Voices carry a wealth of information about the speaker, like their age or how
they are feeling, or indeed, whom they are speaking to. However, the systematic
exploration of how individuals differ in their ability to perceive voices has been slow, not least of all because standardised tests to assess voice perception abilities easily and quickly were not available.
This thesis presents a short, computerised standardised voice test, the Bangor
Voice Matching Test (BVMT). This short voice test has been constructed as a tool to assess a wide range of voice identity perception ability levels. Test construction was carried out using Item Response Theory to identify the most suitable items for this purpose. We found a large inter-individual variability in performance on the BVMT in a sample of healthy adults. Performance on the BVMT was only weakly to moderately correlated with auditory working memory and general auditory perception skills, pointing to voice perception as a distinct auditory ability.
Two subsequent behavioural experiments explored the links between interindividual differences in voice identity perception ability (as measured by the BVMT) and inter-individual differences in the size of the voice identity adaptation effects. Both correlated positively, and voice identity aftereffect sizes were not related to more general auditory abilities or a control adaptation condition (adaptation to musical harmony). These findings again highlight voice perception as a distinct auditory ability for which adaptive voice coding mechanisms play an important functional role.In a functional magnetic resonance imaging (fMRI) study, possible links between neural activation patterns for physical features of familiar voices and the BVMT as an independent behavioural voice perception correlate were investigated. Bilateral temporal voice areas (TVAs) as well as bilateral putamen coded physical differences between consecutive vocal morphs. BVMT performance correlated negatively with the size of this carry-over effect in insulae and right supramarginal gyrus. The recruitment of these areas might reflect a compensatory mechanism when
voice perception is challenging.
The findings of this thesis provide the foundations for the systematic research
of individual differences in identity perception by the sound of a person’s voice. It also opens new and exciting avenues for the exploration of how individual differences in identity perception link to other aspects of voice perception (e.g. affect), and how individual differences in voice perception relate to other, visual domains of person perception such as face perception.