Sylwia Hyniewska


Institution : University of Bath
Research field : Expression of emotions, perception of emotions in others, face, gaze, dynamics of emotions, integration of modalities, EMG
Discipline : Psychology / affective computing

NCCR position : PhD Student
NCCR PI : Susanne Kaiser

In 2013 I defended my PhD at the University of Geneva and at Télécom ParisTech.

To read about my university background in Psychology see my lab's webpage and to read about my lab in Paris see

In the first stage of my PhD project short audio-visual clips presenting a person in a face-to-face context with another human were evaluated. The study consisted of a forced-choice questionnaire based on the appraisal theory's items and on the attribution of emotional labels. Videos were coded with the FACS technique. The appraisal theory is used to predict the intuitive answers of participants to the observed expressions.
The second task was conceived to improve the existing communicational capacities of 'embodied conversational agents' (ECAs). Facial muscle movements were transposed to a virtual agent, Greta (developped by Catherine Pelachaud's team, at Télécom Paristech). The interface of this agent was used as a tool to validate the observations established in the first task. This application of the theory enabled to run perceptive studies to verify and detail some aspects of the results. What is more, the behaviours verified by perceptive tests as expressing some particular mental states were added to the agent's repertory, increasing the number of varied facial behaviours. A system (the Multimodal Sequential Expressions algorithm) was defined in order to enable the generation of emotional expressions that are ordered in time and belonging to different non-verbal communicative channels.