Hearing-Impaired Listeners Show Reduced Attention to High-Frequency Information in the Presence of Low-Frequency Information

Abstract

Many listeners with sensorineural hearing loss have uneven hearing sensitivity across frequencies. This study addressed whether this uneven hearing loss leads to a biasing of attention to different frequency regions. Normal-hearing (NH) and hearing-impaired (HI) listeners performed a pattern discrimination task at two distant center frequencies (CFs): 750 and 3500 Hz. The patterns were sequences of pure tones in which each successive tonal element was randomly selected from one of two possible frequencies surrounding a CF. The stimuli were presented at equal sensation levels to ensure equal audibility. In addition, the frequency separation of the tonal elements within a pattern was adjusted for each listener so that equal pattern discrimination performance was obtained for each CF in quiet. After these adjustments, the pattern discrimination task was performed under conditions in which independent patterns were presented at both CFs simultaneously. The listeners were instructed to attend to the low or high CF before the stimulus (assessing selective attention to frequency with instruction) or after the stimulus (divided attention, assessing inherent frequency biases). NH listeners demonstrated approximately equal performance decrements (re: quiet) between the two CFs. HI listeners demonstrated much larger performance decrements at the 3500 Hz CF than at the 750 Hz CF in combined-presentation conditions for both selective and divided attention conditions, indicating a low-frequency attentional bias that is apparently not under subject control. Surprisingly, the magnitude of this frequency bias was not related to the degree of asymmetry in thresholds at the two CFs.

Perceptual Learning of Noise-Vocoded Speech Under Divided Attention

“…The current study only recruited young listeners with a normal hearing capacity. However, past evidence has shown that listeners with hearing loss have reduced spectrotemporal and spatial acuity (e.g., Bernstein & Oxenham, 2006 ; Cusack et al, 2004 ; Drennan et al, 2003 ) and show lower ability to use these cues for selectively attending to a target signal (e.g., a target talker among competing talkers; e.g., Best et al, 2009 ; Dai et al, 2018 ; Roverud et al, 2020 ). Moreover, perceptual learning seems to diverge in young and old listeners.…”