Neural and Perceptual Mechanisms for Coding Frequency Modulation

NIH RePORTER · NIH · R21 · $127,282 · view on reporter.nih.gov ↗

Abstract

Modulations in frequency and amplitude are fundamental attributes of natural sounds. Frequency modulation (FM) is essential for sound segregation, talker recognition, and the tracking of prosody in speech and music. FM sensitivity degrades with normal aging and hearing loss, thereby restricting the communication abilities of older and hearing-impaired individuals. Importantly, assistive listening devices, such as hearing aids and cochlear implants, have not been successful in restoring FM sensitivity. This limitation is in part due to an incomplete understanding of the neural and perceptual mechanisms underlying FM coding. Human FM perception is best at the low rates (fm < ~5-10 Hz) and carrier frequencies (fc < 4-5 kHz) that are also most relevant for speech and music, with sensitivity worsening at faster rates and/or higher carriers. These rate and carrier-dependent trends are widely believed to reflect dual mechanisms for FM coding: one mechanism based on precise phase-locking to temporal fine structure (TFS; time code), and another based on the fidelity of cochlear filtering across the tonotopic axis and amplitude modulation (AM) processing (envelope/place code). Crucially, slow-rate, low-carrier FM has been treated as the gold standard behavioral measure for monaural TFS coding. The proposed research will use a combination of psychoacoustics, computational modeling, and electroencephalography (EEG) to test the hypothesis that neural and perceptual FM sensitivity in normal- hearing listeners can instead be explained by a unitary neural place code. Aim 1 will provide new insights into the necessity of time coding in FM detection in young, normal-hearing listeners by testing whether classic behavioral evidence for time coding is also found when using novel amplitude-modulated stimuli that provide no TFS cues. Aim 2 uses a rigorous, large-scale approach to test whether perceptual and neural degradations in slow FM processing with age can be accounted for by TFS coding, as currently believed, or whether these changes too can be explained within the unified framework based on place coding. This aim has important clinical implications, as it provides the first large-scale replication and extension of a new EEG measure, the FM following response (FMFR), which has been claimed to reflect individual fidelity of TFS coding to slow FM. The results will provide the preliminary data and foundation for future studies with hearing-impaired listeners to determine how these critical acoustic cues are affected by different types and degrees of hearing loss and how both perceptual and neural responses can be explained within a unifying theoretical framework in order to provide a novel basis for future sensory interventions.

Key facts

NIH application ID
11222335
Project number
7R21DC019409-04
Recipient
UNIVERSITY OF MICHIGAN AT ANN ARBOR
Principal Investigator
Kelly L Whiteford
Activity code
R21
Funding institute
NIH
Fiscal year
2024
Award amount
$127,282
Award type
7
Project period
2021-12-01 → 2025-11-30