Variations of horizontal cortical network structures and their corresponding state space dynamics
see a follow-up publication on Complex dynamics in recurrent cortical networks based on spatially realistic connectivities.
- Nicole Voges, Laurent Perrinet. Variations of horizontal cortical network structures and their corresponding state space dynamics, URL . 2011 abstractNeuronal wiring in the cortex exhibits a complex spatial pattern composed of local and long-range patchy connections [1,2]. Most studies on cortical networks dynamics, however, are either based on purely random wiring or neighborhood couplings [3,4,5]. Assuming an enlarged spatial scale we analyzed the effect of different horizontal connectivities on the 'idle' dynamics of cortical networks. We considered purely random or purely local couplings, i.e. distance dependent connectivities, as well as mixed network architectures that also include spatially clustered projections. Our networks consisted of ~50.000 conductance based integrate-and-fire neurons, spatially embedded in a 2D sheet of cortex with a side length of five millimeters. Network dynamics were simulated with NEST/PyNN . Focusing on the regularity and synchrony in neuronal spiking, we compared the spatio-temporal activity patterns and the phases spaces of such network architectures. Different dynamical states (e.g. synchronous regular or asynchronous irregular firing) occurred, in dependence of the input rate and the relation between exc. and inh. synaptic strengths [3,4,5]. Assuming distance-dependent synaptic delays led to a first set of changes in the phase space of random networks . In addition, we found that networks including local coupling exhibited higher firing rates, sharper transitions, as well as various types of complex network activities. In contrast to random networks, models with distance dependent connectivity architectures exhibited wave propagations. In order to capture such activity patterns we computed a delay-dependent correlation coefficient. The effect of including patchy projections, however, was not detectable from our analysis. Furthermore, to account for stability, we applied spatially restricted activity injections. Depending on the network architecture, the dynamics changed from a low activity state to high firing rates, and then switched back or not. This work was supported by EU Grant 15879 (FACETS).1. Binzegger, Douglas, Martin (2007) J. of Neurosci, 27(45):12242-12254.2. Voges, Schüz, Aertsen, Rotter (2010) Prog Neurobiol, 92(3): 277-292.3. Kumar, Schrader, Aertsen, Rotter (2008) Neural Computation, 20: 1-43.4. Voges & Perrinet (2009) J Phys Paris 104: 51-60.5. Brunel (2000) J. Comput. Neurosci 8(3):183-208.6. Gewaltig MO, Diesmann M: Nest. Scholarpedia, 2(4):1430..
All material (c) L. Perrinet. Please check the copyright notice.
This work was supported by European integrated project FP6-015879, "FACETS".