The computational complexity of rationalizing boundedly rational choice behavior

We determine the computational complexity of various choice models that use multiple rationales to explain observed choice behavior. First, we demonstrate that the notion of rationalizability by K rationales, introduced by Kalai et al. (2002), is NP-complete for K greater than or equal to two. Then, we show that the question of sequential rationalizability by K rationales, introduced by Manzini and Mariotti (2007), is NP-complete for K greater than or equal to three. Finally, we focus on the computational complexity of two models that refine this model of sequential choice behavior. We establish that the model of choice by game trees, from Xu and Zhou (2007), is NP-complete while the status-quo bias model, from Masatlioglu and Ok (2005), can be verified in polynomial time.