With his frequent characterizations of his opponents as “lyin” or “crooked”, Donald Trump’s use of language during his 2016 presidential election campaign was a departure from previous contests. In new research, Stephen M. Utych examines the effects of this sort of emotional, negative language on political decision-making. Through experimental studies, he finds that when such language is used in discussing a topic people rate that topic more negatively. He writes that this negativity can also often spill over to those who are using the negative language.
Whether through his formal speeches, off-the-cuff remarks, or his active Twitter account, observers of American politics have noticed that Donald Trump speaks differently compared to “typical” politicians. Trump frequently characterized his opponents throughout the 2016 election in unflattering terms, ranging from “lyin’” Ted Cruz, to “loser” Jeb Bush, to “crooked” Hillary Clinton. These nicknames go beyond typical insults, as they are laden with words such as lie, lose, or crook that most individuals have strong negative emotional reactions towards. Such words, which I call negative affective language, may serve to bias political decision making among citizens by causing them to feel more negatively than words that are simply negative, without the affective connotations.
In new research, I draw upon multiple theories from social psychology to determine how negative affective language may influence decision-making in politics. Joseph Forgas and colleagues developed a model called the Affect Infusion Model (AIM), to explain how affective or emotional considerations can influence generalized decision making. This model, along with work by Daniel Kahneman and others on dual process information processing strategies, helps to explain the conditions under which affective language can influence political decisions.
There are two key assumptions made by the AIM: first, that generalized mood, or affect, will influence decision making differently depending upon the way individuals process information. Secondly, the AIM assumes that individuals will adopt the information processing strategy that requires the least amount of effort. When information is processed in a simple, superficial way, individuals will use their mood as an additional piece of information, but this will not influence how new information is stored in memory. An example of a simple information processing task is when individuals are presented evidence strongly in favor of one side of an issue, guiding them on what to think about the issue.
However, information processing is not always such a simple endeavor. People are often faced with difficult decisions in politics, such as accusations thrown around between two political candidates, where the “correct” conclusion is difficult, or impossible, to arrive at. In this case, individuals must adopt a more complex information processing strategy. Here, the AIM argues that affect or mood will still influence decisions, but will do so in a more complex manner than it does in easier decisions. When a decision is difficult, individuals often have to search their memories to bring in relevant information and store the new information in their memory. Since individuals struggle to differentiate between facts and mood driven information, when they receive information with negative affective language, they will be likely to store this information in their memory with a negative connotation.
To apply the AIM to political events, I conducted two experimental studies. In each study, I used words coded by the Affective Norms for English Words (ANEW) database as words that most individuals have strong, negative emotional reactions towards. These are words like cancer, death, lie, and criminal that individuals are predisposed to view negatively. I randomly assigned participants in the studies to one of two groups: a group who received a mock news article that was negative, but free of negative affective language, and a group who received similarly negative information that also contained negative affective words. These studies presented articles on different topics, in order to vary the difficulty of the decision making task for participants.
The first study presented participants with a simple decision task. The article was written about a voter identification law, and was generally unsupportive of passing the law. The article featured a report on the law, along with quotes from a fictional politician who opposed the law. This study provided support to the notion that affect can influence decisions in a cognitively simple task. Individuals exposed to negative affective language were significantly less supportive of the law than those who were presented with neutral affective language. As expected, negative affective language biased evaluations negatively overall, as the politician in the article was rated more negatively in the negative affective language group, even though these participants agreed with his position on the law more than those in the neutral language group.
These effects were not all encompassing, however. In line with predictions from the AIM, individuals did not seem to store these negative feelings in their memory. When asked to produce some thoughts about the law, those who were exposed to negative affective language were equally likely to mention a negative opinion of the law as those in the neutral language group. This suggests that, while negative affective language influenced judgments in the short term, there were no long term effects in this simple scenario.
The second study presented a more complex decision task for participants. I presented participants with a mock news article on political mudslinging from two opposing political candidates. One candidate made accusations of corruption against another, who then denied the allegations. Here, it was unclear which politician was telling the truth, making the decision of who to believe difficult for readers.
In this study, affect did not play a role in quick decision making. Participants in the negative affective language group rated each candidate roughly the same as those in the neutral language condition, despite the fact that those who were exposed to negative affective language reported feeling higher levels of negative affect. Here, negative affective language biased decisions in a different way.
As in the first study, I asked participants an open-ended question to list their thoughts on the candidates. Those exposed to negative affective language were more likely to write more in response to these questions, suggesting a deeper information processing style. They were also more likely to mention negative reactions to either candidate.
Taken together, these results suggest that negative affective language may have powerful impacts on political decision making. When exposed to these affective words, individuals tend to have more negative reactions towards all information they received, and the persistence of the effect is dependent upon the difficulty of the decision. This suggests that politicians may face a catch-22 when faced with using negative affective language. While this language will cause citizens to feel more negatively about the person or policy they are criticizing, they are also likely to make them feel more negatively about the person using the language as well.
This article is based on the paper, ‘Negative Affective Language in Politics’, in American Politics Research.
Note: This article gives the views of the author, and not the position of USAPP – American Politics and Policy, nor the London School of Economics.
Shortened URL for this post: http://bit.ly/2nsYqKp
Stephen M. Utych – Boise State University
Stephen M. Utych is an Assistant Professor of Political Science at Boise State University. His research focuses on political psychology, specifically the role of language and emotions in politics. He additionally is interested in how contextual factors about politics and elections influence political attitudes. He earned his Ph.D. from Vanderbilt University, and his B.S. from the Georgia Institute of Technology.