University of California, Los Angeles: Zooming across the political divide

Social psychologists at UCLA have done what seems impossible, at least on the internet: getting liberals and conservatives to have meaningful and congenial political discussions.


The trick? They held these conversations over Zoom, the video conferencing tool that the pandemic has made a household word.


The researchers discovered that most people, when asked to converse face-to-face without the anonymity and influence-chasing offered by social media and other online forums, instinctively connected with each other and found their interaction more pleasant than expected. Participants left with a greater appreciation for others’ views and felt less rigid in their own.


The results, published today in the journal PLOS ONE, suggest that Zoom conversations could help mitigate political polarization, with a few caveats. Whether or not people had an audience, for instance, had a significant influence on the amount of conflict involved in the exchange.


“Most studies about cross-ideological communication are either written retrospectively about past experiences or speculatively, but almost no one has looked at what happens when people actually have the conversation,” said UCLA psychology professor and study author Matthew Lieberman, who noted the difficulty of setting up political conversations under most other experimental conditions.


“To our knowledge, this is the first time that researchers have used Zoom to have these conversations,” he said. “We’re using it as the experimental platform, and we experimentally manipulated whether or not people had an audience on the platform.”


Study authors and UCLA psychology doctoral students Ashley Binnquist and Stephanie Dolbier recruited participants with strong liberal or conservative political opinions from across the United States. They asked recruits what they expected a conversation with their political opposite would be like and how they would feel afterwards.


Most participants, having witnessed the personal attacks and pile-on behavior common on Twitter, Facebook and other platforms, unsurprisingly dreaded these conversations, expected them to be filled with conflict and outrage, and imagined they would feel terrible when the conversation ended. Many who passed the initial screening didn’t even show up for the experiment, the researchers said.


Those who did first had a Zoom conversation with someone who shared their viewpoints. They were next paired with an ideological opposite. In about half of these cross-ideological conversations, the ingroup members the participants had first spoken to — one conservative and one liberal — remained to silently observe the conversation on a hot-button topic.


All the conversations were recorded so that researchers could code the interactions for comparison with participants’ post-conversation assessments. Each conversation addressed one of the following opinions:


In an unintended pregnancy, the father has a right to be involved in deciding about an abortion.
People shouldn’t be forced into the categories of male or female; gender is a spectrum.
Cities should defund the police to combat systemic discrimination.
Colleges should use affirmative action policies when making admissions decisions.
Private businesses should have the right to refuse service when it conflicts with their religious beliefs.

When two people talked privately, discussions began politely and stayed fairly congenial, although several become heated. Some conversations resembled the way people talk about politics at family gatherings like Thanksgiving, where relatives avoid certain topics or pretend to respect or share someone’s point of view to keep the peace, the authors said.


On average, participants reported spending less time in conflict during the conversation and found it more enjoyable, less stressful and less difficult than they predicted. They tended to like their conversation partners more than expected and found them less emotional and more logical in their arguments than anticipated. Even individuals who reported higher levels of conflict rated both the conversation and their partner more positively than they had expected.


However, when participants knew that the ideological ingroup members with whom they had previously spoken were watching, conversations were more conflicted. The presence of these observers appeared to make it harder for participants to open up and find common ground with their ideological opponent. Both participants and researchers rated these conversations more stressful and difficult.


Yet even in these cases, participants still found the conversations more enjoyable and less difficult than they expected and came away with generally positive impressions of their conversation partners — similar to the impressions of participants in private, non-observed conversations.


The research demonstrates that Zoom conversations can be an effective way for people to overcome their biases and begin to establish open dialogue that could help reduce political polarization, the authors said.


Lieberman’s group is currently using near-infrared spectroscopy, in which participants wear a cap with sensors that detect blood oxygen levels, to find out if brain activity synchronizes when participants in cross-ideological Zoom conversations agree, and if it fails to synchronize when they disagree.