Lexical alignment and speaker visibility influence gestural alignment in conversation: Implications for theories of linguistic alignment

People often copy each other’s sentence structures, lexical choices, and co-speech gestures. Such linguistic alignment has been suggested to be motivated by low-level priming as well as high-level collaborative efforts to establish mutual understanding (grounding). Although their roles have been tested for verbal aspects of language use, the extent to which each process drives gestural alignment remains underexplored. Here, we manipulated partner visibility in a dyadic, video-mediated referential communication game ($n = 45$ dyads) and examined (i) whether priming and grounding affect the frequency and form of referentially aligned gestures, and (ii) how alignment in lexical choices relates to alignment in gesture. We found that seeing a partner’s gestures increased the frequency of referentially aligned gestures but not the similarity of their forms, suggesting that semantic priming of the gestured referent (meaning) is stronger or more persistent than motor–spatial priming of the gesture itself (form). We also found a positive association between the frequencies of lexical and gestural alignment, even after accounting for a set of control variables, suggesting that lexical and gestural are causally linked. Lastly, gesture-to-gesture priming and cross-modal influence of lexical alignment on gestural alignment only accounted for about half of the observed referentially aligned gestures, leaving room for other mechanisms and factors to drive gestural alignment, such as shared conceptualization of the referent. Overall, the findings provide evidence for both gesture-to-gesture and cross-modal priming and show that gestural alignment in multimodal language use reflects multiple mechanisms and factors beyond priming and grounding.