Abstract
There is a limit to how much information can be stored in visual working memory (VWM). Previous models of this limitation have focused on the effects of VWM load, finding that behavioral recall of memory items becomes worse as load increases. Similarly, ERP studies examining the neural markers of VWM have focused primarily on load effects, exploring a component that tracks memory load: the contralateral delay activity (CDA). However, there is evidence that load is not the only limiting factor of performance, as behavior is better predicted by attentional resource allocation across items than by load alone. Given that attention can be flexibly distributed across items, we tested whether the CDA reflects the allocation of attention in addition to memory load. In Experiment 1, we examined the CDA during low (one) and high (four) memory loads; critically, in one condition we also manipulated the allocation of attentional resources by presenting a cue indicating that one of four items would be probed on 50% of trials. When one item was prioritized, the CDA was intermediate between the low and high load conditions, suggesting that the CDA was modulated by resource allocation rather than load alone. In Experiment 2, we examined a wider range of cueing probabilities (i.e. 100%, 75%, 25%, or 0% probe likelihood) while holding load constant. Stimuli were also presented along the vertical and horizontal midline to examine ERP components reflecting attentional enhancement (N2pc) and suppression (PD) during encoding. The results revealed that the amplitudes of the N2pc and CDA (but not the PD) increased proportionally with priority, and that these amplitudes predicted individuals’ behavioral precision. Our findings suggest that ERP markers of VWM encoding and maintenance reflect attentional prioritization in addition to load.
Acknowledgement: Natural Sciences and Engineering Research Council of Canada (NSERC) Doctoral Post-Graduate Scholarship, NSERC Discovery Grant [#435945], and the NSERC Research Tools and Instruments Grant [#458707]