Common sense holds that feelings precede and cause various kinds of behaviors: we frown because we feel angry, sit slumped in our chairs because we feel depressed, speak well of a candidate because we feel attracted to her, let our attention wander because we are bored, and so forth. Following William James (1884) original proposal, self-perception theory argues that common sense has the sequence of events exactly backwards: First we act, and the acting creates the feeling. We feel angry because we scowl, depressed because we sit slumped, and are attracted to the candidate because of the speech we have made. In effect, feelings are the perceptions of our actions and the contexts in which they are performed.
The obvious empirical difference between common sense and self-perception theory is in what each would predict if we induced someone to act as if they felt something. Common sense would anticipate no effect of actions on feelings, whereas self-perception theory predicts that acting would lead to feeling. Indeed, if the actions did not produce corresponding feelings, self-perception theory would clearly be wrong. Most of my research for years has tested this kind of prediction, for a wide variety of behaviors and feelings. As self-perception theory predicts, people who are induced to act as if they feel something report actually feeling it, even when they are unaware of how they are acting, or the way in which their feelings arise. This effect has been demonstrated for a wide variety of feelings, and with an even wider variety of behaviors. (See Laird, J.D. & Bresler, C. (1992) The Process of Emotional Feeling: A Self-Perception theory. In M. Clark (Ed.) Emotion: Review of Personality and Social Psychology (Vol 13), 223-234. Newbury Park, CA: Sage. )
One implication of self-perception theory is that cues from a number of different emotional behaviors should combine to produce stronger feelings than the parts alone. That effect has been recently demonstrated: Flack, W.F., Laird, J.D., & Cavallaro, L.A. (1999) Additive effects of facial expressions and postures on emotional feelings. European Journal of Social Psychology 29, 203-217.
People appear to differ in how strongly they respond to their own bodily reactions and behaviors. Some people are very responsive, and feel happy when induced to smile, angry when induced to frown, more sad and less confident when they sit in a slumped posture, more in love when the exchange mutual gaze with a stranger, etc. Others, however, are relatively unaffected by their bodies and behaviors, and instead their emotions are determined by social expectations (see Laird & Bresler, 1992 or Laird & Apostoleris, 1996) for reviews of these and many other similar studies). These differences in response to bodily and behavioral, or "personal" cues as opposed to "situational" cues appear to be general across many domains of feeling. For example, a recent study (Wilcox & Laird, in press, Journal of Research in Personality) found that women who were more responsive to situational cues enjoyed looking at magazine pictures of extremely slender models, apparently because they identified with the models. In contrast, women more responsive to personal cues enjoyed the pictures less, and their self-esteem dropped, apparently because these women used the models as standards of comparison.
We have been applying self-perception analyses to other, non-emotional feelings. In two studies we found that postures affected degrees of confidence:
Laird, J. D., Kuvalanka, K., Grubstein, L., Kim, T. H. & Nagaraja, T. (Submitted) Posture and confidence: standing (and sitting) tall, makes you feel good.
Another recent study found that avoiding the gaze of an interviewer induced feelings of guilt (Schnall,S., Laird, J.D., Campbell, L., Hwang, H., Silverman, S. & Sullivan, D. (2000) More than meets the eye: avoiding gaze makes you feel guilty. Presented at the meeting of the Society for Personality and Social Psychology, Nashville, TN.)
Alex Genov and others have recently found that people who are more responsive to
personal cues detect pain more rapidly in the cold pressor task, and tolerate less pain
than do people more responsive to situational cues. (Genov, et al. Submitted.)
More recently my collaborators and I have been applying self-perception theory to a
number of clinical syndromes, including
Premenstrual tension appears to be a misattribution or misperception, in which bodily states produced by monthly hormonal changes are mistaken as emotional. If so, we should be able to predict which women will experience PMS, and should be able to at least temporarily alleviate it. In two studies we found this to be true.
Schnall, S.,Abrahamson, A., Capron, C. & Laird, J.D. (1997) Self-perception and
pre-menstrual tension. Presented at the meeting of the American Psychological Association,
Chicago, August, 1997) Schnall, S., Abrahamson, A. & Laird, J. D. Premenstrual
tension is an attribution error: An individual differences perspective. (Submitted) (Draft
Paper available on request.)
Our original thought was that the link between expressive behavior and feeling might
have been disrupted in people sufferrring from schizophrenia. That does not seem to
be the case, since the people with schizophrenic diagnoses respond to manipulations of
expressive behavior in much the same way as do normals. Our current thought is that
"flat affect" may be may be an attempt to control feelings through controlling
behavior.
Flack, W. F. Jr., Laird, J. D. & Cavallaro, L. A. (1999) Emotional expression and
feeling in schizophrenia: Effects of expressive behavior on emotional experience. Journal
of Clinical Psychology.
Laird, J.D. & Flack, W.F.Jr. (1996) Emotional Behavior and Feeling in Schizophrenia.
Presented at the meeting of the International Congress of Psychology, Montreal, Canada,
August 17.)
Flack, W.F., Jr., Litz, B.T., Beaudreau, S.A., Ruef, A., Laird, J.D., & Weathers,
F.W. (1997, November). Emotional expressions and feelings in combat-related PTSD.
Presented at the 13th annual meeting of the International Society for Traumatic Stress
Studies, Montreal, Quebec, Canada.;
Flack, W.F., Jr., Litz, B.T., Laird, J.D., Beaudreau, S.A., Ruef, A., & Weathers, F.W.
(1997, November). A self-perception approach to emotion in combat-related PTSD. Presented
at the 13th annual meeting of the International Society for Traumatic Stress Studies,
Montreal, Quebec, Canada.) (Draft paper available on request.)
Self-perception and Self-control/regulation:
A more theoretical direction has been toward trying to understand the role of feelings in behavior more completely. Our current view is that feelings are information which serves as feedback to hierarchically stacked control systems, and therefore at the center of self-control.
Laird, J.D. & Apostoleris, N. (1996) Emotional self-control and self-perception:
Feelings are the solution, not the problem. In R. Harre & W.G. Parrott (Eds.) The
emotions: Social, cultural and physical dimensions. London, Sage.)
Self-perception theory predicts that manipulating expressive behavior should provide
the opportunity for deliberate self-regulation of emotional feelings, and a first study
shows that it does. However, individucal differences strongly affect the relative
efficacy of different emotional control strategies:
Duclos, S. A. & Laird, J. D. (submitted) The Deliberate Control of Emotional Experience Through Control of Expressions.
Simone Schnall conducted a direct test of the effect of "practicing" expressions. People who deliberately adopted expressions reported feeling those emotions more strongly when tested later, and also recalled personal experiences that were consistent in emotional tone with the practiced emotion.
Schnall, S. & Laird, J. D. (1998) The enduring effects of emotional expressions on moods. Presented at the meeting of the American Psychological Association, San Francisco, August.
One criticism of many self-perception experiments is that the participants might have been deliberately reporting feelings they believed the experimenter wanted them to feel. To reduce this possibility, in a recent study participants were given the emotional Stroop test while they were adopting emotional expressions. When expressions and words on the Stroop list were related, interference was greater than when expressions and words were different. A paper reporting these results will be presented at APA this summer, Schnall, S., Dhavale, D., Diriwaechter, R. & Laird, J. D. Stroop inteference and emotions induced by expressions.