Jeunghwan Choi¹ (), Jeremy Wolfe^2,3, Ava Mitra², Nathan Trinkl², Sang Chul Chong^1,4; ¹Graduate Program in Cognitive Science, Yonsei University, ²Brigham and Women’s Hospital, ³Harvard Medical School, ⁴Department of Psychology, Yonsei University

In visual search, people typically find targets more quickly in repeated displays than in novel ones, an effect known as contextual cueing (Chun & Jiang, 1998). For example, you may find your phone more quickly in your own room than in a stranger’s room even if you don’t explicitly recall the phone’s location. Two mechanisms have been proposed to account for this effect: attentional guidance and response facilitation. Attentional guidance posits that repeated displays facilitate the search, guiding observers to the target so they encounter fewer distractors. Response facilitation holds that the search is unchanged but the speed of target identification is increased. Here, we add a third factor. The accuracy of target identification is improved by repetition. People sometimes fail to detect the target although it is clearly in view (known as a Look But Fail To See (LBFTS) error, Wolfe et al., 2022). Observers can also make LBFTS fixations where they fixate near the target, fail to identify it, and only find it on a later fixation. In the present study, we investigated whether participants make fewer LBFTS fixations and errors in repeated displays than novel ones. We conducted a contextual cueing experiment while recording participants’ eye movements. Participants were asked to find a T shaped target among L shaped distractors. Half of the displays were repeatedly presented throughout the entire experiment. We found faster reaction times (RTs) and lower LBFTS fixation and error rates in the repeated condition than in the novel one. Moreover, we obtained the same result in a previous contextual cueing study (Choi & Chong, 2020) when we reanalyzed those data. These results suggest that people can identify the target more accurately in the repeated displays and this more accurate target detection contributes to faster RTs in the contextual cueing paradigm.

Acknowledgements: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT)(RS-2022-NR070542) and NSF 2146617 & NIH-NEI EY017001.