functioning of the HMI accordingly, mood recognition may conceivably have a beneficial effect. It is in a way a logical extension of the principles to design interfaces for everyone ( Newell, 1993), although extending the scope of normal and disabled users to include calm and agitated users as well. Another possibility is that the interface also can show or represent emotions, i.e., be an affective interface. This could be seen as an extension, although a significant one, of adaptive interfaces. It would therefore also comprise the risks of adaptive interfaces, and possibly amplify them. The main risk is that two systems that reciprocally adapt to each other very easily run the risk of going into an unstable region. If the response to a recognised affect is inappropriate, which in control terms means that the damping of the system is inadequate, then this may all too easily lead to an amplification of deviations, rather than the neutralisation and homeostasis that is really wanted ( Maruyama, 1963).
For this reason alone, it is highly uncertain whether there is anything gained by having an affective interface, although there may be a potential advantage in having one that can recognise affects. The question to be pondered by researchers in HMI (and HCI) is therefore not whether an affective interface, in either sense, is possible or technologically feasible, but whether it is required. Once affective interfaces are available, and this will undoubtedly happen at some time, they may join the many other technological solutions that are in search of a problem. It would be nice if, for once, a genuine need was established first. At the moment, it is my belief that the virtue of human-human communication are exaggerated, and that the risks of affective interfaces need to be understood better. This will require a sorely needed revision of the dominating paradigms for IJMI.
Abelson, R. P. ( 1963). Computer simulation of "hot" cognition. In S. S. Tomkins & S Messick (Eds.), Computer simulation of personality. New York: Wiley.
Attneave, F. ( 1959). Applications of information theory to psychology: A summary of basic concepts, methods, and results. New York: Holt, Rinehart & Winston.
Conant, R. C. & Ashby, W. R. ( 1970). "Every good regulator of a system must be a model of that system". International Journal of Systems Science, 1( 2), 89-97.
Edwards, W. ( 1954). "The theory of decision making". Psychological Bulletin, 51, 380-417.