Seminar Series
Home → Seminar Series → Presentation

ICS import...
ICal Link

Wednesday March 7, 2018 -- Engineering proprioception: learning to use artificial sensation to guide movements

SMBB 2650, 12:45 pm

speaker photo

Speaker: Maria Dadarlat , Department of Physiology, University of California, San Francisc

Presentation Abstract:

One of the primary applications of brain-machine interfaces (BMIs) is to restore some degree of motor control to individuals who have lost the ability to move their limbs. Despite great progress in recording technology and decoding algorithms, BMI-based movements remain slow, choppy, and unnatural, in part because the feedback on which existing BMIs rely is, unlike natural motor control, exclusively visual. Indeed, amputees and the paralyzed lack an afferent channel for the somatosensory feedback--touch and proprioception--that plays such an important role in smooth and graceful movements.

Is it, then, possible to provide usable information about the external world directly to the cortex of animals, i.e. bypassing their sensory organs altogether? How? And how is sensory information in the cortex affected by a patientís self-movement? I provide positive answers to these questions with recent experimental results. I show in particular that monkeys can learn to move to invisible targets based solely on information provided by direct electrical stimulation of the cortex, and ultimately optimally integrate visual feedback with this artificial sensory feedback. I then discuss neural coding strategies used by the cortex to increase the information content of sensory signals during self-movement in mice. I conclude by proposing a set of experiments to study the development of sensorimotor integration.

Faculty Host: Chris Butson