Distractor filtering is affected by local and global distractor probability, emerges very rapidly but is resistant to extinction

Abstract

Effects of statistical learning (SL) of distractor location have been shown to persist when the probabilities of distractor occurrence are equalized across different locations in a so-called extinction phase. Here, we asked whether lingering effects of SL are still observed when a true extinction phase, during which the distractor is completely omitted, is implemented. The results showed that, once established, the effects of SL of distractor location do survive the true extinction phase, indicating that the pattern of suppression in the saliency map is encoded in a form of long-lasting memory. Quite unexpectedly, we also found that the amount of filtering implemented at a given location is not only dictated by the specific rate of distractor occurrence at that location, as previously found, but also by the global distractor probability. We therefore suggest that the visual attention system could be more or less (implicitly) prone to suppression as a function of how often the distractor is encountered overall, and that this suppressive bias affects the degree of suppression at the specific distractor-probability location. Finally, our results showed that the effects of SL of distractor location can appear much more rapidly than has been previously documented, requiring a few trials to become manifest. Hence, SL of distractor location appears to have an asymmetrical rate of learning during acquisition and extinction, while the amount of suppression exerted at a specific distractor location is modulated by distractor contextual probabilistic information.

Impaired selection of a previously ignored singleton: Evidence for salience map plastic changes

“…The experiment was constructed using PsychoPy3 Version 2020.1.3 software (Peirce et al, 2019), and because of the COVID-19 pandemic restrictions experiments were run online using the Pavlovia web hosting service (Open Science Tools Limited, Nottingham, UK). However, previous studies have shown that similar online experiments replicated the pattern of results obtained with experiments conducted in the lab (Duncan & Theeuwes, 2020;Valsecchi & Turatto, 2021). The search arrays consisted of six red or green shapes (circles or diamonds) containing horizontal or vertical bars (Figure 1a).…”

“…The sample size (18 observers) was determined to match the one that produced a reliable target position effect in the extinction phase of Experiment 3 of our previous study (Valsecchi & Turatto, 2021). Since in the current study the exact task in the most relevant comparison (colour singleton search post-training) differed from the one (shape singleton search) used in the previous study, we also computed the observed power for the target position effect in the test phase using G*Power 3.1.9.7 (Faul et al, 2007).…”

Does it help to expect distraction? Attentional capture is attenuated by high distractor frequency but not by trial-to-trial predictability.

“…This strengthens the inference that the observed attenuation of attentional capture is in truth caused by enhanced distractor suppression mechanisms. Other recent evidence supporting the conclusion that distractor frequency per se is a factor that modulates capture comes from the study by Valsecchi and Turatto (2021). This work demonstrated that distractor filtering is affected by both local and global distractor probability; when leaving the frequency of distractor occurrence at a particular location unchanged, increasing the overall distractor frequency in the search task (by increasing the frequency at other locations) leads to less capture at that particular location (Valsecchi & Turatto, 2021).…”

“…When confronted with many distractors, on the other hand, nonspecific suppression is rapidly in place. The observed time-course suggests that probabilistic expectations regarding distractor occurrence are formed very rapidly and speaks to the remarkable flexibility of second-order suppression, a flexibility that had so far only been reported for learned suppression based on predictable distractor characteristics (e.g., Valsecchi & Turatto, 2021; Wang & Theeuwes, 2020; Won et al, 2021).…”

“…flexibility that had so far only been reported for learned suppression based on predictable distractor characteristics (e.g., Valsecchi & Turatto, 2021;Won et al, 2021).…”