Abstract
The aims of the present study were to examine whether familiarity with a masker improves word recognition in speech masking situations and whether there are age-related differences in the effects of masker cuing. Thirty-two older listeners (range = 59–74; mean age = 66.41 years) with high-frequency hearing loss and 32 younger normal-hearing listeners (range = 21–28; mean age = 23.73) participated in this study, all of whom spoke Chinese as their first language. Two experiments were conducted and 16 younger and 16 older listeners were used in each experiment. The masking speech with different content from target speech with syntactically correct but semantically meaningless was a continuous recording of meaningless Chinese sentences spoken by two talkers. The masker level was adjusted to produce signal-to-masker ratios of -12, -8, -4, and 0 dB for the younger participants and -8, -4, 0, and 4 dB for the older participants. Under masker-priming conditions, a priming sentence, spoken by the masker talkers, was presented in quiet three times before a target sentence was presented together with a masker sentence 4 s later. In Experiment 1, using same-sentence masker-priming (identical to the masker sentence), the masker-priming improved the identification of the target sentence for both age groups compared to when no priming was provided. However, the amount of masking release was less in the older adults than in the younger adults. In Experiment 2, two kinds of primes were considered: same-sentence masker-priming, and different-sentence masker-priming (different from the masker sentence in content for each keyword). The results of Experiment 2 showed that both kinds of primes improved the identification of the targets for both age groups. However, the release from speech masking in both priming conditions was less in the older adults than in the younger adults, and the release from speech masking in both age groups was greater with same-sentence masker-priming than with different-sentence masker-priming. These results suggest that both the voice and content cues of a masker could be used to release target speech from maskers in noisy listening conditions. Furthermore, there was an age-related decline in masker-priming-induced release from speech masking.
Highlights
In everyday scenarios, individuals are often faced with the difficulty of interpreting the speech of one person while other people are speaking simultaneously (Cherry, 1953)
We address whether prior familiarity with a masker affects word recognition in speech masking situations among younger and older adults and whether there are agerelated differences in the abilities of listeners to utilize masking speech cues for target speech identification
The results showed that the average value of the threshold μ (50% correct threshold) for identification of the whole target sentence was lower when the masker prime was presented three times than when it was presented two times, regardless of whether the delay between the masker prime and the speech presented in the masker was 3000 ms (−1.0 dB vs. −0.7 dB) or 4000 ms (−1.1 dB vs. −0.9 dB) under different-sentence masker-priming conditions
Summary
Individuals are often faced with the difficulty of interpreting the speech of one person while other people are speaking simultaneously (Cherry, 1953). Two main factors are thought to contribute to this difficulty: energetic masking (Leek et al, 1991) and informational masking (Freyman et al, 1999; Arbogast et al, 2002; Kidd et al, 2005; Schneider et al, 2007; Agus et al, 2009; Helfer and Freyman, 2009). Energetic masking occurs at the auditory periphery, when components of the speech signal in some time–frequency region are rendered inaudible because of swamping by the masker (Leek et al, 1991; Kidd et al, 1994, 1998; Freyman et al, 1999), such that the response of the peripheral neurons to the target is suppressed by that evoked by the masker. Steady noise is considered to produce primarily energetic masking [even though Stone et al (2011, 2012) recently demonstrated that the inherent random amplitude fluctuations of such a masker impede speech identification, presumably due to modulation masking], while a speech masker produces both energetic and informational masking
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
