Distinct High-Gamma Signals in Primate Prefrontal Cortex Differentiate Cue Information, Preference Revision, and...

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Computational Neuroscience

Background:

• Understanding prefrontal cortex (PFC) signals in decision-making requires analyzing neural activity's spectral and temporal structure.
• The role of high-gamma (Hγ) power in local field potentials (LFPs) for differentiating decision variables like reward expectation remains unclear.

Purpose of the Study:

• To characterize how Hγ power in PFC subregions evolves over time to support reward-guided decision-making.
• To investigate Hγ activity's role in decoding preference revision, reward expectation, cue value, and prediction errors.

Main Methods:

• Recorded LFPs from anterior cingulate cortex (ACC), dorsolateral PFC (DLPFC), and orbitofrontal cortex (OFC) in macaques during a multicue reward decision task.
• Extracted Hγ power using sliding windows and employed support vector machine classifiers to decode various decision variables.

Main Results:

• Hγ responses differentiated preference-reversal from confirmation trials across PFC subregions.
• Cue value and position were decoded across PFC, with regional specializations (OFC for value, DLPFC for position).
• Hγ activity distinguished expected reward levels (85% accuracy in ACC) and signaled unexpected reward omission (78% accuracy in OFC).

Conclusions:

• Hγ activity in PFC subregions robustly differentiates multiple decision variables with distinct temporal dynamics.
• Hγ power serves as a reliable marker for dynamic, reward-guided decision processes in the brain.