NTU's Fully-funded PhD Studentship Scheme 2023
Project ID: S3 18
As humans, we form judgements about other people just by looking at their faces. Some of these judgements can be formed very quickly and accurately, such as age and gender. Other judgements can be more difficult, such as how healthy we perceive other people to be, or judgements about their personalities. Some judgements can also be subjective, such as whether we like the other person. Psychologists have long theorised about which of these judgements contain a kernel of truth, and which are purely in the eye of the beholder. Now, modern techniques such as 3D imaging, Geometric Morphometric Methods, FACS coding, texture analysis, and psychometric scales allow us to perform rigorously controlled experiments and build mathematical models to determine the level of accuracy of different judgements, and to identify which shape, texture, colour and motion cues we use to form these judgements.
The project will address the following questions: 1) What aspects of facial appearance and motion affect judgements of personality, health, and likeability; 2) How accurate are these judgements of health and personality traits? 3) How can we best train artificial intelligence to accurately estimate these health and personality traits from photographs, videos, and 3D scans of human faces?
You will use machine learning and specialist software to analyse photographs, videos, and 3D scans of people’s faces to predict data you have collected about their health and personality. By showing these photographs, videos and 3D scans to observers in laboratory-based cognitive and behavioural experiments, you will collect data about the way different aspects of appearance and motion influence the judgements that observers form from looking at faces. By comparing these judgements with the health and personality data, you will determine how accurate these judgements are.
By answering these questions, this research will answer an age-old question about what makes people “judge a book by its cover”, which has implications for how people achieve successful social interactions. The results may also be useful in applications such as remote diagnosis and human-machine interaction.
There will also be an opportunity to work with collaborators in Asia to examine cross-cultural similarities and differences in perceptions.
This project would suit a student with interests in person perception, social interaction, and evolutionary approaches to behaviour. Good statistical skills are required, and experience of the statistical package R would be an advantage.
Dr Ian Stephen (DOS)
Dr Eithne Kavanagh
Prof Bridget Waller
For the eligibility criteria, visit our studentship application page.
How to apply
To make an application, please visit our studentship application page.
Fees and funding
This is part of NTU's 2023 fully-funded PhD Studentship Scheme.
Guidance and support
Application guidance can be found on our studentship application page.