Boundary extension

Already interesting in itself, Unreal Nature's "Hypocritic art" post today also had me thinking back to one here in July. Ray Girvan's comment on the psychological factors influencing interpretation of a photograph included awareness of a subconsciously hypothesised 360° "boundary extension". The elements within the photograph produced an imagined continuity of logically induced landscape elements behind him.

Fascinating, in both cases.

Returning to Unreal Nature's post, I can't help speculating that the visual boundary extension effect arises from the different ways in which vision and depiction work. Vision has a central area of attention focus surrounded by a ellipse of peripheral vision to give context. Depiction does not – it concentrates on simulating the central focus of attention, then cuts off at the frame. It seems likely that our memory of depicted scenes includes an subconscious spurious addition of missing peripheral vision.

Later (2008-11-01-1249): Ray Girvan comments...

The experiment mentioned in the Science Daily piece is interesting in a number of ways. The procedure described - showing two similar views briefly interrupted by a mask of randomish visuals - is more or less a textbook scenario for inducing "change blindness", where you can't perceive what has altered (see the links at "Best illusions"). I wonder whether that was intentional; if not, it's introducing a major extraneous factor into the task of comparing the images.

And later still (2008-11-01-1932): Ray again...

Ah - I just found the paper. ... They don't mention change blindness explicitly, but it is in the references, so they're clearly aware of it.

Intraub, C.A.D.H., "Transsaccadic Representation of Layout: What Is the Time Course of Boundary Extension?" in Journal of Experimental Psychology: Human Perception and Performance, 2008. 34(3): p. 543–555.