Modularization dramatically increases stability of oscillating attractor networks
Friday 27 November 2009
to 12:00 at
Mikael Lundqvist (Computational Biology and Neurocomputing / CSC / KTH)
Attractor neural networks are thought to underlie working memory functions in the cerebral cortex. Several such models have been proposed that successfully reproduce firing properties of neurons recorded from monkeys performing working memory tasks. However, the regular temporal structure of spike trains in these models is often incompatible with experimental data. Here, we show that the in vivo observations of bistable activity with irregular firing at the single cell level can be achieved in a large-scale network model with a modular structure in terms of hypercolumns. Despite high irregularity of individual spike trains, the model shows population oscillations in the beta and gamma band in ground and active states respectively. Irregular firing typically emerges in a high-conductance regime of balanced excitation and inhibition. Population oscillations can produce such a regime, but in previous models only a non-coding ground state was oscillatory. Due to the modular network structure comprising several connected hypercolumns, the oscillatory regime is much more stable in our network and also the active state is oscillatory. The model therefore maintains oscillatory and irregular firing also in the memory retrieval state without fine-tuning. It provides a novel mechanistic view of how irregular firing emerges in cortical populations as they go from beta to gamma oscillations during memory retrieval.