Skip to main content
NCBI home page
Autism & Developmental Language Impairments logoLink to Autism & Developmental Language Impairments
. 2026 Sep 14;11:23969415261485091. doi: 10.1177/23969415261485091

Multimodal Communication in Autism Is Reorganized, Not Reduced: Evidence From Speech and Gesture Across Four Languages

Armita Ghobadi 1,✉, Pauline Wolfer 2, Stephanie Durrleman 2
PMCID: PMC13583284  PMID: 42756667

Abstract

Speech and gesture form an integrated communicative system, yet how this system operates in autism, and whether it varies across languages, remains unclear. We examined narrative production in 88 children (44 autistic, 44 neurotypical [NT]), matched on age, socioeconomic status, and nonverbal IQ, acquiring English, French, German, or Spanish, using a structured storytelling task (Autism Diagnostic Observation Schedule, ADOS-2 Monkey Story). Speech and gesture were analyzed along parallel dimensions of amount, diversity, and complexity. Autistic children produced comparable speech amount and lexical diversity to NT peers but differed in complexity. Critically, autistic children produced gestures more frequently, with greater diversity and complexity, including gestures conveying additional semantic information. This pattern indicates that gesture contributes meaningful information alongside speech. Cross-linguistic analyses revealed limited variation in speech but robust language-related differences in gesture across both groups. These findings are consistent with a reorganization of multimodal communication in autism rather than a reduction, highlighting gesture as an integral component of communicative expression across languages.

Keywords: Autism, autistic children, multimodal communication, gesture, narrative production, cross-linguistic development

Introduction

Speech and gesture form an integrated communicative system in which meaning is distributed across modalities (Colonnesi et al., 2010; Goldin-Meadow, 2005; Hostetter, 2011; Kita, 2009). From early childhood, children produce gestures alongside speech, using gesture both to reinforce spoken content and to convey information not expressed in speech alone. Together, speech and gesture support reference, event representation, and message organization, making both modalities essential for understanding communicative competence.

How this multimodal system operates in autism 1 remains incompletely understood. Early differences in gesture and language have been documented in autistic infants and young children (Özçalışkan et al., 2016, 2017; Ramos-Cabo et al., 2019; Shumway & Wetherby, 2009). However, findings in older children are mixed (see McKern et al., 2023 for a full review). Some studies report differences in gesture frequency, diversity, type or coordination with speech (Bartak et al., 1975; McKern et al., 2023; Özçalışkan et al., 2016; Shumway & Wetherby, 2009), whereas others report comparable levels of multimodal production when autistic and neurotypical (NT) children are evaluated under similar task demands (Attwood et al., 1988; de Marchena & Eigsti, 2010; Tantam et al., 1993). These mixed findings leave unresolved whether multimodal communication in autism differs primarily in overall production or in how communicative meaning is distributed across speech and gesture.

Language context represents an additional limitation. Most studies of multimodal communication in autism have examined single-language samples (de Marchena & Eigsti, 2010; Özçalışkan et al., 2016; Shumway & Wetherby, 2009), despite evidence that linguistic structure systematically shapes gesture production in NT speakers (Kendon, 1992; Kita, 2009; McNeill, 1992). As a result, it remains unclear whether multimodal communication in autism reflects language-specific patterns or broader, cross-linguistic organizational principles.

Examining speech and gesture together provides a more complete understanding of communicative expression in autism. Gesture can contribute meaningful information alongside speech and may support communicative clarity, particularly in contexts where spoken language alone may not fully convey intended meaning (Hostetter & Alibali, 2008; Kita & Özyürek, 2003). Understanding how autistic children use gesture in relation to speech is therefore essential for accurately characterizing their communicative abilities.

To address these gaps, the present study examines speech and gesture production in autistic and NT children aged 4 to 10 years across four languages: English, French, Spanish, and German. Using a structured narrative task and a shared analytic framework, we analyze speech and gesture across parallel dimensions of amount, diversity, and complexity. This approach allows direct comparison across modalities, diagnostic groups, and languages, providing new insight into how autistic children express and organize communicative meaning across speech and gesture.

Speech–Gesture Integration: Theoretical Frameworks

Speech and gesture form an integrated communicative system in which meaning is distributed across modalities (Goldin-Meadow & Alibali, 2013; Kita, 2009). Converging theoretical frameworks propose that gesture contributes meaningful information alongside speech, either by expressing complementary aspects of a shared communicative representation or by supporting the organization and retrieval of linguistic information during communication (Hostetter & Alibali, 2008; Kita & Özyürek, 2003; McNeill, 1992, 2005).

This multimodal perspective is particularly relevant for understanding communication in autism. Because gesture may convey information not fully expressed in speech, evaluating speech alone may underestimate autistic children's communicative abilities, whereas examining speech and gesture together provides a more complete characterization of how meaning is expressed and organized. Evaluating speech–gesture integration can therefore clarify whether differences in multimodal communication reflect reduced production or differences in how information is distributed across modalities, which is essential for accurately characterizing communicative competence and informing assessment and intervention.

Speech Production in Neurotypical and Autistic Development

Speech production reflects children's ability to organize and express communicative meaning through language (Bates et al., 1994; Berman & Slobin, 1994; Brown, 1973). Across development, children show systematic increases in speech amount, lexical diversity, and structural complexity. Those dimensions capture how effectively communicative information is expressed and organized (Huttenlocher et al., 1991; Miller & Chapman, 1981; Rice et al., 2010). Autistic children show substantial variability along these same dimensions, reflecting differences in how linguistic and communicative processes support meaning expression across contexts (Bacon et al., 2014; Kover et al., 2014; Shumway & Wetherby, 2009; Tager-Flusberg & Anderson, 1991). These dimensions provide a useful framework for directly comparing communicative expression in autistic and NT children.

Amount of Speech Production

Children show reliable increases in speech across development, reflected in greater word production and extended discourse (Brown, 1973; Huttenlocher et al., 1991). Speech amount is shaped by linguistic and cognitive capacities, including receptive language and working memory, and increases further in narrative contexts that place greater demands on reference tracking and event organization (Adams & Gathercole, 2000; Berman & Slobin, 1994; Peterson & McCabe, 1987). Speech output in autistic children varies similarly as a function of communicative context, with structured narrative tasks often supporting increased verbal production (Evans & Craig, 1992; Kover et al., 2014). Within comparable contexts, verbally fluent autistic and NT children often produce similar amounts of speech, indicating that speech amount reflects communicative demands and language ability rather than diagnostic status alone (Loveland et al., 1990; Norbury & Bishop, 2003).

Diversity of Speech Production

Lexical diversity increases as children expand their productive vocabulary and their ability to express a wider range of referents, actions, and relationships (Berman & Slobin, 1994; Hoff, 2006). Narrative production places particularly strong demands on lexical diversity because children must describe multiple characters and events across extended discourse (Westerveld & Moran, 2011). In autistic children, lexical diversity reflects variability in how referential and semantic information is expressed. Some studies report differences in referential expression between autistic and NT children during narrative production (Norbury & Bishop, 2003), whereas others report comparable lexical diversity between groups when language ability is controlled (Bang et al., 2013; Eigsti et al., 2007). These findings indicate that lexical diversity reflects individual language profiles rather than uniform group differences.

Complexity of Speech Production

Speech becomes increasingly complex as children acquire morphosyntactic structures that allow them to express relations among agents, actions, and events (Brown, 1973; Rice et al., 2010). Narrative contexts reliably elicit more complex speech than less structured interactions, reflecting increased communicative and organizational demands (Berman & Slobin, 1994; Hickmann, 2003). Speech complexity in autistic children is influenced by similar contextual and task demands, with narrative tasks eliciting more elaborated utterances than conversational contexts (Kover et al., 2014). Some studies report comparable utterance length between autistic and NT children when matched on language ability (Diehl et al., 2006), whereas others report group differences in morphosyntactic structure (Eigsti et al., 2007). This variability indicates that complexity findings depend on how closely groups are matched on underlying language ability.

Gesture Production in Neurotypical and Autistic Development

Gesture emerges early in development and plays a central role in communicative expression, often preceding and supporting spoken language (Bates, 1976; Bates et al., 1979; Rowe & Goldin-Meadow, 2009). Early gesture use predicts later language outcomes, highlighting its importance as part of the developing communicative system (Rowe et al., 2009). Autistic children show diverse developmental patterns in gesture production, reflecting variability in how gesture contributes to communicative expression across contexts (Adamson et al., 2009; Özçalışkan et al., 2016; Ramos-Cabo et al., 2019). Establishing how gesture develops, and how autistic children use gesture, provides a foundation for directly comparing its communicative role across groups.

Amount of Gesture Production

Gesture frequency increases across development and is shaped by linguistic growth and communicative demands (Colonnesi et al., 2010; Nicoladis & Jiang, 2018; Rowe & Goldin-Meadow, 2009). Narrative contexts reliably elicit higher gesture rates because they require organizing events and maintaining reference across discourse (Nicoladis & Jiang, 2018). Gesture frequency in autistic children varies across developmental stage and communicative context. Studies of young autistic children report differences in gesture frequency and use of specific gesture types relative to NT peers (Bartak et al., 1975; McKern et al., 2023; Özçalışkan et al., 2016; Shumway & Wetherby, 2009), whereas studies of older autistic children often report gesture rates comparable to NT peers in structured and conversational contexts (Attwood et al., 1988; de Marchena & Eigsti, 2010; Tantam et al., 1993). These findings indicate that gesture production reflects communicative demands and developmental profile rather than a uniform reduction in autism.

Diversity of Gesture Production

Gesture diversity refers to the range of meanings conveyed through gesture, including different referents and actions (Rowe & Goldin-Meadow, 2009). Children who express a broader range of meanings in gesture show stronger language development, highlighting the role of gesture in expanding communicative expression (Iverson & Goldin-Meadow, 2005; Rowe & Goldin-Meadow, 2009). Research on gesture diversity in autistic children remains limited, as most studies have focused on gesture frequency or broad gesture categories rather than the range of meanings conveyed. As a result, it remains unclear how autistic and NT children differ in the diversity of meanings expressed through gesture, particularly in narrative contexts.

Complexity of Gesture Production

Gesture complexity reflects how gesture contributes to meaning alongside speech. Children produce gesture–speech combinations in which gesture reinforces or supplements spoken information, sometimes conveying meanings not fully expressed in speech (Butcher & Goldin-Meadow, 2000; Özçalışkan & Goldin-Meadow, 2005a, 2005b). With development, gesture–speech coordination becomes more integrated, particularly in narrative contexts where gesture supports reference tracking and event representation (Colletta et al., 2015; Dimitrova & Özçalışkan, 2022). Gesture complexity in autistic children has primarily been examined through this same gesture–speech coordination. Autistic children produce supplementary gestures that convey information not expressed in speech, and these combinations are associated with language development (Özçalışkan et al., 2016), while other studies report gestures conveying meanings distinct from or independent of speech (de Marchena & Eigsti, 2010; So et al., 2013). These findings indicate that gesture contributes meaningful semantic information in both groups, though direct comparisons of complexity between autistic and NT children remain limited.

Cross-Linguistic Differences in Neurotypical and Autistic Speech and Gesture Production

Research shows that speech and gesture production vary across languages, even in comparable narrative tasks (, Berman & Slobin, 1994; Colletta et al., 2015; Hickmann, 2003; Nicoladis et al., 2009). Cross-linguistic studies demonstrate systematic differences in speech production, including variation in event packaging, clause structure, and reference management (Berman & Slobin, 1994; Hickmann, 2003). However, few studies have applied standardized quantitative measures, such as word token counts, lexical diversity, or morphosyntactic complexity, within a shared analytic framework. Gesture production also varies across languages. Some studies report cross-linguistic differences in the composition of children's gesture repertoires, such as Italian children relying more heavily on representational gestures while American children relied predominantly on deictic gestures (Iverson et al., 2008), and French, English, and Italian children differing in rates of cospeech gesture production (Colletta et al., 2015), whereas monolingual French- and English-speaking children have been found to gesture at comparable rates when telling narratives (Nicoladis et al., 2009).

Although speech and gesture have been examined in autistic children, most studies have focused on single-language samples, typically English, and have not directly tested cross-linguistic patterns of multimodal communication (de Marchena & Eigsti, 2010; Özçalışkan et al., 2016a; Shumway & Wetherby, 2009). Some studies include other languages, such as French, Mandarin, Hebrew, or Italian, but these investigations have been conducted independently and lack the shared tasks and analytic frameworks needed for direct comparison (Colletta et al., 2015; Losh & Capps, 2003; Norbury & Bishop, 2003). As a result, it remains unclear whether multimodal communication in autism varies systematically across languages, or whether the language-related patterns established in NT development extend to autistic populations.

Together, these findings indicate that speech and gesture reflect both universal communicative processes and language-specific organization. Establishing cross-linguistic patterns using shared methods is essential for distinguishing diagnosis-related differences from language-related variation and has direct relevance for improving communication assessment and interpretation in multilingual clinical and educational settings.

Current Study

The present study addresses these gaps by examining speech and gesture production in autistic and NT children across four languages: English, French, Spanish, and German. Using a structured narrative task and a shared analytic framework, we evaluate speech and gesture across parallel dimensions of amount, diversity, and complexity. This approach allows direct comparison across diagnostic groups and languages and provides a comprehensive characterization of how communicative meaning is expressed across modalities.

The study addresses three questions: first, is diagnostic group associated with the amount, diversity, and complexity of speech and gesture? Second, do these patterns remain consistent across languages, or do they vary as a function of cross-linguistic context? Third, how are speech and gesture production patterns associated within individuals, and does this association differ across diagnostic groups and languages?

By addressing these questions, the present study provides new insight into how multimodal communication is organized in autism and contributes to a more complete understanding of autistic children's communicative profiles across languages.

Method

Participants

The sample included 44 children 2 with ASD (Mage = 102 months, SD = 27; seven females) and 44 NT children (Mage = 96 months, SD = 29; 23 females) acquiring English, French, Spanish, or German (see Supplemental Tables S1 and S2 3 ). Groups were comparable in age, socioeconomic status (SES), and visuospatial working memory, as measured by an adaptation of the Corsi Block Tapping Task. 4 Groups differed in gender distribution 5 and receptive language abilities. The groups also differed in receptive grammar (TROG) and receptive vocabulary (PPVT), reflecting differences observed in the range and distribution of linguistic abilities in autistic compared to NT populations, while recognizing the substantial heterogeneity and variability in language profiles across autistic individuals (Silleresi, 2023). 6

Participants’ Recruitment, Eligibility, and Inclusion Criteria

Participants were recruited through flyers, clinician and hospital outreach, and websites serving autistic communities. Families received monetary compensation 7 and provided informed consent before participation. Eligibility 8 was determined using caregiver questionnaires. Children were eligible if they were between 3 and 12 years of age, demonstrated comprehension in one study language (English, French, Spanish, or German), and had typical or corrected-to-typical hearing and vision. NT children had no suspicion or diagnosis of ASD or other developmental conditions. No child in either group had reported visual or auditory impairments. All children in the ASD group had a formal clinical diagnosis of autism spectrum disorder based on DSM-5 or earlier DSM criteria (American Psychiatric Association, 2022) or International Classification of Diseases, 11th Revision criteria (World Health Organization, 2024), established before the study by a certified clinician using gold-standard diagnostic instruments, including the Autism Diagnostic Observation Schedule-2 (ADOS-2; Lord et al., 2012) or the autism diagnostic interview-revised (ADI-R; Lord et al., 1994). Caregivers confirmed that children could follow task instructions and remain engaged.

Procedure for Data Collection

The study was approved by the institutional review board. 9 Children participated in an individual session conducted in a quiet room in their target language by a native or highly proficient speaker. Caregivers provided written informed consent and completed questionnaires on language exposure, developmental history, and demographics. Children completed standardized assessments of receptive grammar, receptive vocabulary, and visuospatial working memory. Receptive grammar was assessed using TROG-2 (Bishop, 2003) or validated French, German, and Spanish adaptations (Supplemental Appendix S1). Receptive vocabulary was assessed using PPVT (Dunn et al., 2003) or validated language-specific adaptations. Visuospatial working memory was assessed using a matrix-based task adapted from the Corsi Block-Tapping paradigm (Morales et al., 2013, as implemented in Baumeister et al., 2026). Raw scores were converted to standardized Z-scores using test-specific norms to ensure cross-language comparability. Narrative production was elicited using a six-picture sequence adapted from ADOS-2 materials (Lord et al., 2012), depicting a story involving two monkeys and a stolen coconut (Figure 1). Children were asked to describe the events without gesture instructions. Standardized prompts were used when needed and occurred more frequently in the autistic group (β = 1.14, z = 2.25, p = .02). All sessions were video recorded for transcription and multimodal coding.

Figure 1.

Figure 1.

Open in a new tab

Example of the six-picture ADOS-2 Monkey Story narrative task. Children from different diagnostic groups and language backgrounds told the story to a native or highly proficient speaker in their test language. ADOS-2 = Autism Diagnostic Observation Schedule-2.

Transcription, Coding, and Reliability

All videos were transcribed using the CHILDES system (MacWhinney, 2000). Speech was segmented into utterances following CHILDES guidelines and coded for amount (total word tokens), diversity (unique word types), and complexity using mean length of utterance in words (MLU; Ghobadi & Özcaliskan, 2026, Ghobadi & Durrleman, 2026; Pınar et al., 2021). Repetitions, false starts, and nonverbal vocalizations were excluded. Gestures were coded independently as communicative hand or body movements (Özçalışkan & Goldin-Meadow, 2005a) and classified as deictic, iconic, emblematic, or beat (McNeill, 1992). Gesture production was coded for amount (total gestures), diversity (distinct referents), and complexity based on the gesture's informational relation to speech 10 (Ghobadi & Özcaliskan, 2026; Ghobadi & Durrleman, 2026; Özçalışkan & Goldin-Meadow, 2005a).

Reliability

Gesture coding reliability was assessed by an independent trained coder blind to study hypotheses, who coded a randomly selected 30% of the recordings. Intercoder agreement was high across all dimensions: 98% (κ = 0.97) for gesture identification, 97% (κ = 0.96) for gesture type, 98% (κ = 0.97) for gesture referents, and 98% (κ = 0.97) for gesture complexity (i.e., whether gestures reinforced or added information relative to speech).

Statistical Analysis

Statistical analyses were conducted separately for speech and gesture production using regression models appropriate to the distributional structure of each outcome, allowing examination of diagnostic group, language, and their interaction. All analyses were performed in Python (Python Software Foundation, 2023).

For speech production, three separate linear mixed-effects models were fitted corresponding to speech amount (i.e., total number of speech tokens), speech diversity (i.e., lexical diversity), and speech complexity (i.e., MLU). For gesture production, separate linear mixed-effects models were fitted for gesture amount and gesture diversity. Model assumptions were evaluated through inspection of outcome distributions, residual distributions, Q–Q plots, and fitted-versus-residual plots. Because gesture complexity was bounded, zero-inflated, and did not meet the normality assumption required for a Gaussian model, it was analyzed using a binomial mixed-effects regression instead 11 . Gesture complexity was operationalized as the number of additive/complementary gestures relative to the total number of gestures produced by each child. Children who produced no gestures were excluded from this analysis because gesture complexity was undefined in the absence of gesture production. Reinforcing gestures were analyzed separately as the proportion of gestures that reinforced or emphasized accompanying speech. In all models, children with ASD were treated as the reference group, such that estimates for NT children reflect deviations relative to this baseline. For analyses examining cross-linguistic effects, English served as the reference language.

Two model specifications were used. Model 1 examined diagnostic group effects. Model 2 examined diagnostic group, language, and their interaction to test whether patterns varied across languages. 12 For the linear mixed-effects models, participant was included as a random intercept. Gesture complexity was modeled separately using a binomial mixed-effects model with a logit link, in which individual gestures were treated as additive/complementary or nonadditive and participant was included as a random intercept to account for clustering of gestures within children. Data preprocessing was conducted using Pandas (McKinney, 2010) and NumPy (Harris et al., 2020), and models were implemented using Statsmodels (Seabold & Perktold, 2010). Descriptive statistics were computed to support interpretation of model results.

Results

The Impact of Diagnosis on Children's Narrative Production

First, we asked whether diagnostic group (ASD vs. NT) is associated with children's multimodal narrative production. For speech production, diagnostic group was not significantly associated with speech amount (β = −10.23, z = −1.61, p = .11) or lexical diversity (β = −2.91, z = −1.62, p = .11), but it did have a significant effect on speech complexity (β = 0.53, z = 2.08, p = .04; see Supplemental Table S3). Descriptively, autistic children produced a higher number of speech tokens than NT children, a pattern mirrored in lexical diversity; they also produced shorter utterances on average than NT children (Supplemental Table S6).

Turning to gesture production, diagnostic group was significantly associated with gesture amount, diversity, and complexity. Specifically, children with ASD produced a greater number of gestures overall (β = −3.98, z = −11.55, p < .001), showed greater gesture diversity (β = −2.36, z = −14.16, p < .001), and produced gestures that were more likely to convey complementary 13 or additional information relative to speech than gestures produced by NT children (β = −0.87, z = −2.85, p < .01). Descriptively, children with ASD produced nearly twice as many gestures as NT children and these gestures were both more diverse and more likely to be additive/complementary (Supplemental Table S6 14 ).

To test whether this multimodal pattern reflects a within-child trade-off between speech and gesture, we examined whether children's MLU was associated with their number of additive/complementary gestures, and whether this relationship differed by diagnostic group. The MLU × diagnostic group interaction itself was not significant (β = 0.31, z = 1.13, p = .26), but the correlation between MLU and additive gesture count was negative and approached significance in the ASD group (r = −.30, p = .05), but not in the NT group (r = .05, p = .76).

Taken together, autistic children produced comparable speech amount and lexical diversity to NT but showed higher gesture amount and diversity. They also produced shorter utterances, which were supplemented by more complementary gestures (Figure 2).

Figure 2.

Figure 2.

Open in a new tab

Distribution of speech and gesture production measures by diagnostic group. The y-axis represents the frequency of each target variable, and the x-axis denotes diagnostic group. Boxplots show speech amount, speech diversity, and speech complexity (top row) and gesture amount, gesture diversity, and gesture complexity (bottom row) for ASD and NT children during narrative production. Boxes represent interquartile ranges, horizontal lines indicate medians, whiskers denote 1.5× the interquartile range, and points indicate individual outliers. Gesture complexity (bottom right) is shown as the percentage of gestures that were additive/complementary, with raw counts (n/N) reported above each bar. ASD = autistic; NT = neurotypical.

The Impact of Diagnosis Status on Children's Narrative Production Across Languages

We next conducted exploratory analyses examining whether the impact of diagnosis on children's multimodal narrative production differed across four languages (English, French, Spanish, and German); given the smaller per-cell sample sizes (n = 11), these results should be interpreted as preliminary rather than confirmatory. Beginning with speech, we did not find a significant main effect of diagnosis or language on speech amount or lexical diversity. However, there was a significant interaction between diagnosis and language for lexical diversity (β = −14.55, z = −2.11, p < .05), specifically for German relative to English (see Supplemental Table S4), indicating that the relation between diagnostic group and lexical diversity differed across languages.

Descriptively, German-speaking children with ASD produced more lexically diverse speech than their NT peers, whereas the opposite pattern was observed in English-speaking children (Supplemental Table S7). A similar pattern to English was observed in French, whereas Spanish showed a pattern similar to German (Supplemental Table S7).

For speech complexity, no significant main effects or diagnosis × language interactions were observed (all ps > .10; see Supplemental Table S4).

Turning to gesture production, a marginal main effect of diagnosis was observed across languages (β = −6.00, z = −2.11, p < .05), with children with ASD producing a greater number of gestures overall (see Supplemental Table S5). No diagnosis × language interactions were observed.

Gesture amount also showed a significant main effect of language (β = −7.36, z = −2.61, p < .001). English-speaking children produced the highest number of gesture tokens across diagnostic groups, followed by French, German, and Spanish (see Supplemental Tables S5 and S7).

Gesture diversity showed no significant effect of diagnosis but a significant main effect of language (β = −3.55, z = −2.29, p = .02). Gesture diversity was highest among English-speaking children in both diagnostic groups, followed by French, German, and Spanish (Supplemental Table S7).

Gesture complexity showed significant main effects of both diagnostic group and language. Across languages, autistic children produced a higher proportion of additive/complementary gestures than NT children (β = −1.15, z = −3.80, p < .001). Language also predicted gesture complexity: English-speaking children produced the highest proportion of complex gestures, followed by French-speaking children, while Spanish- and German-speaking children produced markedly fewer complex gestures overall, with only six complex gestures produced in total across these two language groups combined (Supplemental Table S7). Because of this sparsity, a diagnostic group × language interaction could not be reliably estimated for gesture complexity, and this interaction was not formally tested for this outcome (Figure 3).

Figure 3.

Figure 3.

Open in a new tab

Speech and gesture production by language and diagnostic group during narrative discourse. Top row: speech amount, lexical diversity, and speech complexity (MLU). Bottom row: gesture amount, gesture diversity, and gesture complexity. Boxes represent interquartile ranges, medians, and 1.5× Interquartile range (IQR) whiskers across English, French, Spanish, and German. Gesture complexity was modeled as a binomial outcome (“Statistical Analysis” section) and is shown instead as percentage of additive/complementary gestures, with raw counts annotated above each bar. ASD = autistic (light blue); MLU = mean length of utterance; NT = neurotypical (orange).

Overall, diagnostic group did not interact with language in shaping speech production or most gesture measures. For gesture complexity, the interaction could not be reliably tested due to sparse data in the Spanish and German subsamples; across the languages where it could be assessed, autistic children consistently produced a higher proportion of complex gestures than NT children. Although gesture amount, diversity, and complexity all varied by language, these differences did not alter the overall multimodal pattern distinguishing diagnostic groups.

Discussion

The Impact of Diagnosis on Children's Narrative Production

The first goal of this study was to examine whether diagnostic group relates to children's multimodal narrative production. In speech, autistic and NT children did not differ in amount or lexical diversity, indicating comparable narrative quantity and vocabulary range. However, autistic children produced shorter utterances on average (MLU), suggesting differences in how linguistic information was organized rather than how much was produced.

This pattern aligns with prior work showing variability in syntactic and discourse organization among verbally fluent autistic children, particularly in contexts requiring planning, reference tracking, and event integration (Diehl et al., 2006; Eigsti et al., 2006; Frith et al., 2003; Norbury & Bishop, 2003). Narrative production requires organizing events sequentially and maintaining reference across characters (Berman & Slobin, 1994; Hickmann, 2003), processes linked to morphosyntactic structure. Consistent with this interpretation, autistic children in our sample showed lower receptive grammar scores, a known predictor of syntactic production (Eigsti et al., 2007; Rice et al., 2006). These findings indicate that group differences emerged primarily in utterance structure rather than overall verbal output.

In gesture, autistic children produced more gestures, expressed a broader range of referential meanings, and more often conveyed complementary information relative to speech (Figure 2). Gestures clarified referents and supported event interpretation, indicating that gesture functioned as an active communicative resource. Because groups were comparable in visuospatial working memory, increased gesture production likely reflects differences in how gesture was recruited during narrative communication rather than visuospatial capacity (Cook et al., 2008; Hostetter & Alibali, 2008).

Taken together, this pattern is consistent with a reorganization of multimodal narrative communication in autism rather than a reduction. Speech conveyed comparable content and vocabulary range, while gesture contributed greater referential and semantic support (Figure 4).

Figure 4.

Figure 4.

Open in a new tab

Children with ASD produced more complex gestures that added or clarified information relative to speech, such as pointing to the second monkey while saying “this one stole the coconut,” or pointing to the sad monkey while saying “look at him!”. ASD = autistic.

Speech and Gesture in Children With ASD

These findings provide insight into how speech and gesture jointly contribute to narrative production in autistic children. Theoretical accounts differ in whether speech and gesture arise from a unified representational system (McNeill, 1992, 2005) or reflect interacting but partially distinct processes, with gesture supporting information organization and expression (Kita, 2000; Krauss et al., 2000; Kita & Özyürek, 2003; Hostetter & Alibali, 2008).

The present findings are more consistent with complementary accounts. Autistic children produced a greater proportion of gestures conveying information not fully expressed in speech, including clarifying referents, distinguishing characters, and conveying spatial or event-related information. These gestures contributed semantic content rather than mirroring speech, indicating that speech and gesture function as coordinated but partially distinct modalities. Communicative expression in autistic children cannot therefore be fully characterized through speech alone, as gesture represents an integral component of narrative meaning. These findings suggest that gesture contributes systematically to communicative expression in autism.

The Impact of Diagnosis on Children's Narrative Production Across Languages

We found no overall effects of diagnosis or diagnosis × language interactions for speech amount or complexity, indicating stable speech production across languages in autistic and NT children. The only exception involved lexical diversity: German-speaking autistic children produced more lexically diverse speech than English-speaking autistic children, whereas NT children showed the opposite pattern.

Age and language experience likely contributed to these differences. German-speaking autistic children were older on average than English-speaking autistic children, and lexical diversity increases with age as vocabulary and narrative skills develop (Berman & Slobin, 1994; Hoff, 2013). In contrast, English-speaking NT children were more often monolingual and showed higher receptive vocabulary scores, consistent with evidence that monolingual children often demonstrate greater lexical diversity within a single language than bilingual peers (Hoff et al., 2012).

Clear cross-linguistic differences emerged for gesture production. English-speaking children produced more gestures, greater gesture diversity, and more complementary gestures than children speaking French, Spanish, or German. Language effects were strongest for gesture amount, diversity, and complexity, while diagnosis remained associated with gesture complexity across languages: autistic children produced a higher proportion of additive/complementary gestures than NT children regardless of language. No diagnosis × language interaction was observed for gesture amount or diversity. For gesture complexity, this interaction could not be reliably tested, as Spanish- and German-speaking children together produced very few complex gestures; whether the diagnosis effect on gesture complexity is itself stable across languages therefore remains an open question. With approximately 11 children per language × diagnosis cell, these analyses were in any case underpowered to detect all but large interactions, so the general absence of significant diagnosis × language interactions should be interpreted as inconclusive rather than as evidence that gesture production is uniformly unaffected by diagnosis across languages.

Speech and Gesture Across Languages in Children With ASD

These cross-linguistic findings offer initial insight into how speech and gesture operate together in autistic children. While speech production remained relatively stable across languages, gesture production varied systematically in amount, diversity, and complexity. This dissociation is consistent with gesture functioning as a flexible communicative resource responsive to language-specific demands, though the present data cannot rule out alternative explanations—for example, that gesture patterns reflect typological differences in gestural conventions or speech–gesture timing across the four language communities rather than any within-child adaptation.

These patterns were observed in both autistic and NT children, suggesting that sensitivity to cross-linguistic variation is a feature of multimodal communication more broadly, including in autism. These findings are consistent with theoretical frameworks that conceptualize speech and gesture as coordinated but partially distinct systems (Kita & Özyürek, 2003; Hostetter & Alibali, 2008). For gesture amount and diversity, we found no evidence that the group difference varied by language; with only 11 children per language × diagnosis cell, however, these analyses were exploratory, and larger per-language samples will be needed to establish whether the group difference holds consistently across languages. For gesture complexity specifically, autistic children produced more additive/complementary gestures than NT children regardless of language, indicating that this difference was not confined to a single language, even though its magnitude across languages could not be formally tested.

Together, these findings indicate that gesture is an integral component of multimodal communication in autism and that this pattern holds across typologically diverse languages (see Ghobadi & Durrleman, 2026; Wolfer et al., 2025, for similar conclusions).

Clinical and Educational Implications for Understanding and Supporting Autistic Communication

The present findings have important implications for evaluating and supporting communication in autistic children. Autistic children conveyed meaningful narrative information through both speech and gesture, with gesture often contributing information not fully expressed in speech. Assessments that focus exclusively on speech may therefore underestimate communicative ability, whereas including gesture provides a more complete characterization of communicative expression and language use.

This multimodal profile was observed across English, French, Spanish, and German, suggesting that gesture may function as a communicative resource across languages, though the modest number of children per language warrants some caution before generalizing this pattern to other populations or languages.

These findings also have implications for intervention and education. Gesture provided an accessible means for autistic children to organize and convey meaning, often contributing complementary information alongside speech. Recognizing gesture as part of a coordinated communicative system may help clinicians, speech-language therapists, and educators interpret communicative intent and support more effective interactions.

Several limitations should be noted. The sample was drawn primarily from Western, highly educated, and relatively high-socioeconomic backgrounds, which may limit generalizability, and the structured narrative task may not fully capture communication in everyday interaction. The cross-linguistic comparisons are likewise tentative: each language contributed only 11 children per diagnostic group, so these analyses were exploratory and offer a preliminary rather than definitive account of how the pattern varies across languages, with larger per-language samples needed to test language-specific effects directly. Gesture amount was measured as a raw count rather than a rate controlling for speech output; because gesture-frequency comparisons in the broader literature are sensitive to this choice (McKern et al., 2023), future work should test whether the present pattern holds when gesture production is normalized to words spoken. Finally, each of these outcomes was summarized as a single observation per child, so the random-intercept specification, while appropriate, could capture between-child but not within-child variability; this, together with the modest overall sample, is a general limitation of the modeling approach across analyses. A further limitation is more central to our interpretation: the present data cannot establish reorganization as a within-child mechanism. The lower speech complexity and the higher rate of additive/complementary gestures in autistic children are group-level effects, and the cross-modal analysis revealed no reliable within-child association between them (see “The Impact of Diagnosis on Children’s Narrative Production” section); the reorganization account should therefore be read as consistent with the data rather than demonstrated by them. Future research should examine multimodal communication in more diverse populations and naturalistic contexts and test the within-child reorganization hypothesis directly. Despite these limitations, the findings demonstrate that multimodal communication in autistic children is systematically organized and responsive to cross-linguistic variation. Gesture functioned as a meaningful communicative resource that complemented speech and supported narrative expression. These results underscore the importance of examining speech and gesture together to fully characterize autistic children's communicative abilities.

Supplemental Material

sj-docx-1-dli-10.1177_23969415261485091 - Supplemental material for Multimodal Communication in Autism Is Reorganized, Not Reduced: Evidence From Speech and Gesture Across Four Languages
sj-docx-1-dli-10.1177_23969415261485091.docx (36.6KB, docx)

Supplemental material, sj-docx-1-dli-10.1177_23969415261485091 for Multimodal Communication in Autism Is Reorganized, Not Reduced: Evidence From Speech and Gesture Across Four Languages by Armita Ghobadi, Pauline Wolfer and Stephanie Durrleman in Autism & Developmental Language Impairments

Acknowledgment

We also thank the following research assistants and students for their support with data coding and project coordination: Angela Yun, Cindy Kim, Darwin Kol, Alexander Jose Aponte, Franziska Baumeister, Gabriela Costa, Leandra Quick, Lea Mizrachi Gazal, Lucia Leguerin, and Natalia Jankovic. Their dedication and careful work were essential to the completion of this study.

1.

Both person-first and identity-first language are used when referring to individuals diagnosed with autism, in recognition of the diverse language preferences within the autistic community (Kenny et al., 2016).

2.

Sample size was determined using an a priori power analysis (G*Power 3.1; Faul et al., 2007). A sample of 44 participants per group provided approximately 80% to 85% power to detect medium effects (R2 = .20, α = .05). This power analysis did not extend to the cross-linguistic analyses, which involved smaller subsamples (n = 11 per language × diagnosis cell); these analyses and their interpretation are therefore treated as exploratory throughout.

3.

All tables are available in the Supplemental Materials (Supplemental Tables S1–S8).

4.

Visuospatial working memory was assessed using a matrix-based task modeled on the Frog Matrix Task (Morales et al., 2013), which assesses the same underlying construct as the Corsi Block-Tapping paradigm.

5.

There were fewer girls represented in the autistic group, consistent with the higher prevalence of autism in boys than girls in the general population, with epidemiological estimates indicating approximately a 3:1 male-to-female ratio (Maenner, 2023).

6.

Because multilingualism is a common phenomenon (Grosjean, 2010), bilingual and monolingual status could not be fully matched across diagnostic groups. Overall, 33% of children with ASD and 55% of NT children were bilingual. The distribution of bilingualism also varied by language: German-speaking children were similarly divided between monolinguals and bilinguals, English- and French-speaking samples included a higher proportion of monolingual children, and the Spanish-speaking sample included a larger proportion of bilingual children (see Supplemental Table S1).

7.

Participants received a gift voucher valued at USD 35, CAD 60, or CHF 35 as compensation for their participation.

8.

Exclusion criteria included epilepsy, traumatic brain injury, or deafness. Co-occurring developmental language disorder or ADHD did not result in exclusion, given their prevalence in autistic populations (Bishop, 2010; Durrleman & Delage, 2016; Lord et al., 2004; Tager-Flusberg, 2006).

9.

This study was approved by the Swiss Association of Research Ethics Committees (Swissethics; Project ID-2022-00878), the Institutional Review Board of the University of Connecticut (USA), the Research Ethics Board Office at McGill University (Canada), the Psychology Research Ethics Committee of the University of Edinburgh (UK), the Institutional Review Board of Emerson College Boston (USA), and the Research Ethics Committee of the Unió Catalana Hospitals Foundation (Spain). All procedures complied with the Declaration of Helsinki.

10.

Full coding definitions and examples are provided in Supplemental Appendix S2.

11.

For gesture complexity, initial inspection indicated that the distribution was non-normal, with some children producing no additive gestures and both groups showing medians at or near zero. We therefore did not model this outcome as a Gaussian proportion. Instead, we used a binomial mixed-effects model with a logit link, treating each gesture as additive/complementary or nonadditive. This approach is appropriate for a bounded count outcome and does not require normally distributed residuals. To evaluate whether the diagnostic group effect was driven by extreme observations, we also conducted a leave-one-out sensitivity analysis in which each autistic child was excluded one at a time and the model was refit. The results of this model and sensitivity analysis are reported in “The Impacts of Diagnosis on Children’s Narrative Production” section.

12.

For speech amount, speech diversity, speech complexity, gesture amount, gesture diversity, and reinforcing gestures, Model 1 examined diagnostic group effects:

Outcomei = β0 + β1(Groupi) + εi

Model 2 examined diagnostic group, language, and their interaction:

Outcomei = β0 + β1(Groupi) + β2(Languagei) + β3(Groupi × Languagei) + εi

For these models, Outcomei represents the participant-level speech or gesture outcome for child i, β0 denotes the intercept, β terms represent fixed effects, and εi represents residual error.

Gesture complexity was modeled separately using a binomial mixed-effects model with a logit link. For gesture complexity, Model 1 was specified as:
logit[P(Additiveij=1)]=β0+β1(Groupi)+ui
Model 2 was specified as:
logit[P(Additiveij=1)]=β0+β1(Groupi)+β2(Languagei)+β3(Groupi×Languagei)+ui
where Additiveij indicates whether gesture j produced by child i was additive/complementary rather than nonadditive, β0 denotes the intercept, β terms represent fixed effects, and ui represents a random intercept for child. For gesture complexity, β3 could not be reliably estimated due to sparse data in the Spanish and German subsamples (see “The Impact of Diagnosis status on Children’s Narrative Production Across Languages” section) and is reported as inestimable rather than as a formal test of interaction.
13.

As shown in Figure 2, gesture complexity was right skewed with a floor effect in both groups, motivating a binomial rather than Gaussian model (“Statistical Analysis” section).

14.

For gesture complexity, descriptives are reported as the number and percentage of additive/complementary gestures out of all gestures produced, rather than as means and standard deviations, to align with the binomial model used for this bounded count outcome.

Footnotes

Ethical Considerations: This study was approved by the Swiss Association of Research Ethics Committees (Swissethics; Project ID-2022-00878), the Institutional Review Board of the University of Connecticut (USA), the Research Ethics Board Office at McGill University (Canada), the Psychology Research Ethics Committee of the University of Edinburgh (UK), the Institutional Review Board of Emerson College Boston (USA), and the Research Ethics Committee of the Unió Catalana Hospitals Foundation (Spain). All procedures complied with the Declaration of Helsinki.

Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article. This work was supported by the Swiss National Science Foundation (SNSF) awarded to Stephanie Durrleman (grant PR00P1_193104/1).

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement: The quantitative data for the study are available on the Open Science Framework (OSF) and can be accessed using the following view-only link.

Supplemental Material: Supplemental material for this article is available online.

References

  1. Adams A. M., Gathercole S. E. (2000). Limitations in working memory: Implications for language development. Journal of Child Language, 27(1), 95–117. 10.1080/136828200247278 [DOI] [PubMed] [Google Scholar]
  2. Adamson L. B., Bakeman R., Deckner D. F., Nelson P. B. (2009). From interactions to conversations: The development of joint engagement during early childhood. Child Development, 80(6), 1661–1677. 10.1111/cdev.12189 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. American Psychiatric Association. (2022). Diagnostic and Statistical Manual of Mental Disorders (5th ed., text rev.). American Psychiatric Publishing. [Google Scholar]
  4. Attwood A., Frith U., Hermelin B. (1988). The understanding and use of interpersonal gestures by autistic and Down’s syndrome children. Journal of Autism and Developmental Disorders, 18(2), 241–257. 10.1007/BF02211950 [DOI] [PubMed] [Google Scholar]
  5. Bacon E. C., Dufek S., Schreibman L., Stahmer A. C., Pierce K., Courchesne E. (2014). Measuring Outcome in an Early Intervention Program for Toddlers with Autism Spectrum Disorder: Use of a Curriculum-Based Assessment. Autism Research and Treatment, 2014, 1–9. 10.1155/2014/964704 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bang J., Burns J., Nadig A. (2013). Brief report: Conveying subjective experience in conversation: Production of mental state terms and personal narratives in individuals with high-functioning autism. Journal of Autism and Developmental Disorders, 43(7), 1732–1740. 10.1007/s10803-012-1716-4 [DOI] [PubMed] [Google Scholar]
  7. Bartak L., Rutter M., Cox A. (1975). A comparative study of infantile autism and specific developmental receptive language disorder. British Journal of Psychiatry, 126 (2), 127–145. 10.1192/bjp.126.2.127 [DOI] [PubMed] [Google Scholar]
  8. Bates E. (1976). Language and context: The acquisition of pragmatics. Academic Press. [Google Scholar]
  9. Bates E., Benigni L., Bretherton I., Camaioni L., Volterra V. (1979). The emergence of symbols: Cognition and communication in infancy. Academic Press. [Google Scholar]
  10. Bates E., Marchman V., Thal D., Fenson L., Dale P., Reznick J. S., Reilly J., Hartung J. (1994). Developmental and stylistic variation in the composition of early vocabulary. Journal of Child Language, 21(1), 85–123. 10.1017/S0305000900008680 [DOI] [PubMed] [Google Scholar]
  11. Baumeister F., Wolfer P., Vila Borrellas E., Solaimani E., Eigsti I.-M., Durrleman S. (2026). Associations between second-language proficiency and executive functions in autistic and neurotypical children. Open Mind, 10, 613–640. 10.1162/OPMI_a_00349 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Berman R. A., Slobin D. I. (1994). Relating events in narrative: A crosslinguistic developmental study. Lawrence Erlbaum Associates. [Google Scholar]
  13. Bishop D. V. (2010). Which neurodevelopmental disorders get researched and why? PLoS ONE, 5(11), e15112. 10.1371/journal.pone.0015112 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Bishop D. V. M. (2003). Test for reception of grammar (2nd ed.). Pearson Assessment. [Google Scholar]
  15. Brown R. (1973). A first language: The early stages. Harvard University Press. [Google Scholar]
  16. Butcher C., Goldin-Meadow S. (2000). Gesture and the transition from one- to two-word speech: When hand and mouth come together. In McNeill D. (Ed.), Language and gesture (pp. 235–258). Cambridge University Press. [Google Scholar]
  17. Colletta J. M., Guidetti M., Capirci O., Cristilli C., Demir O. E., Kunene-Nicolas R. N., Levine S. (2015). Effects of age and language on co-speech gesture production: An investigation of French, American, and Italian children’s narratives. Journal of Child Language, 42(1), 122–145. 10.1017/S0305000913000585 [DOI] [PubMed] [Google Scholar]
  18. Colonnesi C., Stams G. J. J. M., Koster I., Noom M. J. (2010). The relation between pointing and language development: A meta-analysis. Developmental Review, 30(4), 352–366. 10.1016/j.dr.2010.10.001 [DOI] [Google Scholar]
  19. Cook S. W., Mitchell Z., Goldin-Meadow S. (2008). Gesturing makes learning last. Cognition, 106(2), 1047–1058. 10.1016/j.cognition.2007.04.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. de Marchena A., Eigsti I. M. (2010). Conversational gestures in autism spectrum disorder. Autism Research, 3(6), 311–322. 10.1002/aur.159 [DOI] [PubMed] [Google Scholar]
  21. Diehl J. J., Bennetto L., Young E. C. (2006). Story recall and narrative coherence of high- functioning children with autism spectrum disorders. Journal of Abnormal Child Psychology, 34(1), 87–102. 10.1007/s10802-005-9003-x [DOI] [PubMed] [Google Scholar]
  22. Dimitrova Nevena, Özçalışkan Şeyda. (2022). Identifying Patterns of Similarities and Differences between Gesture Production and Comprehension in Autism and Typical Development. Journal of Nonverbal Behavior, 46(2), 173–196. 10.1007/s10919-021-00394-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Dunn L. M., Dunn D. M., Bulheller S. (2003). Peabody picture vocabulary test (PPVT). Swets Test Services. [Google Scholar]
  24. Durrleman S., Delage H. (2016). Autism spectrum disorder and specific language impairment: Overlaps in syntactic profiles. Language Acquisition, 23(4), 361–386. 10.1080/10489223.2016.1179741 [DOI] [Google Scholar]
  25. Eigsti I. M., Bennetto L., Dadlani M. B. (2007). Beyond pragmatics: Morphosyntactic development in autism. Journal of Autism and Developmental Disorders, 37(6), 1007–1023. 10.1007/s10803-006-0239-2 [DOI] [PubMed] [Google Scholar]
  26. Eigsti I.-M., Zayas V., Mischel W., Shoda Y., Ayduk O., Dadlani M. B., Davidson M. C., Aber J. L., Casey B. J. (2006). Predicting cognitive control from preschool to late adolescence and young adulthood. Psychological Science, 17(6), 478–484. 10.1111/j.1467-9280.2006.01732.x [DOI] [PubMed] [Google Scholar]
  27. Evans J. L., Craig H. K. (1992). Language sample collection and analysis: Interview compared to free play assessment contexts. Journal of Speech, Language, and Hearing Research, 35(2), 343–353. 10.1044/jshr.3502.343 [DOI] [PubMed] [Google Scholar]
  28. Faul F., Erdfelder E., Lang A.-G., Buchner A. (2007). G*Power 3. Behavior Research Methods, 39(2), 175–191. 10.3758/BF03193146 [DOI] [PubMed] [Google Scholar]
  29. Frith U, Hill E. L, Tager-Flusberg Helen, Joseph Robert M. (2003). Identifying neurocognitive phenotypes in autism. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 358(1430), 303–314. 10.1098/rstb.2002.1198 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Ghobadi A., Durrleman S. (2026). Gesture, not speech, distinguishes autistic and neurotypical children's narratives across toy and human listener contexts [Preprint]. SSRN. 10.2139/ssrn.7173698 [DOI] [Google Scholar]
  31. Ghobadi A., Özçalışkan Ş. (2026). Patterns of speech and gesture production in the communications of bilinguals and monolinguals: Do speakers’ proficiency and discourse context matter?. Language and Cognition, 18. 10.1017/langcog.2026.10071 [DOI] [Google Scholar]
  32. Goldin-Meadow S. (2005). Hearing gesture: How our hands help us think. Harvard University Press. [Google Scholar]
  33. Goldin-Meadow S., Alibali M. W. (2013). Gesture’s role in speaking, learning, and creating language. Annual Review of Psychology, 64(1), 257–283. 10.1146/annurev-psych-113011-143802 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Grosjean F. (2010). Bilingual: Life and reality. Harvard University Press. [Google Scholar]
  35. Harris C. R., Millman K. J., van der Walt S. J., Gommers R., Virtanen P., Cournapeau D., Wieser E., Taylor J., Berg S., Smith N. J., Kern R., Picus M., Hoyer S., van Kerkwijk M. H., Brett M., Haldane A., del Río J. F., Wiebe M., Peterson P., Gérard-Marchant P., Sheppard K., Reddy T., Weckesser W., Abbasi H., Gohlke C., Oliphant T. E. (2020). Array programming with NumPy. Nature, 585, 357–362. 10.1038/s41586-020-2649-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Hickmann M. (2003). Children’s discourse: Person, space, and time across languages. Cambridge University Press. [Google Scholar]
  37. Hoff E. (2006). How social contexts support and shape language development. Developmental Review, 26(1), 55–88. 10.1016/j.dr.2005.11.002 [DOI] [Google Scholar]
  38. Hoff E. (2013). Language development (5th ed.). Wadsworth. [Google Scholar]
  39. Hoff E., Core C., Place S., Rumiche R., Señor M., Parra M. (2012). Dual language exposure and early bilingual development. Journal of Child Language, 39(1), 1–27. 10.1017/S0305000910000759 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Hostetter A. B. (2011). When do gestures communicate? A meta-analysis. Psychological Bulletin, 137(2), 297–315. 10.1037/a0022128 [DOI] [PubMed] [Google Scholar]
  41. Hostetter A. B., Alibali M. W. (2008). Visible embodiment: Gestures as simulated action. Psychonomic Bulletin & Review, 15(3), 495–514. 10.3758/PBR.15.3.495 [DOI] [PubMed] [Google Scholar]
  42. Huttenlocher J., Haight W., Bryk A., Seltzer M., Lyons T. (1991). Early vocabulary growth: Relation to language input and gender. Developmental Psychology, 27(2), 236–248. [Google Scholar]
  43. Iverson J. M., Capirci O., Volterra V., Goldin-Meadow S. (2008). Learning to talk in a gesture-rich world: Early communication in Italian vs. American children. First Language, 28(2), 164–181. 10.1177/0142723707087736 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Iverson J. M., Goldin-Meadow S. (2005). Gesture paves the way for language development. Psychological Science, 16(5), 367–371. 10.1111/j.0956-7976.2005.01542.x [DOI] [PubMed] [Google Scholar]
  45. Kendon A. (1992). Some recent work from Italy on quotable gestures (“emblems”). Journal of Linguistic Anthropology, 2(1), 72–93. 10.1525/jlin.1992.2.1.92 [DOI] [Google Scholar]
  46. Kenny L., Hattersley C., Molins B., Buckley C., Povey C., Pellicano E. (2016). Which terms should be used to describe autism? Perspectives from the UK autism community. Autism, 20(4), 442–462. 10.1177/1362361315588200 [DOI] [PubMed] [Google Scholar]
  47. Kita S. (2000). How representational gestures help speaking. Language and Gesture, 1, 162–185. 10.1017/CBO9780511620850.011 [DOI] [Google Scholar]
  48. Kita S. (2009). Cross-cultural variation of speech-accompanying gesture. Language and Cognitive Processes, 24(2), 145–167. 10.1080/01690960802586188 [DOI] [Google Scholar]
  49. Kita S., Özyürek A. (2003). What does cross-linguistic variation in semantic coordination of speech and gesture reveal? Journal of Memory and Language, 48(1), 16–32. 10.1016/S0749-596X(02)00505-3 [DOI] [Google Scholar]
  50. Kover S. T., Haebig E., Oakes A., McDuffie A., Hagerman R. J., Abbeduto L. (2014). Sentence comprehension in boys with autism spectrum disorder. American Journal of Speech-Language Pathology, 23(3), 385–394. 10.1044/2014_AJSLP-13-0073 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Krauss R. M., Chen Y., Gottesman R. F. (2000). Lexical gestures and lexical access: A process model. Language and Gesture, 2(261), 261–283. 10.1017/CBO9780511620850.017 [DOI] [Google Scholar]
  52. Lord C., Rutter M., DiLavore P. C., Risi S., Gotham K., Bishop S. L. (2012). Autism diagnostic observation schedule (2nd ed.). Western Psychological Services. [Google Scholar]
  53. Lord C., Rutter M., Le Couteur A. (1994). Autism Diagnostic Interview-Revised: A revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. Journal of Autism and Developmental Disorders, 24(5), 659–685. 10.1007/BF02172145 [DOI] [PubMed] [Google Scholar]
  54. Lord C., Shulman C., DiLavore P. (2004). Regression and word loss in autistic spectrum disorders. Journal of Child Psychology and Psychiatry, 45(5), 936–955. 10.1111/jcpp.2004.45.issue-5 [DOI] [PubMed] [Google Scholar]
  55. Losh M., Capps L. (2003). Narrative ability in high-functioning children with autism or Asperger’s syndrome. Journal of Autism and Developmental Disorders, 33(3), 239–251. 10.1023/A:1024446215446 [DOI] [PubMed] [Google Scholar]
  56. Loveland K. A., McEvoy R. E., Tunali B., Kelley M. L. (1990). Narrative story telling in autism and down’s syndrome. British Journal of Developmental Psychology, 8(1), 9–23. 10.1111/j.2044-835X.1990.tb00818.x [DOI] [Google Scholar]
  57. MacWhinney B. (2000). The CHILDES Project: Tools for analyzing talk (3rd ed.). Lawrence Erlbaum Associates. [Google Scholar]
  58. Maenner M. J. (2023). Prevalence and characteristics of autism spectrum disorder among children aged 8 years—Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2020. MMWR. Surveillance Summaries, 72(2), 1–14. 10.15585/mmwr.ss7202a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. McKern N., Dargue N., Sweller N. (2023). Comparing gesture frequency between autistic and neurotypical individuals: A systematic review and meta-analysis. Psychological Bulletin, 149(11-12), 724–745. 10.1037/bul0000408 [DOI] [PubMed] [Google Scholar]
  60. McKinney W. (2010). Data structures for statistical computing in python. In van der Walt S., Millman J. (Eds.), Proceedings of the 9th python in science conference (pp. 56–61). [Google Scholar]
  61. McNeill D. (1992). Hand and mind: What gestures reveal about thought. University of Chicago Press. [Google Scholar]
  62. McNeill D. (2005). Gesture and thought. University of Chicago Press. [Google Scholar]
  63. Miller J. F., Chapman R. S. (1981). The relation between age and mean length of utterance in morphemes. Journal of speech. Language, and Hearing Research, 24(2), 154–161. 10.1044/jshr.2402.154 [DOI] [PubMed] [Google Scholar]
  64. Morales J., Calvo A., Bialystok E. (2013). Working memory development in monolingual and bilingual children. Journal of Experimental Child Psychology, 114(2), 187–202. 10.1016/j.jecp.2012.09.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Nicoladis E., Jiang Z. (2018). Language and cognitive predictors of lexical selection in storytelling for monolingual and sequential bilingual children. Journal of Cognition and Development, 19(4), 413–430. 10.1080/15248372.2018.1483370 [DOI] [Google Scholar]
  66. Nicoladis E., Pika S., Marentette P. (2009). Do French–English Bilingual Children Gesture More Than Monolingual Children?. Journal of Psycholinguistic Research, 38(6), 573–585. 10.1007/s10936-009-9121-7 [DOI] [PubMed] [Google Scholar]
  67. Norbury C. F., Bishop D. V. (2003). Narrative skills of children with communication impairments. International Journal of Language & Communication Disorders, 38(3), 287–313. 10.1080/136820310000108133 [DOI] [PubMed] [Google Scholar]
  68. Özçalışkan Ş, Adamson L. B., Dimitrova N. (2016). Early deictic but no other gestures predict later vocabulary in both typical development and autism. Autism, 20(6), 754–763. 10.1177/1362361315605921 [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Özçalışkan Ş, Adamson L. B., Dimitrova N., Baumann S. (2017). Early gesture provides a helping hand to spoken vocabulary development for children with autism, down syndrome and typical development. Journal of Cognition and Development, 18(3), 325–337. 10.1080/15248372.2017.1329735 [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Özçalışkan Ş, Goldin-Meadow S. (2005a). Gesture is at the cutting edge of early language development. Cognition, 96(3), B101–B113. 10.1016/j.cognition.2005.01.001 [DOI] [PubMed] [Google Scholar]
  71. Özçalışkan Ş, Goldin-Meadow S. (2005b). Do parents lead their children by the hand? Journal of Child Language, 32(3), 481–505. 10.1017/S0305000905007002 [DOI] [PubMed] [Google Scholar]
  72. Peterson C., McCabe A. (1987). The connective ‘and’: Do older children use it less as they learn other connectives? Journal of Child Language, 14(2), 375–381. 10.1017/S0305000900012988 [DOI] [PubMed] [Google Scholar]
  73. Pınar E., Öztürk S., Ketrez N., Özçalişkan Ş. (2021). Parental speech and gesture input to girls versus boys in singletons and twins. Journal of Nonverbal Behavior, 45, 297–318. 10.1007/s10919-020-00356-w [DOI] [Google Scholar]
  74. Python Software Foundation. (2023). Python (Version 3.x) [Software].
  75. Ramos-Cabo S., Vulchanov V., Vulchanova M. (2019). Gesture and language trajectories in early development: An overview from the autism spectrum disorder perspective. Frontiers in Psychology, 10, 1211. 10.3389/fpsyg.2019.01211 [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Rice M. L., Redmond S. M., Hoffman L. (2006). Mean length of utterance in children with specific language impairment and in younger control children shows concurrent validity and stable and parallel growth trajectories. Journal of Speech, Language, and Hearing Research: JSLHR, 49(4), 793–808. 10.1044/1092-4388(2006/056) [DOI] [PubMed] [Google Scholar]
  77. Rice M. L., Smolik F., Perpich D., Thompson T., Rytting N., Blossom M. (2010). Mean length of utterance levels in 6-month intervals for children 3 to 9 years with and without language impairments. Journal of Speech, Language, and Hearing Research : JSLHR, 53(2), 333–349. 10.1044/1092-4388(2009/08-0183) [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Rowe M. L., Goldin-Meadow S. (2009). Early gesture predicts later language learning. Developmental Science, 12(1), 182–187. 10.1111/j.1467-7687.2008.00764.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  79. Seabold S., Perktold J. (2010). Statsmodels: Econometric and statistical modeling with Python. In van der Walt S., Millman J. (Eds.), Proceedings of the 9th Python in Science Conference (SciPy 2010) (pp. 92–96). [Google Scholar]
  80. Shumway S., Wetherby A. M. (2009). Communicative acts of children with autism spectrum disorders in the second year of life. Journal of Speech. Language, and Hearing Research, 52(5), 1139–1156. 10.1044/1092-4388(2009/07-0280) [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Silleresi S. (2023). Developmental profiles in autism spectrum disorder: Theoretical and methodological implications. John Benjamins Publishing Company. [Google Scholar]
  82. So W. C., Kita S., Goldin-Meadow S. (2013). Communication without speech: Gesture and language in typically developing children and children with autism. Journal of Child Language, 40(2), 1–25. 10.1017/S0305000912000578 [DOI] [PubMed] [Google Scholar]
  83. Tager-Flusberg H. (2006). Defining language phenotypes in autism. Clinical Neuroscience Research, 6(3-4), 219–224. 10.1016/j.cnr.2006.06.007 [DOI] [Google Scholar]
  84. Tager-Flusberg H., Anderson M. (1991). The development of contingent discourse ability in autistic children. Journal of Child Psychology and Psychiatry, 32(7), 1123–1134. 10.1111/j.1469-7610.1991.tb00353.x [DOI] [PubMed] [Google Scholar]
  85. Tantam D., Monaghan L., Nicholson H., Stirling J. (1993). Autistic children’s ability to interpret faces: A research note. Journal of Child Psychology and Psychiatry, 34(5), 785–790. 10.1111/j.1469-7610.1989.tb00274.x [DOI] [PubMed] [Google Scholar]
  86. Westerveld M. F., Moran C. A. (2011). Expository language skills of young school-age children. Language, Speech, and Hearing Services in Schools, 42(2), 182–193. 10.1044/0161-1461(2010/10-0044) [DOI] [PubMed] [Google Scholar]
  87. Wolfer P., Baumeister F., Cohen D., Dimitrova N., Solaimani E., Durrleman S. (2025). Co-speech gesture comprehension in autistic children. Journal of Child Language, 53(4), 1–22. 10.1017/S0305000925000157 [DOI] [PubMed] [Google Scholar]
  88. World Health Organization. (2024). Clinical Descriptions and Diagnostic Requirements for ICD- 11 Mental, Behavioural and Neurodevelopmental Disorders. World Health Organization. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

sj-docx-1-dli-10.1177_23969415261485091 - Supplemental material for Multimodal Communication in Autism Is Reorganized, Not Reduced: Evidence From Speech and Gesture Across Four Languages
sj-docx-1-dli-10.1177_23969415261485091.docx (36.6KB, docx)

Supplemental material, sj-docx-1-dli-10.1177_23969415261485091 for Multimodal Communication in Autism Is Reorganized, Not Reduced: Evidence From Speech and Gesture Across Four Languages by Armita Ghobadi, Pauline Wolfer and Stephanie Durrleman in Autism & Developmental Language Impairments

Articles from Autism & Developmental Language Impairments are provided here courtesy of SAGE Publications

RESOURCES