Walking while performing visual, auditory and crossmodal tasks produces oscillations entrained to the gait cycle

Poster Presentation: Tuesday, May 21, 2024, 2:45 – 6:45 pm, Pavilion
Session: Action: Locomotor, flow, steering

David Alais1 (), Gabriel Clouston1, Matthew Davidson1; 1University of Sydney, Australia

Although we are inherently active beings, most knowledge of perception comes from experiments that restrict natural movement and thus ignore the perception-action loop. Recent evidence from our lab (Davidson, Nature Comms, 2023) has demonstrated that during walking – one of our most common everyday actions – performance on a simple visual detection task cycles through a good and bad phase with every step. Here we extend this line of work to test the influence of walking on more complex visual, auditory and crossmodal tasks. In a series of visual experiments, we observe oscillations of both sensitivity and criterion in a two-alternative forced-choice visual discrimination task, as well as oscillations in perception of (visual) numerosity. In auditory experiments, we demonstrate that performance on an auditory detection task is also entrained to the rhythm of the step-cycle, suggesting stride-cycle oscillations are a domain-general phenomenon. Finally, using an audio-visual temporal synchrony task, we observe that the width of the synchrony window broadens and narrows during the step-cycle. Together, these results demonstrate that gait-related oscillations in performance generalise from simple visual detection tasks to visual discrimination tasks, and to auditory and audio-visual tasks. In most observers, these modulations occur at 1 cycle per step although some participants exhibit a modulation at 2 cycles per step. These findings are clear examples of action modulating perception and show that perception cannot be fully understood without studying the perception-action loop. Studying perception while active provides conditions closer to those for which vision evolved and will likely lead to new insights not possible in static lab environments.