The Influence of Statistical Regularities on the Spatial Congruency Bias

Thumbnail Image



Journal Title

Journal ISSN

Volume Title


The Ohio State University

Research Projects

Organizational Units

Journal Issue


Introduction/Background: During object recognition, visual features such as shape and location must be integrated. Previous studies investigating object feature binding posit that location plays a unique role in the binding process. Further, a "spatial congruency bias," or a tendency to report objects as the same identity if they are presented at the same location, was discovered (Golomb et al, 2014, JEP:Gen), supporting a privileged role for location in object recognition. The current study investigated the persistence of this bias. Methods: While maintaining central fixation on a screen, subjects were sequentially presented with two objects in the periphery. The objects were novel-object morphs of either the same or different identity, and they appeared at either the same or different location. Subjects were then asked whether the objects were of the same or different identity. To see if we could override the spatial congruency bias, we biased the statistical regularities in the opposite direction, such that 75% of same location trials included different identity objects, and 75% of different location trials included same identity objects. If the spatial congruency bias were still observed after this manipulation, it would indicate that the role of location in object perception is preserved despite contradictory associations. Results and Conclusions: A calculation of d' and response bias revealed that subjects were indeed still more biased to report objects as the same identity when they appeared in the same location compared to different location trials. These results indicate the preservation of the spatial congruency bias even with conflicting statistical regularities, further reinforcing the dominant role of spatial information during object feature binding.



Object recognition, Statistical learning, Vision