Changes in the stimulus-preceding negativity and lateralized readiness potential during reinforcement learning

Area of Science:

• Cognitive Neuroscience
• Neuroscience of Learning
• Motor Skill Acquisition

Background:

• Dopamine-dependent anticipatory processes are crucial for learning from rewards and feedback in reinforcement learning.
• Understanding outcome anticipation is key to elucidating the neural mechanisms underlying learning.
• Slow cortical potentials (SCPs) are electrophysiological measures sensitive to cognitive processes.

Purpose of the Study:

• To examine variations in slow cortical potentials during motor-skill acquisition.
• To investigate how feedback delay influences neural activity related to outcome anticipation.
• To test hypotheses about the role of the stimulus-preceding negativity (SPN) in outcome prediction.

Main Methods:

• Healthy young adults learned timed key press sequences with delayed feedback (2.5s or 8s).
• Electroencephalography (EEG) was used to record slow cortical potentials, including the stimulus-preceding negativity (SPN).
• Analysis focused on changes in SPN and lateralized readiness potential (LRP) amplitudes across learning trials.

Main Results:

• The SPN amplitude decreased across trials during the 2.5s feedback delay, contradicting the outcome prediction hypothesis.
• The SPN amplitude increased across trials during the 8s feedback delay.
• A reversed-polarity lateralized readiness potential (LRP) was observed at the end of the 8s prefeedback interval, potentially indicating action representation maintenance.

Conclusions:

• The findings suggest that the SPN's role in outcome anticipation may depend on the learning system engaged (habit vs. episodic memory).
• The observed LRP may provide the first direct psychophysiological evidence for the 'eligibility trace' construct in reinforcement learning models.
• This study offers novel insights into the neural dynamics of motor learning and reward anticipation under different delay conditions.