Goldin-Meadow & Brentari (G-M&B) emphasize technology's utility in revealing common features and differences in sign language versus co-speech gesture. We propose a complementary approach: the study of gesture use in individuals with language and communication deficits, particularly autism spectrum disorder (ASD). In most populations exhibiting language impairment, affected individuals compensate for speech difficulties via increases in gesture; this is observed in expressive language delay (Thal et al. Reference Thal, Tobias and Morrison1991), Down syndrome (Stefanini et al. Reference Stefanini, Caselli and Volterra2007), Williams syndrome (Bello et al. Reference Bello, Capirci and Volterra2004), and Specific Language Impairment (Mainela-Arnold et al. Reference Mainela-Arnold, Evans and Alibali2006).

In sharp contrast, individuals with ASD have broad deficits in social-communication skills including salient and striking differences in gesture production. Deficits in nonverbal communication (including, but not limited to, gesture) are required for a diagnosis of ASD (American Psychiatric Association 2013), and are codified on gold-standard ASD diagnostic measures and screeners such as the Autism Diagnostic Observation Schedule (Lord et al. Reference Lord, Rutter, DiLavore, Risi, Gotham and Bishop2012), the Autism Diagnostic Interview (Lord et al. Reference Lord, Rutter and LeCouteur1994), and the Modified Checklist for Autism in Toddlers (Robins et al. Reference Robins, Casagrande, Barton, Chen, Dumont-Mathieu and Fein2014); on these measures, the absence or infrequency of gesture is rated as symptomatic.

Gesture impairments have been part and parcel of ASD symptomatology since the very earliest descriptions of the disorder (Asperger Reference Asperger1944; Wing Reference Wing1981). Gesture delays, most notably reductions in declarative deictic gestures, have consistently been reported in studies of toddlers with ASD (Luyster et al. Reference Luyster, Lopez and Lord2007; Mundy et al. Reference Mundy, Sigman, Ungerer and Sherman1987; Mundy & Stella Reference Mundy, Stella, Wetherby and Prizant2000) and in high-risk infant siblings of children with ASD (Iverson & Kuhl Reference Iverson and Kuhl1995), who are statistically more likely to themselves develop ASD.

A number of studies have failed to find differences in gesture frequency in older individuals with ASD compared to control groups (Attwood et al. Reference Attwood, Frith and Hermelin1988; Capps et al. Reference Capps, Kehres and Sigman1998; de Marchena & Eigsti Reference de Marchena and Eigsti2010; Garcia-Perez et al. Reference Garcia-Perez, Lee and Hobson2007; Morett et al. Reference Morett, O'Hearn, Luna and Ghuman2016), with some exceptions as noted below (de Marchena & Eigsti, in press). Research on ASD thus offers an opportunity to investigate the relationship between gesture and language production. As discussed by G-M&B, studies of gesture timing suggest systematicity in the amount of time between the onset of a gesture stroke and of the relevant word in speech (Nobe Reference Nobe and McNeill2000).

Narrative data from teens with ASD, however, suggest that even when utterance and gesture rates are comparable to those of typically developing peers, and even when language and gesture share semantics, the stroke phase of iconic gestures is asynchronous, such that gestures both precede and follow the relevant speech (de Marchena & Eigsti Reference de Marchena and Eigsti2010). Furthermore, naïve raters are sensitive to the degree of speech-gesture asynchrony: The more asynchronous the gestures, the more “difficult to follow” they rate videorecorded narratives, across diagnostic groups. These findings demonstrate not only that gesture and speech are intricately coordinated, but also their coordination and timing matters for communicative quality.

Currently, our group is evaluating the ability of teens with ASD to deliberately synchronize their gestures with highly rhythmic speech (nursery rhymes), in order to discover whether the asynchrony reflects a broad motor difficulty with smooth coordination across multiple motor systems (Bhat et al. Reference Bhat, Galloway and Landa2012), or instead, resides with the communicative function of gesture. The mechanism relating gesture impairments to broader communication skills (and deficits) remains undefined; evidence that gesture-vocalization coordination develops slowly in ASD (Parlade & Iverson Reference Parlade and Iverson2015) suggests that, regardless of mechanism, synchrony differences are apparent early in life and are part of a larger suite of communication delays. The work described here, and by G-M&B, suggests that speech-gesture synchrony is central to communication development more broadly (e.g., Özyürek et al. Reference Özyürek, Willems, Kita and Hagoort2007).

In addition to illuminating subtle qualitative aspects of gesture, research on gestures in ASD can highlight the aspects of gesture that are integral to language production within a syntactically well-formed system versus the aspects that arise as part of social interaction – similar to the distinction between sign language and gestures as addressed in the target article. Children with ASD may have a reduced gestural repertoire (Capps et al. Reference Capps, Kehres and Sigman1998; Wetherby & Prutting Reference Wetherby and Prutting1984), such that they produce a similar number of gestures but do not use their gestures to fulfill as many communicative functions, even at 9–12 months (Colgan et al. Reference Colgan, Lanter, McComish, Watson, Crais and Baranek2006). Compared to gestures serving a purely sociocommunicative function, individuals with ASD produce relatively more gestures to serve cognitive functions, such as self-monitoring (de Marchena & Eigsti Reference de Marchena and Eigsti2014)

A small body of research characterizes the errors in sign language produced by fluent speakers of American Sign Language who also have ASD. Despite the iconic transparency of signed pronouns, which are points to oneself or others, signing children with ASD tend to avoid producing pronouns, instead spelling out names or using name signs, even to refer to themselves (Shield et al. Reference Shield, Meier and Tager-Flusberg2015). This research has the potential to reveal the role of iconicity and self-representation in sign language and delays in development.

Gestures help speakers develop better internal representations of what they are attempting to communicate (Kita Reference Kita and McNeill2000). Accordingly, gestural differences in ASD may illuminate both language acquisition in general, and fundamental aspects of cognition and representation in that population.