Abstract
Interruptions are a common occurrence in our daily life. Previous research has shown that interruptions negatively impact human performance (Froughi et al., 2014; Nicholas & Cohen, 2016). However, research on the impact of interruptions during visual search is sparse. In three experiments, we examined how interruptions affect performance on visual search tasks and whether interruption cost is modulated by task difficulty. Participants counted the number of Landolt C targets of a specific orientation in a display of 150 items. After search, participants reported the number of targets they identified. Throughout the experiments, search was interrupted on certain trials by a set of math problems. In Experiment One, working memory load was varied during search by asking participants to count targets in one or three separate colors. In Experiment Two, WM load was held constant and the interruption task was either easy math problems (single digit) or hard math problems (double digit). In both experiments, response times (RT) were significantly longer during interrupted trials. Contrary to previous research in other domains, there were no accuracy costs for interrupted trials in either experiment. The effect of interruption on RT was not modulated by the difficulty of the search task or the interruptions. In Experiment Three, we investigated whether these results might be due to a speed-accuracy tradeoff by placing a 30-minute limit on task completion time. Thus, participants were discouraged from spending extra time on interrupted trials in order to avoid making mistakes. However, Experiment Three showed the same pattern of results observed in prior experiments. Overall, these results suggest interruptions lead to a general time cost during visual search, which cannot be explained by a speed-accuracy tradeoff.