Abstract
According to the coarse-to-fine framework of visual search (e.g., Over et al., 2007), eye movements are initially rapid and distributed for scanning the coarse information of a scene, but as search progresses, they become slower and local for processing detailed information. Previous studies, however, examined the transition towards fine-scale processing when the target was present, leaving the transition in target-absent trials unclarified. The current study investigated how visual processing changes when people terminate target-absent visual search. We hypothesized that towards the end of search, target-absent trials should involve coarse processing as many distractors are simultaneously rejected compared to target-present trials where target selection would involve fine processing. In Experiment 1, while participants were searching for a uniquely oriented target among heterogeneous distractors, their eye movements were tracked. Here, distractor heterogeneity was manipulated into two conditions to vary ease of distractor rejection. Results showed that at the search onset, there was an increase in saccadic amplitudes along with short fixation durations which indicate coarse processing. The increasing trend of saccadic amplitudes was stronger when the distractors were less heterogeneous. Towards search termination, regardless of the extent of distractor heterogeneity, target-absent trials showed larger, increasing pattern of saccadic amplitudes with shorter fixation durations, compared to target-present trials. This suggests that the coarse processing remained until the end when no target was found. This termination effect was further investigated with reaction time in Experiment 2. We found that search duration of the current trial was shorter when preceded by a correct target-absent trial than by a correct target-present trial. This implies that target-absent trials benefit subsequent searches because of the maintained coarse processing mode which can be used when both terminating target-absent trials and starting new trials. Altogether, the current study indicates that rejecting multiple distractors brings coarse-scale processing when terminating target-absent trials.
Acknowledgement: This work was supported by National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (NRF-2016R1A2B4016171).