Goal-directed action modulates audiovisual multisensory perception

Multisensory integration is often thought to occur automatically when sensory cues are spatially and temporally congruent. However, emerging evidence suggests that integration is dynamically shaped by attention, task demands, and action. The present study investigated whether unimodal visual, bimodal congruent (i.e., audiovisual), and bimodal incongruent (i.e., sound-induced flash illusion) perception was modulated by the spatial proximity between a cue and a target location (Experiment 1) and by the actual execution of a goal-directed action (Experiment 2). Participants performed pointing movements toward visual targets while unimodal, bimodal, and sound-induced flash illusion stimuli (i.e., fission and fusion) were presented either at the target location or at adjacent non-target locations. Upon completion of the action, participants reported how many flashes they perceived (i.e., perceptual accuracy). Experiment 1 revealed a spatial gradient in perceptual accuracy, as evidenced by higher bimodal accuracy at target versus non-target close and far locations. Fusion illusion susceptibility was higher (i.e., lower accuracy) at the target versus non-target close (i.e., higher accuracy) condition. In Experiment 2, the modulation of multisensory perception at the target location was observed only during movement versus a no-movement condition. Consistent with neurophysiological evidence suggesting distinct underlying mechanisms, only the fusion illusion was sensitive to spatial and motoric modulation, while the fission illusion remained stable across both experiments. These findings demonstrate that multisensory perception is not uniformly distributed across space or action state but instead may be modulated at behaviorally relevant locations during voluntary actions. OPEN PRACTICES STATEMENT: The raw data used to generate the conclusions of this work will be made available upon reasonable request to the corresponding author, at tristan.loria@wsu.edu. A reasonable request should identify the purpose for which the data will be used and specify the data or variables being requested. Requests for purposes such as verification or reproduction of the reported findings, secondary scientific analyses, methodological evaluation, or other legitimate research or educational uses will generally be considered reasonable. Data sharing will be subject to applicable ethical, privacy, participant-consent, and institutional requirements, and data will be de-identified as appropriate prior to sharing.