Abstract
Contextual information can provide an important source in predicting relevant goals for behavioral orienting. For example, in visual search, detection of a target is faster when a layout of nontarget items is repeatedly encountered, as compared to search within novel item arrangements. These results show that contextual invariances can implicitly guide attention to the target location (contextual cueing; Chun & Jiang, Cogn. Psychol., 1998). Here, we explored how contextual cueing can adapt to environmental changes. A series of experiments were performed in which observers learned to associate a given repeated context with a given target in an initial learning phase. Then, in a subsequent test phase, location (and task) changes were introduced to the target. The results showed that contextual cueing could not compensate for unpredictable changes of the target location. By contrast, robust adaptation occurred when a change was predictable (i.e., learnable), suggesting that a given learned context can be “remapped” and adjusted to novel requirements (Conci, Sun & Müller, Psychol. Res., 2011). Subsequent experiments demonstrated that target location changes are only effectively remapped when both the initial and the future target positions are stable and remain predictable across the entire experiment. Otherwise, contextual remapping fails, demonstrating that target locations mutually interfere with each other. In sum, this pattern of results suggests that multiple, predictable target locations can be learned and associated in parallel with a given repeated context, allowing the flexible adaptation of previously learned contingencies to novel task demands.