Age alters audio-visual speech perception: Identification and asynchrony perception during young and middle adulthood
MetadataVis full innførsel
- Institutt for psykologi 
Research has shown considerable change in audio-visual speech perception across the lifespan, in particular during childhood and old adulthood. However, changes between young and middle adulthood have received relatively little attention. The intention of the current project was to contribute to a more comprehensive lifespan understanding of audio-visual speech perception by investigating changes in this age interval. The three studies included explored age differences in audio-visual speech perception from two different perspectives: audio-visual identification and audio-visual asynchrony perception. Audio-visual speech identification relates to the ability to use the multitude of auditory and visual speech information to recognize and label speech segments, such as stop consonants in the case of this study. Audio-visual asynchrony perception in contrast, is a more basic ability, reflecting the temporal thresholds for whether the auditory and the visual speech signals are integrated into a single audio-visual percept. All the experiments included have one thing in common: the assessment of age-related experience through comparison of young and middle-aged participants. All studies also used speech stimuli of the same complexity, namely syllables composed of different stop consonants trailed by a vowel. The first experiment explored age differences in audio-visual asynchrony perception. Participants made simultaneity judgements, that is, decided whether the auditory and visual signals were synchronous or asynchronous for audio-visual stimuli ranging from 440 millisecond audio-lead to 440 millisecond visual-lead. Whereas visual-lead sensitivity was unaffected by age, middle-aged adults were significantly more sensitive to audio-lead asynchronies than young adults. It was suggested that in contrast to visual-lead asynchrony, the natural temporal predictability of audio-lead asynchronies makes audio-lead sensitivity more susceptible to the predictability dependent perceptual learning. Hence, that age affected audio-lead asynchrony perception specifically suggests a contribution of audio-visual speech experience, especially since age-related sensory and cognitive differences were assessed to be minor. The second experiment was a continuation of the first, and in addition to using a similar simultaneity judgement task, employed a measurement of cognitive processing speed. Cognitive processing speed was assessed through recognition reaction time tasks where participants responded as quickly as possible to the syllable /ba/ amid presentations of /da/ and /ga/ syllables. The first aim of the experiment was to evaluate the relationship between cognitive processing speed and audio-visual asynchrony perception. The second aim was to assess the effect of age on this relationship. The results confirmed the notion put forward in the first experiment, that young and middle-aged adults have similar cognitive processing speed. The result of the young adults also confirmed the proposed relationship between cognitive processing speed and audio-visual asynchrony perception, that is, increasing reaction time was associated with increasing tolerance for audio-visual asynchrony. However, the lack of such a correlation for the middle-aged adults implied that age-related audio-visual experience mitigates the influence of cognitive processing speed on audio-visual asynchrony perception. It was suggested that middle-aged adults’ audio-visual experience may facilitate selective attention, thus reducing the influence cognitive processing strain by focusing on the most relevant audio-visual speech cues. The third experiment investigated age and gender differences in audio-visual speech identification. Audio-visual benefit and visual influence were used to measure the visual contribution to audio-visual speech identification. Audio-visual benefit measured proficiency in auditory-visual integration, and revealed no age and gender differences in performance. Visual influence on the other hand quantified the individual contribution of auditory and visual speech cues in audio-visual speech identification, and whereas no gender difference was observed for the young adults, the middle-aged females relied significantly more on visual cues in audio-visual speech identification than middle-aged males. It was proposed that the gender difference in visually influenced responses in middle adulthood is a result of an experience-related shift in the audio-visual perceptual strategy of females. Because females were observed to be better speech-readers in both age groups, it was suggested that speechreading proficiency in young adulthood could gradually shift females’ audio-visual perceptual strategy towards more visual reliance in middle-adulthood. Collectively, the three studies show that audio-visual speech perception is susceptible to change between young and middle-adulthood. Together with previous research on childhood and older adulthood, these findings underpin the flexibility and ever-evolving nature of audio-visual speech perception across the lifespan. That cognitive processing speed and gender interact with age-related audio-visual experience exemplifies the value of a lifespan perspective when exploring other influences on audio-visual speech perception.