Affective Computing


                             FaceReader © Noldus Information Technology 2012


This is a 1 year project aiming to investigate ways of accurately gathering physiological data such as ECG, Galvanic Skin Response (sweat), and face recognition and aggregating these data sources to try and determine the emotion being experienced by participants whilst using computers. The Affective Computing (AC) project aims to design a computer system that responds to real-time changes in user affect (e.g., happiness, sadness, etc), as represented by speech, facial expressions, and physiological signals.

It is designed to investigate and explore the significant potential of AC for applications in areas benefiting from deeper insight into the user experience (such as game design, online shopping portal refinement, etc.) beyond traditional methods such as interviews and questionnaires. The project is focusing on the exploration of AC and its challenges to meet real world user needs and has established links with commercial partners such as Failbetter Games and Noldus to maintain a close connection between the research and user/real world requirements.


In order to address the complexity associated with AC and its (real world) application, the project follows a multi-disciplinary strategy to establish the foundations for AC research as part of Horizon. In particular, the project focuses on the combination of information sources both from multiple heterogeneous physiological sensors as well as individual-specific and contextual information. The project is centered around the creation of an AC test bed or ‘hot seat’ which in turn provides the platform for multiple prototypes which will support the research efforts in terms of information fusion and evaluation. The test bed and prototypes incorporate strong input from design in order to ensure good usability/user friendliness as well as to maximise their impact for dissemination activities and the investigation for the deployment potential of AC as part of real world applications.


Take a look at our Affective Computing Video



Christian Wagner – Principal Investigator

Partners: Noldus, Failbetter Games