Speech, gait, and brain dynamics during natural conversation in motion

Abstract

Communication connects us to others and immerses us in our environment. Mobile EEG and hearing aid-embedded motion sensors allow investigation of communi-cation and movement in real-world contexts. Here we reliably record and model neural responses to natural conversational speech while participants are walking, demonstrating the feasibility of combining EEG with whole-body movement moni-toring. Participants (N = 61) performed focused listening and natural conversation while walking indoors and outdoors. Motion sensors, EEG, and audio recordings captured behavioural, neural, and motor dynamics. Walking speed varied with en-vironment and task: faster outdoors and slower during speaking. Temporal align-ment between gait and speech revealed coordinated rhythmic interactions, stronger during speaking than listening. Auditory temporal response functions were preserved across listening tasks. Gait TRFs displayed lateralized and temporal dy-namics consistent across tasks. These results provide insight into the coordinated interplay of speech, brain activity, and movement. This highlights the potential of multimodal approaches in ecologically valid settings.