Mutual Information Analysis of Neural Representations of Speech in Noise in the Aging Midbrain
Date:
Abstract
Young adults with normal hearing can understand speech in the presence of a competing speaker incredibly well. However, the ability to understand speech in challenging conditions deteriorates with age. Older adults typically have problems understanding speech in noise, even with clinically normal hearing. Earlier studies have shown that age- related deficits in speech understanding affect the frequency following response (FFR) in midbrain. However, it remains an open question whether the aging midbrain can nevertheless process sufficient information for auditory cortex to compensate. To investigate these issues we analyzed speech-driven FFR, for speech masked by competing speech, using mutual information, to estimate the amount of stimulus information is contained in the FFR. Our results show first a wide-band informational loss in both amplitude and phase of FFR caused by aging. Second, the loss of information is more severe in higher frequency bands in older listeners. Third, the information decreases as noise level increases for both age groups, but the older adults benefit from meaningless background compared to meaningful background, while the younger do not. The results reveal a clear foreground informational loss in midbrain caused by aging, which is more severe when the background noise is meaningful to them than it is meaningless. Download Poster