Correlations between prefrontal neurons form a small world network that optimizes the generation of multineuron sequences of activity

Sequential patterns of prefrontal activity are believed to mediate important behaviors, e.g. working memory, but it remains unclear exactly how they are generated. In accordance with previous studies of cortical circuits, we found that prefrontal microcircuits in young adult mice spontaneously generate many more stereotyped sequences of activity than expected by chance. However, the key question of whether these sequences depend on a specific functional organization within the cortical microcircuit, or emerge simply as a byproduct of random interactions between neurons, remains unanswered. We observed that correlations between prefrontal neurons do follow a specific functional organization; they have a small world topology. However, until now it has not been possible to directly link small world topologies to specific circuit functions, e.g., sequence generation. Therefore, we developed a novel analysis to address this issue. Specifically, we constructed surrogate datasets that have identical levels of network activity at every point in time, but nevertheless represent various network topologies. We call this method SHuffling Activity to Rearrange Correlations (SHARC). We found that only surrogate datasets based on the actual small world functional organization of prefrontal microcircuits were able to reproduce the levels of sequences observed in actual data. As expected, small world datasets contained many more sequences than surrogate datasets with randomly arranged correlations. Surprisingly, small world datasets also outperformed datasets in which correlations were maximally clustered. Thus, the small-world functional organization of cortical microcircuits, which effectively balances the random and maximally clustered regimes, is optimal for producing stereotyped sequential patterns of activity.

from Physiology via xlomafota13 on Inoreader http://ift.tt/1RPD4zD
via IFTTT