2023 Distinguished Career Award Winner

Friday, April 21: 17:00 – 18:00

Abstract

“The cerebellum – prediction in motor control and cognition”

The role(s) of the cerebellum are still uncertain. A prominent theory is that the cerebellum holds a predictive internal model of the sensory-motor system. This is a crucial component in the process of state estimation, combining information from descending motor commands and ascending sensory afferent signals to predict the outcome of actions, helping determine the current state of the motor system. Without state estimation, feedback delays in sensory pathways would degrade performance. State estimate is also likely to underpin coordinated actions, again overcoming feedback delays to allow synchronicity of different effectors. This role would explain the contribution the cerebellum makes to control action, as well as its obvious importance for and dependence on learning and adaptation.  Inaccuracy in state estimates would lead to hypometria and dyscoordination like that observed in cerebellar ataxia. The fundamental role of prediction in state estimation also suggests a wider role for the cerebellum in non-motor domains.

In this lecture I describe studies from my laboratory that support this theory, using laboratory-based motor tasks to explore movement control and coordination. Our motor studies have included testing the effects of artificially extending visual feedback delays, adaptation tasks, and recording, inactivating, and imaging cerebellar activity. We have also used transcranial magnetic and electrical stimulation to disrupt its operations, in movement and in cognitive tasks. I will end with some recent experiments testing novel “event related” methods of TDCS to enhance motor adaptation.

Abstract

“Understanding how the brain controls movement through neural manifolds”

The activity of populations of single neurons underlying behavior is well captured by relatively few population-wide patterns. Intriguingly, the study of these activity patterns—the “latent dynamics”—has shed light into questions about cognition, motor control, and learning that had remained elusive when focusing on the activity of individual neurons.

In this talk, I will give an overview of our work to understand the neural basis of motor control and learning from this perspective. This research is based on the hypothesis that the latent dynamics arise from “neural manifolds” that reflect fundamental biophysical constraints on circuit function. Under this assumption, first I will show that animals generate the same latent dynamics as they perform the same covert or overt behavior on different days, which we can uncover even when we record from different neurons. Second, I will show how adopting this framework reveals large similarities in neural manifolds underlying a variety of reaching and wrist manipulation tasks, even if the properties of single neurons change dramatically across them. Third, I will provide evidence that adopting this framework helps identify region-specific contributions to motor adaptation. Finally, I will present recent results showing that even if each animal has a brain that is unique, individuals from the same species that are engaged in the same behavior share preserved latent dynamics.

Thus, the study of neural manifolds and their associated latent dynamics provides insights into how individual animals both consistently and flexibly perform a variety of behaviors, and may enable principled studies across groups of individuals, and even comparative studies across different species.