概率学习的两种认知过程研究

Abstract

目的：在重复预测任务中，决策者倾向于根据事件发生的概率匹配他们的预测频率，该行为违背了理性决策理论的最优化假设。以往的研究结论支持了匹配概率行为的两种假说：其一是基于简单的联结学习过程，即重复成功/回避失败的策略；其二是来自于更老练的模式搜索策略，比如赌徒谬误信念。本研究尝试利用记忆辅助操纵实现对认知过程的分离。方法：招募50名大学生，在重复预测任务中增加动态窗口反馈，更新近期出现的事件，窗口容量设置为3和9两个条件(各25人)。结果：1) 对连续事件敏感的行为模式表现出正近因效应(倾向于重复最近一次事件)和负近因(倾向于预测相反事件)的共存；2) 与低容量条件相比，高容量条件下负近因效应持续的时间(试次数)更长；3) 随着实验的进行，负近因效应逐渐被正近因所取代。结论：本研究采用新的实验操纵手段，为分离概率学习的认知过程提供了方法上的改良，为匹配概率行为的两种假说提供了直接的证据。 Probability matching is a classical behavioral anomaly observed in lab and field studies. When asked to make predictions of two mutually exclusive random events from Bernoulli process re-peatedly, people tend to match their predictive frequencies to observed event frequencies, which violate maximization assumption in rational decision theory. Previous studies suggest there are two independent psychological processes that lead to this behavior: Associative learning, pattern searching. Previous researchers’ attempts to distinguishing these two processes are failed because of the confounding of latent variables. Our study tried to use a cognitive auxiliary procedure, a dynamic feedback window, to extend short-term memory capacity of recent events. By manipulating the length of this window (at the level of 3 and 9), we can separate cognitive process of probability learning. The results showed that: First, by checking how responses are sensitive to streaks/runs of events, we find that the recency curve showed a complex wavy form: A mixture of the positive and negative recency effect. Secondly, as subjects gained experiences, negative part of the curve is gradually replaced by the positive effect, suggesting that the process of pattern searching is overturned a process of associative learning process. Third, the pattern of recency curve also changed when window length is increased (from 3 to 9): at the high-probability-event end of the curve, the shifting from decreasing to increasing occurred early in the condition of 3, but latter in the condition of 9; but at the low-probability-event end, the opposite is true. This result implies that the process of searching patterns may embed in the strategies of encoding and inferencing of streaks of events, which are sensitive to task situations. In conclusion, by using a dynamic feedback design to manipulate short-term memory, our studies separated cognitive process during probability learning, provided new evidence to associative learning and pattern searching hypothesis of probability learning. These are both meaningful from theoretical and methodological views.