The role of singleton salience in second-order suppression

Abstract

To process the various stimuli in the real world effectively, it is essential to suppress salient distractions. Recent studies employing majority search paradigms found that people can suppress the capture of salient color singleton distractors even without knowing their color, which proved that suppressing a salient singleton was not only based on first-order information (e.g., a specific color) but also on second-order features (e.g., a unique color) or global salience information (e.g., a feature singleton). However, it remains unknown whether variations in salience influence the attention suppression effect in the higher-order suppression mechanism. In the current study, we explored the effect of salience on suppression by adopting the majority search paradigm and varying the singleton salience in different ways (set size, color contrast, and luminance contrast). Across four experiments, Experiment 1 showed that manipulating relative singleton salience through set size did not affect higher-order suppression, as reflected by comparable search facilitation across set sizes. Experiments 2 and 3 employed the capture-probe paradigm and demonstrated that participants can proactively suppress singleton distractors based on second-order salience. Experiment 4 employed a psychophysical oddball detection task to further validate the effectiveness of the salience manipulation and used eye tracking to examine whether oculomotor suppression differed between high- and low-salience singleton distractors. Furthermore, by manipulating the luminance contrast of singletons, participants had a stronger suppression effect on low-salience singletons compared to high-salience singletons. We concluded that the level of salience is a potential factor in influencing the attentional suppression effect.

The materials and datafiles of this article have been publicly available at https://osf.io/f9qd8. This study’s design and hypotheses were preregistered at https://osf.io/fk8jy.