IBYME   02675
INSTITUTO DE BIOLOGIA Y MEDICINA EXPERIMENTAL
Unidad Ejecutora - UE
congresos y reuniones científicas
Título:
Correlation structure reshaping during response selection in the rat prefrontal cortex.
Autor/es:
MININNI C. J; ZANUTTO S; LEW S. E
Lugar:
San Diego California, USA
Reunión:
Congreso; Annual Meeting of the Society for Neuroscience; 2013
Institución organizadora:
Society for Neuroscience.
Resumen:
Correlation structure reshapes during response selection in the rat Prefrontal Cortex Camilo J. Mininni1, B. Silvano Zanutto1,2, Sergio E. Lew2 1.Instituto de Biología y Medicina Experimental (IByME-CONICET); 2.Instituto de Ingeniería Biomédica (FI-UBA) The Prefrontal Cortex (PFC) and Ventral Tegmental Area (VTA) are key brain regions for understanding goal directed behavior. The PFC is proposed to be a decision making structure, integrating sensory information and initiating behavioral responses, while the VTA is involved in the processing of reward information, as well as information about reward-predicting stimuli. In the present work, 4 Long Evans rats were trained in a GO/NOGO sound discrimination task under a head-restrained paradigm. Neural population activity in the PFC and the VTA was recorded simultaneously during the discrimination task. A pool of 248 single cells (95 from PFC and 153 from VTA) were selected for analysis, and the tone-related activity of each area was assessed. After performance reached 80%, firing rates showed different dynamics between areas and GO or NOGO trials. Both PFC and VTA neurons steadily increased their firing rate during tone presentation in GO trials, while the NOGO trials showed a transient increase, returning then to their baseline level. On the other hand, paired correlations between PFC neurons changed during the decision period. During GO trials correlations increased significantly whereas in NOGO trials correlations decreased. In the same line, normalized variance showed more variability in NOGO trails than in GO ones. In order to understand the collective activity of PFC neurons, we built one Ising model for GO trials and another for NOGO trials, whose main parameters (local field and dipole interactions) were derived from PFC data. In terms of entropy of the model, we found that 1) in GO trials the model walks along a more reduced set of states than in NOGO trials, 2) Lowering 50% the temperature in the model for NOGO trial makes both, GO and NOGO models equivalent. In these terms, the NOGO action would require a PFC with a broader repertoire of possible states, which correlate with an increase in VTA firing. Our results bring to light the interplay between first order (firing rates) and second order (pairwise correlations) statistics of the activity at the PFC during the time of response selection.