Disentangling affective and perceptual processing in emotion recognition: Applications to health and affective computing.

The goal of this project is to disentangle affective and perceptual processing in emotion recognition and apply this understanding to improving human-robot interaction with affective robots.
Description of the Project: 

This interdisciplinary project between Psychology and Computer Science seeks to understand the relative contribution of perceptual (visual information) and affective processing in emotion recognition of faces by using stimulus sets tailored to participants. This project will advance both theory, such as the double empathy problem in people with autism and the facial feedback hypothesis, as well as support the development of personalised health behaviour interventions. In addition to being important for effective social communication, facial emotion recognition is disrupted in a number of conditions (e.g., depression, autism and Parkinson’s) and is increasingly important in the development of artificially intelligent systems (affective computing) including social robots (e.g., Furhat robot). Behavioural performance will be measured for different facial expressions across a range of experimental paradigms. Recording of participants’ eye movements as well as quantification and classification of production of facial expressions with participants from different populations (e.g., autistic traits) will also take place.  

Resources required: 
Robot for expressive interaction, e.g. the FurHat
Project number: 
240019
First Supervisor: 
University: 
Heriot-Watt University
First supervisor university: 
Heriot-Watt University
Essential skills and knowledge: 
Machine learning
Desirable skills and knowledge: 
ROS, interest in dialogue and NLP Background in Psychology
References: 

Delicato, L. S. (2020) A robust method for measuring an individual’s sensitivity to facial expressions. Attention, Perception & Psychophysics. 82 (6): 2924 – 2936 doi: 10.3758/s13414-020-02043-w