Abstract
Daily objects are generally composed of many sub-objects or components. We psychophysically examined the relationship between sensitivity to the main object and the sub-objects that comprise the main object. The main object was the frame of a cube made of 12 small cylinders, and the sub-objects were 4 cylinders placed in parallel (parallel object) and 4 cylinders placed as a square (square object). Those objects are determined in the virtual reality (VR) space and had a 3-D impression. The participants observed those objects as a 2-D image on a flat display or stereoscopic VR image through a head-mounted display. The overall orientation of those objects varied from 0 to 90 degrees. Each object was embedded in many randomly placed single cylinders in the VR space. To estimate detection sensitivity toward each object, the participants were asked to detect the objects from the cylinders as quickly and precisely as possible. In the VR presentation, the participants can freely move the head and body, while in the flat display condition, a head rest was used so that the head did not move. Detection sensitivity was the highest for the main cube and the parallel objects, but only when they were vertically placed. When those objects were oriented at 45 degrees or horizontal, or for the square object of any angle, detection sensitivity was lower. Therefore, the vertical structure in the main object was found to be the key for detection. Those characteristics were prominent when the objects were presented in the 3-D VR space than in the flat display. These results indicate that the detection sensitivity of 3-D objects would depend on the vertical component of the object in the allocentric reference of frame, especially when the viewer is moving the head and body.