Cognitive control and associative processes in a process dissociation model of the Stroop task with application to list-wide and item-specific proportion congruent effects

在一项开创性的研究中，Lindsay和Jacoby（1994）应用过程解离过程来解离经典Stroop任务中的单词阅读和颜色命名过程。为了验证估计的颜色命名参数（C参数）和单词阅读参数（W参数），Lindsay和Jacoby操纵了颜色原型性（仅影响C参数）和一致和不一致刺激的比例（仅影响W参数）。对于比例操纵，当Stroop刺激大多不一致时，干扰效应较小；一个结果模式最初归因于在单词阅读阻碍任务执行的情况下（即，当刺激大多不一致时），减少单词阅读过程的实验范围内的认知控制策略。然而，一种新的解释将比例一致效应归因于联想学习机制（Schmidt & Besner, 2008）。这一解释假设，当刺激频繁出现时，特定刺激与特定反应的联系就会加强（即，比例一致性操纵实际上可能是一种权变操纵）。鉴于这种对比例一致性效应的新解释，林赛和雅各比对W参数的解释显然是模棱两可的：只要比例一致性效应背后的过程不清楚，就不能得出由比例一致性操纵引起的W参数变化反映了阅读过程对反应影响的变化的结论。在W参数背后的过程尚不清楚的情况下，不能简单地通过比例一致性调节W参数来得出比例一致性效应代表了被调节的阅读过程的认知控制策略。为了打破这一恶性循环，我们计划利用可分离的W参数来研究W参数的基础过程以及比例一致效应的基础过程。有计划的实验将操纵认知控制和联想学习。这些操作对一致和不一致刺激W参数的影响将有助于我们评估认知控制和联想过程对单词阅读参数和比例一致效应的相对影响。

In a seminal study, Lindsay and Jacoby (1994) applied the process dissociation procedure to dissociate word-reading and color-naming processes in the classical Stroop task. To validate the estimated color-naming parameter (C parameter) and word-reading parameter (W parameter), Lindsay and Jacoby manipulated color prototypicality (affecting only the C parameter) and the proportion of congruent and incongruent stimuli (affecting only the W parameter). For the proportion manipulation, smaller interference effects are typically obtained when Stroop stimuli are mostly incongruent; a result pattern that was initially attributed to experiment-wide cognitive control strategies of reduced word-reading processes in case that word reading hinders task performance (i.e., when stimuli are mostly incongruent). However, a new account ascribes proportion congruent effects to an associative-learning mechanism (Schmidt & Besner, 2008). This account postulates that the link of a particular stimulus with a particular response is strengthened when the stimulus is frequently shown (i.e., a proportion congruent manipulation might in fact be a contingency manipulation). Given this new interpretation of proportion congruent effects, it is apparent that Lindsay and Jacoby's interpretation of the W parameter is ambiguous: As long as processes underlying proportion congruent effects are not clear, it cannot be concluded that a change in W parameters caused by a proportion congruent manipulation reflects a change of the impact of word-reading processes on responses. And as long as the processes underlying W parameters are not clear, it cannot be concluded that proportion congruent effects represent cognitive control strategies of modulated word-reading processes by simply showing that varying proportion congruence modulates W parameters. To break out of this vicious circle, we plan to investigate the processes underlying W parameters as well as the processes underlying proportion congruent effects with the help of separable W parameters for congruent and incongruent stimuli. Planned experiments will manipulate cognitive control and associative learning. The impact of these manipulations on the W parameters of congruent and incongruent stimuli will help us to assess the relative influence of cognitive control and associative processes on word-reading parameters and on proportion congruent effects.