This article is for use only
of students
enrolled in GPS 520: Advanced Social Psychology for purposes associated with
this course and may not be retained or further disseminated.
Fiske,
S. T. (1995). Social cognition. In A. Tesser (Ed.) Advanced Social Psychology, NY:
McGraw-Hill.
SCHEMAS
What Are
Schemas?
A schema is a
"cognitive structure that represents knowledge about a concept or type of
stimulus, including its attributes and the relations among those
attributes" (Fiske & Taylor, 1991, p. 98). Schemas are basically
preconceptions or theories, in this case about the social world. So, presumably
you have a concept of a warm person: What are warm people like? What do warm
people do? "They are friendly." What else? "Thoughtful."
"Helpful." Do they say "Hi" to you in the hallway?
"Yes." You have some preconceptions about what a warm person is like.
So, the term schema is just a fancy way of saying that. There are a lot of
elements to this preconception, but the elements are connected to each other in
some fashion, and they influence each other's meaning. They form a theory about
what warm people are like.
What Kinds Are
There?
Different types of
schemas have been studied (Taylor & Crocker, 1981). We just described what
might be called a person schema, that is, a schema for warm people.
People have schemas for all kinds of personality traits, such as what it means
to be irritable, outgoing, or conventional (Cantor & Mischel, 1979). People
also have schemas for social goals, such as ingratiation, revenge, and
helpfulness (R. C. Anderson & Pichert, 1978; Owens, Bower, & Black,
1979; Zadny & Gerard, 1974; see Pervin, 1989, for a collection of related
work). Person schemas also include people in situations, such as a shy person at
a party or a practical person at a meeting (Cantor, Mischel, & Schwartz,
1982).
Self-schemas
form the core of our self-concept; people have schemas for themselves. For
example, do you consider yourself independent or dependent? Why? If one has a
well-developed self-concept on a dimension, one is schematic on that dimension
(Markus, 1977; Markus & Wurf, 1987). People's schemas for -themselves are
much more complicated than their schemas for other people, as you might imagine
(Linville, 1982a; Lord, 1980). That has some interesting implications, such as
increasing one's ability to remember information considered self-relevant (for a
review, see Kihlstrom, Cantor, Albright, Chew, Klein, & Niedenthal, 1988).
People also differ in the chronic accessibility of different traits. That is,
for certain people, certain traits are easily activated and readily applied to
self and others (Higgins, King, & Mavin, 1982; Higgins, 1989).
Role
schemas contain our understanding of the behaviors and attributes expected
of people in particular social positions. For example, you may have learned what
the role of graduate student means, how that is different from being an
undergraduate. Which one routinely calls professors by their first names? And
for which one does Saturday night become a night when you can get a lot of work
done? And graduate students are the only people who spend more money on books
than on food. Graduate students are people undergoing an initiation to become
part of the community of scholars. If you saw a graduate student coming to class
in a new fur coat and a BMW, you'd wonder what was going on with this person.
The point is that one has some conceptions about what graduate students are
like. One important distinction is between achieved roles, such as graduate
student, chain gang member, or zookeeper (sometimes "graduate student"
seems to include the other two); achieved roles are those we acquire through
some activity. In contrast, ascribed roles are those given to us automatically
(e.g., age, race, gender). Schemas for roles are equivalent to stereotypes,
people's expectations about people who fall into particular social categories.
Allport (1954b) described them as the "nouns that cut slices." Of
course, there are many different explanations for stereotypes besides schemas,
but role schemas are one way to think about stereotypes.
Finally, we
have schemas for certain kinds of events. To imagine an event without a schema,
consider the following:
The
procedure is actually quite simple. First, you arrange things into different
groups. Of course, one pile may be sufficient depending on how much there is
to do ... It is important not to overdo things. That is, it is better to do
too few things at once than too many. In the short run this may not seem
important, but complications can easily arise. A mistake can be expensive as
well. At first, the whole procedure will seem complicated. Soon, however, it
will become just another facet of life (Bransford & Johnson, 1972, p.
722).
Reading this event
sequence without the proper script makes it all seem rather disconnected and
mysterious. Doubtless, it would be difficult to remember the whole passage a few
minutes from now. Event schemas contain appropriate sequences of events
in social situations. What happens when one goes to a restaurant: get seated,
look at a menu, order something, eat, pay, and leave. It is a predictable
sequence, almost a script (Abelson, 1981; Schank & Abelson, 1977). Recently,
the understanding of event sequences has been applied to phenomena as varied as
theft, defecting from one's country, underdogs winning, stressful life events,
and jury decision making. By the way, the mystery procedure described earlier
was washing clothes. Now does it seem a more familiar and memorable script?
Organized
prior knowledge or preconceptions--schemas of all types smooth our information
management and social experiences. The point is that people seek simplicity and
good-enough accuracy, good-enough understanding of the world around them, and
schemas are guides. Having these preconceptions helps one to understand things
with relative efficiency and accuracy.
What Do Schemas
Do?
Pat woke up
feeling sick again and wondered if she really was pregnant. How would she tell
the professor she had been seeing? And the money was another problem.... Pat
went into the kitchen, took the pot out of the cabinet, put in some mix, made
some coffee, looked at the coffee, and decided to add some milk and sugar.
After that, Pat got dressed and went to the doctor. After arriving at the
office, Pat checked in with the receptionist, and then saw the nurse, who went
through the usual procedures. Pat stepped on the scale, and the nurse recorded
Pat's weight. The doctor entered the room and examined the results of the
procedures, smiled, and said, "Well, it seems that all my expectations
have been confirmed." Then Pat left the office. Pat arrived at the
lecture hall and decided to sit in the front row. Pat walked down the aisle,
sat down. The professor went to the podium and began the lecture right away.
All through the lecture, Pat just could not seem to concentrate on what was
being said. The lecture seemed really long. But, finally, it ended. The
professor was surrounded by people after class, so Pat quickly left. Later
that afternoon, Pat went to a department cocktail party and looked around the
room to see who was there. Pat went over to the professor, wanting to talk to
him, but feeling kind of nervous about what to say. A group of people began to
play some games. Pat went over and had some refreshments. The hors doeuvres
were good, but Pat did not feel very interested in talking to some of the
other people at the party. After awhile, Pat decided to leave (after Owens,
Bower, & Black, 1979).
Suppose
Pat and the professor had had a chance to have a conversation. How do you
suppose that Pat would have been feeling under those circumstances? What are
some emotions that come to mind? "Anxious." "Uncomfortable."
Do you think Pat and the professor might have made arrangements to meet to talk
later, or would they have been able to take care of what they had to talk about
right there at the party? "Definitely later." Is Pat glad when Pat
discovers that the scale has gone up recently? "Hardly."
Now
consider the same story with a different lead-in replacing the first sentences
up through the ellipses ( .... ): "Pat woke up wondering how much weight he
had gained so far. His football coach had told him he could start in the game
Saturday only if he gained enough weight and passed the chemistry test. The
pressure was intense. . . ." Now reread the rest of the original story
after the ellipses.
The
meaning changes rather dramatically, doesn't it? How is football Pat feeling
about the professor? What are his goals for talking to the professor at the
party? How does this Pat feel about the scales? When we have two different Pats,
it becomes rather a different story, doesn't it? This experiment, conducted by
Gordon Bower and his colleagues (Owens, Bower, & Black, 1979), presented
versions of essentially this story to undergraduates, with a third receiving the
pregnant lead-in for a character named Nancy, a third receiving the football
lead-in for a character named Jack, and a third receiving no lead-in. Half an
hour later, they asked subjects to recall the action sequences, as close to
verbatim as possible. Subjects given a lead-in problem remembered more episodes,
more often in correct order, and displayed many more intrusions of recall than
did control subjects. The memory intrusions included thematic interpretations of
ambiguous statements (for Nancy, the nurse's "usual procedures" became
"pregnancy tests"), as well as inferences about motives and feelings
(at the party, Nancy "was feeling miserable"). This research pointed
out that we have schemas for would-be athletes and for unwanted pregnancies, and
they shape how we encode, remember, and judge the information that follows.
Schemas
Direct Attention and Guide the Encoding of Schema-Relevant Information.
What did you notice in the first story? Given the pregnant Pat Schema, one is
particularly likely to notice that the doctor goes through the "usual
procedures," which one may encode as a gynecological examination. For
would-be athlete Pat, that interpretation is unlikely. Indeed, the "usual
procedures" is not a relevant phrase and probably rather meaningless in
that context. It is not something that one would notice. But would-be athlete
Pat getting on the scale is something noteworthy, because this poor person is
trying to gain some weight. Pat's dealing with food is probably significant in
both cases but in different ways. Pat's noticing whether there are a lot of
other people around the professor takes on a different significance in the two
cases. So, what one notices and how one interprets what one does notice are both
affected by the schema applied to the particular situation.
People's
encoding processes balance the schema against the data, neither completely
relying on the schema and completely ignoring the data nor completely allowing
the data to speak for themselves without any influence of the schema. On the
other hand, schemas do have a major impact, in several respects: Schemas often
are cued by visually prominent physical features (e.g., age, race, gender, dress
indicating social class) or immediately provided labels (e.g., occupation). From
the earliest moments of perception, schemas affect how quickly people notice,
what people notice, and how people interpret what they notice. For example,
racial labels automatically can cue stereotypic associations within milliseconds
(Devine, 1989). Schemas, because they stem from categorization processes, tend
to minimize differences among instances of the same schema ("They all look
alike") and maximize differences between groups (for a review, see Mullen
& Hu, 1989). Schemas provide simple concepts of outgroup members (e.g.,
Linville, 1982b). And schemas encourage stereotypic interpretations of behavior
(e.g., Darley & Gross, 1983). When people do not attend to inconsistency
(information that may dispute their schemas), their subsequent processing is
biased toward the schema (Hilton, Klein, & von Hippel, 1991; White &
Carlston, 1983).
On
the other hand, people sometimes do pay close attention to schema inconsistent
evidence (e.g., Brewer, Dull, & Lui, 1981). When people do attend to
inconsistency, their processing is more idiosyncratic and (for some people) more
responsive to the individual confronting them (Fiske, 1992,1993b). People do
balance schema-based processes, i.e., prior theories, against the data they
encounter (for reviews, see Fiske & Neuberg, 1990; Higgins & Bargh,
1987). In the next section, we will consider the circumstances under which
people do and do not remember the inconsistent information they notice.
Schemas
Also Guide Memory.
Think back to pregnant Pat. Some story items were more related to pregnancy, and
some were more related to football. Afterward, without benefit of knowing the
alternative opening to the story, people who get the pregnancy lead-in remember
more of the pregnancy items, and people who get the football lead-in remember
more of those (Owens, Bower, & Black, 1979). In general, when people have
very well developed schemas, they remember things that are consistent with the
schema (Fiske & Neuberg, 1990; Higgins & Bargh, 1987; Ruble &
Stangor, 1986). In contrast, if people are just beginning to develop a theory or
an impression of a particular person, then people notice both the consistent and
the inconsistent information. People may then remember the inconsistent
information, as a consequence of the effort to integrate it into a developing
impression (Hastie & Kumar, 1979; Srull, 1981; Srull,& Wyer, 1989).
Without a strong prior theory, people remember the inconsistencies, but with
strong prior theories, they do not recall as much inconsistent information (for
other conditions of the inconsistency advantage, see Stangor & McMillan,
1992). In any case, people remember schema relevant information (both consistent
and inconsistent) better than irrelevant information.
Schemas
Influence Judgment.
People sometimes make judgments based on what they can remember, particularly if
the judgment is unanticipated (Hastie & Park, 1986). However, many judgments
are made on-line, as the information is received (Bassih, 1989). For example,
people have immediate evaluative responses to other people they meet; they know
rapidly whom they like and dislike, as the initial encounter unfolds. When we
meet someone who reminds us of someone else we once knew, the new person elicits
the same emotional response as the old one; this might be called
schema-triggered affect or, in psychoanalytic terms, transference (Andersen
& Cole, 1990; Fiske, 1982; Westen, 1988). Sometimes the newly met person
does not remind us of anyone in particular but instead comes from a category
(occupational, ethnic, or otherwise), to which we have attached a strong
affective tag (Dijker, 1987; Fiske, 1982; Fiske & Pavelchak, 1986; see
Devine, Chapter 12). The most powerful type of category is "us" versus
"them," a primitive intergroup divide that precipitates in-group
favoritism (Brewer, 1979; Mullen, Brown, & Smith, 1992). Schemas can allow
simplified, polarized judgments and evaluations (Linville, 1982a, b; but also
see Judd & Park, 1988, Marques, Yzerbyt, & Leyens, 1988; Millar &
Tesser, 1986). In short, schemas shape judgments in various and important ways.
Schemas
versus Evidence.
Schemas have far-reaching effects on attention, memory, and judgment, no doubt
about it. But data themselves also influence people's thinking. People's use of
schemas versus evidence depends on, for example, the fit between the schema and
the relevant data (Fiske, Neuberg, Beattie, & Milberg, 1987) and the
perceived diagnosticity of the data (Hilton & Fein, 1989; Leyens, Yzerbyt,
& Schadron, 1992).
People's
trade-off between schemas and data also depends on motivation, as described by
several current theories (Brewer, 1988; Fiske & Neuberg, 1990; Gollwitzer,
1990; Hilton & Darley, 1991; Kruglanski, 1989a; Ruble, 1994; Snyder, 1992;
Stangor & Ford, 1992). Under some conditions, people conserve mental
resources and rely on a prior schema or expectancy, namely, when they are
oriented toward efficiency, speed, action, or interaction, or when they are
distracted, anxious, or defensive. Such conditions all promote category-based
processes. When people are overloaded, they are more likely to form superficial
judgments. Under other conditions, people are concerned with accuracy and
examine the data more carefully, namely, when they are outcome-dependent,
subordinate, stigmatized, accountable, depressed, or instructed to be accurate;
these conditions promote data-driven or piecemeal processes (for reviews, see
Fiske & Taylor, 1991, Chap. 5; Fiske, 1992,1993a, 1993b).
This
resolution-when people use schemas and when they use data-did not come easily.
Indeed, it has been a recurring problem in social perception. Initially, as
discussed earlier (see "Where Do These Ideas Come From?"), person
perception researchers faced the apparent contradiction between the
psychological plausibility of schema-like Gestalt models and the predictive
value of algebraic models that seem to emphasize data. Later, as another example
of theory-driven versus data-driven processes, social cognition researchers
faced apparent contradictions between people's use of stereotypes and their
neglect of population base rates (see below; Locksley, Hepburn, & Ortiz,
1982). Researchers wondered, so how can that be? How can both kinds of models be
right? Is one wrong, and the other one's right, or what is going on?2
Basically,
the bottom line is that both approaches are right; it just depends on the
circumstances. One way to think about it is that there is a continuum between
the two. On the one end are schema-based or categorical kinds of impressions
that are the more holistic, integrated impressions that we form about people.
And on the other end are more individuating and attribute-oriented impressions,
elemental kinds of impressions, piecemeal kinds of impressions. So, we have a
continuum between the schematic kinds of impressions and the elemental,
algebraic kinds of impressions. In between are some intermediate kinds of
processes, including subtyping and self-reference (for more detail, see Fiske
& Pavelchak, 1986; Fiske & Neuberg, 1990).
Basically,
I would argue, the normal, default option is to go with the schema, the
category, the preconception, the theory. That is what my friend did in the
shopping mall. In the first nanoseconds, the automatic reaction was to go with
the schema (see Devine, Chapter 12). People's immediate reaction is to go with
the category unless they have more time, attention, and motivation to be more
careful. Then, in that case, they may move down the continuum toward the more
individuated, personalized, piecemeal kinds of responses where they take in each
piece of information about the person and think hard about it as an individual
piece of information. So what this means is that the more motivated one is, the
more attention that one can pay to a person, the less schema-based is one's
impression, all else being equal. You can think of this as "How do you
overcome stereotypes?" How do you get people to go beyond their initial
schemas and stereotypes, moving into a more individuated impression of somebody
else? The answer is: attention and motivation. It takes more time to form a
fully attribute-oriented, elemental kind of impression of somebody. That is why,
when you buy a car, you may in fact buy a car by saying, "What kind of
mileage? What does it look like?" and so on, because that is a very
involving and highly motivated kind of decision. Suppose you are trying to pick
somebody to advise your dissertation: That also might be a more considered pros
and cons kind of decision. So, when it is really important to people, they can
make those kinds of decisions. But most of the time it is the initial, knee-jerk
reaction, which is to go with our schemas, to go with our prior categories.
Conclusion
I
hope this section has convinced you that people use schemas for good enough,
very rapid understanding. Schemas can have a big impact, yet people also can
engage in more individually tailored understanding of other people. There are a
couple of different processes, and both of them operate, depending on
information and motivation.
The
schema section has discussed what might be called one of the major elements of
social cognition. Are the basic elements the individual items of the impression,
or are they somehow the whole schema or configuration? What are the building
blocks? What are some of the elements of making sense of other people? Schemas
are clearly one. I would also argue that attributions are another element in
making sense of other people, People are causal agents (think back to
differences between perceiving people and perceiving objects), so it is
important for us to figure out why they do what they do and to predict what they
will do in the future. And that is why attribution theory has become such an
important part of social cognition research, because causal explanation is so
important in perceiving people and not so important in perceiving nonsocial
objects (see Gilbert, Chapter 4). When attributions are made to personal
dispositions, we then have available all the information stored in the relevant
trait schema.
Schemas
and attributions are fundamental building blocks of social cognition. As
suggested earlier, one can consider causal attributions and schemas as some of
the elements of social cognition. In effect, these are the contents on which we
operate when we form impressions. Next, we consider three major processes that
operate on schemas and attributions. What are the processes by which we make
sense of other people? The chapter next turns to these information processes:
attention, memory, and inference. They are ordered as if they make a sequence,
but we will rapidly see that there is a constant interplay among them.