Proceedings of the Annual Meeting of the Cognitive Science Society

The importance of verbal and visuospatial working memory(WM) for co-speech gesture comprehension was tested in twoexperiments using the dual task paradigm. Healthy, college-aged participants encoded either a dot locations in a grid(Experiment 1), or a series of digits (Experiment 2), andrehearsed them as they performed a discourse comprehensiontask. The discourse comprehension task involved watching avideo of a man describing household objects, and judgingwhich of two words probes was most related to the video.Following the discourse comprehension task, participantsrecalled either the verbally or visuo-spatially encodedinformation. In both experiments, performance on thediscourse comprehension task was faster when gesturalinformation was congruent with the speech than when it wasincongruent. Moreover, performance on the discoursecomprehension task was impacted both by increasing the loadon the visuospatial WM system (Experiment 1) and the verbalWM system (Experiment 2). However, in both studies effectsof WM load and gesture congruency were additive,suggesting they were independent.