Amplitude modulation detection with a short-duration carrier: Effects of a precursor and hearing loss

Abstract

This study tests the hypothesis that amplitude modulation (AM) detection will be better under conditions where basilar membrane (BM) response growth is expected to be linear rather than compressive. This hypothesis was tested by (1) comparing AM detection for a tonal carrier as a function of carrier level for subjects with and without cochlear hearing impairment (HI), and by (2) comparing AM detection for carriers presented with and without an ipsilateral notched-noise precursor, under the assumption that the precursor linearizes BM responses. Average AM detection thresholds were approximately 5 dB better for subjects with HI than for subjects with normal hearing (NH) at moderate-level carriers. Average AM detection for low-to-moderate level carriers was approximately 2 dB better with the precursor than without the precursor for subjects with NH, whereas precursor effects were absent or smaller for subjects with HI. Although effect sizes were small and individual differences were noted, group differences are consistent with better AM detection for conditions where BM responses are less compressive due to cochlear hearing loss or due to a reduction in cochlear gain. These findings suggest the auditory system may quickly adjust to the local soundscape to increase effective AM depth and improve signal-to-noise ratios.

Adaptation to Noise in Human Speech Recognition Depends on Noise-Level Statistics and Fast Dynamic-Range Compression

“…The better recognition scores at mid-levels may have occurred because basilar membrane compression reduced the noise level fluctuations more at mid-levels, and thus induced greater neural DRA to the noise statistics and better speech-innoise recognition. Hearing-impaired (HI) listeners show less or no noise adaptation in word recognition (at least for babble maskers) (Ben-David et al, 2012) or AM detection (Jennings et al, 2018). Because hearing loss is typically associated with combined loss or dysfunction of cochlear IHCs and OHCs (Liberman and Dodds, 1984;Lopez-Poveda and Johannesen, 2012;Johannesen et al, 2014), and the MOCR is thought to have antimasking effects (for review, see Guinan, 2006;Lopez-Poveda, 2018), the conventional view is that OHC loss would impair noise adaptation because OHCs are necessary for the MOCR to exert its antimasking effects (Jennings et al, 2018;for review, see Lopez-Poveda, 2018).…”

“…Hearing-impaired (HI) listeners show less or no noise adaptation in word recognition (at least for babble maskers) (Ben-David et al, 2012) or AM detection (Jennings et al, 2018). Because hearing loss is typically associated with combined loss or dysfunction of cochlear IHCs and OHCs (Liberman and Dodds, 1984;Lopez-Poveda and Johannesen, 2012;Johannesen et al, 2014), and the MOCR is thought to have antimasking effects (for review, see Guinan, 2006;Lopez-Poveda, 2018), the conventional view is that OHC loss would impair noise adaptation because OHCs are necessary for the MOCR to exert its antimasking effects (Jennings et al, 2018;for review, see Lopez-Poveda, 2018). If this were the case, however, it would be hard to explain why HI listeners show less or no noise adaptation while cochlear-implant users show normal adaptation even though the two listener groups are expected to have impaired or absent MOCR effects (Marrufo-Pérez et al, 2018.…”

Adaptation to noise in amplitude modulation detection without the medial olivocochlear reflex

“…Similarly, the sensitivity of NH listeners to amplitude modulation (AM) in quiet or in noise improves when the AM carrier is preceded by a few hundred milliseconds of steady-state noise (Almishaal et al, 2017;Jennings et al, 2018;Marrufo-P erez et al, 2018b;Sheft and Yost, 1990;Viemeister, 1979). Several physiological mechanisms may underlie these improvements, including mechanisms mediated by the medial olivocochlear reflex (MOCR).…”

“…Some efferent fibers terminate in the bodies of outer hair cells (OHCs) in the cochlea (Warr and Guinan, 1979), and may be activated reflexively by ipsilateral and/or contralateral sounds (reviewed by Lopez-Poveda, 2018;Guinan, 1996). Activation of the MOCR could facilitate the detection of AM in quiet when the AM carrier is preceded by noise (Almishaal et al, 2017;Jennings et al, 2018;Marrufo-P erez et al, 2018b). The idea is that the preceding noise activates the ipsilateral and/or the contralateral MOCR.…”

“…The difference was greater at mid than at lower or higher AM carrier levels, consistent with the precursor activating the ipsilateral MOCR, linearizing the BM responses at moderate-to-high levels (where the BM response is more compressive), and facilitating AM detection. Jennings et al (2018) reported that hearing-impaired (HI) listeners (1) had better AM detection thresholds than NH listeners when a short AM carrier with a moderate level (around 65 dB SPL) was presented in quiet, 1 but (2) showed smaller improvements in thresholds than NH listeners did when a precursor was used. Those results are consistent with HI listeners having less compressive BM responses and a smaller change in BM compression as a result of MOCR activation.…”