Culture Experience and Perception

123
Recognizing the Perceptual World
indicating a place to eat. Similarly, perhaps you remember a time when an obviously
incompetent referee incorrectly called a penalty on your favorite sports team. You knew
the call was wrong because you clearly saw the other team’s player at fault. But suppose you had been cheering for that other team. The chances are good that you would
have seen the referee’s call as the right one.
Top-Down and Bottom-Up Processing Together
doing
2
learn
by
WHAT DOES IT LOOK LIKE
TO YOU? Some people see
the Devil’s face in this photo
of smoke from the World
Trade Center under attack on September
11, 2001. This perception results partly
from top-down processes, such as knowledge about the evil of the attack, that
make it easier to see something demonic.
People who don’t expect to see a face
in the smoke—or whose cultural background doesn’t include the “devil”
concept—may not see one until that
interpretation is suggested. To check that
possibility, show this photo to people
from various religions and cultures (cover
the caption). Ask if they see anything in
the smoke, but don’t immediately tell
them what to look for. How many, and
which ones, identified a demonic face
before you mentioned it?
FIGURE
3.27
Interaction of Top-Down and
Bottom-Up Processing
Which obscured line do you
find easier to read: the one on
the top or the one on the bottom? Top-down processing should help
you read the obscured text on the top line.
However, in the bottom line, the words are
not related, so top-down processing cannot operate.
doing
2
learn
by
Bottom-up and top-down processing usually work together to help us recognize the
perceptual world. This interaction is beautifully illustrated by the process of reading.
When the quality of the raw stimulus on the page is poor, as in Figure 3.27, top-down
processes compensate to make continued reading possible. They allow you to fill in
where words are not well perceived and processed, thus giving you a general idea of
the meaning of the text.
You can fill in the gaps because the world is redundant; it provides multiple clues
about what is going on. So even if you lose or miss one stimulus in a pattern, other
clues can help you recognize the pattern. There is so much redundancy in written language, for instance, that many of the words and letters you see are not needed. Fo- exmp-e, y-u c-n r-ad -hi- se-te-ce -it- ev-ry -hi-d l-tt-r m-ss-ng. Similarly, vision in three
dimensions normally provides multiple cues to depth, making recognition of distance
easy and clear. It is when many of these cues are eliminated that ambiguous stimuli
create the sorts of depth illusions discussed earlier.
In hearing, too, top-down processing can compensate for ambiguous stimuli. In one
experiment, participants heard strings of five words in meaningless order, such as “wet
brought who socks some.” There was so much background noise, however, that only
about 75 percent of the words could be recognized (Miller, Heise, & Lichten, 1951).
The words were then read to a second group of participants in a meaningful order (e.g.,
“who brought some wet socks”). The second group was able to recognize almost all of
the words, even under the same noisy conditions. In fact, it took twice as much noise
to reduce their performance to the level of the first group. Why? When the words were
in meaningless order, only bottom-up processing was available. Recognizing one word
was no help in identifying the next. Meaningful sentences, however, provided a more
familiar context and allowed for some top-down processing. Hearing one word helped
the listener make a reasonable guess (based on knowledge and experience) about the
others. (See “In Review: Mechanisms of Pattern Recognition” for a summary of bottomup and top-down processing.)
Culture, Experience, and Perception
We have been talking as if all aspects of perception work or fail in the same way for everyone, everywhere. The truth is, though, that virtually all perceptual abilities are shaped to
some extent by the sensory experiences we have or have not had (Kitayama et al., 2003).
For example, people are better at judging the size and distance of familiar objects than
of unfamiliar ones. Size and shape constancy, too, depend partly on the knowledge and
experience that tell us that most solid objects do not suddenly change their size or shape.
The experience-based nature of perception can also be seen in brightness constancy: You
perceive charcoal to be darker than a sheet of writing paper partly because, no matter
how much light the charcoal reflects, you know charcoal is black. Experience even teaches
us when to ignore certain stimulus cues (Yang & Kubovy, 1999). To fully experience the
depth portrayed in a painting, for example, you have to ignore ridges, scratches, dust, or
other texture cues from the canvas that would remind you of its flatness. And the next
in review
124
Chapter 3 Sensation and Perception
MECHANISMS OF PAT TERN RECOGNITION
Mechanism
Description
Example
Bottom-up processing
Raw sensations from the eye or the ear are
analyzed into basic features, such as edges,
color, or movement; these features are then
recombined at higher brain centers, where they
are compared with stored information about
objects or sounds.
You recognize a dog as a dog because its
features—four legs, barking,
panting—match your perceptual category
for “dog.”
Top-down processing
Knowledge of the world and experience in
perceiving allow people to make inferences
about the identity of stimuli, even when the
quality of raw sensory information is low.
On a dark night, a small, vaguely seen blob
pulling on the end of a leash is recognized as
a dog because the stimulus occurs at a
location where we would expect a dog to be.
?
1. Your ability to read a battered old sign that has some letters missing is a result of
processing.
2. When stimulus features match the stimuli we are looking for,
takes place.
3. Schemas can create a
that makes us more likely to perceive stimuli in a particular way.
Source: Hudson (1960).
FIGURE
3.28
Culture and Depth Cues
People in various cultures were shown
drawings like these and asked to judge
which animal is closer to the hunter. Those
in cultures that provide lots of experience
with pictured depth cues choose the antelope, which is at the same distance from
the viewer as the hunter. Those in cultures
less familiar with such cues may choose
the elephant, which, though closer on the
page, is more distant when depth cues are
considered.
time you are watching TV, notice the reflections of objects in the room that appear on
the screen. You have learned to ignore these reflections, so it will take a little effort to perceive them and a lot of effort to focus on them for long.
What if you hadn’t had a chance to learn or practice these perceptual skills? One
way to explore this question is through case studies of people who had been blind for
decades and then had surgery that restored their sight. It turns out that these people
can immediately recognize simple objects and perceive movement, but they usually have
problems with other aspects of perception (Gregory, 2005). For example, M.M. had
been blind from early childhood. When his vision was restored in his forties, he
adjusted well overall, but he still has difficulty with depth perception and object recognition (Fine et al., 2003). Often, as people move toward or away from him, they appear
to shrink or inflate. Identifying common objects can be difficult for him, and faces pose
a particular challenge. To recognize individuals, he depends on features such as hair
length or eyebrow shape. M.M. has trouble, too, distinguishing male faces from female
ones and great difficulty recognizing the meaning of facial expressions. He is also
unable to experience many of the perceptual illusions shown in this chapter, such as
the closure illusion in Figure 3.20(D) or the size illusions in Figure 3.24.
For the rest of us, too, the ability to experience perceptual illusions depends on
our sensory history. So people who grow up in significantly different sensory environments are likely to have noticeably different perceptual experiences. For example, the size illusion shown in Figure 3.24(A) is strongest in the “carpentered world,”
where seeing straight lines is an everyday experience (Leibowitz et al., 1969).
Responses to illusions such as this one are not as strong for people from rural Africa
and other places in which the visual environment contains more irregular and
curved lines than straight ones (Coren & Girgus, 1978). Similarly, responses to depth
cues in pictures and paintings differ in cultures that do and do not use such images
to represent reality. People in the Me’n or the Nupa cultures of Africa, who have little experience with pictorial representation, have a more difficult time judging distances shown in pictures than do people in picture-oriented cultures (see Figure
3.28). These individuals also tend to have a harder time sorting pictures of threedimensional objects into categories, even though they can easily sort the objects
themselves (Derogowski, 1989). And residents of dense tropical rain forests, where
most objects are seen over relatively short distances, may have some difficulty when
asked to judge the distance of remote objects on an open plain (Turnbull, 1961). In
other words, although the structure and principles of human perceptual systems
tend to create generally similar views of the world for all of us, our perception of