Visual search can be facilitated by learning of statistical regularities, such as when repeatedly encountering a spatial layout of search items that speed the detection of the target (contextual cuing). While contextual cuing is usually thought to improve attentional guidance in finding the target, some studies have also suggested that the benefit for repeated search layouts arises from more efficient response-related processing, after the target has been localized. To test this latter account, the current study systematically varied the difficulty of the required target decision response by varying the number of possible response alternatives (while keeping attentional guidance constant). The results from Experiment 1 showed that contextual cuing was smaller with four than two possible response alternatives, thus indicating that a more difficult response decision actually hampered context-based learning. Moreover, Experiment 2 and 3 showed that even when providing additional, “learnable” (i.e. repeating and thus predictable) response cues did also not enhance contextual cuing, neither with two nor four possible response alternatives. This pattern of results thus suggests that contextual cuing does not facilitate processing after locating a to-be-searched for target, even when additional regularities are provided. On the contrary, the response selection actually appears to interfere with contextual cuing.