Synaptic plasticity in a cortical microcircuit model: different - PowerPoint PPT Presentation

Synaptic plasticity in a cortical microcircuit model: different scenarios Renan O. Shimoura, Antonio C. Roque Physics Department, FFCLRP, University of São Paulo, Ribeirão Preto, SP, Brazil Laboratório de Sistema Neurais (SisNe) renanshimoura@usp.br

Introduction

Primary visual cortex • As the entire neocortex, V1 is anatomically divided into six layers, where each layer has different types and numbers of neurons. • Synaptic plasticity is thought to be the underlying mechanism behind learning and memory. • There are neurons in V1 which its response is selective to angular orientation.

Goal • How does the synaptic plasticity affect the orientation selectivity of the network?

Methods

The network 10,000 neurons ~ 5 million synapses Excitatory neurons Excitatory Poissonian synapses noise (8 Hz) Inhibitory neurons Inhibitory synapses Excitatory/inhibitory ratio = 4:1 Potjans TC, Diesmann M (2014).

Stochastic neuron model (GL model) Synaptic increment w ij i j

Izhikevich (2007). Galves, A., Löcherbach, E. (2013).

Asymmetric spike-timing-dependent plasticity (STDP) rule %∆w t pre -t post * Song S, Miller KD, Abbott LF (2000). Competitive Hebbian learning through spike- timing-dependent synaptic plasticity.

Simulations - Duration of simulation: 10000 ms; - 1st: Poissonian spike trains applied as background with 8 Hz; - 2nd: visual stimuli applied at L4 as angular oriented pulses; ∗ ) 𝐽 𝑓𝑦𝑢,𝑗 = 𝐽 ∙ cos (𝜄 𝐽 − 𝜄 𝑗 Orientation selectivity index (OSI): 2 + ( 𝜄 𝑔 𝜄 𝑔 𝑗 𝜄 𝑑𝑝𝑡 2𝜄 𝑗 𝜄 𝑡𝑓𝑜 2𝜄 )² 𝑃𝑇𝐽 𝑗 = 𝜄 𝑔 𝑗 (𝜄) OSI = 0 → The neuron fires for any stimuli. OSI = 1 → The neuron fires preferentially to one angle.

Preliminary results

Control (no STDP) With STDP L23e L23e L23i L23i L4e L4e L4i L4i L5e L5e L5i L5i L6e L6e L6i L6i

Orientation Selectivity Control STDP Index (OSI) L23 L23 #n: 1.86 % #n: 15.20 % L4 L4 #n: 13.51 % #n: 0.56 %

Orientation Selectivity Control STDP Index (OSI) L5 L5 #n: 0.00 % #n: 20.51 % L6 L6 #n: 6.55 % #n: 4.84 %

Partial conclusion • In the first case, the network with STDP higher average frequency; • STDP can improve the orientation selectivity in this network.

References

References  Potjans TC, Diesmann M (2014). The cell-type specific cortical microcircuit: relating structure and activity in a full-scale spiking network model. Cereb. Cortex , 24;785-806.  Izhikevich EM (2007). Dynamical Systems in Neuroscience: The Geometry of Excitability and Bursting. MIT Press, Cambridge , MA.  Galves, A., Löcherbach, E. (2013). Infinite systems of interacting chains with memory of variable length: a stochastic model for biological neural nets. J. Stat. Phys. 151:896-921.  Song S, Miller KD, Abbott LF (2000). Competitive Hebbian learning through spike-timing-dependent synaptic plasticity. Nat Neurosci 3(9):919-926.

Control (no STDP) V ( mV)

With STDP V ( mV)

Control STDP