The Efficacy of Short-term Gated Audiovisual Speech Training for Improving Auditory Sentence Identification in Noise in Elderly Hearing Aid Users

Abstract

This study aimed to examine the efficacy and maintenance of short-term (one-session) gated audiovisual speech training for improving auditory sentence identification in noise in experienced elderly hearing-aid users. Twenty-five hearing aid users (16 men and 9 women), with an average age of 70.8 years, were randomly divided into an experimental (audiovisual training, n = 14) and a control (auditory training, n = 11) group. Participants underwent gated speech identification tasks comprising Swedish consonants and words presented at 65 dB sound pressure level with a 0 dB signal-to-noise ratio (steady-state broadband noise), in audiovisual or auditory-only training conditions. The Hearing-in-Noise Test was employed to measure participants’ auditory sentence identification in noise before the training (pre-test), promptly after training (post-test), and 1 month after training (one-month follow-up). The results showed that audiovisual training improved auditory sentence identification in noise promptly after the training (post-test vs. pre-test scores); furthermore, this improvement was maintained 1 month after the training (one-month follow-up vs. pre-test scores). Such improvement was not observed in the control group, neither promptly after the training nor at the one-month follow-up. However, no significant between-groups difference nor an interaction between groups and session was observed. Conclusion: Audiovisual training may be considered in aural rehabilitation of hearing aid users to improve listening capabilities in noisy conditions. However, the lack of a significant between-groups effect (audiovisual vs. auditory) or an interaction between group and session calls for further research.

Perceptual Doping: An Audiovisual Facilitation Effect on Auditory Speech Processing, From Phonetic Feature Extraction to Sentence Identification in Noise

“…If the V signal is not available, the AV simulation feeds back to A brain areas and facilitates the voice (A) identification of that person. Note that in our previous articles (Moradi et al 2013, 2017b; Lidestam et al 2014), the speakers in the gated tasks and the HINT were different. Our previous research indicates that the perceptual doping effect is independent of speaker idiosyncrasy, as only one exposure to AV speech stimuli (with the same or different speakers in the initial exposure and following the A speech task) is sufficient to obtain subsequent improvement in A speech identification.…”

“…The extent to which the type of listening conditions (e.g., clear versus distorted) would affect the perceptual doping effect on word identification in sentences requires further research. Moradi et al (2017b) suggested that background noise (relative to a silent listening condition) may enhance the perceptual doping effect by focusing listeners’ attention (or cognitive effort) on the A and V components of a speech item, to form a coherent AV speech item from an incoming speech signal so as to map it to corresponding phonological or lexical representations. The data from gated consonants, gated vowels, and vowel duration discrimination tasks in the present study indicate a perceptual doping effect even in silence.…”

“…However, the efficiency of those A training program has been reported to be low and not very robust (see Henshaw & Ferguson 2013; Ferguson et al 2014). We predict that the association of congruent V speech cues with an A training program will boost their effects, even with shorter training sessions than those for A training, in terms of improving the listening capabilities of people with hearing loss (see Moradi et al 2017b).…”

“…Moradi et al (2017b) used the gated AV and A speech materials used in Lidestam et al (2014), but recruited elderly hearing aid users, to evaluate the efficiency of gated AV speech training and the maintenance of training-related improvements in subsequent A HINT performance. Their results generalized the findings by Lidestam et al by showing that short AV speech exposure (of around 35 min) to gated consonants and words promptly improved A HINT performance (post-test training versus pretest training); furthermore, this improvement was maintained 1 month after the AV speech training (1-month follow-up versus pretest training).…”

“…Moradi et al (2017b) coined the term “perceptual doping” to explain how prior AV speech exposure subsequently facilitates the A identification of speech stimuli. According to this hypothesis, prior AV speech exposure can recalibrate (or retune) phonological and lexical maps in the mental lexicon, such that the maps become more distinct and easily accessible without effort .…”