Relationship between working memory capacity and contingent involuntary orienting

Abstract

Individual differences in working memory capacity are related to the ability to control attention; where less working memory capacity is associated with less attentional control. A well-known demonstration of attentional control is contingent involuntary orienting (Folk, Remington, & Johnston, 1992), which is the finding that visuospatial attention is captured by a salient peripheral cue, but only if the cue is featurally similar to the target of search and not if the cue is featurally dissimilar. This control of attention and ability to resist shifting attention towards an irrelevant visual cue is believed to be moderated by attentional settings for target-specific features. This study establishes that working memory capacity is related to contingent involuntary orienting of attention. Lower working memory capacity was associated with larger stimulus-driven cueing effects at short cue-to-target onset asynchronies, than at longer asynchronies, suggesting working memory capacity was associated with the ability to control covert orienting of attention at early processing stages.

Keywords:

• Attentional capture
• Attentional control
• Executive attention
• Working memory capacity

Acknowledgements

Data related to this work were presented at the 49th and 50th annual meetings of the Psychonomic Society. We thank Antony Delliturri for assistance with data collection.

Notes

¹This procedure to create high-span and low-span groups differs from that used in other studies (e.g., Conway et al., 2001; Kane et al., 2001). In those studies, a large pool of subjects is measured on the OSPAN task. The 25th and 75th percentiles are determined and random samples are selected from among subjects scoring above the 75th percentile and below the 25th percentile. Such a procedure ensures equal distribution of individual differences from high-span and low-span populations into the respective samples. Because subjects in this study were randomly selected, individual differences in the population should already be equally distributed within the sample. Hence, there should be little difference between the high-span and low-span groups.