Attribution and Interpersonal Perception (p.163)
The fundamental distinction b/w situational and dispositional causes of behavior is the heart and soul of
virtually all attribution theories. Attribution theories describe how ppl think about each other.
The logic of attribution
Attribution equation or Lewinian equation B= S+D
o It acknowledges the fact that a person’s behavior (B) is thought to be a joint function of the situation
the person is in (S) and the person’s unique predispositions to act (D)
o Virtually no one believes that human behavior is entirely the product of situational forces, and
virtually no one believes that it is entirely the product of dispositions.
o The equation suggests that inner and outer forces combine to produce human behavior
o The Lewinian equation is merely meant to illustrate the dynamic relation b/w the 3 attributional
elements. These relations constitute the core logic of all attribution theories
o If there is one “decision rule” that is common to all attribution theories it is this An observer should
not conclude that a person has a unique predisposition to behave when the person does exactly what
the situation demands.
o In other words, an observer should not assume the D is positive or negative when S and B are equal.
This statement is called the discounting principle.
o The principle suggest that when we try to estimate a person’s dispositions, the person’s behavior
should be discounted, or ignored, when it is precisely the sort of behavior that the situation
o Such behavior is said to be nondiagnostic; that is, it tell us nothing about the person’s unique and
enduring tendencies to behave.
The correspondence bias
o The discounting principle is a simple rule that tells us how we can estimate the dispositions of others
by watching their behavior.
o The correspondence bias which is the tendency to conclude that a person has a disposition that
corresponds to his or her behavior even when that behavior is attributable to the situation.
o Scores of experiments show the correspondence bias to be one of the most reliable and robust
ﬁndings in the annals of research on human attribution. It is also called fundamental attribution
error. The ppl attribute behavior to dispositions even when there is every reason not to do so
The causes of correspondence bias
o There are 2 sorts of considerations:
o The damning objections suggests that some funny quirk of the experimental setting
caused subjects to make judgments that they would not normally make, and the results tell
us more about the particulars of social psychology experiments than about the way ppl
o Interesting objections suggests general psychological mechanisms that might cause ppl to
make dispositional attributions when they should not.
-after many years of work, social psychologist have assembled a list of 4 factors that they believe cause the
o Humans tend to see the world as they wish to see it. Reality keeps us from straying too far from the facts,
but even beliefs that are ﬁrmly grounded in reality can be lightly shifted in the direction of desire.
o One reason why ppl may show the correspondence bias is that they prefer to make dispositional
attributions. Why such preference? Because dispositional attributions give us a sense of control, and they
do this in 2 ways
o A dispositional worldview to chase away the gloomthis worldview is not a well-articulated
philosophy of human behavior but, rather, a general sense that ppl do what they do because
of the kinds of ppl they are, and that as such, whatever happens to them is pretty mush their
This worldview is instilled in us in subtle ways by culture that has hundreds of
words to describe the diff dispositions of ppl and virtually none to describe the
casual power of situations.
o If behavior is a product of inner forces, then not only can ppl control their own lives but,
moreover, they can predict the ways in which others will do so. The ability to predict the behavior of an entity allows one to control the effects of the entity upon oneself. To predict
something one must have a decent theory.
In a sense dispositions serve some of the same functions that scientiﬁc theories
do: they give us a simple way of thinking about a multitude of past observations,
and they allow us to predict what we will observe in the future. Dispositional
attributions give us a “handle” on others’ they afford us a sense of predictive
Playing a game with a person should increase one’s desire to predict that person’s
behavior, after all, behavior is much more important when it affects us personally
than when is doesn’t
o Both our general dispositionist worldview and our need to predict a speciﬁc person’s actions
can cause us to lean toward dispositional explanations of behavior.
o Our ambitions and fears, our hopes and our hungers, can indeed impair our ability to draw accurate
inferences about others. But even without these needs, there are plenty of ways for attributional analysis
to be derailed.
o Another cause of correspondence bias is the tendency for ppl to underestimate the power of situations.
Ppl’s estimates of situational power are not always accurate, and research suggests that the inaccuracies
tend to be of a particular sort underestimations rather than overestimations of situational power. There
are 2 reasons for why ppl constantly underestimate the strength of situations.
o The ﬁrst is that situations are often invisible. Correspondence bias occurs because ppl often
don’t realize that there is a constraining situation to be considered.
o The second is related to the ﬁrst. When ppl do not recognize the presence of situational
forces, they cannot be expected to subtract them out of their behavioral observations. But
when ppl recognize the presence of such forces , they still may underestimate the capacity
of those forces to alter behavior.
Why is there a disparity b/w what we think we should do and what we actually do
in such a situation? Because when we are asked how we would behave in a certain
situation, our psychological construal (or mental picture) of that situation does not
usually include all its intricate details.
Underestimation of situational power is the road to correspondence bias.
o Estimating the S Lewinian equation is not that easy
o Behavior, on the other hand has an “in your face” quality. It is palpable and dynamic, bombarding our
senses with its noises and colors and aromas. It is tangible, conspicuous, and concrete.
o Therefore it is generally easy to estimate the B, however estimating B is not always an easy task
because action identiﬁcation, like any other form of inference, can go wrong.
o Factors that determine the accuracy of our action identiﬁcation
o Our prior knowledge or our expectations exert a particularly strong inﬂuence on what we
perceive others to do.
o These expectations need not be conscious to have their effects
Ex. Our beliefs about a nurse’s job enable us to see that he is “helping”.
o Oftentimes we don’t even know we have expectations until they are violated and we ﬁnd out train of
thought curiously derailed.
o Perceptual assimilation someone (ex. Psychiatrist) “read in” to the behavior what he expected to see
o It is the nature of perception that we often see things as conforming more to our expectations than
they actually do. This fact can pave the way for correspondence bias.
o Trope’s two-stage model of attribution reminds us that action identiﬁcation precedes attributional
inferences; in other words , before we can ask why a person behaved a certain way, we must ﬁrst
know what the person is doing.
o Having information about the situational constraints on a target can increase the accuracy of our
attributions (by preventing us from underestimating S) it can also decrease the accuracy of our
identiﬁcations (by causing us to overestimate B). As such, we can end up making reasonable
attributions about something the target didn’t really do.
o As the two-stage model predicts, good, solid information can have negative as well as positive
consequences; it can promote, as well as prevent, correspondence bias (ex. Of unquestionably calm
and ambiguous facial expression)
o The effect of a situation may be subtracted out during attribution, but the two-stage model shows
that this effect can sneak in during identiﬁcation. Failing to use information
o When we talk about estimating B and S, we are talking metaphorically about what ppl know. But we
haven’t said anything about how to use what they know. Having and using information are not the same
thing. Ppl can estimate S and B well and still fall pray to the correspondence bias— because they cannot
always use the estimates they have made.
o Homuncular functionalism which essentially means that psychologist make theoretical progress by