Study On Emotion Driven Behavior Selection Mechanism Of Artificial Life

There are two main problems on the traditional (excluding emotion) Behavior Selection Mechanism (BSM) of artificial life.â‘ Under uncertainty conditions, BSM can't achieve the autonomous behavior selection, devoid of the adaptability for its existing environment.â‘¡Under same conditions, BSM's behavioral responses are completely same, absence of the diversity of them. The successful application of emotion mechanism to man-machine interaction and reinforcement learning, excites researchers engaging the research on BSM to apply it to their fields in order to solve these two problems.According to the viewpoints on working principle of emotion mechanism in BSM, emotion-related behavior selection mechanisms are divided into three types: emotion-affected mechanism, emotion-emerged mechanism, and emotion-driven mechanism. After reviewing carefully the gained results of these three kinds of emotion-related BSM(EBSM), three problems have been founded.â‘ The EBSMs show a systematic default that they have not included the basic elements, which BSM should have had.â‘¡The architectures of EBSM and emotion models are different, there are no successions to each other.â‘¢The present EBSMs and emotion models are designed, there are not the corresponding academic supports, on relationship between emotion and behavior, for them, from ethnology, psychology or economics. These existing problems explain that the fundamental problem is, at present, that there are not elementary theories of EBSM, which is the result of other emotion-related disciplines having not discovering clearly emotion mechanism. Therefore, the research topic of this paper is to build up an elementary theories system of EBSM, the innovative results of the research are followings.â‘ Emotion schema function has been set up to discover the functional characteristic of relationship between emotion and resource. Through careful analyses of the emotion-related research results of psychology and economics, energy essence and behavior driving function attribute of emotion, resource-obtained meaning attribute of emotion color, are elucidated, emotion essence of the following concepts is discovered: motivation, (inner) driving force, utility, subjective value. Based on Prospect Theory, therefore, emotion schema function is built up. â‘¡The elementary principle of emotion-driven BSM has been founded. The theoretic foundation of EBSM in ethnology is founded, by introducing emotion mechanism to classical psycho-hydraulic pressure model in ethnology to resolve its two defaults of addition between heterogeneous dimensions and motivation intensity being independent of artificial life's living circumstances. Applying emotion mechanism to the expected utility family behavior selection theories in economics, the constraint of mathematical expected principle to them has been broken in principle. The single mapping relation between living resources and behaviors in ethnology is testified by Mental Accounting phenomena in economics, thus, the single mapping relation between behavior and emotion schema function is set up. Being dead against the unlimited generalization of the limited rationality principle, emotion rationality principle is put forward. Thus, based on these three-research works, the elementary principle of emotion-driven BSM has been founded to lay an academic foundation for the settlement of two problems that aritifical life body can't achieve the autonomous behavior selection under incertainy conditions and its behavior response is"distorted". Through the application of present theory to the settlements for Allais paradox, Exchange paradox, and St. Petersburg paradox, the emotion-driven behavior selection principle is validated preliminarily.â‘¢The elementary process of emotion organising behaviour has been discovered. Particular emotion-driven BSMs are built up to confirm the proposed thought. By applying the universal normal cloud model theory of artificial intelligence with uncertainty, ability schema function is established; consequently, the competitive and cooperative emotion-driven BSMs are constructed. And, applying the competitive emotion-driven BSM to fight mechanism, an emotion-driven fight BSM is established, which discovers the three laws appeared in animal's fight behaviors: upgrading characteristic, effect of wining/losing, challenge preference. The simulations of them in an uncertainty artificial enviorment constructed on Swarm, one of standard platform of artificial life, validate the present EBSMs.â‘£As the advanced validation of emotion-driven BSM principle, substituting emotion payoff for utility payoff, the thought of emotion-driven BSM is introduced to game behabvior research to propose an emotion-driven game principle of artificial life. Taking the modified ultimatum game for a example, the related equilibrium are gained by game analysis, which explain the rationality of familiar gaming policies:"Catch the ball before the bound","Be insatiable","Winding path leads to sequestered place","Try peaceful means before resorting to force"that traditional game theory have not explained. Not only the analysis testifies emotion-driven gaming thought of artificial life, but also illuminates that outspreading the in-depth research on it will be meaningful.During the course of research, the following jobs are also accomplished. Making use of the nerve-biological mechanism of emotion production and experience, the correctness of"Heart Governing Mind"theory of herbalist doctor psychology is explained. Therefore, it is hypothesized that the heart is the primary organ sensing emotions, which has emotion-driven BSM had an organic carrier. Moreover, according to the research results of relationship between"body and mind"diseases and emotions, emotion-working mechanism in the course of natural selection is discovered, deepening Darwin's views of the problem. Centering on the efficiency of resource-gaining behaviors, the logical relation among different emotions is discovered, which sets forth the general picture of emotion experienced by an anima in its living activities.In conclusion, by our reseach efforts, the primary theoretical framework of emotion-driven behavior selection mechanism of artificial life has been built up which has included the necessary elements and had the emotion-related theoretical supports in ethnology, emotional psychology and economics, which lays a foundation for the settlement of achieving the autonomous behavior selection in uncertainty conditions, and the construction of real artificial life body.