In a signal detection theory (SDT) approach to associative learning, one assumes that, when a subject is exposed to a flow of stimuli, an association is created between the internal representations of a cue and of an outcome, allowing the representation of the cue to activate the representation of the outcome. The outcome activation is a random variable drawn from a Gaussian distribution with mean m (sensitivity to the contingency) and standard deviation d (variability in outcome activation). Depending on whether the outcome activation is above or below various decision thresholds, the participant perceives either a negative, a null, or a positive contingency between the cue and the outcome. This study presents a detailed SDT analysis of the performance of four participants on whom data in a contingency assessment task were collected almost daily during several months. Parameters from the SDT model proved relatively stable over time, except if feedback was provided to the subject. In that case, for some participants but not all, the sensitivity increased. The decision criteria were also affected. Some of these changes endured despite the discontinuation of feedback. The variability in outcome activation was not affected by the feedback.

在用于关联学习的信号检测理论（SDT）方法中，人们假设，当受试者暴露于刺激流中时，会在提示的内部表示和结果之间建立关联，从而允许提示的表示。激活结果表示。结果激活是从高斯分布中得出的随机变量，平均值为m（对偶发事件的敏感度），标准差为d（结果激活的可变性）。根据结果​​激活是高于还是低于各种决策阈值，参与者会感知提示与结果之间的否定，空或正或有偶然性。这项研究对四个参与者的表现进行了详细的SDT分析，这些参与者在几个月中几乎每天都收集了应急评估任务中的数据。事实证明，SDT模型的参数在一段时间内相对稳定，除非向受试者提供了反馈。在这种情况下，对于某些参与者但不是全部参与者，敏感度增加了。决策标准也受到影响。尽管反馈中断，但其中一些变化仍然存在。结果激活的可变性不受反馈的影响。