Abstract
[Purpose] When we move, the movement changes the viewpoint. Despite such viewpoint changes, we perceive the objects and object layouts unchanged in the space. This indicates that our visual system has viewpoint independent representations of objects/layouts. We examined whether the spatial representations can be obtained implicitly and whether there is a viewpoint dependent representations that is independent on self-motion. [Experiment] We adopted contextual cuing effect (CCE) to investigate implicit learning of spatial layouts. The CCE is a learning effect of spatial layout in visual search displays and is known to be an implicit learning effect. Without noticing the repetitions, visual search performance increases by repeating a visual search task. We examined whether the CCE is obtained with viewpoint changes due to self-motion. Visual search displays were presented on a head mount display so that the viewpoint of the display and the head position were controlled independently. We compared the CCE between with and without self-motion in a visual search task where the stimulus viewpoint was changed and also where it was not. [Results] We found the CCE with self-motion when the display changed according to the self-motion with the viewpoint change of 40° while no CCE with that of 90°. When the display changed without self-motion, however, the CCE disappeared with 40° change. This indicates that there is an implicit learning effect in the spatial coordinates but with a limited range of viewpoint changes. The CCE was also found when retinal image was kept with viewpoint changes. This indicates that viewpoint specific presentations of retinal information, which is obtained implicitly, are maintained independently of viewpoint independent spatial representations. [Discussion] Our results suggest that object layouts are learned implicitly in two different processes: one works in the spatial coordinates and the other in the retinal coordinates.
Meeting abstract presented at VSS 2012