More recently the orbitofrontal cortex has been shown to be critical to signaling of outcome expectancies—signals concerning the characteristics, features, and specific value of particular outcomes that are predicted by cues (and perhaps responses; though see Ostlund and Balleine 2007b) in the environment (Schoenbaum and Roesch 2005). Here we will argue that this function provides a better explanation for the role of the orbitofrontal cortex in adaptive behavior than either of the established hypotheses. First, we will review data from reversal learning tasks, showing that orbitofrontal cortex is critical for changing behavior in the face of unexpected outcomes. Then, we will provide a brief overview of the two dominant hypotheses, followed by data that directly contradict both accounts. Thereafter, we will review more recent evidence that orbitofrontal cortex is critical to signaling information about expected outcomes. As we will show, these signals are prominent in the neural activity and BOLD response in orbitofrontal cortex, and their role in guiding behavior is evident in deficits caused by orbitofrontal damage in a variety of behavioral settings in which outcomes must be used to guide normal behavior, even when contingencies are not changing. We will suggest that these same signals are also necessary for the detection of prediction errors when contingencies are changing, thereby facilitating changes in associative representations in other brain areas and, ultimately, behavior. Finally, we will also suggest that expectancy signals in orbitofrontal cortex might also impact early sensory regions, optimizing behavior through context-dependent firing and recall of environmental cues predicting future reward.
Copyright © 2011 by Taylor and Francis Group, LLC.