NEURAL CORRELATES OF LEARNING IN THE PREFRONTAL CORTEX OF THE MONKEY - A PREDICTIVE MODEL

The principles underlying the organization and operation of the prefro ntal cortex have been addressed by neural network modeling. The involv ement of the prefrontal cortex in the temporal organization of behavio r can be defined by processing units that switch between two stable st ates of activity (bistable behavior) in response to synaptic inputs. L ong-term representation of programs requiring short-term memory can re sult from activity-dependent modifications of the synaptic transmissio n controlling the bistable behavior. After learning, the sustained act ivity of a given neuron represents the selective memorization of a pas t event, the selective anticipation of a future event, and the predict ability of reinforcement. A simulated neural network illustrates the a bilities of the model (1) to learn, via a natural step-by-step trainin g protocol, the paradigmatic task (delayed response) used for testing prefrontal neurons in primates, (2) to display the same categories of neuronal activities, and (3) to predict how they change during learnin g. In agreement with experimental data, two main types of activity con tribute to the adaptive properties of the network. The first is transi ent activity time-lacked to events of the task and its profile remains constant during successive training stages. The second is sustained a ctivity that undergoes nonmonotonic changes with changes in reward con tingency that occur during the transition between stages.