Theory of correlation in a network with synaptic depression

Area of Science:

• Computational neuroscience
• Theoretical neuroscience

Background:

• Synaptic depression influences neuronal mean responses and response variability correlations.
• Existing theories of neural correlation in spiking neuron models do not account for synaptic depression.

Purpose of the Study:

• To extend theoretical frameworks of neural correlation to include short-term synaptic depression.
• To analytically calculate neural correlations in a network model with synaptic depression.

Main Methods:

• Developed a theoretical framework for spiking neuron models incorporating short-term synaptic depression.
• Applied mean-field theory to analyze neural correlations.
• Utilized a ring attractor network with Mexican-hat connectivity as a model for the primary visual cortex.

Main Results:

• Synaptic depression was found to reduce neural correlations within the network.
• The reduction in neural correlation due to synaptic depression may enhance sensory coding.
• The study provides a theoretical basis for investigating the impact of altered network interactions on response functions.

Conclusions:

• Short-term synaptic depression plays a significant role in modulating neural correlations.
• Reduced neural correlation driven by synaptic depression could be a mechanism for improved sensory information processing.
• This work advances theoretical neuroscience by incorporating synaptic dynamics into correlation analysis.