Reclaiming Personal Relevance in Political Research on Emotions

• In Event: How Do You Feel about That? The Role of Emotion in Politics

Abstract

The tension between affect and reason has long been a touchstone of Western intellectual thought. While a mainstay of political psychology, scholarly attention to the affective dimension of politics is undergoing a renaissance across the political and communication sciences. However, what these studies often overlook is the role of personal relevance in triggering emotional response and subsequently, political behavior.

Prior research suggests that for emotions to become activated or conscious, stimuli must first be perceived as relevant to one’s life, aspirations, or goals. Our survey of the literature finds that the vast majority of political experiments on emotions do not explicitly test the extent to which political stimuli are relevant for participants. Thus, we argue that research may be neglecting a key explanatory factor in the study of emotions and political behavior. We ask:

RQ: What is the role of personal relevance in mediating emotional response and political expression?

Theory

We find general support across several theories of emotion that stimuli must be perceived as relevant to trigger an emotional response. Cognitive Appraisal Theory, for example, argues that emotions are generated based on congruence or incongruence with one’s personal goal relevance (Lazarus, 1981). From Affective Intelligence Theory (AIT), Brader (2005) defines emotions as the physiological and mental dispositions triggered by situations of perceived significance. Supporting this notion, political experiments suggest that emotions might not affect political behavior without citizens’ perceiving a political stimulus as personally relevant or without a relatively high level of political interest (Huddy & Gunnthorsdottir, 2000; Jones et al., 2012).

By contrast, research from economic and environmental behavior suggests that when personal relevance is operationalized as a financial consequence, emotions should only influence political behavior when the cost of a decision is low (Westbrook & Braver, 2015). We seek to test the competing expectations of AIT and this low-cost hypotheses (LCH) (Diekmann & Preisendörfer, 2003).

Method

We design a large-scale online voting experiment (N=1,200) where US participants can a) participate in a majority vote on whether a $2000 donation goes to the Democratic or Republican party, or b) choose to not exercise this influence and receive a financial bonus instead. ­­­­Different from traditional survey experiments in political psychology, our experiment expects participants to be most likely to reveal their true political preferences if they are properly incentivized to do so (Murphy et al., 2005).

The experiment contains two incentivized tasks. The first measures Party Preference by letting participants vote on which political party should receive the $2000 donation. The second constructs our main dependent variable, the value of Party Preference Expression, by offering participants the option to withdraw their vote and receive personal compensation instead.

For treatments, we use a 2x2 between-subject design. The first treatment factor (Emotion) splits participants into groups that view a video designed to induce fear or enthusiasm before we measure Party Preference Expression. The second treatment factor (Cost of Expression) randomly assigns participants to receive either a $1 or $5 financial bonus, which they can either keep or use to buy tickets that will decide the majority vote. Thus, we ask participants to place a monetary value on their personal participation in influencing the large donation to a political party. Whether participants are assigned to the $1 or $5 comprises our measure of personal relevance. We presume that since participants take online tasks for money, a $5 bonus makes the decision more personally relevant than being assigned to the $1 group.
­
Following AIT’s expectations for political mobilization, we hypothesize:

(H1): Party Preference Expression is higher in treatments with enthusiasm than fear.

Based on the LCH, we expect:

(H2): Party Preference Expression is low in high-cost treatments.

Regarding personal relevance, we answer the study’s overarching research question by measuring the difference between the emotional treatments across the high-cost and low-cost treatment conditions. Given that AIT and the LCH offer competing expectations, we keep this posed as an exploratory research question.

Contributions
The study contributes to political psychology in three ways. First, we place a conceptual spotlight on the role of personal relevance in eliciting emotions. Second, we assess whether the expectations of AIT hold in methodological designs that do not rely on self-reports. Third, our design may yield evidence for an ‘affective tipping point’ (Redlawsk et al., 2010), where emotions only affect participants’ political behavior up to a point, whereafter financial incentives (i.e., rational choice) take over.

Authors

• Mike Farjam, Lund University

• Michael Bossetta, Lund University, Sweden

• Amy Funck, Rutgers University, New Brunswick