Max Planck Institute for Dynamics and Self-Organization -- Department for Nonlinear Dynamics and Network Dynamics Group
Personal tools
Log in

BCCN AG-Seminar

Tuesday, 21.11.2006 16 c.t.

Neural dynamics and computations in decision making process

by Prof. Dr. Tomoki Fukai
from Neural Circuit Theory, RIKEN Brain Science Institute, Japan

Contact person: J. Michael Herrmann

Location

Seminarraum Haus 2, 4. Stock (Bunsenstr.)

Abstract

Animals or humans often encounter such situations in which they must choose their behavioral responses to be made in near or distant future. Such a decision is made through continuous and bidirectional interactions between the environment surrounding the brain and its internal state or dynamical processes. To make a decision, the brain must analyze pieces of information given externally, the past experiences in a similar situation, possible behavioral responses, and predicted outcomes of the individual responses. I will discuss two different aspects of decision making, presenting results of our recent attempts to model the neural mechanisms of decision processes. The first part is devoted to the neural mechanism of animal’s choice behavior among various options. Typical psychological experiments of alternative choice tasks assign different reinforcement schedules or probabilities to different choice options. In many such situations, the subject’s choice behavior exhibits the so-called matching law. We discuss how this behavior may appear from the temporal difference algorithm of reinforcement learning. In fact, such a system often (but not always) exhibits the matching behavior even in a situation in which the matching behavior is only sub-optimal. In the second part, I will briefly discuss the neural mechanism that typically determines the timing of behavioral responses in decision making. Here, temporal integration of externally or internally driven information plays a crucial role, and several models have already been proposed in literature. I demonstrate how a stochastic recurrent network of spiking neurons with a binary response property can perform the temporal integration of synaptic input. We test the predictions of this model in the graded activities recorded from the monkey anterior cingulate cortex.

back to overview