Mechanisms of object-based correspondence

• Applicant: Dr. Elisabeth Hein
• Subject Area: General, Cognitive and Mathematical Psychology
• Term: since 2015
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 280139525

The visual system is able to create and maintain stable representations of the environment – e.g., maintaining the identity of elements even as they move and change through space and time – despite the fact that the information at the retina with regard to this structure is ambiguous. For a long time, the dominant view in the field was that the visual system establishes correspondence between elements, a fundamental component of this scene structure, based mainly on spatiotemporal proximity. This factor is thought to then determine the motion energy between the elements and thus establish correspondence at a relatively low level of visual processing. In contrast to this motion-based view of correspondence, during the first project phase we found in a series of behavioral experiments clear evidence that higher-level factors also play an important role in determining correspondence and therefore that correspondence can be solved at higher, object-based processing levels. We showed, for example, that visual attention influenced correspondence. Moreover, we demonstrated that perceived attributes of objects, attributes that are abstracted from the retinal features of objects and interpreted corresponding to the contextural cues in which they are embedded, influenced correspondence. Finally, we showed that correspondence is influenced by the spatiotemporal history of elements, such that identical stimuli with different histories could result in different perceived correspondences. While confirming the influence of higher-order processes in the establishment of correspondence, our studies also consistently showed that low-level motion-based information affects correspondence as well. The aim of this second project phase is therefore to gain new insights into how these higher-level object-based and lower-level motion-based correspondence mechanisms interact with each other, and what conditions determine which, or both, are activated. In particular, we plan to investigate whether depending on the reliability and complexity of the perceptual input one or the other correspondence mechanism might dominate the correspondence solution and whether different correspondence solutions can be learned. We also want to examine whether the object-based correspondence mechanism might be developed only later in life looking at how children solve correspondence under different conditions. Finally, we will investigate whether the distinction between object-based and motion-based correspondence is particular for the visual domain or whether it generalizes to other modalities. We expect that this series of experiments will offer new insights into the interplay of motion- and object-based correspondence mechanisms and thus will extend our current understanding of human object perception.

• DFG Programme: Research Grants
• International Connection: United Kingdom, USA
• Cooperation Partners: Tessa Dekker, Ph.D.; Professorin Dr. Cathleen M. Moore
• Co-Investigator: Professorin Dr. Bettina Rolke