Abstract
When we look at an object, we assign it a place in our spatial surroundings. Simultaneously, we process details of the object such as color, shape, curves, and edges. However, what if the object moves from one place to another? Do we have to re-bind all of these features to the object, or are we able to keep the traits intact? Recent research has shown that an object’s identity is bound to its location, resulting in a “spatial congruency bias” (Golomb et al, 2014, JEP:Gen). When two objects appear consecutively, we are more likely to deem them as identical if they appear in the same location, versus if the second object appeared in a different location. In the current experiment, subjects saw the first object for 500ms, then watched it move to a new location on-screen. A second object then immediately appeared either at the location where movement ended, the original starting location, or a control location. Subjects were asked to indicate whether the second object was the same (identical) or different (slightly altered) from the first. The “congruency bias” was found to remain at the starting location – subjects were more likely to deem the two objects as the same identity if the second object was presented at the original location of the first object. This suggests that features are bound to the original object location and may not automatically update with movement. However, in a follow up experiment, when the first object remained visible for 250ms at the end location after movement (allowing time to re-encode the object at the new location), the congruency bias was found at both the original starting location and the end location of the movement. These data suggest that features bound to a location may need time to re-bind following object movement.
Meeting abstract presented at VSS 2015