Can reactive emotions diminish the perceived social power of an angry person?

Abstract

An evolving line of research has found that expression of emotions is used by observers to appraise the social power of the person expressing the emotion. Research indicates that people who express anger tend to be perceived as having a higher degree of social power than those who express sadness. Recently research has found that the perceived social power of the person who expresses the emotion depends also, to some degree, on the response of the addressee of this expression, i.e., on the reactive emotion of the addressee. Specifically, when the initial expression is anger and the addressee of this anger responds with neutrality or anger, the angry person is perceived as less powerful than when the response to this anger is fear or sadness. Nevertheless, it appears that anger still leads to higher perceived social power irrespective of whether the reactive emotion shown by the addressee of this anger is anger or neutrality.

The effect of reactive emotions is explained by the fact that the emotional reaction of the addressee is an additional source of information about the social power of the first person. Thus, the emotional response is perceived as a confirmation or disconfirmation of the power of the first expresser, as reflected by the expression of anger. In other words, the expression of anger can be seen as a claim that the person is powerful, and the response to this anger as a counterclaim.

The main goal of the present research is to explore whether and under what conditions a reactive emotion to anger diminishes the perceived social power of the person expressing anger. To this end, the research examines the possibility that contempt, happiness, or both, as reactive emotions, have a more potent effect on the perceived social power of a person expressing anger than do anger and neutrality. This is because both emotions are expected to pose a greater challenge to the claim of high social power suggested by anger. The study compared the perceived power of a person expressing anger (associated with high social power) with that of a person expressing sadness (associated with low social power), as a function of reactive emotion. Thus, one research question was whether any of these reactive expressions of emotion in a dyadic exchange can negate the perceived power of a person expressing anger. A related goal of the present study was to assess the contribution of reactive emotions to the perception of social power by comparing its effect with a condition in which the addressee does not respond to the expression of anger by the first person. Previous research failed to answer this question. An assessment of the degree to which a certain reactive emotion affects the perceived social power requires adding such a condition as a benchmark.

Interactions involving expressions of emotions exchanged between the partners to the interaction are dynamic by nature, both because expressions are exchanged in the course of the interaction and because the expressions evolve and change over time. To assess the contribution of such dynamics to the interaction, the dynamic nature of the interaction was gradually introduced across the studies. This made it possible to assess in a controlled manner the contribution of the dynamic nature of the social interaction to the judgments in question, above and beyond the effect of the emotions exchanged between the people in the interaction. It also enabled the sets of studies to gradually approximate the interaction that may be witnessed by observers in real life.

Study 1 explored the perceived social power of a person expressing anger or sadness as a function of the reactive emotions of the addressees of these emotions. Still photos depicting the sequence of exchanged expressions, step by step, focusing on each expresser individually, were used to represent the interaction. Study 2 used the same approach, with videos of the expressions replacing still photos. Finally, Study 3 used videos showing both expressers in the interaction. In this study, the intensity of the expressions was reduced to better match expressions in real life interactions. All three studies also assessed the extent to which the effect of the reactive emotions on the perceived social power of the first expresser can be explained by such effect serving as perceived (dis)confirmation of the claim of social power suggested by the expressions of the first person.

Findings show, consistently across all three studies, that both the angry and the sad person were perceived as having a higher degree of social power when the response to these expressions was one of fear than when no reactive emotion was seen. As expected, the effect of fear was mediated by the perceived confirmation of social power.

Despite the fact that high-power reactive expressions were generally perceived as a disconfirmation of the anger expresser’s claimed social power, the effects of these emotions were inconsistent and even contradictory across the studies. In Study 2, happiness led to a decrease of perceived social power of the angry person compared to a situation in which no reactive emotion was seen. By contrast, happiness had the opposite effect in Study 3 in response to a sad expresser. Similarly, a reactive emotion of anger caused the sad person to seem to have a greater degree of social power than when no reactive emotion was seen. These effects were not mediated by perceived confirmation of social power. As noted above, the intensity of the emotions in Study 3 was reduced to better match real-life expressions of emotions. This suggests that the signal value of certain emotions depends, at least to some extent, in addition to the fact that they are witnessed as part of a social interaction, on their intensity and/or other factors associated with the context of their perception. Finally, to answer the main question of the present research, across the studies, the angry person was always seen to be more powerful socially than the sad person, even when each of these expressions was challenged by the reactive emotion of the addressee.

Overall, the research shows that the social signal value of expressions of emotion changes to some extent as a function of the emotional response of the addressee. The social signal value of emotions does not stand alone but must be understood in the fuller context of the interaction. The effect of reactive emotions was clearest when the prototypical expressions were shown as dynamic videos, in Study 3, because it allowed participants to see clearly and label the facial expressions, as reflected in the participants’ judgments. Finally, the findings also indicate that expressions of anger are a potent signal of social power. Even if the reaction of the addressee to the anger of the first person suggests that they reject the first person’s claim of power, observers still tend to see the angry person as having rather high power. When the targets of the anger responded with an expression that confirmed the high power, the angry person seemed more powerful. Overall, this suggests that observes tend not to ignore the meaning of claims of high social power suggested by expressions of anger. This is possibly because ignoring claims of high social power is potentially costlier than ignoring signals disconfirming it. In other words, ignoring the possibility that someone is socially powerful is more risky than accepting that they are powerful when in reality they are not.

Taken together, the three experiments show that the social signal value of expressions of emotion changes as a function of the emotional response of the addressee. Nevertheless, the degree of this change depends on the type of response. Thus, the social signal value of emotions does not stand alone but must be understood in the fuller context of the interaction.