Search Papers | Poster Sessions | All Posters
Poster A55 in Poster Session A - Tuesday, August 6, 2024, 4:15 – 6:15 pm, Johnson Ice Rink
On the generative mechanisms underlying the cortical tracking of natural speech
Edmund Lalor1 (), Andre Palacios Duran1, Aaron Nidiffer1; 1University of Rochester
Low frequency cortical activity tracks the dynamics of natural speech. However, the mechanisms that produce this tracking are debated. One theory proposes that intrinsic cortical oscillations entrain to the rhythms of speech in an anticipatory manner. Meanwhile, a second theory assumes that neural measures of speech processing reflect transient evoked responses. Here, we attempt to reconcile these theories. We leverage the fact that, when you regress neurophysiological data against (say) the amplitude envelope of speech, you obtain a temporal response function (TRF) that that can reliably predict responses to novel speech stimuli. We then ask: can the existence of TRFs be explained as deriving from the entrainment of an ongoing oscillation? We do this by driving two oscillatory models with speech stimuli, attempting to fit TRFs to the resulting simulated brain activity, and then assessing whether such simulated brain activity can be predicted using the resulting TRF. We find that both models could produce TRFs with predictive power. However, one model is biologically implausible, and the second model produces simulated neural activity and TRFs with highly atypical characteristics. Nonetheless, this study establishes a framework for resolving an important debate in the field of speech neurophysiology.
Keywords: Speech EEG Oscillations Modeling