Does autism affect gesturing during parent-child interactions in the early school years?

https://doi.org/10.1016/j.rasd.2019.101440Get rights and content

Highlights

  • Older, verbal children with ASD produce similar amounts and types of gestures as expressive vocabulary matched TD peers.

  • Children with ASD primarily use gesture to complement speech, while TD children use gesture to add information to speech.

  • Parents of children with ASD and TD children did not different in the overall amount or frequency of gesture types and gesture-speech combinations.

  • Differences in gesture production in children reflects child’s own communicative intentions, not parental input.

Abstract

Background

Young children with autism spectrum disorder (ASD) have been shown to differ from typically developing (TD) children in their production of gesture, as well as the relationship between gesture and the content of their speech. In this study, we asked whether older children with ASD continue to differ from TD children in the types of gestures and gesture-speech combinations that they produce, and whether these differences reflect differences in parental gesture input.

Method

Our study examined the gestures and speech produced by 39 early school-age children (17 with ASD, 14 boys; Mage = 7;2, and 22 with TD, 13 boys; Mage = 5;4), comparable in expressive vocabulary, and their parents, during a 10-minute play interaction. Gestures were coded for total amount, gesture type (deictic, conventional, or iconic), and gesture-speech relationship (complementary, disambiguating, or supplementary).

Results

Children with ASD were similar to TD children in the amount and types of gestures that they produced, but differed in their gesture-speech combinations, using gesture primarily to complement their speech. Parents did not show any group differences in their production of different types of gestures and gesture-speech combinations. There were no correlations between parent and child gesture patterns.

Conclusions

These findings suggest that differences in children’s gesture use may reflect the child’s own communicative intentions rather than parental input. These finding have important implications in understanding how older children with ASD use gesture in everyday interactive contexts, which can inform intervention or school-based practices to support learning.

Introduction

Gesture plays an important role in the early language acquisition of children with autism spectrum disorder (ASD). Like typically developing (TD) children (Iverson & Goldin-Meadow, 2005; Özçalışkan & Goldin-Meadow, 2005b), young children with ASD use gestures (e.g., point at bike) and gesture-speech combinations (e.g., ‘ride’+point at bike) before they produce similar words and sentences in speech (e.g., ‘bike’, ‘ride bike’; Özçalışkan, Adamson, Dimitrova, & Baumann, 2017; Özçalışkan, Adamson, Dimitrova, & Baumann, 2018). Also like parents of TD children (Iverson, Capirci, Longobardi, & Caselli, 1999; Özçalışkan & Goldin-Meadow, 2005a), parents of young children with ASD provide models for the types of gestures and gesture-speech combinations that their children produce (Özçalışkan et al., 2018). Here we ask whether the gesture-speech system of older children with ASD remains similar to that of TD children as they develop more complex language abilities, and if so, whether children’s gestures reflect the gestural input that they receive from their parents.

TD children use gestures to communicate about referents before they do so with words (e.g., Bates, 1976). Initially they use deictic (e.g., point at book) and give gestures (e.g., extend open palm toward book) to indicate or request referents. These gestures are followed by conventional gestures that express culturally-prescribed meanings (e.g., nodding head for yes) and by a few iconic gestures that convey characteristic actions or features associated with referents (e.g., flapping arms for bird flying; Acredolo & Goodwyn, 1985; Iverson, Capirci, & Caselli, 1994; Özçalışkan & Goldin-Meadow, 2005a, 2011). Importantly, these early gestures—particularly deictic gestures that indicate referents—precede and predict the emergence of similar vocabulary items in children’s speech (Iverson & Goldin-Meadow, 2005).

TD children continue to use gesture even after they begin to produce words, producing gesture-speech combinations. They first produce complementary combinations, in which gesture conveys the same information as speech (e.g., ‘dog’+point at dog), followed by disambiguating combinations in which gesture clarifies a pronominal referent in speech (‘that one’+point at dog) and supplementary combinations, in which gesture adds information not found in speech (‘pat’+point at dog). Importantly, these supplementary gesture-speech combinations precede and predict not only the time of onset but also the content of the word-word combinations TD children eventually produce in speech (Goldin-Meadow & Butcher, 2003; Özçalışkan & Goldin-Meadow, 2005b, 2010).

Even after the onset of two-word speech, TD children continue to use gesture to further expand their communicative repertoires in more extended speech contexts, including gains in their production of iconic gestures that convey unique meanings not found in speech, beginning around ages 2–3 (Özçalışkan, Gentner, & Goldin-Meadow, 2014). Gesture’s contribution to communication continues at the later ages, with 4- to 10-year-old children using gesture to disambiguate or supplement the information in their speech. These gesture-speech combinations precede and predict their emerging abilities in producing narratives and explanations in speech (e.g., Church & Goldin-Meadow, 1986; Colletta, Pellenq, & Guidetti, 2010, 2015; Demir, Levine, & Goldin-Meadow, 2015; Stites & Özçalışkan, 2017; Özçalışkan, 2007).

Compared to their TD peers, children with ASD often show delays in achieving early language milestones (i.e., first words, first sentences; Tager-Flusberg, 2007), with 30% of children with ASD considered to be non- or minimally verbal; thus verbal children with ASD represent a subset of the greater ASD population (Tager-Flusberg & Kasari, 2013). Children with ASD have also been shown to have weaknesses in gesture production early in development (2;0-5;0), particularly for deictic gestures (Mundy, Sigman, & Kasari, 1990), even when compared to TD children and children with Down’s syndrome with similar expressive speech vocabularies (Özçalışkan, Adamson, & Dimitrova, 2016; Toret & Acarlar, 2011). Despite differences in the amount of gesture production, young children with ASD (age 2;6) show a pattern akin to TD children in the types of gestures that they produce, using predominantly deictic and give gestures in their early interactions with their parents (Özçalışkan et al., 2016; Özçalışkan et al., 2018). Young children with ASD (2;6) also indicate referents first in gesture before conveying similar meanings exclusively in speech (e.g., point at cat before saying ‘cat’); thus mirroring the pattern observed in TD children (Özçalışkan et al., 2017).

Like TD children, children with ASD also continue to use gesture after they begin to produce their first words. However, they differ in their production of gesture-speech combinations, producing notably fewer combinations compared to TD children similar in language ability (Özçalışkan et al., 2018). Despite these differences, young children with ASD (2;6) resemble language-comparable TD children (1;6) in the types of combinations that they produce and the way these combinations relate to their emerging abilities in sentence construction: they produce predominantly complementary gesture-speech combinations, followed by supplementary and disambiguating combinations—mirroring the patterns observed in language-comparable TD children (Özçalışkan et al., 2018). Furthermore, the types of semantic relations children with ASD convey in these early supplementary combinations also resemble the ones produced by TD children, primarily conveying simple relations between entities and/or actions (e.g., ‘drink’+point to cup; ‘baby’+point at bottle). More importantly, the majority of children with ASD produce such supplementary combinations at a point when they are not yet producing word-word combinations in speech, suggesting a trajectory similar to TD children in the emergence of different sentence-like constructions across modalities (Özçalışkan et al., 2018).

Research that examines the gesture production of children with ASD beyond the two-word stage is relatively sparse. However, the few existing studies suggest that verbal children with ASD during the earlier (ages 6–12) and later school years (ages 11–16) continue to gesture, but less so than their age or IQ-matched TD peers (Medeiros & Winsler, 2014; Silverman, Eigsti, & Bennetto, 2017; So, Wong, Lui, & Yip, 2015). Research on gesturing at the later ages also presents mixed findings on the similarities children with ASD show to TD children in the types of gestures that they produce. Some studies (de Marchena & Eigsti, 2010; Medeiros & Winsler, 2014; Morett, O’Hearn, Luna, & Ghuman, 2016) suggest that school age children with ASD (ages 6–12) use fewer conventional gestures during narrative or problem-solving tasks as compared to their age-matched TD peers, while others (Wong & So, 2018) found that children with ASD use more iconic gestures than their age- and IQ-matched TD peers during spoken narrative tasks.

Similarly, research on gesture-speech combinations in older children with ASD presents mixed results. Some studies (Morett et al., 2016) suggest that adolescents with ASD (ages 10–20) produce fewer supplementary gesture-speech combinations than TD children matched on age, gender, and verbal IQ, a pattern that becomes more pronounced when the listener is visible to the child compared to a condition where the listener is not visible. Other studies (Wong & So, 2018), however, show that school-age children with ASD (ages 6–12) use similar amounts of disambiguating and supplementary combinations as TD children, but produce more complementary gesture-speech combinations than their TD peers. Recent work has also shown that adolescents with ASD (ages 13–16) show difficulties in gesture use during social-communicative tasks (e.g., narration), a deficit that is not observed during executive function tasks where gesture serves a cognitive function, thus suggesting that lower amounts of gesturing in ASD may be driven by difficulties in social communication (de Marchena & Eigsti, 2014). Less is known, however, about how verbal, school-age children with ASD compare to language-comparable TD peers in their spontaneous production of gesture types and gesture-speech combinations during parent-child interactions, particularly as their speech becomes more complex.

In summary, existing research on the gesture-speech system of children with ASD suggests that gesture is a robust aspect of the language learning process in children with ASD, particularly at the earlier ages (1;6-3;0). Children with ASD, even though they gesture less, nonetheless use gesture to convey their emerging abilities in speech—from first words to first sentences. They also continue to accompany speech with gesture—but still at lower rates—when producing more extended speech types (i.e., narratives) at the early school years. However, less is known about how early school-age children with ASD gesture, especially in comparison to language-comparable TD peers. Such research may shed further light on the patterns of similarities and differences in gesture production in ASD and in TD groups — patterns not driven by differences in speech production. Similarly, earlier work with older children with ASD has focused on speech and gesture production in narrative contexts, leaving gesture production during naturalistic interactions with a familiar partner, such as a parent, relatively unexplored. As such, the study of gesture production in a more naturalistic context might provide a fuller picture of relative weaknesses and strengths in gesture use among school-age children with ASD.

Parents of young TD children (1;0-3;0) gesture frequently when interacting with their children, providing models for the different types of gestures and gesture-speech combinations and tuning their gestures to their child’s communicative needs (Bekken, 1989; Iverson et al., 1999; Özçalışkan & Goldin-Meadow, 2005a). Parents predominantly use deictic gestures, and gestures that complement what they convey in speech, and produce fewer iconic gestures and supplementary gesture-speech combinations (Iverson et al., 1999; Özçalışkan & Goldin-Meadow, 2005a, 2011; see Özçalışkan & Dimitrova, 2013 for a review). Importantly, research on parental gesture input has primarily focused on its role in early language development, leaving parent gesture input to older TD children unexplored.

Research on gesture input to children with ASD also remains relatively sparse. The few existing studies that focused on gesture input to younger children (1;0-3;0) with ASD or at high-risk for ASD show that parents of children with or at high-risk for ASD are comparable to parents of TD children in both the amount and the types of gestures and gesture-speech combinations that they produce (Mitchell, 2015; Özçalışkan et al., 2018). However, additional research has suggested that even if parents do not differ in the types of gestures that they produce, they do show differences in how often they produce different types of gesture. For example, Talbott, Nelson, and Tager-Flusberg (2015) compared the gestural input provided to infants at high vs low risk for ASD and found no difference in the types of gestures produced or their relative frequency. There were differences, however, in how often they produced each gesture type, with mothers of children with high risk for ASD using more of each gesture type than mothers of children with low risk for ASD.

To date, there have been only a few studies that observe parental nonverbal input to school-age children with ASD. Medeiros and Winsler (2014), using a collaborative problem-solving task, found that parents of TD children and children with ASD gestured at the same rate as their children. However, other studies that focused on a broader set of nonverbal parental behaviors (e.g., physical contact, social verbal approaches, directive statements) showed that parents of children with ASD used more high-intensity approaches and directive behaviors, including gestural prompts (Doussard-Roosevelt, Joe, Bazhenova, & Porges, 2003; Wan et al., 2012).

In summary, existing research on parental gesture input to children with ASD remains relatively sparse, and most of it focuses on young children. The overarching finding for the early gesture input is that parents of children with ASD show similar patterns to parents of TD children in the types and amounts of gestures and gesture-speech combinations that they produce. However, there are no studies that specifically focus on gesture input directed to older children with ASD—a point in development where children might be making important gains in their language development. The few existing studies focus on either a broader set of nonverbal behaviors (including but not limited to gesture) or a particular type of task, leaving parents’ gesturing in more spontaneous everyday interactions unexamined. As such, examination of parent gesture input will not only provide important information about the nature of parental nonverbal input at the later ages, but it may also shed new light on how that input might reflect patterns of gesture production in children with more complex language skills.

Previous work has shown that gesture development and speech development go hand-in-hand in younger children with ASD and that parents contribute to this process by serving as models to their children for the types of gestures and gesture-speech combinations. However, we know little about the nature of the gesture-speech system in early school-age children with ASD in interactive contexts, particularly for children with more advanced expressive language skills, and whether parents continue to play a role in the development of the child’s gesture-speech system. In this study, we focus on the speech and gestures produced by 17 early school-age children with ASD and 22 TD children, comparable in language ability, and their parents. The decision to form comparable groups based on language, and not chronological age or IQ, stemmed from our interest in identifying patterns of similarities and differences in gesture production not driven by differences in speech production. We first ask whether early school-age children with ASD differ from language-comparable TD children in the types of gestures and gesture-speech combinations that they produce. Based on previous research with both younger children and adolescents with ASD, we predict that early school-age children with ASD will produce fewer gestures, a difference that will be more pronounced for deictic gestures and supplementary gesture-speech combinations.

We also ask whether the group differences that we observe in child gesture reflect differences in the gestures and gesture-speech combinations that their parents produce. The existing evidence suggests two possibilities: parents of early school-age children with ASD will not differ from parents of TD children, thus following a pattern observed in studies of young children with ASD (Mitchell, 2015; Özçalışkan et al., 2018). An alternative possibility is that parents of early school-age children with ASD might gesture more than parents of TD children, while remaining similar in their proportional use of each gesture and gesture-speech combination type, as a way to provide additional communicative support (e.g., Talbott et al., 2015).

Section snippets

Participants

The participants included 39 children (17 with ASD, 22 TD)—all native English speakers, along with their parents. The gender ratio in the ASD group reflects the approximate 4:1 ratio of boys to girls in the greater population (Baio et al., 2018). The current sample size was based on a similar earlier study that compared gesture use in TD children and children with ASD and found that 16 subjects per group was adequate to detect reliable effects at p < .05, η2 = 0.44 (So et al., 2015). The

Speech

Children did not show group differences for either the amount (i.e., word tokens, F(1, 38) = .35, p = .56, ηp² = .01) or the diversity of the words (i.e., word types, F(1, 38) = .26, p = .61, ηp² = .01) that they produced (see Table 2, upper half), which was by design. The children also did not differ by group in their production of communicative acts with speech, F(1, 38) = .84, p = .37, ηp² = .02. All children in the sample were producing word-word combinations at the time of our observation,

Discussion

Previous work shows that young children with ASD (Mage = 2;6) and with TD (Mage = 1;6), but not their parents, show group differences in their relative production of different types of gestures and gesture-speech combinations (Özçalışkan et al., 2018). In this study, we asked whether the patterns observed for younger children remain similar in older children with ASD (Mage = 7;2) and with TD (Mage = 5;4) and their parents. More specifically, we asked whether older children with ASD continue to

Declaration of Competing Interest

Authors declare they have no conflict of interest.

Acknowledgements

This research was supported by a grant from NSFNSF (National Science Foundation), United States (BCS 1251337) to Özçalışkan and Adamson. We thank Melinda Reed, Alice Guo, and Kadija Hamki for their help in the collection and coding of the data, and Dr. Diana Robins for assistance with participant recruitment.

References (52)

  • E. Bates

    Language and context

    (1976)
  • K. Bekken

    Is there motherese in gesture?

    (1989)
  • B.B. Braddock et al.

    Patterns of gesture use in adolescents with autism spectrum disorder

    American Journal of Speech-Language Pathology

    (2016)
  • L. Capps et al.

    Conversational abilities among children with autism and children with developmental delays

    Autism

    (1998)
  • R.B. Church et al.

    The mismatch between gesture and speech as an index of transitional knowledge

    Cognition

    (1986)
  • J. Colletta et al.

    Effects of age and language on co-speech gesture production: An investigation of French, American, and Italian children’s narratives

    Journal of Child Language

    (2015)
  • A. de Marchena et al.

    Conversational gestures in autism spectrum disorders: Asynchrony but not decreased frequency

    Autism Research

    (2010)
  • A.B. de Marchena et al.

    Context counts: The impact of social context on gesture rate in verbally fluent adolescents with autism spectrum disorder

    Gesture

    (2014)
  • O.E. Demir et al.

    A tale of two hands: Children’s gesture use in narrative production predicts later narrative structure in speech

    Journal of Child Language

    (2015)
  • J.A. Doussard-Roosevelt et al.

    Mother-child interaction in autistic and nonautistic children: Characteristics of maternal approach behaviors and child social responses

    Development and Psychopathology

    (2003)
  • L.M. Dunn et al.

    The peabody picture vocabulary test

    (2007)
  • F. Faul et al.

    GPOWER: A priori-, post hoc-, and compromise power analyses for MS-DOS [computer program]

    (1992)
  • S. Goldin-Meadow et al.

    Pointing toward two-word speech in young children

  • J.M. Iverson et al.

    Gesture paves the way for language development

    Psychological Science

    (2005)
  • A.S. Kaufman et al.

    Kaufman brief intelligence test

    (2004)
  • C. Lord et al.

    The autism diagnostic observation schedule-generic: A standard measure of social and communication deficits associated with the spectrum of autism

    Journal of Autism and Developmental Disorders

    (2000)
  • View full text