In the first part of this talk we introduce Game Theory, its solution concepts and applications. We then discuss the limits of classical game-theoretic predictions and give possible reasons for that. At the end of the first part we present some behavioral and evolutionary models. These models have better predictive power than classical models and are therefore more useful in real life.

In the second part of this talk a joint work with Assoc. Prof. A. Ule is presented. The main topics of our research are reputation and (dis)honesty.  Dishonesty has a negative impact on everyone if and when the truth is revealed, and the consequences can be severe. How dishonesty spreads through the social network? Can we eliminate (or at least reduce) harmful lies and mistrust, and thus inefficiency? We try to answer these kind of questions.

We are also interesting in finding stable strategies, i.e. strategies that persist over time. We present our theoretical model and (partial) results. In order to get some new insights about people behavior (e.g. what strategies do people use), we also ran a pilot experiment. The experimental results are presented in the last part of the talk.