Seeing a speaker’s face helps stream segregation for younger and elderly adults.
Alexandra Jesse, Esther Janse · The Journal of the Acoustical Society of America · 2009
Listening to a speaker while hearing one or more competing speakers in the background can be a challenging task, especially for elderly adults. We show that younger and elderly listeners (above the age of 65) with varying degrees of age-related hearing loss benefit in stream segregation and speech processing from seeing the target speaker talk in addition to hearing an audio mix of a target and a competing speaker. This audiovisual benefit was found for response accuracy and speed in a phoneme monitoring task, where participants indicated by button press the occurrence of given target phonemes in the monitored speech of the target speaker. The audiovisual benefit was larger for younger than for older adults, despite their equivalent performance on auditory-only trials and in an off-line phoneme lip-reading task. Better lip-reading performance, however, predicted a larger audiovisual benefit. The audiovisual benefit was found for both highly visible phonemes (/p/) and poorly visible phonemes (/k/), but was modulated in its size by segmental visibility. The audiovisual benefit therefore arises from local segmental visual information but is also at least partially driven by audiovisual synchrony information that aids in attending to the target speaker.