MIT CogNet, The Brain Sciences ConnectionFrom the MIT Press, Link to Online Catalog
SPARC Communities
Subscriber : Stanford University Libraries » LOG IN

space

Powered By Google 
Advanced Search

 

Building Neural Representations of Habits: Deep Brain Learning and Memory

 Ann Graybiel
  
 

Abstract:
Much of our normal behavior depends on learning how to perform entire action sequences so smoothly that we carry them out almost without conscious effort. This type of learning, known as procedural or habit learning, is critical for maximizing cognitive function. We depend on habits to free us to think and plan and to react to novel events in the environment. Clinical and experimental evidence suggests that our ability to acquire habits depends on the basal ganglia, deep forebrain structures that are interconnected with the frontal cortex in a series of loop circuits. Almost nothing is yet known, however, about the type of neural processing that lets us transform behaviors into habits. Nor do we understand neural activity that allows us to perform complex acts automatically once habits are formed. Our laboratory is focusing on these two issues.

In rats, we are recording chronically from the striatum with multiple tetrodes as the animals undergo training on a T-maze. We find that large and widely distributed changes in neuronal activity patterns occur in the sensorimotor striatum during behavioral acquisition, culminating in task-related activity emphasizing the beginning and the end of the automatized procedure. These new ensemble patterns remain stable over weeks of subsequent performance of the same task, and are accompanied by increases in temporal coordination of spike activity. These results suggest that a dynamic reorganization of population activity of striatal neurons occurs as habit learning proceeds.

In monkeys, we are recording the patterns of neuronal activity in frontal cortex and striatum as monkeys perform sequential saccade tasks under conditions that favor either an attentive, reactive mode of performance or an increasing automization of the saccade performance. We are finding striking modulations of firing frequency of supplementary eye field (SEF) neurons as the monkeys switch from the reactive to the semi-automatic mode of saccade performance. Initial recordings in the striatum of these monkeys suggest that mode-specific modulation patterns occur there as well. These experiments suggest that cortico-basal ganglia loop circuits carry signals reflecting the stepwise release of performance from reactive, attentional demands as performance becomes increasingly.

 
 


© 2010 The MIT Press
MIT Logo