The University of Arizona
NSMA Neural Systems, Memory and Aging Division of ARL

Society for Neuroscience Research Abstracts

Selective expression of neural activity states preceding and following reward in rat medial prefrontal cortex.

*D. R. EUSTON, M. TATSUNO, B. L. MCNAUGHTON

ARL NSMA, Univ. Arizona, Tucson, AZ

The dorosomedial prefrontal cortex (dmPFC) is strongly interconnected with the nucleus accumbens and other areas known to be involved in the evaluation of reward. In addition, the area receives strong dopaminergic innervation from the VTA, which also plays a role in evaluating expected and unexpected reward. In the present work, principal components analysis was used on binned firing rates from a group of 60-120 simultaneously recorded cells from the dmPFC of rats run on a highly familiar, memory-based spatial sequence task. Either medial forebrain stimulation or food was used as reward. The pattern of firing across all cells at a single time slice is referred to as a "state vector". Plots of the distribution of state vectors on the first and second principal components revealed two distinct lobes which were shown to correspond uniquely to the period (100-500 msec) immediately before reward acquisition or immediately after reward. Average state vectors immediately before and after reward are roughly orthogonal, meaning that the group of cells responding to reward approach is distinct from that which respond after reward receipt. For each cell, a selectivity index was computed, indicating the degree to which it responded in the period immediately before or after reward. The distribution of these values showed that a subpopulation of cells where solely responsive during the “before-reward” period. A much smaller population was responsive solely during the after-reward period. While previous studies have noted the presence of cells in dmPFC which respond on approach to reward, the present study shows that reward proximity is a central determinant of the responses of cells in this area and that the region strongly discriminates reward approach and reward departure.

Grant/Other Support: MH046823; NS020331

Keyword (Complete): memory; consolidation; ensemble recording; principal components