Abstract Summary/Description
Gestures and speech form a tightly knit system in first language (L1) production contexts, with gestures either reinforcing or adding new information to speech. However, the link between speech and gesture in second language (L2) production contexts remains less understood, as existing studies—mostly utilizing narrative tasks—have provided inconclusive results. Moreover, limited research has examined how speaker proficiency and task demand impact gesture-speech integration in bilingual speakers, highlighting the need for studies exploring these factors in multimodal bilingual communication. In this study, we investigated the speech and gestures produced by two groups of adult Persian (L1)-English (L2) bilinguals—with either high (n=22) or low (n=22) proficiency in their L2—and compared their productions to those of monolingual English speakers (n=22). Participants engaged in two different prompts: retelling a story ("Please retell the story") and reasoning about the main character's actions ("Tell me what the problem was and how the main character resolved it"). Focusing on English productions, we asked whether language proficiency (native, high proficiency, low proficiency) and task prompt (retelling, reasoning) would affect the amount, diversity, and complexity of speech and gesture production in similar ways. We measured speech amount by the number of words, speech diversity by the number of different word types (e.g., "cat" vs. "bird"), and speech complexity by the mean length of utterance in words (MLU). Similarly, we assessed gesture amount by the number of gestures, gesture diversity by the number of different referents conveyed in gesture (e.g., pointing at a cat vs. a bird), and gesture complexity by the number of supplementary gesture-speech combinations, following earlier work. We analyzed speech and gesture production separately using Generalized Linear Mixed Models (GLMMs) with proficiency levels and task prompts as fixed factors and individual variability as a random effect. For speech, our analysis showed a main effect of task prompt for the retelling prompt (β = 0.42, z = 2.33, p = 0.01). There was no main effect of proficiency on the amount or diversity of speech production, but there was a main effect on speech complexity (β = 0.70, z = 2.61, p = 0.01), with bilinguals with low proficiency producing sentences with lower MLU than the other two groups. Turning to gesture, we found a main effect of task prompt again for the retelling prompt (β = 0.43, z = 2.18, p = 0.02), mirroring the pattern observed in speech. There was no main effect of proficiency on gesture amount or diversity, but there was a main effect on gesture complexity (β = 1.15, z = 2.37, p = 0.02). Specifically, only bilinguals with low proficiency produced more complex gestures, a pattern inverse to their speech production. Our study extends existing theories on gesture's compensatory role in language production. The findings underscore the adaptive strategies speakers employ to maintain effective communication, emphasizing gesture as a crucial resource for conveying complex ideas when speech alone is insufficient. Furthermore, our research highlights the influence of task prompts on gesture use, suggesting that prompt complexity or prompts that require higher reasoning demands can modulate reliance on gestural communication, especially among individuals with limited language proficiency. These results align with other speech-gesture theories observed in speakers with speech difficulties (e.g., children in the early steps of their language learning and adults with aphasia), indicating that when speakers face challenges addressing specific tasks, they utilize hand gestures to compensate. Specifically, bilinguals with low proficiency used more complex hand gestures to offset their speech difficulties, particularly when required to provide detailed explanations. These insights have practical implications for language education and therapy, where encouraging gesture use could facilitate language learning and fluency in second-language speakers. Future research could build on these findings by exploring more complex prompts (e.g., argumentative, role-playing, etc.) or examining how gestures interact with other non-verbal modalities to support communication in bilinguals, particularly focusing on low-proficiency groups.
