Neuro-cognitive role of acoustic edges during speech perception

• Applicant: Dr. Yulia Oganian
• Subject Area: Biological Psychology and Cognitive Neuroscience
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 541322282

Speech is the primary channel of human communication, a spectrally complex and temporally variable signal. It is widely accepted that speech comprehension relies on the amplitude envelope of speech to identify syllabic units in the continuous acoustic signal. Yet, it is not clear how exactly the amplitude envelope contributes to this process. A dominant view suggests that troughs in the envelope correspond to syllable boundaries, thus by locating troughs in the envelope, it is possible to chunk the speech stream into syllables. However, in my recent work I showed that human speech cortex does not represent the entire amplitude envelope of speech, nor envelope troughs. Rather, the speech cortex only retains a discrete representation of local peaks in the first temporal derivative of the speech envelope, marking acoustic onset edges, which we coined peakRate events. Further, I found that in natural speech, peakRate events correspond to vowel onsets in syllables – thus they are an internal marker of syllable structure, and do not mark the boundaries of syllables. Given these findings, the role of the speech envelope in speech perception is again unclear. With this proposal I propose to probe multiple possibilities for the role of peakRate events in speech perception. I propose 1) to separate neural markers of acoustic edges from neural processing of perceived syllabic onsets; 2) to test whether acoustic edges support the analysis of the spectral content of speech at times of consonant to vowel transitions; 3) to evaluate whether acoustic edges are used as anterograde and retrograde cues to delineate a temporal window of integration for each syllable. Each project in this proposal investigates one of these hypotheses. Together, the results of this project will advance our understanding of a central enigma in speech perception: The role of amplitude modulations for the analysis of speech spectral content. Furthermore, they will address the role of endogenous cortical oscillations in speech perception.

• DFG Programme: Research Grants