Information filtering by synchronous spikes in a neural population

Area of Science:

• Computational Neuroscience
• Neural Coding
• Signal Processing

Background:

• Neurons encode time-dependent sensory stimuli through spike trains.
• Understanding information transfer in noisy neural populations is crucial.

Purpose of the Study:

• To quantify information encoded in synchronous activity of noisy neurons.
• To analyze frequency-dependent information transfer in neural populations.
• To elucidate mechanisms behind information filtering in neural systems.

Main Methods:

• Developed a mathematical framework to calculate coherence between synchronous neural output and common stimulus.
• Modeled populations of uncoupled, noisy neurons driven by a broadband signal.
• Compared Poisson neuron and Leaky Integrate-and-Fire (LIF) neuron models.

Main Results:

• Poisson neuron models exhibited low-pass coherence, limiting high-frequency information transfer.
• LIF neuron models showed a coherence peak at high frequencies, aligning with experimental data.
• Identified the mechanism responsible for the shift in coherence maximum in LIF models.

Conclusions:

• Synchronous activity in LIF neurons can enhance high-frequency information encoding.
• The findings offer insights into neural information processing and filtering mechanisms.
• Results have implications for understanding sensory processing in biological systems.