ABSTRACT
Searching for items in one’s environment often includes considerable reliance on semantic knowledge. The present study examines the importance of semantic information in visual and memory search, especially with respect to whether the items reside in long-term or working memory. In Experiment 1, participants engaged in hybrid visual memory search for items that were either highly familiar or novel. Importantly, the relatively large number of targets in this hybrid search task necessitated that targets be stored in some form of long-term memory. We found that search for familiar objects was more efficient than search for novel objects. In Experiment 2, we investigated search for familiar versus novel objects when the number of targets was low enough to be stored in working memory. We also manipulated how often participants in Experiment 2 were required to update their target (every trial vs. every block) in order to control for target templates that were stored in long-term memory as a result of repeated exposure over trials. We found no differences in search efficiency for familiar versus novel objects when templates were stored in working memory. Our results suggest that while semantic information may provide additional individuating features that are useful for object recognition in hybrid search, this information could be irrelevant or even distracting when searching for targets stored in working memory.
Acknowledgements
We thank Amber Harrington and Janeshia Adams for their assistance in data collection.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 While there has been considerable debate regarding the distinction between WM and LTM, there is neuroscientific evidence supporting the view that WM and LTM are not separate systems. A review by Ruchkin, Grafman, Cameron, and Berndt (Citation2003) advocates that there is no need to posit the existence of separate short-term storage systems, as event related potential (ERP) evidence suggests that long-term memory systems are capable of supporting the representational basis for working memory.
2 The participant repeatedly took out her mobile phone, despite instructions to attend fully to the experiment.