Abstract
Recent research suggests that, when search targets are difficult to discriminate from distractors, fewer items can be processed within a single fixation, reflecting a narrowing of the Functional Viewing Field (FVF, Young & Hulleman, 2013; Hulleman & Olivers, 2016). Discrimination difficulty is also at the core of the low-prevalence effect (LPE) in visual search (Wolfe & Van Wert, 2010): Observers often miss rare targets, despite directly fixating them (e.g., Hout et al., 2015; Peltier & Becker, 2016). Across three experiments, we examined the hypothesis that discrimination difficulty, induced by prevalence manipulations, should affect the size of the FVF. In Experiments 1 and 2, observers searched for simple stimuli presented in circular arrays at varying eccentricies off-center. Target frequency was manipulated across three “invisible” blocks in each experiment, appearing in 25%, 50%, or 75% of trials. We measured the probability of directing the first fixation (FF) to the target as an index of relative FVF size. In both experiments, the probability of FF decreased as arrays appeared further from central fixation. Unexpectedly, the probability of FF was more likely during LP, relative to high-prevalence, search, particularly when arrays were close to central fixation. In Experiment 3, we eliminated eye movements by presenting search items in Rapid Serial Visual Presentation (RSVP) search streams. Peripheral distractors appeared at varying eccentricities in 20% of the RSVP trials, allowing us to use detection and identification as an index of FVF size. Although a classic LPE emerged, observers were more likely to miss peripheral distractors during high-prevalence search, replicating the narrowed FVF observed in Experiments 1 and 2. Together, these results show that FVF size is reduced as target-prevalence increases, suggesting that top-down, preparatory attention mechanisms mediate the attentional window.