Conceptualizing and Testing the Affective Component of Affective Polarization

• In Event: What Is Affective Polarization and How Should We Measure It?

Abstract

Affective polarization, we are frequently told by politicians, pundits, journalists, and many of our colleagues, is a major threat to democratic stability. This concept, which suggests that we should look at polarization through the lens of social identity, borrows heavily from psychological research. However, researchers have largely ignored the emotions literature central to psychology. Consequently, researchers have mostly focused on one dimension of affect using one type of measurement. While political scientists have made some headway in understanding the causes and consequences of this phenomenon, the concept of affective polarization has yet to be explicated. In particular, affective polarization researchers do not have an agreed-upon definition and operationalization of affect. This has led to conceptual slippage; for instance, people use terms like negative partisanship, pernicious polarization, affective polarization, and social polarization interchangeably. Clear conceptualization is necessary to understand and solve this societal problem. Understanding if and how polarization can be affective allows us to clearly identify and target its microfoundations.

In this paper, we introduce a theoretical model suggesting that affective polarization is captured better by turning to the theory of core affect. Affect is conceptualized using valence (ranging from positive to negative) and arousal (ranging from low intensity to high intensity). Core affect can be captured using self-reports, but also using psychophysiological measures. We will present results from a large laboratory experiment conducted in the United States (N=200). Participants were recruited from a diverse community sample – in terms of partisanship as well as race and other socio-economic background variables - in Philadelphia. The study was conducted in 2018 but only analyzed now due to the pandemic and other life events. In the study, participants were in the pre-test asked about their partisanship and the strength of their partisanship. In the second stage of the study, we connected people to biosensors that capture both arousal and valence. Arousal was measured by placing electrodes on two finger tips that capture changes in skin conductance levels, while negative valence was captured with activity of the corrugator major – a muscle above the eyebrow. In the image-viewing task, participants were shown a series of prominent politicians from the Republican and Democratic party. After seeing each image, they were asked to rate the self-reported valence and arousal towards these politicians, while physiological responses were measured during exposure to the image. We are currently analyzing the data. In line with extensive research on self-reports, we find that people report more negative valence towards out-party politicians compared to in-party politicians. At APSA, we will present the full results and explore if and to what extent the politicians also evoke physiological responses. Moreover, based upon recent developments in neuroscience, we will explore if and to what extent physiological and self-reported affect are aligned with each other. This project improves the understanding of the construct of study, which can be helpful when theorizing about the causes and consequences of affective polarization and when designing interventions.

Authors

• Kevin Arceneaux, Sciences Po

• Bert Bakker, University of Amsterdam