Mechanisms Of Feature-location Binding In Visual Working Memory

Properties of visual objects can be divided into featural properties and spatiotemporal properties. Correspondingly, visual working memory can be divided into two subsystems, which separately encode and maintain featural information and spatiotemporal information. How to integrate the two separate streams of information into unified object representation becomes a great challenge to visual working memory. The present study investigated the mechanism of binding between featural information and the basic spatiotemporal information-location information. The study mainly adopts the research paradigm of visual working memory basing on moving objects.The present study consists of three parts with seven experiments in total. In the first part, the author investigated the influence of motion on feature-location binding (Experiment 1), the memory of feature information under the dynamic condition (Experiment 2), and the distinction of multiple objects under the dynamic condition (Experiment 3). On the basis of these experiments, the author discussed whether there is a consolidation process in feature-location binding. In the second part, the author manipulated the attention level by adding demanding verbal rehearsal task (Experiment 4) or backward counting task (Experiment 5) as a secondary task to the binding task, so as to investigate the role of attention in the feature-location binding in visual working memory. In the third part, the author investigated the influence of feature type on feature-location binding, through comparing the memory performance of binding between location and different types of feature under the dynamic and static conditions (Experiment 6 & Experiment 7).The main findings of this study are as follows:(1) Motion does not influence the encoding and maintenance of feature, but weakens the binding between feature and location, indicating that motion impairs the consolidation process in feature-location binding.(2) The feature-location bindings under the dynamic and static conditions are equally impaired by the attention-demanding task, indicating that attention plays the same role in the binding under the two states. (3) The type of features influences feature-location binding. The binding between detailed feature and location is more severely impaired by motion, compared with the binding between basic feature and location, suggesting that the binding between detailed feature and location depends more heavily on consolidation process.On the basis of the findings of the present study and previous research, the author proposed a reentry model of feature-location, which explains the binding between location and different types of feature under both dynamic and static states in a unified theoretical framework.