Augustinova Maria, Silvert Laetitia, Spatola Nicolas, Ferrand Ludovic
Université Clermont Auvergne, CNRS, LAPSCO, F-63000 Clermont-Ferrand, France; Normandie Université, UNIROUEN, CRFDP, F-76000 Rouen, France.
Université Clermont Auvergne, CNRS, LAPSCO, F-63000 Clermont-Ferrand, France.
Acta Psychol (Amst). 2018 Sep;189:54-62. doi: 10.1016/j.actpsy.2017.03.009. Epub 2017 Apr 11.
The aim of this paper is to extend the so-called semantic Stroop paradigm (Neely & Kahan, 2001) - which already successfully distinguishes between the contribution of the semantic vs. response conflict to Stroop interference - so that it can take account of and capture the separate contribution of task conflict. In line with this idea, the Stroop interference observed using the aforementioned paradigm with both short and long RSIs (500 vs. 2000ms) did indeed reflect the specific contribution of the task, semantic and response conflicts. However, the contribution of task conflict (as opposed to the semantic and response conflicts) failed to reach significance when the semantic Stroop paradigm was administered with manual (Experiment 1) as opposed to vocal responses (Experiment 2). These experiments further tested the extent to which the specific contribution of the different conflicts can be influenced by the increased cognitive control induced by a short (vs. long) RSI. The corresponding empirical evidence runs contrary to the assumption that the reduction of overall Stroop interference by a short (vs. long) RSI is due to the reduced contribution of the task (Parris, 2014) and/or semantic (De Jong, Berendsen, & Cools, 1999) conflicts. Indeed, in neither experiment was the contribution of these conflicts reduced by a short RSI. In both experiments, this manipulation only reduced the contribution of the response conflict to the overall Stroop interference (e.g., Augustinova & Ferrand, 2014b). Thus these different results clearly indicate that Stroop interference is a composite phenomenon involving both automatic and controlled processes. The somewhat obvious conclusion of this paper is that these processes are more successfully integrated within multi-stage accounts than within the historically favored single-stage response competition accounts that still dominate current psychological research and practice.
本文的目的是扩展所谓的语义斯特鲁普范式（Neely & Kahan，2001）——该范式已成功区分语义冲突与反应冲突对斯特鲁普干扰的贡献——以便它能够考虑并捕捉任务冲突的单独贡献。基于这一想法，使用上述范式在短和长反应间隔（500毫秒对2000毫秒）下观察到的斯特鲁普干扰确实反映了任务、语义和反应冲突的特定贡献。然而，当语义斯特鲁普范式采用手动反应（实验1）而非语音反应（实验2）时，任务冲突（与语义和反应冲突相对）的贡献未达到显著水平。这些实验进一步测试了不同冲突的特定贡献在多大程度上会受到短（与长）反应间隔所引发的认知控制增强的影响。相应的实证证据与以下假设相反：短（与长）反应间隔导致的整体斯特鲁普干扰减少是由于任务（Parris，2014）和/或语义（De Jong，Berendsen，& Cools，1999）冲突的贡献减少。事实上，在两个实验中，这些冲突的贡献都没有因短反应间隔而减少。在两个实验中，这种操作仅减少了反应冲突对整体斯特鲁普干扰的贡献（例如，Augustinova & Ferrand，2014b）。因此，这些不同的结果清楚地表明，斯特鲁普干扰是一种涉及自动和控制过程的复合现象。本文有点显而易见的结论是，与在当前心理学研究和实践中仍占主导地位的、历史上备受青睐的单阶段反应竞争理论相比，这些过程在多阶段理论中能更成功地整合。
