Abstract
The concept of objects is fundamental to cognition and is defined by a consistent set of sensory properties and physical affordances. Although it is unknown how the abstract concept of an object emerges, most accounts assume that visual or haptic boundaries are crucial in this process. Here, we tested an alternative hypothesis that boundaries are not essential but simply reflect a more fundamental principle: consistent visual or haptic statistical properties. Using a novel visuo-haptic statistical learning paradigm, we familiarised participants with objects defined solely by across-scene statistics provided either visually or through physical interactions. We then tested them on both a visual familiarity and a haptic pulling task, thus measuring both within-modality learning and across-modality generalisation. Participants showed strong within-modality learning and 'zero-shot' across-modality generalisation which were highly correlated. Our results demonstrate that humans can segment scenes into objects, without any explicit boundary cues, using purely statistical information.
Highlights
The coherent organization of information across different modalities is crucial for efficiently interacting with the world and lies at the heart of the concept of what defines an object (Amedi et al, 2001; Pascual-Leone and Hamilton, 2001; Streri and Spelke, 1988)
Boundary cues were uninformative with regard to the object identities, and instead participants could only rely on the statistical contingencies among the shapes that we created in either the visual or the haptic modality during an exposure phase
We examined how the information extracted from the visual or haptic statistics affected performance on both a visual familiarity and a haptic pulling test, measuring within-modality learning as well as across-modality generalisation of statistical information
Summary
The coherent organization of information across different modalities is crucial for efficiently interacting with the world and lies at the heart of the concept of what defines an object (Amedi et al, 2001; Pascual-Leone and Hamilton, 2001; Streri and Spelke, 1988). Classic theories place visual boundaries, edges and contrast transitions, at the very core of the process by which we segment objects from the environment (Marr, 1982; Peterson, 1994; Riesenhuber and Poggio, 1999; Spelke, 1990; von der Heydt et al, 1984; Zhou et al, 2000). Such boundary cues are insufficient for successful segmentation alone as they can both lead to false object boundaries within objects (e.g. the stripes of a zebra) and miss boundaries between objects (e.g. the illusory contours of the Kanizsa triangle).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
