Abstract
People find targets faster in repeated contexts than in novel ones (contextual cueing; Chun & Jiang, 1998). It is well known that contextual information guides attention to the target location. In addition, some studies found that it is difficult to adapt to the change of a target location in learned contexts (Manginelli & Pollmann, 2009; Zellin et al., 2014). However, the identity of the target was the same throughout experiments in their studies. We investigated whether the change in target identity enables people to learn new contextual information about a new target location in learned contexts as well as initial learning. To change the identity of a target, we used real-object images as stimuli. The task was to click the target object among 16 objects in the array using a mouse. There were two phases in Experiment, and the target identity differed across phases. At the beginning of Experiment, participants were informed the identity of a first target and the possibility of a target identity change after the first phase. The identity of a new target for the second phase was announced after the first phase. Contexts were either repeated or newly generated. Specifically, the locations of distractors in the repeated contexts were maintained throughout Experiment, whereas in the novel contexts they were randomly changed for each trial. We found significant contextual cueing effects in both the first target and second target phases. Furthermore, the size of these effects and their speed of learning was similar between the two phases. These results suggest that people can learn and use the new contextual information even in the learned contexts without interference, when the identity of the target was changed. Thus, different target identities enable people to form two separate associations with the two locations of different targets for a given context.