Abstract
There are two types of change detection, the experience of seeing dynamic change and the detection of completed change. Seeing dynamic change may require sustained attention to a changing area, while the detection of completed change may not (Hollingworth, 2008). This indicates that the former is accomplished by a location-based comparison while the latter involves both location-based and object-based comparisons. The seeing process is dominant when pre- and post-change stimuli are presented with a brief interval, while the detection process is dominant when they are presented with a long interval. This study used the flicker paradigm (each image was presented for 250 ms) in order to focus on the processing of the comparisons involved in seeing change. We used original images in which 8 objects were arranged in a circle and change images in which one of the objects was replaced by another object. We manipulated the interval between two images (200 ms and 1000 ms) and the shift of the image (shift and no-shift). In the shift condition, the objects in the change image were presented in a completely different location from the original image, which disturbs only the location-based comparison. The results showed that in the 200 ms interval condition, where seeing dynamic change is dominant, the number of alternations of the original and change image, which is proportional to the reaction time, before the change detection was larger in the shift condition than in the no-shift condition. In the 1000 ms interval condition, where the detection of completed change is dominant, there was no difference between the shift and no-shift conditions. This result indicates that sustained attention is involved in the experience of seeing dynamic change, but is not involved in the detection of completed change.