Should socials get a say during political debates?

The main purpose of a political debate is (arguably) to get an idea of where politicians stand on various issues. In order to do that, though, viewers have to be able to attend to and understand what speakers are saying.

Though recent such debates in the UK have tended to be relatively low-key in terms of set design and on-screen visuals, our American counterparts are fond introducing additional context to debates in the form of layered graphics. Colourful overlays showing worms and tweets have been included in recent years — presumably to broaden viewer's understanding of both the issues being discussed, and wider public opinion.

But, according to recent research from Freddie Jennings and colleagues published in the Journal of Visual Political Communications, debates which include distracting visuals may be undermining their capacity to do so.

For their study, the team asked 116 former graduates and staff from university departments of communication, journalism, and political science to watch clips from the 2015–2016 Democratic and Republican primary debates. Most of these clips featured a live Twitter feed (complete with jibes from politicians of other affiliations) to the right of the screen, as well as trend lines along the bottom of the frame, indicating which candidate was getting more mentions on social media.

A content analysis of these videos provided data on the content of tweets shown, how long it was on screen, and information about who posted it. This was then analysed alongside eye-tracking data from the participants, in order to see which on-screen elements drew their attention the most, and how long for. Following the clips, participants were asked to recall and attribute a number of quotes from what they'd just seen.

Through their explorations, the team found that extraneous visuals can distract from the debate at hand: in fact, participants overall spent more than 10% of their viewing time checking out on-screen tweets. This distraction also led them to retain less information from candidates, with those who most attended to tweets being more likely to struggle recalling and attributing quotes to various speakers.

Tweets could also add a level of scepticism to proceedings. Time spent attending to the social feed was negatively associated with measures of persuasiveness, suggesting that distraction — or perhaps the counterarguments and disputes presented in selected tweets — detracted from the effectiveness of speakers' political rhetoric.

In addition, the team suggested that these graphics may activate partisan identities, perhaps causing viewers to dig in their heels with their particular affiliation. This could leading them to further disregard the persuasive efforts of opponents.

These effects were stronger for Democratic primary debates, and while the authors aren't certain why this is, they suggest that it may be due to Republican candidates more effectively taking advantage of the tweet feed. Ultimately, posts from Republican politicians appeared more during Democratic debates than posts from Democrats did during Republican ones.

As the sample of mainly White, mainly republican, and mainly female, it's tricky to say whether or not these effects would look similar for other demographics. This could be explored in future research efforts, perhaps alongside investigation of how prior familiarity with candidates could influence results.

Regardless of the specifics, however, the authors emphasise the continued need for political debates, citing previous research showing that (at least in their ideal form) they're a useful tool to build political literacy skills, foster perspective taking, and mitigate polarisation. The modern addition of social media analysis, they conclude, flies in the face of these efforts.

So, while the idea of having your incisive take-down of a particular policy broadcast to the nation sounds cathartic, take solace that not having that opportunity is helping to keep everyone focused on the debate itself.

Read the paper in full:

Jennings, F. J., Bouchillon, B., Bramlett, J. C., Eubanks, A. D., Stewart, P. A., & Miller, J. M. (2023). Visual overload: The influence of broadcast social media visuals on televised debate viewing outcomes. Journal of Visual Political Communication. Journal of Visual Political Communication, 10(2), 151–172. https://doi.org/10.1386/jvpc_00029_1