The Conjunction Error and the Representativeness Heuristic
The Conjunction Error and the Representativeness Heuristic
复制标题
连接错误和代表性启发法
DOI：
10.1521/soco.1985.3.3.266
复制
发表时间：
1985
期刊：
影响因子：
1.9
通讯作者：
G. Wells
中科院分区：
文献类型：
--
作者：
G. Wells
在复合事件结合两个不具有代表性（不太可能）的事件、两个具有代表性的事件或一个具有代表性的事件和一个不具有代表性的事件的条件下，测试了其中复合事件被判断为比单独考虑的其元素之一更可能的连接错误。总共使用了12个问题，每个条件下重复四次，144名受试者每人解决两个问题。正如代表性启发式（Tversky & Kahneman，1983）所预测的那样，通过配对两个非代表性事件（平均7%的错误），几乎完全消除了连接错误，而将代表性事件与非代表性事件配对产生了高比率的连接错误（平均= 72%的错误）。配对两个有代表性的事件产生了一个中等的连接错误率（平均= 24%的错误）。结果被解释为是一致的代表性启发式观点的连接错误，并支持特沃斯基和卡尼曼的索赔th…
The conjunction error, wherein a compound event is judged to be more likely than is one of its elements considered alone, was tested under conditions in which the compound event joined two unrepresentative (unlikely) events, two representative events, or one representative and one unrepresentative event. A total of 12 problems were used, four replications in each condition, and each of 144 subjects worked two problems. As predicted by the representativeness heuristic (Tversky & Kahneman, 1983), the conjunction error was eliminated almost totally by pairing two unrepresentative events (average 7% errors), whereas the pairing of a representative with an unrepresentative event produced a high rate of the conjunction error (average = 72% errors). Pairing two representative events produced a moderate rate of the conjunction error (average = 24% errors). The results are interpreted as being consistent with the representativeness-heuristic view of the conjunction error and support Tversky and Kahneman's claim th...