Oyuncular öğrenmeyi öğreniyorlar mı? değişen sayıda hareketli sabit toplam oyunlarından örnekler

This thesis investigates the learning behaviour of individuals in strategic environmentsthat have different complexity levels. A new experiment is conductedin which ascending or descending series of constant sum games areplayed by subjects and the experimental data including both stated beliefsand actual plays are used to estimate which learning model explains the subjects'behaviour best within and across these games. Taking learning rulesthat model the opponent as a learning agent and heterogeneity of the populationinto consideration, the estimation results support that people switchlearning rules across games and use different models in different games. Thisgame-dependency is confirmed by both action, beliefs and the joint estimations.Although their likelihoods vary from game to game, best response touniform beliefs and reinforcement learning are the most commonly used learningrules in the four games considered in the experiment, while fictitious playand iterations on that are rare instances observed only in estimation by statedbeliefs. Despite the change across games, there is no signicant link betweencomplexity of the game and the cognitive hierarchy of learning models. Beliefstatements and best response behaviour also differ across games as we observepeople making smoother guesses in large action games and more dispersedbeliefs statements in small action games. Inconsistency between actions andstated beliefs is stronger in large action games. The evidence strongly supportsthat learning and belief formation are both game-dependent.