Sylvain Baillet

@sylvainbaillet.bsky.social

Director, Research & Innovation, University of Montreal Health Centre (CHUM), and Director CHUM Research Centre (CRCHUM) | Professor, Neuroscience @ University of Montreal.

“Earning Its Place: AI in Healthcare,” my keynote at AI26 — the Lamarr Institute Conference in Bonn earlier this week. AI should enter healthcare not by hype but by improving what matters for patients and health systems: access, quality, equity, and the patient experience. @crchum.bsky.social

BildBildBild

Congrats to @maxlevinson.bsky.social for pulling this off in the middle of 2020's lockdown and the shortage of liquid helium in 🇨🇦. Grateful to our 🇬🇧 colleagues who saved us with grace and wit and more.

Max Levinson@maxlevinson.bsky.social · 3mo ago

NEW PAPER🚨👀 in Science Advances with a wonderful transatlantic team. (incl @sylvainbaillet.bsky.social @olejensen.bsky.social @katduecker.bsky.social) Using human MEG, we found lower-order and higher-order neural processes that underlie illusory transitions in conscious visual perception. 🧵👇

Exciting news from the chairs of Biomag 2026, Prof. Jiahong Gao and Prof. Huan Luo — the conference website is now live: biomag2026.scimeeting.cn The meetings take place in Beijing, 23–25 August 2026. Save the date and start thinking about ideas for posters and symposia! Please share with colleagues

BIOMAG 2025,在线会议,会议直播平台,网络会议直播,美迪康会务通,学术会议管理系统,会议活动管理,注册签到管理,会场幻灯片传输转播,PPT传输,会议网络直播,电子壁报管理系统,医学会,会员管理系统

biomag2026.scimeeting.cn

🍾 First published work of 2026 from the lab. Congrats to recently graduated Dr Niloofar Gharesi on her study of reward prediction error processes under the co-supervision of Prof John Kalaska at @umontreal-en.bsky.social. www.biorxiv.org/content/10.6... @crchum.bsky.social & @theneuro.bsky.social

Neural Signatures of Post-Decision Outcome Expectation and Evaluation in Human Sensorimotor Choice Behavior

The concept of embodied sensorimotor decision-making proposes that processes implicated in evaluating sensory inputs and selecting appropriate motor actions unfold partly in cortical regions traditionally associated with movement planning and execution. Reinforcement learning models emphasize the role of reward prediction error (RPE) in optimizing action selection based on decision outcome feedback. However, most evidence for the existence of RPE signals locates them in midline frontal and parietal cortex, and comes from tasks with externally manipulated reward probabilities that create artificial prediction errors. Whether RPE signals are expressed in human cortical motor areas during deterministic (non-probabilistic) tasks remains unclear, and would provide further support for embodied decision-making. We used magnetoencephalography (MEG) to study post-decision neural dynamics in a color discrimination task in selected cortical regions of interest (ROIs). Participants had to press buttons with their left or right index finger in response to checkerboard stimuli with different levels of color evidence for the correct choice. Outcomes were fully determined by participants' choices. Delayed auditory feedback veridically indicated whether their hand choice was correct or not. We observed a robust beta-band (15-29 Hz) rebound after correct outcome feedback, strongest in ventral and dorsal premotor, anterior cingulate and superior parietal ROIs as well as occipital and auditory ROIs, and weakest in the primary motor and somatosensory ROIs. Critically, the rebound magnitude after correct feedback scaled inversely with color evidence strength and associated decision error rates. It was minimal in strong-evidence trials (~0.1% errors) and maximal in weak-evidence trials (~34% errors), resembling a context-sensitive positive RPE signal that was strongest when a correct outcome was least expected. Alpha-band (8-12 Hz) post-feedback rebound increases in weak evidence trials were not as strong as in the beta band and appeared mainly in occipital, superior parietal and posterior cingulate ROIs. After the decision but before feedback, both beta and alpha band power showed sensitivity to the level of sensory evidence on which the decisions had been based, with reduced post-movement rebound or enhanced suppression in trials with weak evidence, suggestive of internally generated outcome expectations. Pre-feedback alpha rebound suppression was strongest in occipital, superior parietal and posterior cingulate ROIs. Pre-feedback beta rebound suppression was not as strong. Together, these findings reveal distinct beta- and alpha-band dynamics that reflect internal pre-feedback outcome expectations and feedback-driven RPE-like outcome assessments. They support distributed cortical mechanisms, including premotor, parietal and cingulate regions, in reward expectation, outcome evaluation, and adaptive control, highlighting a role for motor and associative cortices in embodied decision-making, performance monitoring, and flexible behavior under uncertainty. ### Competing Interest Statement The authors have declared no competing interest. Fonds de recherche du Québec, https://ror.org/00w3qhf76, FRQ 318042, DOI: https://doi.org/10.69777/318042 NSERC-CREATE: Complex Dynamics of Brain and Behaviour program Canadian Institutes of Health Research, https://ror.org/01gavpb45, CIHR MOP-97944, MOP-142220 [J.K., S.B.]

biorxiv.org

Thank you to @aaas.org for the opportunity to lead a new editorial pod at Science Advances on Systems, Computational, Neuro-AI & Neurotechnology (SCANN). Top opportunity to highlight outstanding work at the intersection of neuroscience, computation, and emerging technologies. Send your papers now!

Bild