ABSTRACT
Multitalker maskers or other maskers with time-varying characteristics are often used to measure speech recognition because they are spectrally and temporally similar to speech maskers in the environment. Most maskers with time-varying characteristics are less effective than continuous noise with an equivalent spectrum, due to an improved signal-to-noise ratio that occurs for brief periods of time. The chapter examines the additivity of multitalker and continuous maskers in young and aged subjects with normal hearing by measuring pure-tone thresholds and word recognition in individual and combined maskers. Older normal-hearing subjects are included to test the hypothesis that multitalker maskers provide an exceptionally difficult listening task for aged listeners. The practice of combining multitalker and continuous maskers to equate thresholds of normal-hearing and hearing-impaired subjects should be applied with caution. Given that speech recognition is usually better in fluctuating than in continuous maskers, this somewhat paradoxical result has been attributed to the perceptual similarity between multitalker maskers and the speech signal.
