EMOTION-ORIENTED COMPUTING: STATE of the ART and KEY CHALLENGES

1. Overview Data suggest that less than 10% of human life is completely unemotional. The rest involves emotion of some sort (http://emotion-research.net/deliverables/D3e%20final.pdf ), but probably less than a quarter consists of archetypal ‘fullblown emotions’ (brief episodes dominated by strong feelings). Most of it is coloured by ‘pervasive emotion’ (feelings, expressive behaviours, evaluations, and so on, that are inherent in most human activities). ‘Emotion-oriented computing’ is about technology that takes account of both types. Picard’s ‘Affective Computing’ in 1997 [1] established the field, and several EC projects made substantial contributions (NECA, ERMIS, SAFIRA, etc ). The EC Network of Excellence HUMAINE now brings together many of the most active teams (and its website provides much of the background material cited here). Emotion-oriented computing is intrinsically complex because emotional life is multifaceted. It involves affect (feelings and physical changes associated with them); cognition and conation (shifts in perception, judgment, and selection of action); interaction (registering other people’s emotional status and conveying emotion to them); personality; culture; ethics; and much more. The research challenge is huge because genuinely satisfying systems need to deal with all those facets in a co-ordinated way (see section 2). The obvious application of emotion-oriented computing is as part of the general drive to let machines interface with humans as richly as humans interface with each other. Since emotion pervades natural human interaction, emotion-oriented computing is as fundamental to the drive as speech recognition and synthesis. Its importance is likely to become clearer as allied technologies mature, and the lack of emotional ‘intelligence’ in otherwise humanlike systems becomes seriously anomalous. People already experience the effect to some extent in call centres that give the same bright message however long the user has been waiting. There is no single ‘killer’ application for systems that support truly natural communication with humans, but it is hard to doubt that demand for them would be massive. However, the timescale is long – decades – because there are very large amounts of fundamental research to be done. It is becoming possible to identify real short term applications for the capacities we have, though. They are reviewed in section 3. Setting the right balance between exploiting limited contemporary technology and making fundamental advances is a key challenge.