Abstract
Visual attention can be guided by statistical regularities in the environment that people implicitly learn from past experience. Moreover, a perceptually salient element is able to automatically capture attention gaining processing priority through a bottom-up attentional control (AC) mechanism, independently of the volitional goals of the observer. The aim of our study was to investigate the interaction between attentional biases developed through statistical learning (SL) and bottom-up signals, to directly test whether the former might shape the attentional selection/suppression of salient items or whether its effects are blocked by a prevalence of saliency signals. By using the same visual search task, we manipulated target frequency (high vs. low) across spatial locations while, in some trials, we introduced a salient stimulus that differed from all the other elements in the array in terms of colour. In Experiment 1 the salient item was an irrelevant distractor, whereas in Experiments 2 and 3 it was the target. In the latter experiment, we also recorded the EEG activity, focusing on N2pc, i.e. an ERP markers related to target selection. Our results revealed that both SL and target-related saliency signals significantly and independently enhanced behavioural performance for targets at high (vs. low) frequency locations and for salient (vs. non-salient) targets, respectively. Furthermore, the benefit of bottom-up AC was confirmed by a larger N2pc, albeit in this case the effect was only evident for salient targets at the low-frequency location. In contrast, when a saliency signal was associated to an irrelevant distractor, participants showed longer RTs compared to when the display was homogeneous, again without any behavioural interaction with SL effects. Overall our findings suggest that SL and bottom-up AC jointly (and mainly independently) contribute to establishing attentional priority for specific spatial locations