Individual Differences in Auditory Training Benefits for Hearing Aid Users

Abstract

The present study aimed to examine whether individual differences in baseline speech perception could serve as predictors for the effectiveness and generalization of auditory training (AT) to non-trained tasks. Twelve adults, aged 60-75 years with bilateral hearing loss, completed a two-month, home-based, computerized AT program, involving sessions four times per week. Training tasks included the identification of vowel frontal, height, manner of articulation, point of articulation, voicing, and open-set consonant-vowel-consonant (CVC) words. Non-trained speech perception tests were conducted one month before AT, prior to training, after one and two months of training, and during a two-month follow-up. The results showed that one month of AT improved performance in most trained tasks, with generalization observed in the CVC words test and HeBio sentences with speech-shaped noise (SSN). No evidence of spontaneous learning or added benefit from an extra month of training was found. Most importantly, baseline speech perception predicted improvements in both training and post-training generalization tasks. This emphasizes the significance of adopting an individualized approach when determining the potential effectiveness of AT, applicable in both clinical and research contexts.

Keywords: auditory training; generalization; individual differences; speech perception.

Figure 1

Illustration of the study procedure. T = test session; CVC = consonant–vowel–consonant words; DIN = digit-in-noise test; HeBio sentences test (Reprinted/adapted with permission from Ref. [36]).

Figure 2

Accuracy scores and statistical analyses across all training sessions for the six training tasks: (a) identification of frontal vowel; (b) identification of vowel height; (c) identification of articulation manner; (d) identification of point of articulation; (e) identification of voicing; and (f) identification of open set CVC words.

Figure 3

The mean, median, and interquartile range in each testing session of (a) HeBio sentences with SSN (SRTn); (b) HeBio sentences with 4TBN (SRTn); (c) CVC words in quiet (accuracy); (d) DIN (SRTn). Participants’ individual scores are represented by circles; the mean is presented by the x-sign.

Figure 4

The change in the six training tasks according to participants’ baseline scores in the speech perception (generalization) tasks: (a) HeBio sentences with SSN (SRTn); (b) HeBio sentences with 4TBN (SRTn); (c) CVC words in quiet (accuracy); (d) DIN (SRTn). Bold regression analysis results indicate a significant relationship.

Figure 5

The change in the speech perception (generalization) tasks according to participants’ baseline scores in the speech perception (generalization) tasks: (a) HeBio sentences with SSN (SRTn); (b) HeBio sentences with 4TBN (SRTn); (c) CVC words in quiet (accuracy); (d) DIN (SRTn). Bold regression analysis results indicate a significant relationship.