Psychoacoustic and neurophysiological research suggests that the human auditory system creates multiple representations of sound, differing in such parameters as spectral and temporal resolution, as well as in their sensitivity to more complex spectrotemporal features. The current program of research examined the auditory representation used for perception of vowel quality. A priming method was developed in which perception of vowel targets is facilitated when the targets are preceded by acoustically-matched nonspeech stimuli. By manipulating the acoustic parameters of these nonspeech ''prime'' tones, it is possible to determine the role of these parameters in the auditory stages of vowel processing. In the first experiment (Chapter 1), the duration of the primes was varied in order to test the temporal resolution of the vowels' auditory representation. Priming was greatest at the shortest duration tested (25 ms), indicating a relatively high temporal resolution. Next, the frequency of the primes was varied (Chapter 2). A broad frequency tuning was found, suggesting a very coarse spectral resolution. Finally, the composition of the primes was varied (Chapter 3), in order to explore the representation of complex spectral features. Primes consisting of two tones, matched to both F1 and F2 of the target vowels, elicited a greater priming effect than primes matched to F1 or F2 alone, suggesting a nonlinear form of combination sensitivity which might underlie the perception of vowels' formant patterns.
Wallace, Andrew Benedict,
"The Auditory Representation of Time and Frequency in Vowel Quality Perception"
(2010).
Cognitive Sciences Theses and Dissertations.
Brown Digital Repository. Brown University Library.
https://doi.org/10.7301/Z0MW2FCP
