John P. O’Doherty, California Institute of Technology
The Brain is a Mixture of Experts: How the Brain Allocates Control as a Function of the Reliability of Predictions
It has long been suggested that human behavior can be understood as reflecting the contributions of multiple systems that cooperate or compete for the control of behavior. Here I propose that the brain can be thought of as a “Mixture of Experts” in which different expert systems propose strategies for action. I will then discuss at a theoretical level how the brain determines which system should control behavior at any one moment in time. I will argue that this is accomplished by keeping track of the precision of the predictions within each system, and by allocating control over behavior in a manner that is proportional to the relative reliability of those predictions. fMRI and neurostimulation studies suggest a specific contribution of the anterior prefrontal cortex in this process. Such a mechanism is necessary and sufficient to allocate control over behavior without necessitating explicit considerations about cognitive effort or computational cost. I will review results from the study of different expert systems in both experiential and social-learning domains that hints at the possibility that this reliability-based control mechanism may be domain general, exerting control over many different expert systems simultaneously in order to produce sophisticated behavior.