I use my space not yours: Use of gesture space for referential identification among children with autism spectrum disorders

https://doi.org/10.1016/j.rasd.2016.03.005Get rights and content

Highlights

  • Gestures identify referents when they are coherent with the established locations.

  • Children with ASD have the ability to produce referent-identifying gestures (RIG).

  • These children are as capable as TD children in producing RIG in spoken narratives.

  • But this ability may be limited when it is necessary to take others perspectives.

  • This ability is related to children’s communication and social interaction skills.

Abstract

Background

Speakers move their hands (gesture) when they talk. They indicate non-present referents by gesturing those referents in the same locations as the previous gestures for the same referents (spatially-consistent referent-identifying gestures). We examined whether children with autism spectrum disorders (ASD), whose gesture development is slow, produce this kind of gesture in two circumstances: (1) demonstration task in which they had to follow experimenters’ established referents and spatial locations; and (2) spoken narrative task in which they created their own referents and locations.

Method

School-aged children with ASD and typically-developing children (TD) demonstrated daily-life activities after the experimenter had orally described and gestured the spatial locations of non-present objects involved (demonstration task) and enacted the characters of a story from a picture (spoken narrative task).

Results

ASD children produced fewer spatially-consistent referent-identifying gestures than TD children in the demonstration task. When not following experimenter’s locations, they established their own gestural locations and kept them consistent. They produced a comparable number of spatially-consistent referent-identifying gestures to their TD counterparts in the spoken narrative task. The ability to produce spatially-consistent referent-identifying gestures was negatively correlated to communication and social interaction skills in the demonstration task in both groups. However, such relationship was not found in the spoken narrative task in ASD children.

Conclusion

ASD children have the ability to use consistent gestural locations to represent referents, but this ability may be limited when they have to follow others’ perspectives. Communication and social interaction abilities are crucial when taking others’ perspectives.

Introduction

Speakers not only use linguistic devices (nouns, pronouns, or zero anaphora) but also produce gestures to indicate referents. Co-speech gestures are spontaneous hand movements produced when speaking (McNeill, 1992, McNeill, 2005. They serve multiple functions including but not limited to expressing culture-specific meanings (e.g., producing a THUMB-UP gesture to praise others), making requests (e.g., POINTING to a cookie to request the cookie), describing dimensions of objects (e.g., producing a ROUND gesture to describe the shape of a cake), and representing concrete objects (i.e., referents) in a real or virtual environment (e.g., POINTING to John while he was standing at the speaker’s right; POINTING to the empty space at the right to represent John again, who has left the room). The present study focuses on the last function of gesture—referential identification.

Gesture conveys visual-spatial information (Cassell and McNeill, 1991, McNeill, 1992, McNeill, 2005). Therefore, speakers can easily exploit the visual-spatial nature of gesture to indicate non-present referents by forming associations between the spatial locations of their gestures and the corresponding non-present entities (Gullberg, 1998, Gullberg, 2003, Gullberg, 2006; So, Coppola, Licciardello, & Goldin-Meadow, 2005; So, Kita, & Goldin-Meadow, 2009; Yoshioka, 2008). Specifically, they produce gestures in non-neutral spatial locations (i.e., left, right, center, top, or bottom relative to the location of the speaker) when representing non-present referents. These gestures are called “spatially modulated gestures” (see discussion of spatially modulated signs in Senghas & Coppola, 2001). A spatially modulated gesture, which is produced in the same location as previous spatially modulated gestures for the same referents is considered spatially consistent. We call this kind of spatially modulated gesture as spatially-consistent referent-identifying gesture in this paper.1 Locations of spatially-consistent referent-identifying gestures are often maintained throughout the discourse (Gullberg, 1998, Gullberg, 2006). That is, both speaker and listener(s) produce gestures at the same locations when the same referents are mentioned later on in the discourse. Below is an example of spatially modulated and spatially-consistent referent-identifying gestures. Imagine a speaker is telling a story about a cat. He might say:

There is a cat [right hand points to the left] and there is a tower in front of it [right hand points to the center]. The cat flies to the tower [right hand moves from the left to the center]. However, the tower later collapses [right hand flips outward at the right] because the cat is heavy.

In this example, the speaker initially assigned the left-hand space to the cat and center to the tower. These two gestures are spatially modulated. The third spatially modulated gesture, MOVE, is a spatially-consistent referent-identifying gesture because its direction of movement is consistent with the previously established spatial locations of the cat and tower. The fourth spatially modulated gesture, COLLAPSE, is not a spatially-consistent referent-identifying gesture as its location is not consistent with the initial spatial set-up of the tower. Establishing a cohesive use of gesture space is crucial for communicating ideas during a conversation. Listeners can derive the meaning of spatially-consistent referent-identifying gestures by integrating them with the accompanying speech and forming an association between the non-present referents and their corresponding spatial locations. In this way, speakers and listeners can build a conversational common ground.

Previous research has found that adult speakers use gestures to identify referents and maintain spatial locations throughout discourse. Typically developing (TD) children start locating referents in abstract space as early as seven to eight years old (McNeill, 1992) and use them frequently from 10 or 11 years old (Cassell and McNeill, 1991, Sekine and Furuyama, 2010). The question of interest is whether children with autism spectrum disorders (ASD) can produce spatially-consistent referent-identifying gestures in discourse. Previous research has shown that children with ASD have slow understanding and development of gestures (Mastrogiuseppe, Capirci, Cuva, & Venuti, 2015; Medeiros & Winsler, 2014; Stone, Ousley, Yoder, Hogan, & Hepburn, 1997). It is a robust finding that children with ASD are particularly impaired in “proto-declarative” pointing gestures, a type of gesture that elicits joint attention (e.g., a child points to a toy car in order to direct his father’s attention to it; Baron-Cohen, 1989; Carpenter, Pennington, & Rogers, 2002; Wetherby & Prizant, 2002). However, their ability to generate proto-imperative pointing—a type of gesture used in making requests—is relatively sparse (e.g., a child points to a cookie to ask his mother to give him one; Baron-Cohen, 1989, Wetherby and Prizant, 2002). Studies have also found that children with ASD have difficulties in producing markers, which carry culturally specific meaning (e.g., the raised thumb for hitchhiking; So, Wong et al., 2015); iconic gestures, which depict actions or attributes associated with objects (e.g., index and middle fingers wriggling for walking); and speech beats, which do not involve semantic meanings but follow the rhythm of co-occurring speech (e.g., index finger flipping outward) (Charman, Drew, Baird, & Baird, 2003; Luyster, Lopez, & Lord, 2007; Wetherby et al., 2004).

However, Capps, Kehres, and Sigman (1998) found that children with ASD were as likely as children with developmental delays to enact activities with gestures in the course of a 6-minute conversation about schools, friends, and vacations. Yet their study did not have a neuro-typical comparison group. In addition, the language age of children with ASD was well below their mental and chronological ages. Therefore, it was not clear whether children with ASD in their study did not have a delay in gesture production. On the contrary, other studies have not found a delay in gesture production in children with ASD. For example, Attwood, Frith, and Hermelin (1988) reported that children with ASD, children with Down’s syndrome, and TD children produced comparable numbers of markers when interacting with other children. In an experiment, So, So, Lui et al., 2015 did not find significant differences between children with ASD and TD children in the numbers of iconic gestures and abstract deictic gestures (gestures that point to abstract locations associated with entities) before and after controlling for the amount of speech. So and colleagues reported similar findings in structured interactions between children and their caregivers: children with ASD produced similar numbers of iconic gestures, speech beats, and deictic gestures (except markers) as TD children (So, Wong et al., 2015).

While the production of some types of gesture may not be delayed in children with ASD, a recent study by So, Lui et al. (2015) found that these children may make less use of gestures to communicate about non-present objects in a modified version of the demonstration task in ADOS-2 (Lord, Rutter, Risi, Gotham, & Bishop, 2012). In that study, an experimenter working with six- to 12-year-old children with ASD and age- and IQ-matched TD children orally described daily-life activities that involved non-present objects and assigned those objects to specific spatial locations with gestures. For example, in one of the trials, the experimenter said with gestures:

Pretend that this is a dining Table and you are going to have your lunch. Pretend that you put your chopsticks here [index finger drawing a rectangle on the right side of the table], your bowl of food here [index finger drawing a rectangle in the center of the table], your spoon here [index finger drawing a rectangle on the left side of the table], and your napkin here [index finger drawing a rectangle at the bottom center of the table]. Now you demonstrate to me how you eat.

Both groups of children were expected to follow the locations assigned by the experimenter (e.g., both hands moved up at the center of Table to pretend to pick up a bowl and moved up at bottom center to pretend to pick up a napkin). By doing so, children could establish a cohesive use of gesture space for referential identification in discourse. However, So and colleagues’ findings showed that children with ASD gestured at the specified locations less often than their TD peers. Instead, they gestured in their frontal space and indicated referents in wrong locations (e.g., both hands moved up at the bottom center of Table to pretend to pick up a bowl and moved up at the center as to pretend to pick up a napkin). These findings suggest that children with ASD may have difficulties in producing spatially-consistent referent-identifying gestures that represent particular non-present objects.

However, there are two questions that remain unaddressed. First, it is not clear whether there are individual variations in producing spatially-consistent referent-identifying gestures among children with AS So, Wong et al. (2015) reported that verbal memory, spatial memory, and eductive ability (as measured by Raven’s Colored Progressive Matrices (RCPM; Raven, 1936) were positively correlated with the ability to produce spatially-consistent referent-identifying gestures in children with ASD. However, it is not certain whether production of spatially-consistent referent-identifying gesture is correlated to severity of autism symptoms, especially impairments in communication and social interactions. Indeed, some individuals with ASD are more responsive to shared knowledge than others (de Marchena & Eigsti, 2016; Nadig, Vivanti, & Ozonoff, 2009; Perner & Leekam, 1986). Thus, it is possible that children who have less severe impairments in communication and social interactions would be more able to produce spatially-consistent referent-identifying gestures than those who have more severe impairments.

The first objective of the present study was to explore whether the severity of autism symptoms might influence the use of spatially-consistent referent-identifying gestures among children with ASD. Previous research has shown that scores on the socialization domain of the Vineland Adaptive Behavior Scales and the severity of the communication symptom on the Autism Diagnostic Observation Schedule were negatively related to the ability of individuals with ASD to represent others’ mental states (Tager-Flusberg, 2003). Therefore, we hypothesize that children with ASD, who have more severe symptoms in the social interaction and communication domains than their TD peers, would have trouble assuming other people’s perspectives and maintaining spatial locations, thus resulting in difficulties producing spatially-consistent referent-identifying gestures. Similar to So, Wong et al. (2015), we also investigated the correlation between verbal and spatial memory and production of spatially-consistent referent-identifying gestures.

Second, So and colleagues’ work leaves open the question of whether children with ASD show deficits in producing spatially-consistent referent-identifying gestures in other circumstances. One of the possible reasons why the children with ASD in So, Lui et al. (2015) study produced fewer spatially-consistent referent-identifying gestures is that such children have difficulty in taking on the experimenter’s perspective, which prevents them from understanding his/her intention in assigning specific locations for particular non-present objects and then establishing a conversational background with him/her. Previous research has shown that it is challenging for individuals with ASD to take on the perspectives of listeners and tailor their speech to listeners’ needs. For example, children with ASD tend to give their listeners more information than is necessary (Volden, Magill-Evans, Goulden, & Clarke, 2007). They are also less likely than their TD peers to modify information in their speech based on shared common ground when explaining to others how a toy works (Perner & Leekam, 1986). Similarly, school-age children with ASD have difficulty incorporating listeners’ needs into utterance planning when specifying an object from a visual array (Nadig et al., 2009). A recent study by de Marchena and Eigsti (2016) reported that TD adolescents provide shorter narrations when their interlocutors have shared knowledge, but adolescents with ASD did not show this pattern. The children with ASD in So, Lui et al. (2015) study may thus have had problems taking on the experimenters’ perspectives and following established spatial locations in their gestures. Perspective-taking is a crucial skill for social interactions. Deficits in perspective-taking result in difficulties representing other people’s mental states. Thus, children would fail to understand that spatial locations, which represent particular non-present objects, were previously assigned by the experimenter and should be maintained in order to establish a coherent discourse. AS a result, such children gesture at different spatial locations from those previously established.

Then would children with ASD be able to produce spatially-consistent referent-identifying gestures if they were not required to take others’ perspectives and to establish common ground with them? The second objective of the present study was to address this question. In addition to administering a demonstration task as in So and colleagues’ study, we asked children with ASD to orally narrate a story (i.e., a spoken narrative task). Children were encouraged to generate their own stories. Thus, unlike in the demonstration task, in the spoken narrative task children did not need to follow the locations established by experimenters; rather they could create the referents and establish their locations. We examined whether children with ASD are as equally able as TD children to represent non-present story characters in consistent gesture locations. We predict that children with ASD are equally able to produce spatially-consistent referent-identifying gestures when telling a story, but that they have difficulty producing this kind of gesture in the demonstration task, after controlling for the language and cognitive skills of both groups of children.

Section snippets

Participants

Twenty-eight Cantonese-speaking children aged six to 12 participated in this study. Fourteen (all males) had been diagnosed with an ASD or autistic disorder by psychiatrists or pediatricians based on the DSM-IV (American Psychiatric Association, 2000) and the other 14 (six males) were age- and IQ-matched TD children. Table 1 shows the descriptive statistics for chronological age, ADOS™-2, and intelligence score in ASD and TD children. There was no significant mean age difference between ASD and

Results

This study examines whether the severity of autism symptoms (especially social interaction and communication impairments) and cognitive abilities are associated with the production of spatially-consistent referent-identifying gestures in the demonstration task. It also examines whether children with ASD are as capable as their TD peers of producing spatially-consistent referent-identifying gestures in the spoken narrative task, in which they do not need to take the perspectives of

Discussion

To summarize, we found that children with ASD produced spatially-consistent referent-identifying gestures less often and scored lower than TD children in the demonstration task, thus replicating the results of So, Lui et al. (2015). In addition, communication and social interaction skills were negatively correlated to the production of spatially-consistent referent-identifying gestures in the demonstration task, suggesting that the ability to maintain consistent spatial locations established by

Acknowledgements

This research has been fully supported by grants from the Chinese University of Hong Kong (Project no. CUHK4930017; CUHK4058023). We acknowledge the help of our research assistants Rachel Au, Sheera Chan, Maggie Chan, Hiu-Man Lavender Chiu, and Aretha Chui. Special thanks go to all of the children and their parents for their help and dedication to education.

References (69)

  • M.A. Bono et al.

    Relations among joint attention: amount of intervention and language gain in autism

    Journal of Autism and Developmental Disorders

    (2004)
  • J. Boucher et al.

    Memory in autistic spectrum disorder

    Psychological Bulletin

    (2012)
  • D.M. Bowler et al.

    Binding of multiple features in memory by high-functioning adults with autism spectrum disorder

    Journal of Autism and Developmental Disorders

    (2014)
  • D.M. Bowler et al.

    Brief Report: the role of task support in the spatial and temporal source memory of adults with Autism Spectrum Disorder

    Journal of Autism and Developmental Disorders

    (2015)
  • L. Camaioni et al.

    A longitudinal examination of the communicative gestures deficit in young children with autism

    Journal of Autism and Developmental Disorders

    (1997)
  • L. Capps et al.

    Conversational abilities among children with autism and children with developmental delays

    Autism

    (1998)
  • M. Carpenter et al.

    Interrelations among social-cognitive skills in young children with autism

    Journal of Autism and Developmental Disorders

    (2002)
  • J. Cassell

    A framework for gesture generation and interpretation

  • J. Cassell et al.

    Non-verbal imagery and the poetics of prose?

    Poetics Today

    (1991)
  • A.S. Chan et al.

    Music training improves verbal memory

    Nature

    (1998)
  • A.S. Chan et al.

    Neuropsychological sequelae of patients treated with microsurgical clipping or endovascular embolization for anterior communicating artery aneurysm

    European Neurology

    (2002)
  • T. Charman et al.

    Brief report: prompted pretend play in Autism

    Journal of Autism and Developmental Disorders

    (1997)
  • T. Charman et al.

    Measuring early language development in preschool children with autism spectrum disorder using the MacArthur Communicative Development Inventory (Infant Form) ?

    Journal of Child Language

    (2003)
  • M.-C. Cheung et al.

    Cognitive function of patients with nasopharyngeal carcinoma with and without temporal lobe radionecrosis

    Archives of Neurology

    (2000)
  • A. Farrant et al.

    Source monitoring by children with autism

    Journal of Autism and Developmental Disorders

    (1998)
  • P. Garber et al.

    Gesture offers insight into Problem-solving in adults and children

    Cognitive Science

    (2006)
  • J. Gilliam

    GARS-2: gilliam autism rating scale

    (2006)
  • S. Goldin-Meadow et al.

    Explaining math: gesturing lightens the load?

    Psychological Science

    (2001)
  • M. Gullberg

    Gesture as a communication strategy in second language discourse: a study of learners of French and Swedish

    (1998)
  • M. Gullberg

    Gestures, referents, and anaphoric linkage in learner varieties

  • M. Gullberg

    Handling discourse: gestures, reference tracking, and communication strategies in early L2

    Language Learning

    (2006)
  • T.C. Gunter et al.

    Inconsistent use of gesture space during abstract pointing impairs language comprehension

    Frontiers in Psychology

    (2015)
  • Y.C. Ho et al.

    Music training improves verbal but not visual memory: cross-sectional and longitudinal explorations in children?

    Neuropsychology

    (2003)
  • C. Jarrold

    A review of research into pretend play in autism

    Autism

    (2003)
  • View full text