Chapter 7 Correspondence Bias
Fundamental attribution error/correspondence bias perceivers blame behavior on the
person doing the behaving often at the expense of considering real, external pressures
that forces that person to behave this way.
Underestimate effects of the situation
Overestimate the effects of personal disposition
EVIDENCE FOR CORRESPONDENCE BIAS
The Jones and Harris Attitude Attribution Paradigm
Read study from lecture first simple to understand from there.
Participants rated the essayist’s true attitude as corresponding to the essay, regardless of
whether there was choice. That is, participants should have adjusted their rating of the
person’s true attitudes to account for the fact that he/she was forced to write a proCastro
essay. But they did not; they failed to consider the situation.
• Is correspondence bias an artifact of methodology?
1 criticism is that the research participants were being implicitly told by the
experimenter to make correspondent inferences. That is, norms of communication
and interaction dictate that people do not typically provide other people with
totally irrelevant information.
Therefore if research participants were given an essay read about an authority
figure, were told that it was constructed by a student, and were informed that the
task was to determine how well it reflected the true attitude of the student, the
implicit communication norm at work would be that the information being
provided was probably useful as an indicator of the person’s true attitude. If not,
the experimenters would not have provided this information.
2 criticism the structure of the essay only those who support Castro can
construct such clear and expert arguments in favor of his regime. Participants saw
the nochoice condition actually to be the one where the essayist had plenty of
The Ross, Amabile, and Steinmetz Quiz Show Paradigm
Primary hypothesis of correspondent inference theory is that as situational
constraint increases, people make fewer correspondent inferences.
Again read the slide first for procedure explanation. Found that observers did not take into account the pressure of the situation when
rating either the knowledge of the contestant or the questioner. Rating the
quizmaster more knowledgeable than the contestant.
More dramatically, the person unable to answer the questions specifically drawn
from the idiosyncratic knowledge of another person was seen as generally less
intelligent, despite the fact that it was clearly the situation that made the person
unable to answer the questions. The correspondence bias thus emerged, again, in a
totally different research paradigm. Since observers could have been assigned the
role of questioner or contestant, they should have been able to imagine what it
would be like to be in that role; remarkably, they did not make these
considerations when evaluating others.
The Snyder and Frankel Sex Tape Paradigm
Asked male participants to watch a videotape without sound of a female target person
being interviewed by a strange male.
Some were led to think that the interview topic was somewhat bland and
Others were told that the woman being asked to discuss her sex life.
The assumption was that the sex tape would be seen as more anxietyprovoking,
and that, despite the woman’s showing no difference in how anxious she appeared
from one tape to the next, the participants would assume that her level of anxiety
as a person would be higher when the topic concerned sex.
Results showed that rather than discounting trait anxiety as a cause for her
feelings and behavior, participants instead made correspondent inferences to
behaviors that were not observed, but merely implied by the situation. They
assumed that the woman on the sex tape was more anxious than the woman on the
Conclude that correspondence bias is a real phenomenon, not an artifact of
THE ACTOROBSERVER DIFFERNCE
Jones and Nisbett assumptions:
Although as observers of others we have a tendency to underutilize information
the power of the situation and overutilize inferences about the disposition of
others, such a bias does not occur when we observers become actors.
When we are judging our own actions, correspondence bias disappears. Most obvious fact is that actors are able to see their own internal states; they have
selfknowledge and insight.
When judging others, we are forced to infer their intentions, goals, emotions, and
so on their observed behaviors. These behaviors are salient to us and are seen
inextricably linked to the actors.
• Differences in consistency and distinctiveness information exist for actors and
Observers cannot see how the actors behave in different situations with either the
same or other entities. Casual data are extended over time and are not available to
Actor’s knowledge of their behavior in other situations and at other times should
make them potentially excellent judge of the “covariation” between their behavior
and the various possible situational and dispositional cause that drive their
If actors detect consistency over time, they can infer dispositional causes, if they
detect variability, they can infer situational causes more readily than observers.
• Different aspects of the environment that are salient
Jones and Nisbett (1971) noted that for actors and observers, different types of
information are salient and have an impact on their interpretations of the reasons a
behavior has been performed.
For actors the situation is salient, they are not typically observing themselves
when acting, but are focused outward on the environment.
Linked to figure and ground Gestalt psychology one is figure for the observer
while it becomes ground for the actor.
If people were to take the perspective of another, the corresponding bias should
disappear. However, the existence of the corresponding bias suggests that perspective taking
is difficult; perhaps not the default approach perceivers take to viewing the social
Perspective taking and the fundamental attribution error
Storms (1973) had observers physically alter their point of view so that they observed the
very same physical stimuli that actors observed.
If behavior was videotaped so that observers saw it from the perspective of the
actors and the actors watched themselves as if they were observers. The actor
observer difference should be reversed.
Actors would make more dispositional attributions to their own behavior and
observers would take the situation into account.
The behavior in this experiment was a conversation between two target actors.
Two observers watched the interaction, and each was instructed to observe one of
the two actors/targets. Observer 1 watchedActor 1, Observer 2 watchedActor 2,
Actor 1 was focused on Actor 2, andActor 2 was focused onActor 1. The
interaction was video- taped with one camera onActor 1 and one camera onActor
2. Participants then watched the videotape of the interaction, and the experimenter
manipulated the per- spective they saw while watching. One group saw the same
perspective they had in the live version of the interaction, so that Observer 1
watchedActor 1, Observer 2 watchedActor 2,Actor 1 focused onActor 2, and
Actor 2 focused onActor 1.Another group had the perspective reversed, so that
Observer 1 now saw Actor 2 (who had previously served as the situation forActor
1), andActor 1 now saw him/herself (rather than the situation, which he/she had
been focused on originally). Similarly, Observer 2 now saw Actor 1 andActor 2
now saw him/herself I SUGGEST THATYOU GO TO PAGE 279 TO LOOK
In summary, people are capable of taking the perspective of others when that
perspective is physically presented to them and when they are explicitly instructed
to take the perspective of others.
Perspective taking then allows the perceivers to form a more complete and
accurate attribution of the causes for behavior. However, the evidence also seems
to suggest that although perspective taking is an important ingredient in forming
accurate attributions, taking the perspective of others is not something perceivers
do quite naturally or spontaneously.
It might strike you as odd that even in a situation such as the quiz show paradigm,
where the power of the situational force seems so obvious and clear, perspective
taking appears not to occur, with the resulting negative consequences (the
perceivers form an unwarranted negative, and dispositional, attribution). People
still ignore the obvious fact that the quizmaster is making up questions based on
his/her own knowledge base, and the contestant is completely at the mercy of being asked difficult questions about something at which the quizmaster is
This might simply suggest that perhaps more is contributing to the tendency to
overattribute actors’behavior to their disposition than merely the perspective that
THE CAUSES OFTHE CORRESPONDING BIAS
1. Situations lack salience and therefore go unnoticed by perceivers
- Once perceivers have attended to a salient behavior, the role of the situation in
causing that behavior goes relatively unnoticed.
2. Underestimating the impact of the situation
- Perceivers may notice the situation and still exhibit correspondence bias,
because they have underestimated the role that the situation plays in shaping
- There are two distinc