David Beniaguev

The paper demonstrates that detailed models of rat pyramidal neurons can perform complex computational tasks using a novel algorithm, highlighting dendritic nonlinearities as key to high-order feature binding.

The paper introduces the Functional Complexity Index (FCI) to quantify neuronal functional complexity and finds human cortical pyramidal neurons are more complex than rat neurons due to morpho-electrical differences.

The paper develops a 'Filter-and-Fire' neuron model that shows increased memory capacity and ability to recognize spatio-temporal patterns through multiple synaptic contacts on dendrites.

The paper trains deep neural networks to model the input-output behavior of cortical pyramidal neurons, revealing their functional complexity as multi-layered processing units.