Abstract
The phenomenon of spatial contextual cueing can be observed by presenting subjects with a mixture of novel and repeated visual search arrays. Over time, subjects implicitly learn that in each repeated array the spatial locations of distractors are correlated with the spatial location of a target, and come to search through repeated arrays faster than novel arrays. Identifying the processing stage wherein this response time benefit arises is a topic of ongoing research and debate. In the first paper reporting contextual cueing (Chun & Jiang, 1998) it was hypothesized that implicit context recognition guided attention to the target location. However, Kunar et al (VSS, 2006) challenged this claim, making the case that contextual cueing exerts its benefit later, at a response selection stage. They introduced response conflict on half of the trials, and surprisingly, contextual cuing was abolished by the response conflict. However, they employed an efficient, pop-out search task to measure the role of a response stage, whereas prior contextual cuing studies have typically used inefficient, serial search tasks. Thus, we sought to replicate their disruption of contextual cueing by response conflict by using serial search rather than pop-out search. Under conditions of serial search, there was robust contextual cueing and no effect of response conflict. These results call into question the inferences made about contextual cueing using a pop-out search task and the general role of response conflict in contextual cueing.