Using EEG to Identify Neural Signatures of Implicit Sequence Learning
Kelsey R. Thompson, Laura Batterink, Ken A. Paller, Paul J. Reber
Implicit learning reflects a reshaping of neural circuits through extensive practiceand occurs largely outside of conscious awareness. The Serial Interception Sequence Learning (SISL) task provides a relatively process-pure measure of implicit motor sequence learning. The task produces little to no explicit recognition of the trained sequence in participants. During the task, participants watch a rapid progression of moving cues and attempt to make precisely timed 4-alternative-forced-choice responses. The cues include a covertly embedded repeating sequence. Participants exhibit improved performance during the repeating sequence compared to novel sequences. Using fMRI, we found reductions in activity during performance of an implicitly learned sequence (Gobel, Parrish, & Reber, 2011). However, the temporal dynamics of neural activity associated with performing a well-practiced sequence have not been previously identified. Using a modified version of the SISL task, participants were trained on a motor sequence and subsequently tested on their sequence knowledge while their EEG was recorded. Brain potentials time-locked to cues were analyzed to reveal distinct responses to the four individual cues. Differences in the ERPs to the same cues within trained and untrained sequences were observed in occipital scalp channels, providing a detailed picture of the neural dynamics supporting implicit learning. The findings indicate that implicit sequence learning produces changes in the way cues comprising a trained sequence are visually processed, which then increases fluidity in response planning and execution.
Click here to view the poster