Pre-stimulus alpha oscillations contain representations of expected visual shapes

Poster Session F - Tuesday, April 16, 2024, 8:00 – 10:00 am EDT, Sheraton Hall ABC
Also presenting in Data Blitz Session 4 - Saturday, April 13, 2024, 1:00 – 2:30 pm EDT, Sheraton Hall EF.

Dorottya Hetenyi¹ (dorottya.hetenyi.21@ucl.ac.uk), Peter Kok¹; ¹Wellcome Centre for Human Neuroimaging, University College London

Our prior knowledge greatly influences how we perceive the world. However, it remains unclear how the brain keeps predictions online prior to stimulus onset. Here, we combined magnetoencephalography (MEG) and decoding techniques to investigate the neural dynamics of pre-stimulus sensory predictions. Participants were engaged in a shape discrimination task, while auditory cues predicted which specific abstract shape would likely appear. We trained a shape decoder on data from a separate localiser run, and applied this decoder to the time window after the predictive auditory cues, but before shape onset, to test for neural representations of expected shapes. Frequency analysis revealed significant oscillatory fluctuations in the pre-stimulus decoded time series, predominantly in the alpha band (10 – 11Hz). We created baseline measurements of alpha power by training the decoder 1) with pseudo-randomised labels and 2) on different shapes than those used in the main experiment, to rule out effects of non-specific alpha oscillations. Furthermore, we found that this stimulus-specific alpha power was linked to expectation effects on behavioural accuracy and on post-stimulus neural shape representations. Previous research has already shown that the raw power of alpha oscillations modulate perception. While our results are consistent with these observations, they additionally demonstrate that alpha fluctuations can contain stimulus-specific contents which predict behavioural performance and sensory encoding. Together, these findings show that sensory predictions are embedded in the alpha frequency band and can modulate perception through their oscillatory dynamics, providing a neural mechanism through which the brain generates and deploys predictions.

Topic Area: PERCEPTION & ACTION: Vision