Erik Bucy's research indicates voters notice a candidate's nonverbal cues as much as a rhetoric.
As the old saying goes, “you never get a second chance to make a first impression.”
When it comes to presidential candidates in nationally televised debates, though, a series of studies by a Texas Tech University professor in the College of Media and Communication are showing the nonverbal repertoires that make up a presidential candidate's communication style are important influencers of voter reaction.
Erik Bucy, a regents professor of strategic communication at Texas Tech, is a popular guest lecturer around the world for his research on nonverbal expressions in political news and presidential debates and how those televised leader displays affect public perceptions of candidates.
Over the past year, Bucy has presented the results to several national associations, spoke at a symposium on nonverbal communication and democracy in Sweden, guest lectured at UCLA and participated in an invited conference sponsored by the C-SPAN Education Foundation and Purdue University.
Some of this work, conducted in collaboration with researchers from the University of Wisconsin (UW), is summarized in a paper entitled “The Power of Television Images in a Social Media Age: Linking Biobehavioral and Computational Approaches via the Second Screen,” soon to be published in The ANNALS of the American Academy of Political and Social Science.
What Bucy and his colleagues discovered is candidates' facial expressions, gestures and voice tone do as much do as much or more to drive public reaction to the debates than what the candidates actually say.
“This frustrates some people who study media and politics because they want the discussion to be all about the issues,” Bucy said. “What we're documenting is, in fact, people respond a lot to behavior. Not everybody pays really that close of attention to elections or knows all their party's positions on the issues, but they can get a sense of the candidates' traits by observing competitive political behavior. And traits are reliable predictors of candidate support.”
Cueing in on the Visual
Bucy focuses on the first televised presidential debates, including the most recent, as the basis for his research – Barack Obama vs. Mitt Romney in 2012 and John F. Kennedy vs. Richard Nixon in 1960.
The better documented of the two debates in terms of nonverbal expressions and the electorate's reaction came in 2012 due to advances in research and the explosion of social media, which has allowed Bucy and his team, along with graduate student assistants Harrison Gong and Riley Davis, to evaluate the impact of candidate communication on a pass scale, and in real time.
Bucy said his research initially focused on visual analysis of candidate behavior to assess whether coding criteria for nonverbal communication developed in his co-authored book, Image Bite Politics, could be gainfully applied to presidential debates. The project started with a detailed content analysis of all four Kennedy-Nixon debates, the moved on to focus groups using footage from 1960 and 2012, to an experiment utilizing eye tracking to discern what facial displays, movements and expressions drew the most attention, or visual fixations.
Using the communications theory of nonverbal expectancy violations, Bucy tracked how sensitive viewers were to what was happening in the debates if there was just a hint of the unexpected in political performance. For example, in the first 2012 debate, voters honed in on President Obama glancing down for lengthy periods, which gave the impression he was either disengaged or dismissive of Romney's viewpoints or arguments.
“When a violation occurs, there is increased visual attention to it as people attempt to figure out what is going on,” Bucy said. “As a consequence of all this scrutiny, usually there is a negative evaluation of the person committing the violation.”
Other notable nonverbal cues include raised eyebrows (indicating surprise), an angry stare, tilting or moving the head side to side (suggesting evasion) and the difference between reassurance and threat as signaled by the amount of teeth showing (happiness often involves the display of upper teeth, accompanies with a smile, while anger or threat displays usually just reveal the lower teeth with a clenched jaw).
Most of these expressions fit into three distinct display types – anger/threat, happiness/reassurance and fear/evasion. The public responds to leaders who exhibit more happiness/reassurance while challengers, as rivals to power, will typically display more anger/threat, Bucy said. But neither are expected to express fear/evasion and are not looked upon favorably if they do.
While Obama was looking down and evidently avoiding eye contact with Romney (a type of evasion) in the first debate, he was much more engaged in the third debate and performed a kind of nonverbal auditing of Romney's statements. In the process, he was able to show more happiness/reassurance. By doing so, the president put Romney on the defensive, Bucy said. In the end, Romney came off with a negative perception.
“To me, this is where the real payoff is with some of this nonverbal coding,” Bucy said. “You can analyze at the words all you want to and write all kinds of stuff, but this is what people pick up on, at least in televised debates. That's what we learned from the eye tracking and focus group analysis.”
Social Media Reaction
As luck would have it, Bucy's research took another step when his friend, UW journalism and mass communication professor Dhavan V. Shah, was invited to give a guest lecture at Texas Tech last spring.
While visiting, both discovered their current research could be mutually beneficial, because Shah was working on Twitter reaction to presidential debates.
“It became one of those Reese's Peanut Butter Cups moments,” Bucy said. “He looked at me like, 'could your visual analysis go with my tweets?' I was like, “could your tweets go with my visual analysis?' We followed up quickly and did some really interesting work that kind of completed the story.”
What Bucy and Shah's combined research showed was, in the first debate, Obama became much more evasive and less reassuring, and Romney became angrier and less neutral over time. In terms of public reaction on social media, Obama started out with a higher number of mentions than Romney in the first debate, but Romney's mentions grew as the debate continued. Obama also began the debate with more positive sentiment than Romney, but by the time it was over, Romney had closed the bag considerably.
According to the study, Romney's anger/threat expressions and attacks on Obama were linked to a greater volume of name mentions on Twitter. In terms of sentiment, anger/threat displays by Romney helped sentiment for Obama grow while sentiment for Romney fell when Obama displayed reassurance. When either candidate attacked or contrasted their record against the other, sentiment improved for the target of the attack.
“As a candidate, it's important to have an appropriate repertoire of nonverbal displays to consistently deliver your message,” Bucy said. “At least in these studies, people appear to respond to facial displays and gestures more than voice tone and rhetorical strategies. Viewers process political communication holistically, and if you look only at the words spoken, you get an incomplete picture of what's going on.”
The First Televised Debate
Much the same could be said about the first televised debate in 1960 between Kennedy and Nixon, though Nixon did as much to hurt his performance going into the debate as he did during his encounter with Kennedy.
Nixon had been in the hospital just days before and insisted on campaigning right up until time for the debate. He'd also banged his knee on the campaign trail and it had become infected, then reportedly banged it again getting out of the car at the debate site. The pain prevented him from crossing his legs while seated. Further compromising his appearance, he also refused to put on any makeup for the camera, using an aftershave-type lotion that showed the scruff in his face.
From visual analysis, Bucy documented Nixon exhibited a much higher rate of blinking than Kennedy and showed a lack of awareness when Kennedy was speaking that the camera might be on him. Nixon displayed more happiness/reassurance overall, but also more aggression and, through blinking and visible perspiration, stress. Though he did much better in the other three debates against Kennedy, Nixon attributed losing the election to his performance in that first debate.
Effect on Voting
Bucy said people don't always want to admit it, but voters are influenced by what they observe in debates and televised coverage of politics. Even if some will acknowledge the candidate's nonverbal communication did have an impact on subsequent evaluations, most will deny any direct influence of visual information.
“It's a fundamental irony of American politics,” Bucy said. “We exalt the word and the issues while ignoring or denigrating the visuals. Yet we're often more influenced by the verbal factors.
“For the most part, if you're dyed in the wool partisan, you know who you are going to vote for. If you're undecided or persuadable, either through weak partisanship or some kind of beef with the party on an issue, then a lot of voters will get mad at the party and be open to defecting when the party doesn't represent that one issue they care about. Nonverbal influence, particularly in those moments, can give someone a feel for a candidate they might be opposed to but looks on paper like they are for what they stand for.”
Bucy said political candidates and their advisers would be smart to realize the importance of communication style to the persuadable segment of the electorate. But there are always traditionalists who will insist that it's the message, not the delivery, that matters most.
“That's what the major, agenda-setting print media still focuses on,” Bucy said. “Print journalism, even in the digital era, is not generally set up to emphasize or replay or obsess about the visual. But for citizens who are paying attention, the traits communicated by nonverbal behavior are consequential.”