The Relative Contributions of Temporal Envelope and Fine Structure to Mandarin Lexical Tone Perception in Auditory Neuropathy Spectrum Disorder

Abstract

Previous studies have demonstrated that temporal envelope (E) is sufficient for speech perception, while fine structure (FS) is important for pitch perception for normal-hearing (NH) listeners. Listeners with sensorineural hearing loss (SNHL) have an impaired ability to use FS in lexical tone perception due to the reduced frequency resolution. Listeners with auditory neuropathy spectrum disorder (ANSD) may have deficits in temporal resolution. Little is known about how such deficits may impact their ability to use E and FS to perceive lexical tone, and whether it is the deficit in temporal resolution or frequency resolution that may lead to more detrimental effects on FS processing in pitch perception. Three experiments were conducted in the present study. Experiment I used the "auditory chimera" technique to investigate how SNHL and ANSD listeners would achieve lexical tone recognition using either the E or the FS cues. Experiment II tested their frequency resolution as measured with their psychophysical tuning curves (PTCs). Experiment III tested their temporal resolution as measured with the temporal gap detection (TGD) threshold. The results showed that the SNHL listeners had reduced frequency selectivity, but intact temporal resolution ability, while the ANSD *Parts of this paper may be published elsewhere. 242S. Wang et al.listeners had degraded temporal resolution ability, but intact frequency selectivity. In comparison with the SNHL listeners, the ANSD listeners had severely degraded ability to process the FS cues and thus their ability to perceive lexical tone mainly depended on the ability to use the E cues. These results suggested that, in comparison with the detrimental impact of the reduced frequency selectivity, the impaired temporal resolution may lead to more degraded FS processing in pitch perception.

The Relative Weight of Temporal Envelope Cues in Different Frequency Regions for Mandarin Disyllabic Word Recognition

“…Temporal information refers to information in speech signals with time-varying wave rates, which can be divided into the temporal envelope (E) below 50 Hz, periodic fluctuations in the range of 50-500 Hz, and temporal fine structure in the range of 500-10,000 Hz (Rosen, 1992). E cues contain temporal modulation information, which is most important for speech perception in quiet conditions, whereas the temporal fine structure can provide information in noisy environments and for tonal and pitch recognition (Smith et al, 2002;Xu and Pfingst, 2003;Moore, 2008;Ardoint and Lorenzi, 2010;Wang et al, 2016). Vocoder studies have shown that E modulation rates of 4-16 Hz are most important for speech intelligibility in quiet (Drullman et al, 1994a,b;Shannon et al, 1995).…”