Triphasic dynamics of stimulus-dependent information flow between single neurons in macaque inferior temporal cortex

Area of Science:

• Neuroscience
• Computational Neuroscience

Background:

• Cortical neuronal connectivity is dynamic, not static.
• Functional coupling strength varies with context.

Purpose of the Study:

• To investigate complex forward-and-back information flow dynamics in cortical circuits.
• To test if Granger causality analysis reveals temporal patterns in neuronal communication.

Main Methods:

• Simultaneous neuronal spike train recordings from macaque inferior temporal (IT) cortex during a visual object discrimination task.
• Granger causality analysis applied to neuronal spike trains.
• Cross-correlation function (CCG) analysis to characterize neuron pair interactions.

Main Results:

• Displaced CCG peaks showed gamma-frequency Granger causality, initially in the forward direction.
• Granger causality exhibited a triphasic pattern: forward, backward, then forward.
• Causality dynamics were independent of firing rate changes and varied with object features.

Conclusions:

• The classical view of functional connectivity needs expansion to include complex bidirectional dynamics.
• Multistage visual object recognition may involve multiphasic directional information flow in IT cortex circuits.