A Logical Framework for Trust - Related Emotions : Formal and Behavioral Results by Manh Hung NGUYEN Co - supervisors

Recently, emotions play a more and more important role in intelligent behavior researching in Artificial Intelligent area, specially in intelligent interaction systems. Purposes of these researches are to construct natural interactions between human and machine as those among humans by assigning computers human-like abilities of observation, interpretation (understanding, representation, reasoning) and generation of affect features, including user’s emotions. This thesis proposes a logical framework which enables computer to represent emotions as well as to reason about emotions based upon some causal attitude which has a tight and natural relation with emotions: trust and distrust mental states. In other terms, from our proposed logical framework, we formalize the concept of some emotions as well as trust and distrust which are composed of their cognitive factors as argumentation of cognitive psychologists. From these formal concepts, we show that there are some relations between trust/distrust and emotions, particularly the effect of trust and distrust on emotions: a success following trust brings about satisfaction and/or gratitude; a failure following trust brings about disappointment and/or anger; a success following distrust brings about relief and/or gratitude; and a failure following distrust brings about fear-confirmed. These relations, after being formalized at the formal level, will also be validated by collecting empirical data, at the behavioral level.