Article Preview
TopIntroduction
Modern information systems for recording people and their interactions (e.g. surveillance systems, mobile communication networks, social robots) collect enormous amount of data which is being processed in time and effort consuming way, yet far from being sufficiently competent to be reliable in performance. Most computing algorithms aim at modeling people as physical objects in dynamic interactions. To achieve better social competence computing systems need to be able to identify the underlying processes guiding human interactions by extracting relevant features for the respective cognitive level of analysis of the perceived information. Apart from identifying other people as physical objects, human perception is capable of identifying signifiers of social signals as well as psychological features that are necessary for adapting to life in the society (Forsyth, 1990).
In the current paper it is proposed to distinguish three levels of cognitive analysis of the observed human behavior. If our system is capable of predicting behavior in response to behavior, it operates on the physical level of description of the observed situations. In human society people make predictions of other peoples’ behaviors in response to social and psychological drives like attitudes and opinions, rather than simply responding to the observed physical drives. If the computing system is capable of predicting behavior in response to attitudes – then it operates on the social level of description of the observed behavior. Whenever people describe other people’s behavior as opinion-driven – then they provide interpretation on the psychological level of description of the observed behavior. The main claims in the current paper are that: a) it is completely possible to increase system intelligence to the level of social interaction understanding by making it capable of predicting behavior in response to attitude; b) at present it is important to make the socially competent computing system adapted to functioning in social situations and capable of introducing structure to various social situations like technology-mediated education (Dimitrova, 2008, Dimitrova & Wagatsuma, 2011, Young et al., 2011), social robotics (Barakova & Lorens, 2010, Barakova & Vanderelst, 2011), smart environments (Deleawe et al., 2010), hospital surveillance systems (Casale et al., 2011) and others. At the same time this is not the case with the efforts to make the computing system psychologically competent in terms of predicting behavior in response to opinion, which is both problematic and, probably, often irrelevant (e.g. Sharkey, 2008). The main focus of the current paper, therefore, is the ability of a computing system to operate on the social level of description of a situation by reading of the relevant cues by especially designed sensors, called social sensors in a way similar to people reading social cues. It will be shown that the task of designing technical systems/devices capable of predicting behavior in response to attitude is a completely achievable present-day task.