当前位置: X-MOL 学术Topics in Cognitive Science › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Beliefs, Actions, and Rationality in Strategical Decisions
Topics in Cognitive Science ( IF 2.9 ) Pub Date : 2021-05-07 , DOI: 10.1111/tops.12534
Zheng Wang 1 , Jerome R Busemeyer 2 , Brahm deBuys 1
Affiliation  

A puzzling finding from research on strategical decision making concerns the effect that predictions have on future actions. Simply stating a prediction about an opponent changes the total probability (pooled over predictions) of a player taking a future action compared to not stating any prediction. This is called an interference effect. We first review five different findings of interference effects from past empirical work using the prisoner's dilemma game. Then we report interference effects obtained from a new experiment in which 493 participants played a six-stage centipede game against a computer agent. During the first stage of the game, the total probability following prediction for cooperation was higher than making a decision alone; during later stages, the total probability following prediction for cooperation was lower than making a decision alone. These interference effects are difficult to explain using traditional economic models, and instead these results suggest turning to a quantum cognition approach to strategic decision making. Toward this end, we develop a belief-action entanglement model that provides a good account of the empirical results.

中文翻译:

战略决策中的信念、行动和理性

战略决策研究的一个令人费解的发现涉及预测对未来行动的影响。与不陈述任何预测相比,简单地陈述关于对手的预测会改变玩家采取未来行动的总概率(汇总预测)。这称为干扰效应。我们首先使用囚徒困境博弈回顾过去实证工作中干扰效应的五种不同发现。然后我们报告了从一项新实验中获得的干扰效应,其中 493 名参与者与计算机代理进行了六阶段的蜈蚣游戏。在博弈的第一阶段,预测后合作的总概率高于单独做出决定的概率;在后期,预测合作后的总概率低于单独做出决定。这些干扰效应很难用传统的经济模型来解释,相反,这些结果表明转向量子认知方法来制定战略决策。为此,我们开发了一个信念-行为纠缠模型,该模型很好地说明了实证结果。
更新日期:2021-05-07
down
wechat
bug