Task goals constrain the alignment in eye-movements and speech during interpersonal coordination

Collaborative task performance is assumed to benefit from interpersonal alignment. Prominent views of language use and social behavior, including the Interactive Alignment Model (Pickering & Garrod, 2004), endorse this idea by building on tasks that require partners to monitor each other’s perspective (e.g., route planning) and positing that behavioral alignment enables task partners to converge conceptually. However, the benefits of alignment to different tasks that may instead need complementarity (e.g., a “divide and conquer” strategy during joint visual search) have yet to be explored. We examine this question directly by manipulating task goals (route planning vs. visual search) as forty dyads work with ten trials involving subway maps while their eye movements and speech were co-registered. In five trials, dyads planned a route from an origin to a destination (route planning); in another five trials, they searched for landmarks sharing some feature (visual search). The interpersonal alignment in the dyads’ eye fixations and word sequences was calculated using Cross Recurrence Quantification Analysis (CRQA). Dyads exhibited more gaze alignment in route planning than visual search across a range of CRQA metrics. When examining the temporal evolution of gaze alignment, we found it to vary across the trial substantially, and its increase influenced accuracy differently across the two tasks. Specifically, in visual search, higher increases in alignment at the end of the trial were associated with accurate performance. When we turned to speech data, we found that dyads exhibited longer and more entropic word sequences in route planning but had lower proportions of overall rates of word recurrence in that task. This finding suggests that the two modalities organize in a complementary fashion to optimize distinct task goals. Altogether, task goals constrain how people coordinate their behavior and provide insights into how collaborating partners distribute their distinctive multimodal behavior.