Younha Collins¹ (), Joseph Schmidt¹; ¹University of Central Florida

Individuals fail to recognize low prevalence (LP) targets, even when fixated directly (Hout et al., 2015), resulting in increased misses and a conservative criterion shift. Likewise, high target prevalence (HP) increases false alarms and shifts criterion liberally (Wolfe & VanWert, 2010). Decades of target-prevalence research focused on long-term-memory (LTM) targets, despite clear differences between LTM and visual-working-memory (VWM) systems (Goldstein & Beck, 2018). The current study examined whether the prevalence effect interacted with memory system by asking participants to search for VWM or LTM targets at low (10%) and high (90%) prevalences. Before the LTM search, participants memorized six targets and their categorical names. At the start of each search, the eye tracker detected central fixation (200ms) before two cues appeared (1s; text cues for LTM and images of real-world objects for VWM), followed by a 1s delay before an 18-item search array appeared (half from each cued category). Replicating prior work, LP produced higher misses than HP, and HP produced higher false alarms than LP. Additionally, the miss rates did not significantly differ between memory systems, suggesting similar low prevalence effects. However, false alarms were more frequent for VWM than LTM targets, suggesting VWM exacerbates the high prevalence effect. Signal detection analyses demonstrate that HP and VWM produced more liberal criteria than LP and LTM, respectively. Sensitivity was positively associated with prevalence for LTM but negatively associated with prevalence for VWM. Criterion correlated with correct-rejection RTs for LTM (replicating prior work), but not for VWM. Sensitivity correlated with correct-rejection RTs in LP but not HP in both memory systems. Taken together, this suggests that each memory system produces independent sensitivity and criterion thresholds and VWM exacerbates the high prevalence effect, demonstrating that there is not a singular prevalence effect, but rather one that interacts with target memory location.