Abstract: Emotions is analyzed by mathematical statistics as an influence factor on behavior interaction in the current behavioral game researches. On the face of the status quo, premising emotion rationality that is the alternative to results rationality, a fresh emotional game computing principle was proposed by building emotion behavior schemata function. The suggested thought was applied to a modified ultimatum game computing, an emotional equilibrium outcome was obtained, and the psychological mechanism of the brave behavior was revealed. Meanwhile, the basic principle of pattern recognition to behavior selection mode was established. According to the constraint conditions of the proposed principle, its Matlab simulation calculation was accomplished. The obtained results are consistent with the behavior laws revealed by the sayings “Not to do anything more than three times”, “He who strikes first gains the advantage” and “Leave adequate leeway”. The rationality and validity of the established emotional game computing system was confirmed. Therefore, it provides a feasible way for emotional game behavior research by the method of non-mathematical statistics.

Key words: behavior computing, game theory, emotion, ultimatum