Navigation and service

Correlations in cortical networks

  • Correlation structure in cortical networks
  • Active decorrelation by recurrent network dynamics
  • Effect of realistic network connectivity

Correlations in the activity of cortical networks have attracted attention in the past for a variety of reasons: The spatio-temporal activity pattern generated by a recurrent neural network can provide a rich dynamical basis which allows readout neurons to produce a variety of responses by tuning the synaptic weights of their inputs. The repertoire of possible responses and the response reliability become maximal if the spike trains of individual neurons are uncorrelated. Theoretical and experimental studies have shown that spike correlations in cortical networks can indeed be smaller than one would expect based on the cortex anatomy. On the other hand, it has been argued that correlated firing, triggered by internal or external events, could be useful for cortical information processing: Spike correlations can modulate the gain of individual neurons and thereby serve as a gating mechanism. Information represented by correlated firing can be reliably sustained and propagated through functional subnetworks. Coherent activity has been suggested as a means to bind elementary representations into more complex objects. From several perspectives, it is therefore important to understand the origin and dynamics of correlations. Previous studies have provided the theoretical tools to describe how correlations are shaped by the interplay between the architecture and dynamics of simple recurrent neural networks. Currently, we exploit and extend this knowledge to study correlations in more realistic cortical network models.

suppression of correlationCopyright: corrtrans manuscript

Own project related publications:

Tetzlaff T., Helias M., Einevoll G.T., Diesmann M., Decorrelation of neural-network activity by inhibitory feedback, submitted

Tetzlaff T., Rotter S., Stark E., Abeles M., Aertsen A., Diesmann M. (2008), Dependence of neuronal correlations on filter characteristics and marginal spike-train statistics, Neural Computation 20(9): 2133-2184

Kriener B., Tetzlaff T., Aertsen A., Diesmann M., Rotter S. (2008), Correlations and population dynamics in cortical networks, Neural Computation 20(9):2185 2226

Tetzlaff T., Buschermöhle M., Geisel T., Diesmann M. (2003), The spread of rate and correlation in stationary cortical networks, Neurocomputing 52-54:949-954