Oscillatory multi-timescale mechanisms underlying multisensory sequence prediction

Peng Wang,Alexander Maÿe,Jonathan Daume,Gui Xue,Andreas K. Engel
DOI: https://doi.org/10.1101/778969
2024-09-23
Abstract:Neural oscillations have been proposed to be involved in predictive processing, however less addressed in the multisensory context. In the present study, we recorded cortical activity using magnetoencephalography during an audiovisual serial prediction task in which participants had to acquire stimulus sequences and to monitor whether subsequent probe items complied with the sequence. In a given trial, either visual or auditory features of the bimodal stimuli were task-relevant. Task-related changes of power and coupling of neural oscillations were analyzed using a data-driven clustering approach. We observed prediction-related theta-band coupling in a network involving the cingulate, premotor, prefrontal, and superior temporal regions. The behavioral performance of participants varied as a function of the phase delay of the coupling between different regions in this network. Two additional clusters were detected in which beta power was stronger if the previous or the following stimuli were the same as the current one in the sequence. These clusters interacted with the theta-band prediction network through cross-frequency coupling. Our results offer novel insights into the multi-timescale neural dynamics underlying multisensory predictions, suggesting that oscillations in multiple frequency ranges, as well as coupling within and across frequencies, may be critical for multisensory sequence processing.
Neuroscience
What problem does this paper attempt to address?