Neuroplasticity of Speech-in-Noise Processing in Older Adults Assessed by Functional Near-Infrared Spectroscopy (fNIRS)
Guangting Mai,Zhizhao Jiang,Xinran Wang,Ilias Tachtsidis,Peter Howell
DOI: https://doi.org/10.1007/s10548-024-01070-2
2024-07-23
Abstract:Functional near-infrared spectroscopy (fNIRS), a non-invasive optical neuroimaging technique that is portable and acoustically silent, has become a promising tool for evaluating auditory brain functions in hearing-vulnerable individuals. This study, for the first time, used fNIRS to evaluate neuroplasticity of speech-in-noise processing in older adults. Ten older adults, most of whom had moderate-to-mild hearing loss, participated in a 4-week speech-in-noise training. Their speech-in-noise performances and fNIRS brain responses to speech (auditory sentences in noise), non-speech (spectrally-rotated speech in noise) and visual (flashing chequerboards) stimuli were evaluated pre- (T0) and post-training (immediately after training, T1; and after a 4-week retention, T2). Behaviourally, speech-in-noise performances were improved after retention (T2 vs. T0) but not immediately after training (T1 vs. T0). Neurally, we intriguingly found brain responses to speech vs. non-speech decreased significantly in the left auditory cortex after retention (T2 vs. T0 and T2 vs. T1) for which we interpret as suppressed processing of background noise during speech listening alongside the significant behavioural improvements. Meanwhile, functional connectivity within and between multiple regions of temporal, parietal and frontal lobes was significantly enhanced in the speech condition after retention (T2 vs. T0). We also found neural changes before the emergence of significant behavioural improvements. Compared to pre-training, responses to speech vs. non-speech in the left frontal/prefrontal cortex were decreased significantly both immediately after training (T1 vs. T0) and retention (T2 vs. T0), reflecting possible alleviation of listening efforts. Finally, connectivity was significantly decreased between auditory and higher-level non-auditory (parietal and frontal) cortices in response to visual stimuli immediately after training (T1 vs. T0), indicating decreased cross-modal takeover of speech-related regions during visual processing. The results thus showed that neuroplasticity can be observed not only at the same time with, but also before, behavioural changes in speech-in-noise perception. To our knowledge, this is the first fNIRS study to evaluate speech-based auditory neuroplasticity in older adults. It thus provides important implications for current research by illustrating the promises of detecting neuroplasticity using fNIRS in hearing-vulnerable individuals.