The contribution of cognition and hearing loss to individual differences in speech intelligibility in a variety of speech-perception-in-noise tests in younger and older adult listeners

Successful Speech-in-Noise (SiN) perception can be difficult, particularly for older listeners. Changes in hearing and in cognition probably both play a role in understanding these difficulties. To fully understand unique and shared contributions of the two, a systematic and theoretically rigorous approach in selecting speech stimuli and cognitive tests is essential. We measured SiN perception in a total of six listening situations that varied in the sematic predictability of the target sound and the extent of informational masking in the background. We selected cognitive tests based on the amodal and verbal components of Baddeley’s model of working memory (central executive (CE), episodic buffer (EB), phonological loop (PL)) and assessed each component by multiple tests (three for CE, and two for EB and PL). We then combined test scores to compute latent variable scores for each component, reducing variation due to surface test characteristics.

To examine potential interaction effects between hearing loss and cognition we tested younger (N=50, age range 18-30 years) and older listeners (N=50, age range 60-85 years). Neither were hearing aid users. Young listeners, who all had pure-tone averages (PTAs) within the clinically normal range (<20dB HL PTA 0.25-4kHz), only showed a small variation in hearing sensitivity. Older listeners, on the other hand, showed more variability in hearing and had a range of hearing levels from normal to mild loss (<40dB HL PTA 0.25-4kHz).

We analysed the behavioural results from the six speech perception tests (3 target types x 2 background types), three latent cognitive variables (CE, EB, PL), hearing sensitivity and age (young, old) in a linear mixed model. Here we concentrate on reporting those results that involve the predictive roles of hearing loss and cognition for speech intelligibility. Hearing loss: The predictive role of hearing level interacted with type of background masker such that it was comparable between younger and older listeners for three-talker babble but was substantially increased for older compared to younger listeners in the presence of signal-modulated noise. Cognition: EB: better scores predicted better speech intelligibility overall, regardless of age group or hearing status. CE: better scores predicted better overall speech intelligibility in older but not younger listeners. PL: better scores predicted better speech perception performance in three-talker babble for younger listeners but worse performance for older listeners.

These results suggest that younger and older adults may employ different listening strategies in order to understand speech in noise in the same situation.