Affective computing is a novel paradigm originally proposed in 1997 by Rosalind Picard from MIT Media Lab in her paramount book. It builds on the results of psychology and medicine and aims at allowing computer systems to detect, use, and express emotions. While at first sight it may look very general from the computer science point of view, in fact it is a constructive and practical approach oriented mainly at improving human-like decision support as well as human-computer interaction.

Basically, affective computing aspires to narrow the communicative gap between the emotional human and emotional challenged computer by developing techniques to recognize and respond to affective states. However, to approach affect-sensitive functionalities in computers systems new methods and techniques should be developed. This new techniques and methods require the collaboration of multidisciplinary groups of different fields such us psychology and neuroscience.

For example, a key contribution of neuroscience in this field is the evidence that the neural substrates of cognition and emotion overlap substantially, mainly in decision-making processes. That is, cognitive processes operate continually throughout the experience of the emotion. This fact emphasizes the importance of affective components in human computer interaction.

Main Hypotheses

1. System adaptation in user-centric computing can be effectively improved by the ability to interpret user affects
2. Existing models of emotions require proper operationalization using artificial intelligence methods
3. To fully model affective states one needs to identify and interpret combined physiological signals and neuronal activity of the brain in decision-making processes.

Research Problem

The principal research problem to be addressed in this project is related to the search and interpretation of relations between physiological signals and neuronal activity of the brain in selected affective states (normal state and in decision-making processes). This analysis will allow not only to find correlations between psychophysiological signals and neural activity but developed a validation method in emotions recognitions based on neural activity.

Affective Context

A novel aspect of the project will include knowledge-based conceptualization of affective context. We use semantic knowledge representation methods to describe the affective context. This makes it both machine processable in the context-aware systems paradigm as well as understandable to the user.

Economic and Societal Impact

1. contributing to a wider adoption of context-aware systems
2. lowering the complexity and cost through the simplified design of such systems
3. improving adaptability of context-aware systems to the affective context of the user