Abstract
Previous research has shown that visual search efficiency is affected when using prior knowledge (e.g., looking for semantically-related objects; Wu et al., 2013, see also Telling et al., 2009) and novel information (e.g., using patterned sequences vs. random sequences to guide attention; Zhao et al., 2013). How might the use of prior knowledge and novel information impact search efficiency when both can be utilized during search? Across two experiments, this study used a visual search paradigm (2-item search array) with known real-life objects (food vs. toys) and statistical patterns with novel rune symbols as proxies for prior knowledge and novel information, respectively. Rune patterns were presented in either structured or random sequences in the locations of upcoming target and distractor objects in the search array. Experiment 1 (behavioral study, n=31) found that participants were marginally faster at detecting items in patterned locations, compared to random locations (replicating Zhao et al., 2013). However, this effect was not present when distractors that were semantically related to the target appeared (i.e., no interaction between use of prior knowledge and novel information). Experiment 2 (ERP with behavioral measures, n=12) revealed that the behavioral results largely replicated those from Experiment 1. Interestingly, the N2pc ERP component (marker of target selection) was twice as large when targets were in the random location, compared to when the targets were in the pattern location, which is the opposite pattern of the behavioral results. Although this N2pc amplitude difference was not statistically significant (p=.287) with 12 participants, data collection is ongoing until the pre-registered number of 20 participants. These results suggest that novel sequences may impact behavior and neural responses to search tasks differently, and in some cases may not interact with prior knowledge.