In the first pages of this book, I promised to give an account of natural vision, not just
snapshot vision but vision that is ambient and ambulatory. Ambient vision is what you
get from looking around at the scenery. Ambulatory vision is what you get from walking
through the countryside.

The standard approach to vision begins with the eye fixed and exposed to a
momentary pattern of stimuli. It then goes on to consider vision with the head fixed
and the eye allowed to explore the pattern by scanning it, that is, by looking at parts
in succession. Each fixation is a glimpse of the pattern comparable to a momentary
exposure and is thus supposed to be analogous to a photographic snapshot taken by a
camera with a shutter. Each successive snapshot is assumed to be transmitted to the
brain. The result of all this is aperture vision, a sequence of snapshots.

The standard approach never gets around to ambient vision with head turning,
and it does not even consider ambulatory vision. The process of perception is supposed
to be localized in the head, not in the muscles, and it begins after the sensory input
reaches the visual projection area of the cerebral cortex. The mind is in the brain.

The ecological approach to visual perception works from the opposite end. It
begins with the flowing array of the observer who walks from one vista to another,
moves around an object of interest, and can approach it for scrutiny, thus extracting
the invariants that underlie the changing perspective structure and seeing the connec-
tions between hidden and unhidden surfaces. This approach next considers the fact of
ambient awareness and explains it by the invariance of the sliding samples of the 360°
array. Only then is the awareness of a single scene considered, the surfaces seen with
the head fixed and the array frozen. The classical puzzles that arise with this kind of
vision are resolved by recognizing that the invariants are weaker and the ambiguities
stronger when the point of observation is motionless. Finally, the kind of visual aware-
ness obtained with the eye fixed and the retina either briefly exposed or made to stay
fixed is considered for what it is, a peculiar result of trying to make the eye work as if
it were a camera at the end of a nerve cable. The visual system continues to operate
at this photographic level, but the constraints imposed on it are so severe that very

-303-