The Influence Of Attention And Similarity On Feature Binding Maintained In Visual Working Memory

It has been found that attention has a very important influence in visual working memory(VWM). Not only encoding and retrieval in VWM requires attention, it seems that objects representation which maintained in VWM also require attention. Memory for Feature and memory for feature binding are disrupted by increased attentional demands. But does memory for feature binding require more attention than memory for Feature? The role of attention in VWM remains controversial; while some evidence has suggested that maintaining multifeature visual objects requires no more attention than memorizing the same number of unifeature visual objects, other evidence has indicated cognitive costs or mnemonic benefits for explicitly attending to feature binding. There are two hypotheses about this issue: attention hypothesis and volatile representation hypothesis. The attention hypothesis( Treisman, 1999) suggests that attention is needed to maintain feature binding in VWM, while the volatile representation hypothesis(Allen, Baddeley, &Hitch, 2006) suggest that feature binding stored in episodic buffer is volatile and easily overwritten, but do not require sustained attention.Adopting change-detect paradigm, we conducted three experiments to explore whether attention is required during the maintenance phase of a VWM task. We used color-shaped objects as memory items. Experiment 1 studied the interfering effects of a to-be-ignored “stimulus suffix” on feature binding. We manipulated the similarity of the object's color to build object representations of different stability. We found that the binding of similar objects was less disrupted by subsequent stimuli. In experiment 2, we inserted an object-based feature report task to consume the object-based attention available for binding during the maintenance phase. The results revealed a more significant impairment for binding than for constituent features for dissimilarobjects; while there was no differential effect of object-based attention for similar objects. In experiment 3, we used ERP technology to further examine the influence of object-based attention and similarity on feature binding. The results showed that for dissimilar objects, the amplitude of NSW(negative slow wave) evoked by feature binding was higher than that evoked by feature. In general, the conclusions are as follows.1. Similarity of objects can help individuals build more stable representation of feature binding in VWM. They need less cognitive resources during the maintaining phase.2. Object-based attention and the stability of representation hypothesis can both work on the maintaining phase, when the representation of feature binding is stable, the representation is less impaired by object-based attention.