Characterizing and dissociating multiple time-varying modulatory computations influencing neuronal activity.

Area of Science:

• Computational Neuroscience
• Systems Neuroscience
• Primate Vision Research

Background:

• Neuronal responses are influenced by numerous cognitive factors, making modeling complex.
• Existing models struggle to capture time-varying modulatory effects on the millisecond timescale.
• Perisaccadic visual responses in primates show complex changes not fully explained by current models.

Purpose of the Study:

• To develop a computational model for dissociating multiple time-varying modulatory effects on neuronal responses.
• To analyze millisecond-scale modulations in extrastriate visual responses during saccades.
• To quantitatively characterize the contribution of individual modulatory sources to perisaccadic responses.

Main Methods:

• Developed a novel extension of the generalized linear model (GLM) framework, termed the sparse-variable GLM.
• Utilized a high spatiotemporal resolution experimental paradigm in nonhuman primates.
• Employed a factorization of the sparse-variable GLM to dissociate and quantify modulatory source contributions.

Main Results:

• The sparse-variable GLM accurately captured millisecond-timescale temporal evolution of neuronal receptive fields during saccades.
• The model successfully dissociated and quantified contributions of multiple sources to perisaccadic visual responses.
• Demonstrated precise tracking of neuronal sensitivity changes around the time of saccades.

Conclusions:

• The novel sparse-variable GLM framework provides a powerful tool for analyzing rapid, complex modulations in neural activity.
• This approach enables quantitative tracking of multiple modulatory influences over time.
• Offers new insights into the neural mechanisms underlying cognitive control of sensory processing during natural behaviors like eye movements.