It may seem that video will be the only effective medium for spreading information online. But a new experiment by MIT researchers shows that video clips have a marginally larger effect on political persuasion than just the written word.
“Our conclusion is that watching a video is no more persuasive than reading a text,” says David Rand, an MIT professor and co-author of a new paper detailing the results of the study.
The study comes amid widespread concern about political misinformation online, including the possibility that technology-enabled “deepfake” videos can easily convince many people to believe false claims.
“Technological advances have created new opportunities for people to falsify video footage, but we still know surprisingly little about how individuals process political video versus text,” said MIT researcher on the paper. Chloe Wittenberg says “Before we can identify strategies to combat the spread of deepfakes, we first need to answer these more basic questions about the role of video in political persuasion.”
The paper, “The (Minimal) Persuasion Advantage of Political Video over Text,” is published today in Proceedings of the National Academy of Science, Co-author Adam J. Berinsky is the Mitsui Professor of Political Science; Rand, Irwin H. Scales Professor and Professor of Management Science and Brain and Cognitive Sciences; Ben Tappin, postdoc in the Human Cooperation Lab; and Chloe Wittenberg, doctoral student in the Department of Political Science.
belief and persuasion
The study works on the difference between the credibility of videos and their persuasion. That is, viewers may find a video believable, but may not change their point of view in response. Alternatively, a video may not seem believable to a large portion of the audience, but may still change the attitude or behavior of the audience.
For example, Rand says, “When you see a stain remover ad, they all have the same format, where some stain gets on the shirt, you put the remover on it, and it goes into the washer.” And hey, look, the stain is gone. So, one question is: do you believe this happened, or was it just a hoax? And the second question is, how much stain remover do you want to buy? The answers should not be closely related.”
To conduct the study, MIT researchers performed a pair of survey experiments involving 7,609 Americans, using the Lucid and Dynata platforms. The first study included 48 advertisements obtained through the Peoria Project, a collection of political material. Survey participants either saw an ad, read the ad copy, or didn’t receive any information. (Each participant did this several times.) For each ad, participants were asked whether the message sounded believable and whether they agreed with its main message. They were then shown a series of questions to measure whether they found the topic personally important and whether they wanted more information.
The second study followed the same format, but included 24 popular video clips about COVID-19 taken from YouTube.
Overall, the results showed that video performed slightly better than text on the reliability front, but had less relative advantage when it came to persuasion. Participants were generally more likely to believe that the events actually occurred when they were shown in a video, not as described in a written transcript. However, the advantage of video over text was only a third as large when it came to changing participants’ attitudes and behavior.
As a further indication of this limited motivational advantage of video versus text, the difference between reading the transcript and viewing the video was as great as the difference between reading the transcript (with participants who received no information) and watching the video. in between.
These differences were surprisingly stable across all groups. For example, in the second study, there were only small differences in the effects observed for political versus non-political messages about COVID-19, suggesting that there are findings in different types of content. The researchers did not find significant differences among respondents based on factors such as age, political partisanship and political knowledge.
“Seeing is believable,” says Berinsky, “but our study shows that just because a video is more believable doesn’t mean it can change people’s minds.”
questions about online behavior
The scholars acknowledge that the study did not exactly replicate the situations in which people consume information online, but they suggest that the main findings provide valuable insights about the relative power of video versus text.
“It’s possible that things are a little different in real life,” Rand says. “It’s possible that when you’re scrolling through your newsfeed, the video captures your attention more than the text. You may be more likely to see it. That doesn’t mean that the video is more natural than the text. The look is more persuasive – just that it has the potential to reach a wider audience.”
That said, the MIT team notes that there are some clear directions for future research in this area — including the question of whether people are more inclined to watch videos than to read material.
“Some people may prefer to watch a video to read the text,” Tappin says. “For example, platforms like TikTok are heavily video-based, and the audience is mostly young adults. Among such audiences, a small persuasive advantage of video over text can grow rapidly because video can reach so many people. Future research may explore these and other ideas.”
The study was supported by funding from Ara, a technology research incubator maintained by Google.
