蒙提霍爾問題中的頓悟式學習

Abstract

研究如何修正人們在特定的困難選擇下的不理性行為在瞭解人的選擇行為上扮演重要的角色。本論文中，我們檢視人們在實驗室玩一個修改後的蒙提霍爾問題時的行為。我們的結果顯示，一個比較簡單的蒙提霍爾問題(100個門)會使人們之後玩比較難的版本(3個門)時選擇最適解(即”換”)的比例顯著地升高到80％。除了那些允許受試者溝通和/或競爭的研究之外，這個結果比起絕大多數過去的文獻結果都更好。此外，從估計個人層級的學習模型結果顯示，大部分人的學習軌跡比較可能是頓悟式學習，而非增強式學習。

Investigating methods to correct irrational behavior in a choice anomaly is central to understanding people’s choice decision. In this paper, we examined subjects’ decision in a modified version of the Monty Hall Problem (MHP) under a laboratory experimental setting. Our results show that an easier (100-door) version of the MHP would significantly raise subjects’ optimal choice (i.e. switch) rate later in the harder one (3-door) up to more than 80%, higher than most of the previous work without subject communication and/or competition. Moreover, results from estimating structural learning models using subject-level data show that the individual learning process is more likely to be an epiphany rather than a gradual one, such as reinforcement learning.