In attempting to specify the cognitive antecedents of emotion, Smith,
Lazarus, and colleagues (e.g., Lazarus & Smith, 1988; Smith & Lazarus,
1990; Smith, Haynes, Lazarus, & Pope, 1993) have advanced a theoretical
distinction between causal attributions and appraisals, and they have hypothesized
that, of the two, appraisals are the more proximal antecedents of emotion.
Briefly, they have proposed that whereas attributions represent an inferential
analysis that contributes to the construction of a cognitive representation
of one's circumstances, appraisals represent an evaluation of the personal
significance of these circumstances as represented in the construal. According
to these investigators, it is this latter evaluation that initiates an
emotional response. Attributions are hypothesized to influence emotions
more indirectly through their effects on appraisal. In other words, appraisals
are hypothesized to mediate the relations between attributions and emotions.
Smith et al. (1993) have reported initial evidence supporting the proposition
that appraisals are more proximal antecedents of emotion than are attributions.
In two studies, involving an examination of six emotions, it was consistently
found that whereas both attributions and appraisals were systematically
associated with the experience of each of the emotions, the relations of
appraisal to emotion were substantially stronger than those of attribution
to emotion. Moreover, the latter relations were completely subsumed by
the former ones.
The present investigation reports the results of two studies that build
upon these findings by testing a specific mediational model of the relations
of attributions and appraisals to experiences of anger and guilt -- two
emotions for which clear attributional models have previously been specified
(e.g., Heider, 1958; Shaver, 1985; Weiner, 1985). In this model, under
unpleasant conditions, anger and guilt are hypothesized to be evoked by
appraisals of other-accountability/blame and self-accountability/blame,
respectively. The appraisals, in turn, are hypothesized to be systematically
influenced by several causal attributions. Specifically, attributions that
the causal locus is external to oneself will tend to produce appraisals
of other-accountability/blame (and hence anger), whereas attributions of
internal locus will tend to produce appraisals of self-accountability/blame
(and hence guilt). However, the relation between attributed locus and accountability
is hypothesized to be moderated by additional attributions, including those
of the degree to which the undesirable outcome was controllable, intended,
and/or foreseeable by the causal locus: attributions of controllability,
intentionality, and foreseeability should strengthen the influence of attributed
locus on accountability/blame, whereas the opposite attributions should
mitigate this influence.
These hypotheses were examined in two studies involving a total of
278 undergraduates. In the first study subjects retrospectively reported
on their relevant attributions, appraisals, and emotions during a variety
of remembered unpleasant experiences, and in the second study subjects
reported on these same variables immediately after receiving their grades
on a midterm examination. Separate path analyses were conducted within
each study to test the mediational models proposed for anger and guilt,
respectively, for a total of four sets of analyses.
Across these analyses, the mediational model received considerable
support. In each case the systematic relations between the attributions
and the emotions were largely mediated through the appraisals of accountability/blame.
These latter appraisals always demonstrated strong and reliable (ps <
.001) direct relations to the emotions in accord with predictions. Moreover,
in every case both the locus of causality and attributions of controllability
were statistically significantly related (all ps < .05, two-tailed)
to the appraisals of accountability/blame as hypothesized. However, once
attributions of controllability had been taken into account, the relations
of foreseeability and intentionality to the appraisals were weak and inconsistent
across analyses.
These findings further validate the conceptual distinction between
attribution and appraisal, and they support the hypothesis that appraisal
is the more proximal antecedent of emotion. In addition, they begin to
specify the ways in which attributions and appraisals are combined in eliciting
an emotional reaction.
References
Heider, F. (1958). The psychology of interpersonal
relations. New York: Wiley.
Lazarus, R. S., & Smith, C. A. (1988). Knowledge
and appraisal in the cognition-emotion relationship. Cognition and Emotion,
2, 281-300.
Shaver, K. G. (1985). The attribution of blame: Causality,
responsibility, and blameworthiness. New York: Springer.
Smith, C. A., Haynes, K. N., Lazarus, R. S., &
Pope, L. K. (1993). In search of the "hot" cognitions: Attributions, appraisals,
and their relation to emotion. Journal of Personality and Social Psychology,
65, 916-929.
Smith, C. A., & Lazarus, R. S. (1990). Emotion
and adaptation. In L. A. Pervin (Ed.). Handbook of personality: Theory
and research (pp. 609-637). New York: Guilford. Weiner, B. (1985). An attributional
theory of achievement motivation and emotion. Psychological Review, 92,
548-573.
