[Re] Non-Additive Coupling Enables Propagation of Synchronous Spiking Activity in Purely Random Networks

A reference implementation of → Non-additive coupling enables propagation of synchronous spiking activity in purely random networks, R.M. Memmesheimer, M. Timme, PLoS Computational Biology, 8(4): e1002384, 2012International audienceDendritic non-linearities increase neurons’ computation capacity, turning them intocomplex computing units [4]. However, network studies are usually based on point neuron models that do not incorporate dendrites and their non-linearities. In contrast, the study replicated here [2] uses a simple point-neuron model that contains an effective description of dendrites by a non-linear summation of its excitatory synaptic input. Due to the simplicity of the model, both large-scale parameter exploration of a medium-sized network, as well as an analytical investigation of its properties are feasible. The original study was based on simulation and analysis code in C and Mathematica, but this code is not publicly available. Here, we replicate the study using the neural simulator Brian 2 [1,6], a simulator based on the Python language that has become a common choice in computational neuroscience [3]. This simulator offers a good trade-off between flexibility and performance and is therefore a suitable choice for this study of a non-standard neuron model