Date/Time:  Friday April 25, 2-4pm EST

In-person location:  46-6011 (Simons Center for the Social Brain)

Zoom link:  https://mit.zoom.us/j/3188129440

Title:  Multi-Object Working Memory and Motion Prediction in the Primate Brain

Abstract:

Primates excel at rapid learning and flexible generalization in many domains.  For example, we can learn how to drive a car after just a few hours of practice, then readily generalize when driving an unfamiliar car.  Rapid learning and flexible generalization are thought to rely on a reusable mental model of the world in terms of objects and how objects move.  However, the mechanisms by which the brain represents scenes of objects and predicts their motion are largely mysterious.  To address this knowledge gap, we studied (i) how the brain represents multi-object scenes, and (ii) how the brain predicts the motion of an object through time.  For both of these questions, we considered long-standing cognitive theories and tested these theories in neural data, namely neural activity recorded from monkeys trained on multi-object representation and prediction tasks.  In this defense I will present the outcomes of this work, focusing primarily on the neural mechanisms of motion prediction.