BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME: Computational principles of motor learning across timescales
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260812T081451Z
UID:tag:localist.com\,2008:EventInstance_42728936372737
DTSTART:20230330T190000Z
DTEND:20230330T200000Z
DESCRIPTION:Humans can adapt existing movements and learn new movements whe
 n exposed to a changing body\, diverse terrain\, robotic interfaces\, and 
 other modifications to the person or the environment. In this talk\, using
  a combination of mathematical models\, simulations\, and empirical eviden
 ce\, I will describe the principles underlying locomotor adaptation and le
 arning across timescales. At short timescales\, humans respond via a robus
 t default feedback controller that maintains stable locomotion\, and at lo
 nger timescales humans slowly change this controller to optimize a perform
 ance metric\, following a negative gradient estimated from intentional exp
 loration. Our model predicts changes in symmetry\, entrainment\, and energ
 y expenditure in multiple natural and human-machine interfacing tasks. I w
 ill also highlight ongoing work on understanding natural motor learning tr
 ajectories that unfold over the timescale of many months. Across tasks and
  timescales\, I will highlight the inductive biases of an optimization-bas
 ed modeling framework that are crucial to capture human motor learning phe
 nomena.
GEO:42.378796;-71.117354
LOCATION:Maxwell Dworkin\, G125
SUMMARY: Computational principles of motor learning across timescales
URL;VALUE=URI:https://events.seas.harvard.edu/event/computational_principle
 s_of_motor_learning_across_timescales
CATEGORIES:Colloquia / Seminar / Lecture
END:VEVENT
END:VCALENDAR