Multitarget Visual Search Flexibly Switches Between Concurrent and Sequential Search Modes

Investigations into people's ability to use multiple working memory representations to concurrently search for targets have led to mixed findings. Although the discourse has predominantly centered around capacity limits in multitarget search, we here propose that people can switch between sequential and concurrent search. In Experiment 1 (n = 16 adults), manual responses and oculomotor behavior revealed that participants could search sequentially, and concurrently for at least two targets, when instructed. In Experiments 2a (n = 16 adults) and 2b (n = 16 adults), participants were free to choose how they searched. Trial-level modeling showed that participants primarily used sequential and concurrent search as specific modes and flexibly adjusted between either mode depending on template set size, template availability, stimulus properties, and individual preference. Our findings stress the dynamic and adaptive nature of visual search. Moreover, understanding that different search modes can be flexibly picked as "tools from the toolbox" may reconcile inconsistencies in prior findings.