We combine the Dimension-Action (DA) model with translational models to account for both the Stroop and the flanker effects. The basic assumption of the model is that there are distinct visual modules, each of which is endowed with both perception and response selection processes. We contrast this model with an alternative widespread view, the standard view, according to which the same response selection processes are shared by all tasks. The two views have different predictions concerning the flanker and Stroop tasks. Seven experiments test these predictions. The first five experiments show that there is a fundamental difference between the typical Stroop and flanker effects. Moreover, these experiments show that words denoting colors can affect print colors only when they are required for naming or when participants verbally mediate the print color task. Experiments 6 and 7 show that an analogous interaction between color and shape exists in the flanker task. These experiments as well as previous studies are consistent with the DA model and the modular view and pose serious difficulties for the standard view. Wider implications of a visual modular architecture are discussed as well.

中文翻译：

---

超越感知的模块化：来自单任务干扰范式的证据

我们将 Dimension-Action (DA) 模型与平移模型相结合，以解决 Stroop 和侧翼效应。该模型的基本假设是存在不同的视觉模块，每个模块都具有感知和响应选择过程。我们将此模型与另一种广泛的观点（标准观点）进行对比，根据该观点，所有任务共享相同的响应选择过程。这两种观点对flanker 和Stroop 任务有不同的预测。七个实验测试了这些预测。前五个实验表明典型的斯特鲁普效应和侧翼效应之间存在根本差异。而且，这些实验表明，只有在命名或参与者口头调解印刷颜色任务时，表示颜色的词才能影响印刷颜色。实验 6 和 7 表明，flanker 任务中存在颜色和形状之间的类似相互作用。这些实验以及之前的研究与 DA 模型和模块化视图一致，对标准视图构成严重困难。还讨论了视觉模块化架构的更广泛含义。