• Two stages or three stages: The time course of feature encoding and binding of dual-feature objects in visual working memory?

    Subjects: Psychology >> Cognitive Psychology submitted time 2024-04-03

    Abstract: Feature binding of visual working memory(VWM) has been a hot topic in cognitive psychology and neuroscience in recent decades. Regarding the time course of objects’ feature encoding and binding in VWM, two controversial viewpoints have been proposed: the two-stage feature binding theory and the three-stage feature processing theory. After careful analysis, it is found that these two different conclusions are mainly based on different experimental materials and the types of features contained in them. Therefore, the feature encoding and binding time of dual-feature objects in VWM need to be further verified. In this study, meaningless objects and meaningful objects composed of two different types of features are selected to explore the specific encoding time course of their features in VWM, and compare the differences between the feature encoding and binding time course of them. Experiment 1 used a mixed design of 3 (memorizing item presentation time: 100ms, 500ms, 1000ms, internal variables of subjects) ×2 (object types: meaningful object, meaningless object, inter-subject variables) ×3 (detection methods: shape change, color change, feature exchange, internal variables of subjects). Its material is a color-shape pattern composed of two basic features. 52 college students were randomly divided into two groups to complete the task of detecting the feature changes of meaningless objects and meaningful objects respectively. The experimental procedure was as follows: First, the subjects were asked to remember the features and binding forms of 4 meaningful or meaningless objects presented on the screen. After 900ms of empty screen, a detection item appeared, and the subjects were asked to press the button as soon as possible to judge whether this detection item had appeared in the memory item on the premise of ensuring correctness. The design and procedure of experiment 2 is the same as that of experiment 1, except that the memory material is changed to a shape-texture pattern composed of one basic feature and one detailed feature. In experiment 3, based on experiment 2, the number of memorized items was reduced to 3, and the third level of presentation time was changed to 1500ms. The number of subjects in experiment 2 and 3 was 52 and 42 respectively. The dependent variable of all experiments was the correct rate of change detection. The results of experiment 1 show that the correct rate of color change is higher than shape change and feature exchange, but there is no difference between shape change and feature exchange. The results of experiment 2 and experiment 3 show that the detection results of shape change are higher than those of texture change and feature exchange, and there is no difference between texture change and feature exchange change. The accuracy of meaningful object was higher than that of meaningless object (1 and 3 of the experiment). With the increase of the time of memorizing items, the results of color-shape object change detection did not change, but the results of shape-texture object decreased significantly. In summary, there is a significant time history effect in encoding two-dimensional dual-feature objects in VWM. In particular, whether the object is composed of two basic features or one basic feature and one detail feature, its process conforms to the two-stage theory of feature binding. From the perspective of time history comparison between objects, the process of feature coding conforms to the three-stage theory of feature processing. These results to a large extent resolve the debate on whether feature encoding and binding in visual working memory is two-stage or three-stage. Object meaningfulness has an effect on feature coding, and the processing of meaningful objects is faster than that of meaningless objects under normal load conditions. In general, objects composed of basic features are processed faster than objects composed of basic features and detailed features.