Encoding and decoding analysis of music perception using intracranial EEG
Bellier, L.; Llorens, A.; Marciano, D.; Schalk, G.; Brunner, P.; Knight, R. T.; Pasley, B. N.
Show abstract
Music perception engages multiple brain regions, however the neural dynamics of this core human experience remains elusive. We applied predictive models to intracranial EEG data from 29 patients listening to a Pink Floyd song. We investigated the relationship between the song spectrogram and the elicited high-frequency activity (70-150Hz), a marker of local neural activity. Encoding models characterized the spectrotemporal receptive fields (STRFs) of each electrode and decoding models estimated the population-level song representation. Both methods confirmed a crucial role of the right superior temporal gyri (STG) in music perception. A component analysis on STRF coefficients highlighted overlapping neural populations tuned to specific musical elements (vocals, lead guitar, rhythm). An ablation analysis on decoding models revealed the presence of unique musical information concentrated in the right STG and more spatially distributed in the left hemisphere. Lastly, we provided the first song reconstruction decoded from human neural activity.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Intrinsic neural timescales in the temporal lobe support an auditory processing hierarchy 96%
- Chimeric music reveals an interaction of pitch and time in electrophysiological signatures of music encoding 96%
- Responses to Visual Speech in Human Posterior Superior Temporal Gyrus Examined with iEEG Deconvolution 96%
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
- Delta/Theta band EEG activity shapes the rhythmic perceptual sampling of auditory scenes 94%
- Audio-visual combination of syllables involves time-sensitive dynamics following from fusion failure 94%
- Online speech synthesis using a chronically implanted brain-computer interface in an individual with ALS 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.