An Outline of a Theory of 
Perception
Knowledge on the part of the 
observer that this display can 
be perceived as a transparent 
rectangle in front of a gray and 
black background may play a 
role in how it is perceived.

What are the general 
characteristics of perception 
with which any theory must 
deal? An adequate theory must 
be able to account for the 
achievement of constancy and 
veridicality, for preferred 
perceptions given an ambiguous 
stimulus, for the effect of 
context and frame of reference, 
for enrichment and completion 
effects and for the role of 
attention. Each of the 
traditional theories deals with 
only some of these features. For 
example, stimulus theory has 
difficulty with the fact that the 
stimulus is often ambiguous 
and that more than one percept 
can arise from it. The necessity 
of perceptual organization also 
raises grave difficulties for this 
approach, as does perceptual 
enrichment. Gestalt theory has 
difficulty with those cases of 
constancy that seem to be based 
on an inferential process (the 
taking-account explanation), 
and it had little to say about 
attention.

    Moreover, both of these 
theories have difficulty with a 
very important fact about 
perception. Often one 
perception seems to depend 
upon another. Earlier I 
suggested that the 
stereokinetic depth effect 
occurs if, and only if, the 
circles are no longer perceived 
to rotate—–that is, they must 
first be seen to maintain their 
orientation before they appear 
to slide around with respect to 
one another. If at first the 
circles are seen to rotate, the 
depth effect will not occur. So 
it is not simply the stimulus 
transformation that leads to 
the depth effect, it is a certain 
prior perception that does so. 
There are many examples of 
this sequence of events, but 
this one will suffice. We see, 
then, that perception is more 
than a direct consequence of a 
stimulus and more than a 
consequence of the 
organization of stimulus 
components.