Research Shows Active Listening Fails to Boost Persuasion

Listening is a good thing, right?

For one, it’s polite — a sign of respect and genuine interest in what someone is saying.

But social scientists and political operatives have also long advanced listening as a strategy. High-quality, non-judgmental listening, the theory goes, can promote persuasion by increasing cognitive processing, reducing defensiveness, and enhancing someone’s affinity toward the person seeking to persuade them.

A new study, published today in Proceedings of the National Academy of Sciences (PNAS) and co-authored by Institution for Social and Policy Studies faculty fellow Josh Kalla, casts doubt on this theory. In a field experiment, the researchers asked trained professional political canvassers to engage in Zoom conversations with study participants about unauthorized immigration to the United States.

In some randomly assigned conversations, the canvassers shared a persuasive narrative while engaging in the types of active listening behaviors touted by earlier studies: nodding along, paraphrasing points the speaker makes, and asking pertinent questions.

While the active listening increased processing and improved perceptions of the persuaders, it did not increase persuasion either immediately after the conversation or five weeks later. The surprising results follow a study Kalla and David Broockman published last year, showing how viewers of Fox News changed their opinions after watching CNN even though they intensely disliked and distrusted the network.

“It’s possible our intuition and prior work on persuasion has overstated the importance of liking the source,” Kalla said.

Kalla and Broockman, an associate professor of political science at the University of California, Berkeley, collaborated on the paper with Erik Santoro, a postdoctoral fellow at Columbia Business School, and Roni Porat, a senior lecturer at The Hebrew University of Jerusalem.

Kalla said the results might indicate an important difference between making a quick decision — when liking the source of information might hold more weight — and when you have more time to listen more closely to the argument, paying less attention to who is providing the information.

“I think the conversations in our experiment are more in this second category,” he said. “These are longer conversations, more personal, about a highly politically sensitive topic. People are slowing down and thinking more, engaging more with the information and the argument and less with the source.”

In 2016, Kalla and Broockman demonstrated that in-person conversations can durably reduce prejudice. The new PNAS paper fits into a program of research investigating what mechanisms might allow these conversations to reduce prejudice, suggesting that sharing non-judgmental persuasive narratives — not the listening behaviors — explain the well-replicated success of door-to-door conversations.

Kalla said he and his colleagues would like to do a version of the study in which people not only ask questions indicative of high-quality listening skills but then tailor follow-up questions based on how the other person responds.

“There is more to effective listening that just listening,” Kalla said. “So maybe this study isn’t necessarily a nail in the coffin of the persuasive power of listening.”