%0 Journal Article %A Benjamin Merkt %A Friedrich Schüßler %A Stefan Rotter %T Propagation of orientation selectivity in a spiking network model of layered primary visual cortex %D 2018 %R 10.1101/425694 %J bioRxiv %P 425694 %X We studied the propagation of orientation selectivity over layers in a model of rodent primary visual cortex in terms of the underlying connectivity. While significant progress has been made in measuring the activity in the different sub-populations as well as the connectivity between those in experiments, a comprehensive theoretical explanation of how network structure and its dynamics are related is still missing. We suggest a model of layered mouse visual cortex by extending the model suggested by Potjans and Diesmann (2014) with thalamic input that has an orientation bias according to Sadeh et al. (2014). We then studied the response properties of the network to such non-homogeneous input. We found that, without further assumptions, the connectivity derived from experimental data leads to layer-specific distributions of orientation selectivity very similar to what has been observed in mouse experiments. Interestingly, the network settles in a dynamical operating point, in which the efficacy of the different projections and their contribution to orientation tuning deviate strongly from the expectations based on the underlying anatomical connectivity. To understand the processes that shape the observed dynamics, it is essential to perform the analysis of the microcircuit on the system level. With that perspective, we find in particular that the difference in tuning of L2/3 and L4 neurons is an immediate, albeit unexpected consequence of the specific network connectivity. Furthermore, we introduce a novel method for predicting the effects of optogenetic stimulation of specific neuronal sub-populations and demonstrate its power in network simulations.Significance Statement Understanding the precise roles of neuronal sub-populations in shaping the activity of neuronal networks is a fundamental objective of neuroscience research. To this end, our work makes three important contributions. First, we show that the experimentally extracted connectivity suffices to explain the degree of selectivity of sub-populations in mouse visual cortex to visual stimulation. Second, we introduce a novel system-level approach for the analysis of input-output relations of recurrent networks, which lead to distinct activity patterns. Third, we present a method for the design of optogenetic experiments that can be used to devise specific stimuli which result in a desired and predictable change of neuronal activity.This work was partially supported by the European Union’s Seventh Framework Programme (FP7/2007-2013) under Grant Agreement 600925 (NeuroSeeker), by the BMBF (grant BFNT 01GQ0830) and DFG (grant EXC 1086). The HPC facilities are funded by the state of Baden-Württemberg through bwHPC and DFG grant INST 39/963-1 FUGG. In addition, our work was promoted by the German Academic Exchange Service (DAAD) and the Carl Zeiss Stiftung. We thank Uwe Grauer from the Bernstein Center Freiburg as well as Bernd Wiebelt and Michael Janczyk from the Freiburg University Computing Center for their assistance with HPC applications. %U https://www.biorxiv.org/content/biorxiv/early/2018/09/25/425694.full.pdf