Adaptation To Speech In Noise Across Age And Signal-to-noise Ratios

BACKGROUND

Understanding speech in background noise is challenging, particularly as we age. Although listeners can rapidly adapt to degraded speech, less is known about short-term adaptation to speech-in-noise. We examined whether a brief period of listening to speech in babble noise results in adaptation, whether the rate of adaptation differs between younger and older adults, and how it is influenced by signal-to-noise ratio (SNR).

METHODS
Eighty-six Hebrew-speaking adults participated. Younger adults (18-35 years) were tested at -5 dB SNR; older adults (58-79 years) were tested at -3, 0, or +3 dB SNR. Participants listened to 30 Hebrew sentences embedded in four-talker babble noise and repeated each sentence aloud. Recognition accuracy across trials was analyzed using a generalized linear mixed model, with sentence number, age/SNR group, and their interaction as fixed effects.

RESULTS
Mean accuracy was 44.8% for younger adults at -5 dB, 44.9% for older adults at -3 dB, 64.5% at 0 dB, and 66.8% at +3 dB. Thus, -3 dB provided the closest performance match between older and younger adults. Accuracy improved significantly across sentences, indicating adaptation. No significant group/SNR × trial interaction was observed. Neither working memory nor musical experience correlated with adaptation rate.

CONCLUSION
Listeners rapidly adapted to Hebrew speech in babble noise with similar adaptation rates across age groups. Among older adults adaptation rate didn`t appear to depend on SNR. This suggests that adaptation may be relatively independent of perceptual challenge as manifested by accuracy differences between different SNRs.