count, the physical environment actually consists of things like surfaces and
objects arranged in space rather than points of color, and this is why perception
is organized as it is. This is the naive realist’s answer, and there is undoubtedly
something to it. Surely evolutionary utility requires that perceptual organiza-
tion reflect structure in the organism’s environment, or at least the part of it
that is relevant to the organism’s survival. Imagine, for example, how much
less useful vision would be if it characteristically misorganized the world. But
although the naive realist’s answer might help explain perceptual organiza-
tion in an evolutionary sense—whyperceptual experience has the structure it
does—it does not explain the mechanisms of organization:howit unfolds in
time during acts of perception. The goal of this chapter is to shed light on these
mechanisms and the stimulus factors that engage them.
TheExperienceError The major difficulty with the view of naive realism is that
the visual system does not have direct access to facts about the environment; it
has access only to facts about the image projected onto the retina. That is, an
organism cannot be presumed to know how the environment is structured ex-
cept through sensory information. The Gestaltists referred to the naive realist’s
approach to the problem of perceptual organization as theexperience errorbe-
cause it arises from the false (and usually implicit) assumption that the struc-
ture of perceptual experience is somehow directly given in the array of light
that falls on the retinal mosaic (Ko ̈hler, 1947). This optic array actually contains
an infinite variety of possible organizations, however, only one of which the
visual system usually achieves.
The confusion that underlies the experience error is typically to suppose that
the starting point for vision is the distal stimulus rather than the proximal
stimulus. This is an easy trap to fall into, since the distal stimulus is an essential
component in the causal chain of events that normally produces visual experi-
ences. It also corresponds to the interpretation the visual system strives to
achieve. Taking the distal stimulus as the starting point for vision, however,
seriously underestimates the difficulty of visual perception because it presup-
poses that certain useful and important information comes ‘‘for free.’’ But the
structure of the environment is more accurately regarded as theresultof visual
perception rather than its starting point. As obvious and fundamental as this
point might seem, now that we are acquainted with the difficulties in trying
to make computers that can ‘‘see,’’ the magnitude of the problem of perceptual
organization was not fully understood until Wertheimer raised it in his seminal
paper in 1923. Indeed, although significant progress has been made in the
intervening years, vision scientists are still uncovering new layers of this im-
portant and pervasive problem.
8.1 Perceptual Grouping
Wertheimer’s initial assault on the problem of perceptual organization was
to study the stimulus factors that affectperceptual grouping:howthevarious
elements in a complex display are perceived as ‘‘going together’’ in one’s per-
ceptual experience. He approached this problem by constructing very simple
arrays of geometric elements and then varying the stimulus relations among
Organizing Objects and Scenes 191