Cortical reliability amid noise and chaos.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Cortical neuron responses to identical stimuli are typically variable, leading to the rate coding hypothesis.
• Spike-time coding has been proposed but its role is limited by unexplored internal network variability.
• Understanding internally generated variability is crucial for deciphering neural coding.

Purpose of the Study:

• To quantify internally generated variability in cortical circuits.
• To investigate the role of noise and network dynamics in neural coding.
• To reconcile the apparent contradiction between network chaos and reliable spike timing.

Main Methods:

• Developed a biophysical model of rat neocortical microcircuitry.
• Incorporated biologically realistic noise sources, focusing on stochastic neurotransmitter release.
• Analyzed network dynamics under varying input conditions.

Main Results:

• Stochastic neurotransmitter release was identified as a key driver of internal variability and chaotic recurrent network dynamics.
• Nonlinear recurrent dynamics were found to transiently overcome chaos with weak thalamocortical inputs.
• Reliable spike times with millisecond precision were supported by the model.

Conclusions:

• Noisy and chaotic dynamics of recurrent cortical microcircuitry are compatible with stimulus-evoked, millisecond spike-time reliability.
• This resolves the long-standing debate between rate coding and spike-time coding in the cortex.
• The study highlights how internal network properties can support precise neural information processing.