The benefit of removing information from working memory: Increasing available cognitive resources or reducing interference?

Removing information from working memory is thought to free up capacity and improve the retention of other information. However, whether this benefit arises from reducing interference from the to-be-forgotten information or from freeing up cognitive resources remains unclear. We examined this by comparing removal immediately following encoding an item (immediate removal), or delayed until after other items have been encoded (delayed removal). Interference theories predict that both types of removal should reduce interference and improve memory performance. In contrast, if removal frees up cognitive resources, the beneficial effect on memory should be greater the earlier it occurs, as the resources can then be allocated to subsequently encoded items. Experiment 1 showed that both immediate and delayed removal failed to reduce interference from the tobe-forgotten items but improved memory for item-location bindings of other items still maintained in working memory. In Experiment 2, removal only facilitated item-location bindings for items encoded afterward. These results suggest that removal frees up working memory capacity by increasing available resources rather than by reducing interference.