Abstract
Attentional capture triggered by an onset distractor is subject to habituation, an ancestral form of plasticity consisting in a response reduction to a repeated irrelevant stimulation. Even if habituation is usually considered a form of non-associative learning, evidence from animal studies has shown that habituation can be specific for the context in which it takes place. In the present work, we used a two-session paradigm to investigate the hypothesis of a context-specific long-term habituation of attentional capture in humans. The task of participants was to report as fast as possible the orientation (left vs. right) of a target line presented in a cued location. On half of trials, the occurrence of the target was preceded by a visual onset distractor appearing in a different location. The key feature of our paradigm was the context manipulation, with the context defined by the background images (naturalistic images in Experiment 1, and abstract-geometrical images in Experiment 2). In both experiments, one group of participants maintained the same background image across the training and the test sessions, whereas the other group was exposed to a background change in the test session. In both experiments, in the training phase we documented an initial attentional capture that decreased gradually with practice, as exposure to the irrelevant distractor continued across blocks. Crucially, in the test phase, a robust recovery of the attentional capture emerged for the group of participants that was exposed to a context change. Our results strongly support the idea that habituation of attentional capture is context-specific, and that a context change leads to a disruption of the model of the habituating distractor, a result suggesting that the distractor filtering mechanisms take into account also the context in which the irrelevant information is presented.