DECISION RULE IN PROBABILISTIC CATEGORIZATION - WHAT IT IS AND HOW IT IS LEARNED
DECISION RULE IN PROBABILISTIC CATEGORIZATION - WHAT IT IS AND HOW IT IS LEARNED
复制标题
DOI：
10.1037/0096-3445.106.4.427
复制
发表时间：
1977-01-01
影响因子：
4.1
通讯作者：
HEALY, AF
中科院分区：
文献类型：
--
作者：
KUBOVY, M;HEALY, AF
采用一个数值决策任务，设计作为一个信号检测的数值模拟，调查用于概率分类的决策规则的性质。两种条件进行了比较，12名付费志愿者参加了6个1小时的会议条件1，和12名参加了3个1小时的会议条件2。在这两种情况下，在每次试验中对5位数的2个分布中的1个进行采样，并且需要S来确定采样的分布。计算每个区组50次试验与最佳拟合静态截止值规则的偏差数。尽管条件1中的偏差明显多于条件2中的偏差，但条件1中的偏差数量与条件2中的偏差数量相比更接近于从最佳可用概率模型预测的偏差数量。此外，条件1和2在其他特性方面相似，例如练习效果和最佳拟合静态截止点的位置。本文的结论是：（a）在数字决策任务中，动态截断规则是个体决策规则的一级近似，（B）文献中已有的动态截断模型--加法算子模型--是不充分的，（c）一个新的动态截断模型--理想学习者模型--也是不充分的;（4）个体在错误后倾向于向规范规定的方向移动他们的截止值，但在正确回答后倾向于随机移动他们的截止值。(47参考）（PsycINFO数据库记录（c）2016阿帕，保留所有权利）
Employed a numerical decision task, designed as a numerical analog of signal detection, to investigate the nature of the decision rule used for probabilistic categorization. Two conditions were compared, with 12 paid volunteers participating in 6 1-hr sessions in Condition 1, and 12 participating in 3 1-hr sessions in Condition 2. In both conditions, 1 of 2 distributions of 5-digit numbers was sampled on each trial, and S was required to determine which distribution was sampled. The numbers of deviations from the best fitting static-cutoff rule were computed for each block of 50 trials. Although the deviations were significantly more numerous in Condition 1 than in Condition 2, in Condition 1 the numbers of deviations were much closer to those in Condition 2 than to the number of deviations predicted from the best available probabilistic model. Furthermore, Conditions 1 and 2 were similar with respect to other characteristics, such as effects of practice and locations of the best fitting static-cutoff point. It is concluded that (a) in a numerical decision task a dynamic-cutoff rule is a good 1st approximation to the decision rule adopted by individuals;(b) the only dynamic-cutoff models previously available in the literature—the additive-operator models—are inadequate;(c) a new dynamic-cutoff model—the ideal-learner model—is also inadequate; and (d) individuals tend to shift their cutoffs in the normatively prescribed direction after errors, but tend to shift their cutoffs randomly after correct responses.(47 ref)(PsycINFO Database Record (c) 2016 APA, all rights reserved)