CHAPTER 8
Depth Perception and the Perception of Events
DENNIS R. PROFFITT AND CORRADO CAUDEK
213DEPTH PERCEPTION 214
Depth Cues: What Is the Information? What Information Is
Actually Used? 214
Integration of Depth Cues: How Is the Effective
Information Combined? 219
Distance Perception 220
Size Perception 222
Geographical Slant Perception 223
EVENT PERCEPTION 224
Perceptual Organization 224
Perceiving Dynamics 226
Perceiving Our Own Motion 227
Visual Control of Posture 228
Perceiving Approaching Objects 228
CONCLUSION 229
REFERENCES 231Our understanding of the perception of depth and events en-
tails a paradox. On the one hand, it seems that there is simply
not enough information to make the achievement possible,
yet on the other hand the amount of information seems to be
overly abundant. This paradox is a consequence of evaluating
information in isolation versus evaluating it in context.
Berkeley (1709) noted that a point in the environment pro-
jects as a point on the retina in a manner that does not vary in
any way with distance. For a point viewed in isolation, this is
true. From this fact, Berkeley concluded that the visual per-
ception of distance, from sight alone, was impossible. Visual
information, he concluded, must be augmented by nonvisual
information. For example, fixating a point with two eyes re-
quires that the eyes converge in a manner that does vary with
distance; thus, proprioceptive information about eye posi-
tions could augment vision to yield an awareness of depth. If
our visual world consisted of a single point viewed in a void,
then depth perception from vision alone would, indeed, be
tough. Fortunately, this is not the natural condition for visual
experience.
As the visual environment increases in complexity, the
amount of information specifying its layout increases. By
adding a second point to the visual scene, additional informa-
tion is created. If the two points are at different depths, then
they will project to different relative locations in the two
eyes, thereby providing information about their relative dis-
tances to each other. If the observer fixates on one point and
moves his or her head sideways, then motion parallax will be
created that gives information about relative depth. Expand-
ing the points into extended forms, placing these forms on
a ground plane, having the forms move, or allowing the ob-
server to move are some of the possible complications that
create information about the depth relationships in the scene.
Complex natural environments provide lots of different
kinds of information, and the perceptual system must com-
bine all of it into the singular set of relationships that is our
experience of the visual world. It is not enough to register
the available information; the information must also be
combined.
From this brief introduction, two fundamental questions
emerge: What is the information provided for perceiving
spatial relationships and how is this information combined
by the perceptual system? We begin our chapter by review-
ing the literature that addresses these questions. (Additional
topics in visual perception are discussed in the chapter by
Kubovy, Gepshtein, and Epstein in this volume.)
There is a third question that we also address: Do peo-
ple perceive spatial layout accurately? The answer to
this question depends upon the criteria used to define accu-
racy. Certainly, people act in the environment as if they rep-
resent its spatial relationships accurately; however, effective
action can often be achieved without this accurate represen-
tation. This issue is developed and discussed throughout
the chapter.