●Repeat after me? Both Children with and without Autism Commonly Align Their Language with That of Their Caregivers

Abstract

Introduction: The role of linguistic repetition in child language is complicated. Echolalia, the persistent repetition of one’s interlocutor’s utterances, has long been associated with Autism Spectrum Disorder (ASD). The literature on echolalia suggests that children with ASD often re-use fragments of language without any communicative or language-learning purpose. At the same time, research on linguistic repetition in typically developing (TD) children and adults increasingly demonstrates that linguistic repetition is an important social activity that helps create shared engagement, common ground and even scaffolds language acquisition. Depending on which literature is consulted, children with ASD, who often have social challenges, might be expected to repeat more or less than typically developing children. Objective: We take a systematic and nuanced approach to linguistic repetition in children with and without ASD: we distinguished exact repetition, rate, and level, as well as lexical, syntactic, and semantic repetition in a large, quasi-naturalistic corpus. We hypothesize that more rigid forms of repetition (exact lexical repetitions) would be more frequent in ASD, and negatively related to individual differences in linguistic, cognitive, and social skills. We hypothesize that more potentially productive forms of repetitions (alignment rate, semantic alignment) would be less frequent in ASD and positively related to the same individual differences.Methods: We analyzed spontaneous speech in 67 parent-child dyads from a longitudinal corpus (30 minutes of play activities at 6 visits over 2 years). We included 32 children diagnosed with ASD and 35 linguistically matched TD children (mean age at recruitment respectively 32.76 and 20.27 months). Alignment was calculated using the ALIGN Python library (Duran et al., 2019). We first contrasted child alignment in actual conversations with that in surrogate pairs formed by a parent and a child from two different dyads. We used Bayesian multilevel zero-and-one-inflated beta regression models. This accounts for the propensity of children to align at all (rate), the proportion of the caregiver’s utterance repeated when the children do align (level), and exact repetitions, as a function of diagnostic group and visit. Lexical repetitions were regressed out from syntactic and semantic repetitions to minimize confounds. Analyses of individual differences include Mullen Expressive Language and Mean Length of Utterance (MLU) as proxies for linguistic skills, Visual Reception for non-linguistic cognitive skill; and Vineland Socialization for socialization skills. Results: We found a small number of exact repetitions in both groups (roughly 1% of utterances across visits); moreover, children with ASD increased their use of repetitions over time while the TD group decreased theirs (at final visit, 1.2% in ASD vs. 0.7% in TD). Partial repetitions were much more frequent: children reused caregivers’ words at high rates regardless of diagnostic group (24% of utterances at first visit). Partial repetitions of words increased in frequency (but not level) over time, and did so faster for TD children (at final visit: 33% for ASD and 40% for TD); and the same happened for partial repetition of syntax (from approximately 20% of utterances in both groups at first visit to 23% for ASD and 43% for TD at last visit) and semantic alignment (from 0.34 semantic similarity at first visit to 0.37 for ASD and 0.44 for TD at last visit). Further, children with richer linguistic abilities also displayed a higher tendency to partially re-use their caregivers’ language (alignment rates and semantic alignment).Discussion: Children with and without ASD commonly re-used the words, syntax, and topics of their caregivers. Furthermore, most repetition was at least potentially productive, with repeated language being re-contextualized and integrated with non-repeated language. Interestingly, children with ASD developed this skill at a slower rate than TD children, contrary to previous findings using less fine-grained methods. Only individual differences in linguistic skills positively related to flexible forms of alignment, further supporting their productive nature. The salience of echolalia in ASD might be partially explained by slight differences in frequency, amplified by lower semantic alignment, persistence over time, and expectations of echolalia. However, more in-depth qualitative and quantitative analyses of how repetitions are used and received in context are needed.

Language‐Specific Constraints on Conversation: Evidence from Danish and Norwegian

“…Interestingly, we see similar effects for entrainment rate and level (both higher in Danish than Norwegian), where previous studies show important differences in their behavior related to the context (Dideriksen et al, 2023) or the development of linguistic and conversational skills (Fusaroli et al, 2023a(Fusaroli et al, , 2023bMisiek, Favre, & Fourtassi, 2020). However, as in previous literature, rate presents bigger variations related to language and context than level does.…”

“…Fully overlapping speech between interlocutors occurred occasionally (e.g., during some backchannels) and was interleaved at the first pause longer than 1 s in the speaker holding the floor. The impact of different definitions of conversational turns on, for example, measures of alignment has been assessed in (Fusaroli, Weed, Rocca, Fein, & Naigles, 2023a, 2023b and showed the robustness of the findings to changes in definition. Relying on videos and transcripts, backchannels and the three kinds of repairs were manually coded.…”

Lexical Alignment is Pervasive Across Contexts in Non‐WEIRD Adult–Child Interactions

“…In this study, we extended previous findings to a semi‐structured naturalistic conversational paradigm encompassing contexts that would more closely represent children's daily conversations. Previous literature has demonstrated that younger children use lexical alignment as a language learning strategy (e.g., Fusaroli et al., 2023b; Rydland & Aukrust, 2005; Strapp & Federico, 2000). In contrast, adults, who already have fully developed linguistic systems, use alignment to facilitate social communication (Huiyang & Min, 2022; Ireland et al., 2011).…”

“…This paper specifically explores lexical alignment, where conversational partners adapt their language and reuse one another's words. The phenomenon of lexical alignment has been observed in adult–adult and adult–child conversations (e.g., Dideriksen et al., 2023; Duran, Paxton, & Fusaroli, 2019; Fusaroli, Weed, Rocca, Fein, & Naigles, 2023b). Both theoretical models and empirical evidence suggest that lexical alignment is crucial for supporting communication success by facilitating cognitive, emotional, and pragmatic functions (e.g., Dideriksen et al., 2023; Nenkova, Gravano, & Hirschberg, 2008; Rasenberg et al., 2022).…”

“…It is speculated that by aligning their lexicons with their children, mothers capture their children's attention, establish shared conversational knowledge, and prepare them to learn new and more sophisticated words. Beyond parental alignment, children's lexical alignment with adults is crucial for predicting their linguistic abilities, suggesting that alignment is an essential component of language and conversational skill development (Fusaroli et al., 2023b).…”