Audiology and Speech Pathology - Theses

Permanent URI for this collection

http://hdl.handle.net/11343/255

Filters

2023 1
1
Reset filters

Settings
Statistics
Citations
End date: 2023 × Start date: 2020 × Type: PhD thesis × Subject: Accented speech perception × Subject: Hearing Loss ×

Search Results

Now showing 1 - 1 of 1
  • Item
    Thumbnail Image
    The processing of Mandarin-accented English by older hearing-impaired adults and their perceptual adaptation to the accent
    Hau, Jutta Alice ( 2023-06)
    ABSTRACT Older, hearing-impaired (OHI) adults have proportionately more trouble hearing speech in adverse listening situations than normally hearing (NH) listeners. When the speech signal is degraded through the presence of background noise, fast speech, or is otherwise unclear, OHI adults have more trouble with speech recognition (Humes and Dubno, 2010) and expend more cognitive resources in working out what has been said than do NH listeners (Tun, McCoy & Wingfield, 2009). This study investigated whether OHI listeners are likewise disproportionately disadvantaged by an unfamiliar accent, namely Mandarin Accented English (MAE), and whether these listeners are able to adapt to the systematicities of the accent. 27 OHI, 27 ONH and 29 YNH listeners were recruited. Cross modal fragment priming (CMFP) was utilised in Experiment 1 to measure the effect of a phonological error often made by speakers of MAE, the lack of reduction of the vowel in unstressed iambic words. Response times and accuracy were compared between the listener groups, between Australian English (AusE) and MAE tokens and between matching and mismatching primes and targets. Talker-independent adaptation was evaluated in a second experiment through an exposure training study design. Listeners were trained either by four MAE talkers or four AusE talkers before listening to sentences presented by a novel MAE talker. Speech recognition was compared between the listener groups and the training accents. In experiment one, the CMFP study, all listener groups were slower and less accurate in their responses to MAE. While the YNH group was faster to respond to all tokens (AusE and MAE) than both the OHI and the ONH groups, they were only significantly more accurate in responding to the AusE tokens, indicating that they too, found MAE speech recognition difficult. The YNH listeners demonstrated significantly faster responses than both older groups in the most cognitively challenging conditions where prime and target did not match. In experiment two, the adaption study, all three groups trained by the MAE talkers showed higher odds of speech recognition than listeners trained by AusE talkers, showing the benefit of exposure to the accent for MAE recognition. The OHI listeners adapted to MAE to the same degree as the NH listeners despite returning lower overall odds of recognising MAE speech. While OHI listeners with mild to moderate hearing loss showed similar accuracy in responses to MAE words and fragment tokens as did the normally hearing listeners, both older listener groups took substantially longer to respond to these tokens, especially in the more cognitively taxing mismatching conditions. This result indicates that older listeners had greater difficulty processing the speech tokens which, however, was not always reflected in their recognition results. In addition, OHI listeners were able to benefit as much from their brief exposure to MAE as did the NH groups. This result suggests that OHI with mild-moderate hearing loss have access to and can exploit the information present in a relatively brief sample of accented speech and generalise this learning to a novel MAE talker. This potential to adapt to the accent of an FAS talker will be important for ensuring smooth communication interactions, particularly when the communication partner is committed to reducing the barriers to communication by, for example, reducing their rate of speech. This study has utilised theories, study designs and methodological techniques developed in the field of psycholinguistics and in particular, has demonstrated the benefit of using RT measurements to learn more about the processing difficulty of FAS as experienced by OHI adults. The study has also indicated the potential for psycholinguistic theories of speech processing to be expanded by examining the language processing of older listeners with an impaired auditory system.