Abstract
Tasks that require visual searches for multiple targets are known to present unique problems from those involving just one target. The Satisfaction of Search (SOS; also called Subsequent Search Misses) paradigm describes one such example of this, whereby searchers are less likely to locate a second target after the first has been identified. While the original hypothesis involved observers prematurely terminating their search once they had been "satisfied" by the detection of one target, recent work has led to a myriad of cognitive and perceptual limitations involved in this unique pattern of flawed visual search. One particularly compelling theory describes a depletion of cognitive resources that prevents searchers from finding a second target. The current analysis uses multilevel modeling to expand this theory of resource depletion in SOS. This analysis began by calculating an intraclass correlation coefficient to determine how much variance in these dual-target errors existed within-participants across timepoints of the experiment (i.e., level 1), and between-participants (i.e., level 2). This unconditional model revealed that the overwhelming majority of variance existed at level 1. The next model used time as a level 1 predictor to demonstrate a significant increase in SOS-errors as the experiment progressed. Furthermore, non-significant random effects (i.e., 𝝉11) for the slope in this model indicated consistent variance across participants in this performance decrement. Previous descriptions of resource depletion in this paradigm have revolved around more acute examples of cognitive resources (e.g., working memory), that are studied in a single trial. However, this model suggests limitations of a chronic cognitive mechanism that depletes over many trials.
Meeting abstract presented at VSS 2018