Distributed Encoding Architecture in Prefrontal Cortex during Abstracted and Embodied Decision Making

Abstract

On one hand, decision-making can be viewed as a process by which individual functions, such as valuation and choice, are abstracted from movement. On the other hand, decision-making can be viewed as an embodied process by which choice and movement are inexorably linked. In either case, neural activity in a range of cortical structures in the primate prefrontal cortex have been implicated in decision-making. I employ both a traditional neuroeconomic paradigm and a novel free-range foraging paradigm to understand the encoding architecture in prefrontal cortices during both abstract and embodied decision-making. First, I show that in an abstracted decision-making paradigm, constituent and higher-level functional computations are not circumscribed to discrete anatomical boundaries. Indeed, the encoding of both feature information and subjective value are distributed across multiple structures. Next, I show that a range of higher-level choice-relevant functions are also computed in a distributed framework that, along the prefrontal medial wall are organized along a ventral-to-dorsal gradient. Last, I show that navigational and foraging task variables in an embodied decision-making paradigm are distributed across prefrontal cortex, organized along a ventral-to-dorsal gradient, and show no evidence of modular functional specialization by neuronal subpopulations. These results strongly support the need for a dramatic shift in the way we view the organization of functional computations in the brain, and thus inform how we might think about targeting interventions for the treatment of neurological and neuropsychiatric disorders.