Do people sound like they look?

Human faces and voices convey an abundance of socially important information. Understanding how we process them is of vital importance for both basic psychological science (developing models of person perception), and for real world forensic and security applications (e.g. witnesses to try and identify perpetrators from their faces or voices, verifying identity from surveillance recordings, or checking passports at border control). However, there is much we do not yet know.

We have all met people for the first time, having heard them speak (e.g. on a phone call, or the radio) but having never seen their face before. Sometimes, when we meet them, they look exactly as we had imagined, but other times they look completely different. Our research shows that some people look and sound more similar than others. That is, people are accurately and consistently able to match some faces and voices easily, while other faces and voices are more difficult to match (Smith, Dunn, Baguley & Stacey, 2016). However, we don’t yet know why this might be: We don’t know which aspects of the face and voice might be being used for identity matching, if everyone uses the same cues, or if we can predict which faces and voices might go together easily, and which faces might not.

We invite applicants to submit proposals on the topic of understanding face and voice identity matching, using an experimental methodology. Proposals could involve investigating the contribution of cues such as masculinity/femininity, dominance, attractiveness, race or sexuality, comparing face and voice ratings, or manipulating facial and vocal features through morphing. Proposals might also involve using eye-tracking methods, the role of cognitive biases, mental imagery, or whether confidence in face-voice matching judgments predicts accuracy.

This project offers the successful applicant the opportunity to develop a range of research skills, from research design, stimulus editing and experiment programming to the application of novel statistical techniques. The successful applicant will be supported by an experienced supervision team (Harriet Smith, Andrew Dunn, and Thom Baguley), specialising in person perception, theoretical and applied experimental psychology and statistical inference. They will be encouraged to develop their network and career opportunities by sharing and publishing their findings, and engaging with internal and external events, collaborators and stakeholders.

Director of Studies: Harriet Smith

NTU co-supervisors: Andrew Dunn, Thom Baguley