BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Optimal control of ballistic movements in a thalamo-cortical circu
 it model - Guillaume Hennequin\, University of Cambridge
DTSTART:20181018T130000Z
DTEND:20181018T140000Z
UID:TALK112417@talks.cam.ac.uk
CONTACT:Alberto Padoan
DESCRIPTION:How does the brain control movement? Experiments suggest that 
 the (pre-)motor cortex behaves like an “engine” whose dynamics drive m
 ovement\, and whose activity must first be initialised into movement-speci
 fic states (“the optimal subspace hypothesis”\, Shenoy & al\, 2013). B
 oth the computational and mechanistic underpinnings of this preparatory pr
 ocess remain poorly understood. Here\, we propose a circuit model for move
 ment preparation and execution. We formalise movement preparation as an op
 timal control problem\, and compute the preparatory input to cortex that d
 rives fastest convergence to preparatory states predicted to yield the cor
 rect motor outcomes. Critically\, we show that optimal control inputs can 
 be realised via feedback in realistic\, thalamo-cortical neural circuit ar
 chitectures\, and that multiple control loops can be flexibly combined to 
 generate movements composed of a few primitives. The model produces natura
 listic patterns of preparatory activity\, and accounts for across-trial va
 riability suppression as well as orthogonality between preparatory and mov
 ement-related activity subspaces.
LOCATION:Cambridge University Engineering Department\, Lecture Room 4
END:VEVENT
END:VCALENDAR