本研究欲探討信任對於對社會動態發展影響及信任機制建立受那些因子影響。採用雙向信任賽局並納入個人差異進行研究分析，總共招募十組受測者，研究發現信任不一定能夠被建立，且每個組別動態發展存在差異，差異的原因受到初始情況、期中互動歷程影響，且兩部分皆受個人因素影響，初始情況是影響後續是否能建立良好合作與互惠關係之重要原因，愈是勤勞、溫順柔和、好動及豪爽的受測者其第一期願意於信任賽局投資之金額愈多；期中互動歷程，研究發現上一期投資金額、投資比率，及投資回報率對受測者而言是最重要的解釋變數，互動歷程相關變數則因人而異。 / In this study, we attempted to find out how trust affected social dynamic development and which factors will influence on the process of building trust mechanism. We recruited 10 groups of subjects and used two-way trust game to analyze it when individual differences were considered. This study showed that trust can’t be built under certain conditions and groups are heterogeneous in their initial conditions and interaction dynamics. This two parts are also influenced by individual differences. Initial condition is the key factor whether if the trust can be built. The more diligent, docile, lively, and forthright subject will offer more in first period. While in interaction dynamics within period, we found that invest, reward and invest ratio are the most important decisive factors to subjects. But variables about interaction dynamics within period depend on individual differences.

Cette thèse de doctorat traite de différentes questions concernant la communication et la transmission d’informations dans le cadre de la théorie des jeux. J’analyse différents dilemmes auxquels peut être confronté un joueur qui envoie des informations. Ces dilemmes correspondent aux questions suivantes: "Devrais-je investir dans un message vérifiable?", "Quand dois-je transmettre mon information?" et "Est-il préférable de ne pas envoyer mon information et uniquement de recueillir l’information des autres?". Cette thèse comprend une introduction et trois chapitres. L’introduction contient une motivation générale pour les trois problèmes que je présente dans cette thèse. Je donne une vue d’ensemble détaillée de tous les chapitres, j’examine la littérature relative au sujet et je la compare à mes résultats. / This Ph.D dissertation addresses different issues concerning communication and information transmission in a game theoretical framework. I analyze different dilemmas that a player who sends information has to deal with. These dilemmas correspond to the following questions: "Should I invest into a verifiable message?", "When should I pass my information?" and "Is it better if I do not send my information, but collect information from others?". This thesis includes an introduction and three chapters. The introduction contains a general motivation for the three different problems that I model in this thesis. I give a detailed overview of all the chapters, survey the related literature and compare it to my results.

This thesis compares forward neural networks with algorithms using game theory on basis of board game Blokus. The theoretical introduction part describes the characteristics of neural networks and work with them. There is also outlined algorithm of game theory. The second part deals about the implementation of players based on the outlined principles  and shortly descriptions GUI of application. In conclusion, the differences between the players  are evaluated on the charts created on the performed tests.