The role of action in the development of visual object recognition

点击翻译按钮获取中文摘要

Processes of visual object recognition play a central role in multiple aspects of human cognition. However, very little is known about the development of these processes beyond the earliest stages of infancy. A long literature suggests a deep and fundamental coordination between vision and action in adults. One way in which action should matter critically to visual development is that one's own actions on objects generates the dynamic visual experiences that one receives about objects. Self- generated actions literally create one's views of objects. In the proposed research, we seek to understand the role of developmental changes in self-generated action in the development of visual object representations and recognition. The participants will be children between 18 and 24 months of age  an age range during which prior work indicates that significant developmental change in object recognition occurs. This age range is also a period of rapid language expansion in typically developing children, and a period in which atypical patterns of language, motor, and more diffuse cognitive development are often first identified. Thus, the study of self-generated dynamic views of object structure will provide a new window into mechanisms and early diagnosis of atypical cognitive development. We focus on 3 aspects of visual object representations, suggested to be important in studies of high level vision in adults, which our preliminary studies indicate have roots in the 18 to 24 month period: 1) Preferred views  the causes and consequences of the emergence of a preference for viewing the planes of objects (sides, front, back, top, and bottom) rather than views of areas between planar surfaces; 2) Axes of elongation  the causes of developmental changes in how children hold and move objects relative to the objects' axes of elongation, and the consequences for children's perception and representation of those objects; and 3) Global structure  the ability to recognize objects from sparse characterizations of their geometric structure. We will use both cross-sectional, and training studies to investigate the roles of action in each of these aspects of typically developing children's visual object recognition.