pervasive develop-
mental disorders. Journal of Autism and Developmental
Disorders, 24, 659–685.
Lord, C., Rutter, M., DiLavore, P.C., & Risi, S. (2002). Autism
Western Psychological Services.
Loveland, K.A., & Landry, S.H. (1986). Joint attention and
language in autism and developmental language delay.
Journalof Autismand Developmental
Luyster, R., Lopez, K., & Lord, C. (2007). Characterizing
communicative development in children referred for Autism
Spectrum Disorders using the MacArthur-Bates Communica-
tive Development Inventory (CDI). Journal of Child Lan-
guage, 34, 623–654.
MacWhinney, B. (2000). The CHILDES Project: Tools for
analyzing talk. Mahwah, NJ: Lawrence Erlbaum Associates.
McCormick, D.A., & Thompson, R.F. (1984). Neuronal responses
of the rabbit cerebellum during acquisition and performance
of a classically conditioned nictating membrane-eyelid
response. Journal of Neuroscience, 4, 2811–2822.
McNeill, D. (1992). Hand and mind: What gestures reveal about
thought. Chicago: University of Chicago Press.
McNeill, D. (2005). Gesture, & thought. Chicago: University of
McPartland, J., Dawson, G., Webb, S.J., Panagiotides, H., &
Carver, L.J. (2004). Event-related brain potentionals reveal
anomalies in temporal processing of faces in autism spectrum
disorder. Journal of Child Psychology and Psychiatry, 45,
Morrel-Samuels, P., & Krauss, R.M. (1992). Word familiarity
predicts the temporal asynchrony of hand gesture and
speech. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 18, 615–623.
Mundy, P., Sigman, M., Ungerer, J., & Sherman, T. (1986).
Defining the social deficits of autism: The contribution of
non-verbal communication measures. Journal of Child
Psychology and Psychiatry, 27, 657–669.
Mundy, P., Sigman, M., & Kasari, C. (1990). A longitudinal study
of joint attention and language development in autistic
children. Journal of Autism and Developmental Disorders,
Nishitani, N., Schurmann, M., Amunts, K., & Hari, R. (2005).
Broca’s region: From action to langauge. Physiology, 20,
Nobe, S. (2000). Where do most spontaneous representational
gestures actually occur with respect to speech? In McNeill D,
editor. Language and gesture. Cambridge: Cambridge University
Press, pp 186–198.
Noldus, L.P.P.J., Trienes, R.J.H., Hendriksen, A.H.M., Jansen, H.,
& Jansen, R.G. (2000). The Observer Video-Pro: New software
for the collection, management, and prsentation of time-
structured data from videotapes and digital media files.
Behavior Research Methods, Instruments, and Computers,
Oberman, L.M., Winkielman, P., & Ramachandran, V.S. (2009).
Slow echo: Facial EMG evidence for the delay of spontaneous,
but not voluntary, emotional mimicry in children with
(ADOS). Los Angeles:
de Marchena and Eigsti/Asynchrony of conversational gestures in ASD321
autism spectrum disorders. Developmental Science, 12,
Otsuka, H., Harada, M., Hisaoka, S., & Nishitani, H. (1999).
Brain metabolites in the hippocampal-amygdala region
and cerebellum in autism: An H-MR spectroscopy study.
Neuroradiology, 41, 517–519.
Robinson, H., et al. (1986). Lower Purkinje cell count in the
cerebella of four autistic subjects: Initial findings of the
UCLA-NSAC Autopsy Research Report. American Journal of
Psychiatry, 143, 862–866.
Robins, D.L., Fein, D., Barton, M.L., & Green, J.A. (2001). The
Modified Checklist for autism in toddlers: An initial study
investigating the early detection of autism and pervasive
developmental disorders. Journal of Autism and Develop-
mental Disorders, 31, 131–144.
Roid, G.H. (2003). Stanford-Binet Intelligence Scales (5th ed.).
Itsaca, IL: Riverside Publishing.
Rutter, M., Bailey, A., & Lord, C. (2003). The Social Commu-
nication Questionnaire. Los Angeles: Western Psychological
Sears, L.L., Finn, P.R., & Steinmetz, J.E. (1994). Abnormal
classical eye-blink conditioning in autism. Journal of Autism
and Developmental Disorders, 24, 737–751.
Silverman, L., Bennetto, L., Campana, E., & Tanenhaus, M.K.
(2010). Speech-and-gesture integration in high-functioning
autism. Cognition, 115, 380–393.
Skipper, J.I., Goldin-Meadow, S., Nusbaum, H.C., & Small, S.L.
(2007). Speech-associated gestures, Broca’s area, and the
human mirror system. Brain and Language, 101, 260–277.
Smith, E.G., & Bennetto, L. (2007). Audiovisual speech integra-
tion and lipreading in autism. Journal of Child Psychology
and Psychiatry, 48, 813–821.
Stefanini, S., Caselli, M.C., Nusbaum, H.C., & Small, S.L. (2007).
Spoken and gestural production in a naming task by young
childen with Down syndrome. Brain and Language, 101,
Tantam, D., Holmes, D., & Cordess, C. (1993). Nonverbal
expression in autism of Asperger type. Journal of Autism
and Developmental Disorders, 23, 111–133.
Thal, D., & Tobias, S. (1992). Communicative gestures in
children with delayed onset speech of oral expressive
vocabulary. Journal of Speech and Hearing Research, 34,
Webb, S.J., Dawson, G., Bernier, R., & Panagiotides, H. (2006).
ERP evidence of atypical face processing in young children
with autism. Journal of Autism and Developmental Disorders,
Wetherby, A.M., & Prutting, C.A. (1984). Profiles of commu-
nicative and cognitive-social abilities in autistic children.
Journal of Speech and Hearing Research, 27, 364–377.
Wing, L. (1981). Asperger’s syndrome: A clinical account.
Psychological Medicine, 11, 115–129.
322 de Marchena and Eigsti/Asynchrony of conversational gestures in ASD