Conference on Cognitive Computational Neuroscience

Both humans and monkeys are skillful at learning their environments. The neural circuits that govern this sophisticated skill include the orbitofrontal cortex and the amygdala. Here, we provide a preliminary report on monkeys playing a simplified version of the game battleship, designed to investigate learning these complex environments, while neural recordings were performed simultaneously in orbitofrontal cortex and amygdala. Monkeys visually uncovered one shape per trial over multiple choices. Monkeys were adept learners, as assessed against the optimal choice that maximizes expected reward. Analysis of a subset of sorted orbitofrontal neurons suggest a role in learning environmental states. One third of neurons signal the entropy over the distribution of possible states, which changed during learning. In addition, the fraction of neurons that represented locations of parts of shapes doubled during learning. Finally, the dimensionality of the population decreased during learning. Our results suggest that the orbitofrontal cortex forms representations of task states during learning by recruiting neurons into coalitions that increases their covariation and consequently decreases the population dimensionality.