从演化论角度为“合取谬误”祛谬

Why Does Evolution Favor Human Cognitive Architecture Tending to Commit “Conjunction Fallacy” ?

"合取谬误"指的是这样一种推理谬误:人们认为单个条件发生的概率要小于多个条件联合发生的概率。由于很多心理学被试都在相关实验中犯下这一谬误,因此就有人认为其普遍存在,进而说明了人类在本性上是非理性的,具有违背概率论的顽固倾向。但本文试图为特定条件下出现的"合取谬误"提供三重合理化辩护。第一,在一定背景知识给定的情况下,若整个合取式从中所得到的被确证度要大于某个合取项从中所得到的被确定度,那对前者的偏好就是合理的;第二,合取式由于往往比合取项包含了更多的信息,因此对采集—狩猎时代人类的生存斗争来说就更具生存指导价值,故而对前者的心理偏好也就可能有着深刻的演化论根源;第三,在赌博中对内在概率较小的合取式的偏好也是合理的。这类偏好虽比对合取项的偏好更难得到满足,但一旦得到满足,其获得的收益也往往更大。

The so-called ＂conjunction fallacy＂ occurs when it is theoretically assumed that the conjunction of two conditions is more probable than a single general one. Since a high ratio of subjects are tending to commit this fallacy in psychological experiments, it may be tempting to assume that human-beings are by nature irrational in the sense that most of them cannot avoid violating simple rules of probability theory in practical reasoning. But perhaps philosophers can still defend human rationality by re-interpreting conjunction fallacy. More specifically, three arguments will be provided in this article for rationalizing human＇s preference of conjunctions rather than conjuncts. First, when certain background information is given, conjunctions partially confirmed by it should be more favorable than conjuncts entirely irrelevant to it. Second, conjunctions are more favorable than conjuncts also from the evolutionary perspective in the sense that the former is routinely more valuable in guiding and forming agents＇ actions leading to their survival and reproduction. Last, the risk of betting on a conjunction （with a low intrinsic probability value） would sometimes be offset by the higher reward of doing it, whereas the reward of betting on a conjunct is usually lower.