Auditory and auditory-visual speech perception and production in noise in younger and older adults

Abstract

The overall aim of the thesis was to investigate spoken communication in adverse conditions using methods that take into account that spoken communication is a highly dynamic and adaptive process, underpinned by interaction and feedback between speech partners. To this end, first I assessed the speech production adaptations of talkers in quiet and in noise, and in different communicative settings, i.e., where the talker and interlocutor were face to face (FTF) or could not see each other (Non-visual) (Chapter 2). Results showed that talkers adapted their speech production to suit the specific communicative environment. Talkers exaggerated their speech productions in noise (Lombard speech) compared to in quiet conditions. Further, in noise, in the FTF condition, talkers exaggerated mouth opening and reduced auditory intensity compared to the non-visual condition. To determine whether these speech production changes affected speech perception, materials drawn from the production study were tested in a speech perception in noise experiment (Chapter 3). The results showed that speech produced in noise provided an additional visual and auditory-visual intelligibility benefit for the perceiver. Following up this finding, I tested older and younger adults to see whether older adults would also show a Lombard speech benefit (Chapter 4 & 5). It was found that older adults were able to benefit from the auditory and auditory-visual speech production changes talkers made in noise. However, the amount of benefit they received depended on the type of noise (i.e., the degree of energetic-masking or informational-masking present in the noise masker), the signal type (i.e., whether the signal is auditory, visual, or auditory-visual) and the type of speech material considered (i.e., vowels or consonants). The results also showed that older adults were significantly poorer at lipreading than younger adults. To investigate a possible cause of the older adults' lipreading problems, I presented time-compressed and time-expanded visual speech stimuli to determine how durational changes affected the lipreading accuracy of older compared to the younger adults (Chapter 6). The results showed that older adults were not disproportionately affected by changes in the durational properties of visual speech, suggesting that factors other than the speed of the visual speech signal determine older adults' reduced lipreading capacity. The final experiment followed up several methodological issues concerning testing speech perception in noise. I examined whether the noise type (i.e., SSN or babble), the degree of lead-in noise, as well as the temporal predictability of the speech signal influenced on speech perception in noise performance (Chapter 7). I found that the degree of energetic- and informational-masking of speech in noise was affected by the amount of lead-in noise before the onset of the speech signal, but not by the predictability of the target speech signal. Taken together, the research presented in this thesis provides an insight into some of the factors that affect how well younger and older adults communicate in adverse conditions.

Date of Award	2014
Original language	English