Behavioral and neuroimaging investigations of perception and action with real-world objects

点击翻译按钮获取中文摘要

The visual system enables us to recognize objects and to act upon them in the real world. Surprisingly, however, almost all behavioral and neuroimaging experiments of human vision have employed pictures of objects rather than real, physical objects. There are important theoretical reasons to expect that the visual system processes images and objects differently, not least of which is that real objects afford actions. After all, one would never attempt to pound a nail with a photo of a hammer! Indeed, our recent NSERC-funded experiments have found differences in the behavioral and neural responses to real objects vs. photos. The long-term goal of my research is to understand how real objects and real actions are processed in the human brain and mind. The short-term goals over the next five years are (1) to further determine how real objects differ from images; and (2) to determine why the processing of real objects and actions differs. The project will investigate three themes and provide training for young investigators, particularly those participating in an international training program (NSERC CREATE-IRTG) between three Canadian universities and two German universities. The first line of research will investigate why real objects are processed differently from images (especially photos). The two most likely explanations are (1) because real objects have depth; and (2) because real objects can be acted upon and those actions have consequences. The second line of research will investigate what new things we can learn about how we perceive objects by using real objects. Real objects have sensible relationships between size and distance (unlike photos, where a car can be the same size as a cat), look different from different viewpoints (unlike photos, which show a single viewpoint), and can change (unlike photos, in which items do not move). Due to these differences, we predict that using real objects will help us better understand the role of size, viewpoint, and motion in object recognition. The third line of research will investigate what new things we can learn about how we perform hand actions toward objects under more realistic situations. We will investigate how grasping differs for real, virtual and photo objects and whether it matters if you see the object directly (without mirrors) and see your own hand move. In addition, we will investigate how actions are used to intercept moving objects and whether this skill is modulated with experience, for example in jugglers. Our work will use state-of-the-art behavioral and neuroimaging techniques to understand how real objects affect visual perception and action and the brain areas involved. This research direction has the potential to bring research in brain and behavior closer to everyday life and make it more relevant for real-world applications (such as neural prosthetics) and the emerging possibilities of virtual reality in science and technology.