Social judgment and attitudes: warmer,
more social, and less conscious
NORBERT SCHWARZ*
University of Michigan, USA
Abstract
Developments in social judgment research during the last two decades have broadened
the explanatory power of the information processing perspective by paying attention to
the social context of human judgment, the importance of `warm' cognition, and the role
of nonconscious processes. The application of social cognition theorizing to the forma-
tion of attitude judgments provided new insights into classic issues of attitude research,
suggesting that attitudes may be fruitfully conceptualized as temporary constructions.
Implications of these developments, open issues, and potentially fruitful avenues for
future research are discussed. Copyright # 2000 John Wiley & Sons, Ltd.
The editors charged the contributors to this millennium series with a daunting task:
`Assess where we are as a ®eld and where we might venture in the new century (and do
so on no more than 40 manuscript pages).' Needless to say, any assessment of where
we are is likely to be controversial and predictions about the future of a creative ®eld
are bound to be wrong. This being said, this article provides an opinionated review of
selected developments in judgment and attitude research and identi®es open issues
that may provide fruitful avenues for future research. Using the information process-
ing paradigm as a backdrop, I ®rst discuss three general developments, namely the
rediscovery of the `social' in social judgment, the renewed interest in `warm' cognition,
and the current fascination with nonconscious processes. Subsequently, I turn to the
implications for attitude research and question the usefulness of the traditional
attitude concept in favor of a mental construal approach.
SOCIAL JUDGMENT AND INFORMATION PROCESSING
Following two decades of research dominated by variations of the cognitive con-
sistency theme, the introduction of the information processing paradigm in the 1970's
Copyright # 2000 John Wiley & Sons, Ltd.
European Journal of Social Psychology
Eur. J. Soc. Psychol. 30, 149±176 (2000)
Agenda 2000
*Correspondence to: Dr Norbert Schwarz, ISR, University of Michigan, Ann Arbor, MI 48106-1248
USA. e-mail: nschwarz@umich.edu
had a liberating in¯uence on social judgment research (for an introduction to the
paradigm see Lachman, Lachman & Butter®eld, 1979). Drawing on general models of
information encoding, storage and retrieval, social psychologists began to address a
broad range of issues within a uni®ed conceptual framework (for reviews see Devine,
Hamilton & Ostrom, 1994; Martin & Clark, 1990; Strack, 1988). On the theoretical
side, the traditional organization of the ®eld by substantive topics gave way to an
increasing emphasis on basic processes that cut across diverse phenomena, as
impressively re¯ected in Social Psychology: Handbook of Basic Principles (Higgins &
Kruglanski, 1996). On the methodological side, the new paradigm's emphasis on
detailed process models changed social psychologists' standards for what counts as
appropriate evidence. As Taylor (1998, p. 48) noted, `it was no longer enough to detail
a theoretical model and use the results as con®rmation of the model. If one stated
what one thought the process was, one had to demonstrate the intervening steps.'
As philosophers and historians of science are aware, however, the metaphors that
guide a given paradigm focus attention on some questions at the expense of others,
even if these `other' questions could, in principle, be fruitfully conceptualized within
the paradigm (see Kuhn, 1962; Root-Bernstein, 1989; Weimer & Palermo, 1973). In
the present case, the computer metaphor underlying the information processing
paradigm stimulated extensive research into the encoding, storage, and retrieval of
information that proved extremely fruitful for social judgment research. Other aspects
of human judgment, however, were less likely to come into focus, although they are
amenable to an information processing analysis.
First, the paradigm's focus on encoding, storage and retrieval processes fostered a
concentration on individuals as isolated information processors. This focus resulted
in a neglect of the social context in which humans do much of their thinking, both in
terms of the immediate interactional context and in terms of individuals' embedded-
ness in a broader cultural context. As a result, `social psychology found itself trans-
formed into a ®eld now mainly concerned not with human social action, but with
human beings as thinkers and information processors about social stimuli' (Forgas,
1981, p. 3; see also Fiske, 1992; Schneider, 1991). But even the study of `human beings
as thinkers and information processors' suered from this neglect, as subsequent
research into situated cognition, conversational aspects of social judgment, and
cultural variations in basic judgmental processes demonstrated.
Second, re¯ecting its historical roots (see Lachman et al., 1979), the paradigm, as
adopted from cognitive psychology, did not invite attention to subjective experiences
and emotional and motivational in¯uences, thus facilitating an emphasis on `cold'
cognition and deliberate inferences at the expense of research into `warm' cognition
and nonconscious processes (for critical discussions see Banaji, Lemm & Carpenter,
in press; Kunda, 1999; Sorrentino & Higgins, 1986; Zajonc, 1980). In the process, the
complex treatment of perceivers' goals and needs that characterized the New Look
(e.g. Bruner, 1957), for example, was reduced to simpler processing goals, like
forming an accurate impression versus remembering the material presented in the
experiment (e.g. Hamilton, Katz & Leirer, 1980; for a critical discussion see Hilton &
Darley, 1991). As Taylor (1998, p. 72) observed, `Although social psychologists were
unwilling to adopt a radial behaviorist view of the person, the willingness to live with
a fundamentally cognitive view of the social being, to the relative exclusion of
motivational, aective, and behavioral processes took root, at least temporarily'
(emphasis added).
150
Norbert Schwarz
Copyright # 2000 John Wiley & Sons, Ltd.
Eur. J. Soc. Psychol. 30, 149±176 (2000)
As social cognition research unfolded, however, all these topics became major foci
of social psychology (and if the reviews of this paper are any indication, some social
cognition researchers may feel that these topics have never been neglected to begin
with). Thus, researchers rediscovered that humans do much of their thinking in a
social context and turned to the exploration of socially situated cognition and the
interplay of cognition and communication in human reasoning. Complementing this
interest in the immediate social context, cultural in¯uences on social judgment are
about to become a `hot' topic. Similarly, the role of moods, emotions, goals and
motivations in human reasoning received considerable attention and the ®eld is
currently experiencing a surge of interest in nonconscious processes. The picture that
emerges from these lines of work is consistent with social psychology's latest
metaphor, that portrays the information processor as a motivated tactician.
This metaphor builds on William James' (1890, p. 333) pragmatic credo that `my
thinking is ®rst and last and always for the sake of my doing'. It emphasizes that
humans have `multiple information processing strategies available, selecting among
them on the basis of goals, motives, needs, and forces in the environment' (Taylor,
1998, p. 50). In fact, even nonconscious, automatic processes are often tuned to
facilitate goal attainment, although the motivated tactician's ability to ¯exibly adjust
cognitive processes to situational requirements is not without limits. Note that this
metaphor does not question the truism that humans are information processors. It
merely highlights that information stands in the service of, and is tuned to meet, the
individual's goals and needs in a given context. Attention to this pragmatic, socially
contextualized, and `warm' nature of human cognition has greatly extended the
explanatory power of the information processing approach beyond the constraints of
the models we initially adopted from cognitive psychology (see Lachman et al., 1979;
Neisser, 1967). In my reading, this extension of the range of questions asked, rather
than the straightforward application of established principles of cognitive psychology
to `social' rather than `nonsocial' objects as targets of judgment, is the key con-
tribution of social cognition research to psychological theorizing. Next, I turn to
selected aspects of these developments.
REDISCOVERING THE `SOCIAL' IN SOCIAL JUDGMENT
If thinking is for doing, we may expect that individuals adjust their cognitive strategies
to the immediate social context. Indeed, a growing body of work demonstrates such
adjustments in response to variables like the accountability of actors (e.g. Tetlock,
1992), their anticipation of audience reactions (e.g. McCann & Higgins, 1992), and
their interdependence and position in a power hierarchy (e.g. Fiske & Depret, 1996).
Unfortunately, a comprehensive review of research into situated cognition is beyond
the scope of this article (see Levine, Resnick & Higgins, 1993; Resnick, Levine &
Teasly, 1991). Instead, I ®rst focus on the social situation that is most relevant to social
judgment research: the research setting in which we test our theories. Subsequently, I
turn to another implication of the notion that thinking is for doing, namely that
cognitive strategies should also be tuned to meet the requirements of the actor's
broader cultural context. Both bodies of research highlight the potential costs of
investigating basic cognitive processes within a very limited social and cultural context.
Social judgment and attitudes
151
Copyright # 2000 John Wiley & Sons, Ltd.
Eur. J. Soc. Psychol. 30, 149±176 (2000)
Situated Cognition in the Lab: The Case of Judgmental Biases
One of the core intellectual contributions of social judgment research is the ident-
i®cation of a wide range of judgmental biases and shortcomings. No matter what we
think about, we often get it wrong: we rely on nondiagnostic individuating informa-
tion at the expense of more diagnostic baserate information, ignore situational
in¯uences in explaining social behavior, are unduly in¯uenced by surface charac-
teristics of the tasks presented to us, are easily misled by suggestive questions, and so
on and on (for reviews see Grin, Gonzalez & Varey, in press; Nisbett & Ross, 1980;
Kahneman, Slovic & Tversky, 1982). After a quarter century of research into
judgmental biases, the good news is: things aren't (quite) that bad. On the one hand,
the biases identi®ed in this research tradition are robust and reliably replicable.
Moreover, the assumed underlying processes have been con®rmed in numerous
studies. On the other hand, however, the errors we make in the laboratory are much
less likely to result in mistakes in daily life than the research ®ndings would seem to
suggest. Instead, the ®eld's treatment of humans as isolated information processors
has resulted in a profound overestimation of the fallibility of human judgment,
providing an involuntary illustration of the power of situated cognition.
At the heart of this discrepancy is a basic misunderstanding about the nature of
communication in a research setting. According to the tacit assumptions that underlie
the conduct of conversation in daily life, `communicated information comes with a
guarantee of relevance' (Sperber & Wilson, 1986, p. vi) and listeners are entitled to
assume that the speaker tries to be informative, truthful, relevant, and clear.
Moreover, listeners interpret the speakers' utterances `on the assumption that they are
trying to live up to these ideals' (Clark & Clark, 1977, p. 122). Bringing these
assumptions to the research situation, participants assume that every contribution of
the researcher is informative, truthful and clearÐand certainly relevant to the aims of
the ongoing conversation (or why else would it be presented?). Unfortunately, they
miss one crucial point. Whereas the researcher is likely to observe conversational
maxims in daily life, he or she is much less likely to do so in the research setting. In
fact, the researcher may provide information that is neither relevant, nor truthful,
informative and clearÐand may have carefully designed the situation to suggest
otherwise. Having no reason to suspect uncooperative behavior, however, partici-
pants are likely to ®nd meaning the researcher's contributions, rendering logically
irrelevant information conversationally relevant.
Supporting this analysis, a growing body of research demonstrates that many of the
apparent biases and shortcomings of human judgment are attenuated or eliminated
when this basic misunderstanding is taken care of (for reviews to Bless, Strack &
Schwarz, 1993; Hilton, 1995; Hilton & Slugoski, in press; Schwarz, 1994, 1996). When
Dostları ilə paylaş: |