Wednesday, November 9th, 2022, from 3:00 to 4:30 p.m., followed by a cocktail.
Université de Montréal, Pavilion Marie-Victorin, Room D-427
The recording of the lecture is available: here
Rhythm Entrainment as Dynamic Bayesian Inference
Abstract: Humans easily perceive, track, and entrain movement to periodic structures underlying complex auditory rhythms. This process has been modeled on the assumption that some part of the brain acts like one or more oscillators that align or resonate with rhythms, but this perspective offers little insight into how rhythm perception might be part of a larger picture of perception, cognitive, and motor processes. Adopting the perspective that much of perception and cognition consists of inference about states of the world that cannot be directly observed, I argue that perceptual entrainment to a rhythm can be viewed as dynamic inference of a hidden state (rhythm phase, and optionally tempo and meter) based on sensory observations (auditory events). When this inference problem is stated formally, solved, and simulated, it mimics human rhythm perception, both at a gross level and in various experimentally identified nuances and illusions. As a model of rhythm perception, it is unique in tracking not only stimulus phase and tempo but also phase and tempo uncertainty, and in predicting from first principles how unfulfilled expectations should warp the perceived passage of time. I will present the work we’ve done on this so far and then gesture towards new directions opened up by this modeling, including some ideas on bridging the conceptual and experimental gap between perceptual and motor entrainment.
Bio: Jonathan recently moved from Boston (where he studied neural modeling at Boston University and did postdoctoral research at Brandeis and MIT) to Hamilton to join the faculty in the McMaster department of Psychology, Neuroscience, and Behaviour. His new research program is focused on timing and rhythm in perception and action, with particular interest in timing-related neural dynamics in the basal ganglia, cerebellum, and supplementary motor area. The questions motivating his work are rooted in his experience as a performing musician. His approach centers on the formulation and simulation of neurophysiological and cognitive models, and draws on dynamical systems theory and the theory of Bayesian cognition. His work will also incorporate psychophysics and EEG experiments, as well as collaborations with experimentalists.
