Abstract
The Low Prevalence Effect (LPE) is the persistent problem of target detection being exceptionally poor when search targets are rare. One method shown to improve LPE is to add a set of target-present trials that provide feedback to the task. We previously showed that distributing these “probe” trials throughout the search task improved accuracy while increasing search times in target-absent trials. However, it is difficult to determine if the improvement in search performance is a product of the feedback provided by the probe trials, an increase in overall target prevalence rate from the inclusion of probe trials, or perhaps both. To address this, we asked participants to search for a T or O among arrays consisting of L and Q distractors. Participants performed two blocks of a low prevalence search task (10% for each target), one control block and the other block containing an additional 50 probe trials, with 80% of the probe trials displaying one target and 20% displaying the other. Additionally, half of our participants received feedback on the probe trials with the other half not receiving feedback on those probe trials –just an increase in target prevalence. Our preliminary results show that while there is an improvement in low prevalence target detection from the probes with feedback when compared to the control block, there is also an improvement observed from the probes without feedback (albeit, not as robust). These findings suggest that while probe feedback is important in reducing the LPE, the increase in target prevalence from the probe trials also plays a crucial role.