Amodal completion

Imagine a cup partly hidden behind a book. You can recognize one continuing cup without seeing the surface behind the book. How does the brain combine the visible edges with what it knows about cups to represent the whole object? This kind of completion does not require a vivid image of the hidden part.[1] [2]

A cup partly hidden behind a book can be represented as a continuing whole even though its hidden surface is not visually experienced. How and when do visible structure and object knowledge combine to produce this amodal completion? A 2024 imaging study compared responses to completions compatible or incompatible with those cues. Both affected a higher visual region involved in object processing, while the tested early visual region showed no corresponding differences. The result identifies influences on measured responses; it does not reveal the complete sequence of causal interactions.[1] [2]

Relevant brain responses are known, but their causal interaction remains incomplete.[1] [2]

Imaging constrains where different influences appear, but does not establish their full causal interaction or one uniquely accepted processing sequence.[1] [2]

Objects overlap in ordinary scenes. Treating the visible pieces as parts of one object, rather than unrelated fragments, connects an incomplete view to a useful account of what is there.[2]

Knowing that an object continues behind another is different from generating a detailed hidden picture. This distinction connects recognition with what information a brain response actually represents: similar responses to a whole and a partly hidden object need not imply that every unseen feature has been reconstructed.[2]