Skip to main content
drawing of an outline of a head with different facing arrows inside

The Power of Contradiction

Despite the conventional wisdom, a consistent message is not always the most persuasive message—a reassuring insight in the time of COVID-19.

Don’t wear a mask; wear a mask. Young people are immune to the virus; the virus can harm young people. Hydroxychloroquine can help manage symptoms; no, it can’t.

The spread of COVID-19 has pushed science to work in real-time. Downstream from this research, public health authorities have been grappling with an uncommon challenge: how should they communicate information as it evolves? Health departments don’t, after all, want to confuse the public by saying one thing one week and another the next.


An article coauthored by Yale SOM’s Taly Reich and Stanford’s Zakary Tormala questions this conventional wisdom. Through a series of five experiments, their research reveals “a contradiction effect” whereby, under the proper conditions, reversing one’s own opinion proves more persuasive than a consistent stance.

“Our studies suggest that contradicting oneself can enhance one’s persuasiveness relative to offering just a single message with no contradiction or even offering multiple messages that are consistent across time,” they write in the Journal of Experimental Social Psychology.

In one of the experiments, participants were asked to imagine that they had a friend interested in education policy. After reading a description of a particular university policy, they were told that their friend either agreed or disagreed with the policy. Next, they read that the same friend raised the issue one week later and approved of the policy; he also gave arguments in its favor. When participants finished reading the scenario, they reported their attitudes toward the policy.

Reich and Tormala found that the friend who contradicted himself—first opposed to the policy, later favoring it—was more persuasive than a friend who argued consistently in the policy’s favor (This was only true in cases when the friend used strong arguments) In this strong argument condition; in write the authors, “the conflicting message tended to outperform the consistent message.”

These results raise two important questions: first, what explains this effect? Second, what are the specific conditions under which contradiction is persuasive?

Reich and Tormala argue that the well-established theory of attributional thinking answers the first question. Unexpected events often trigger people to search for an explanation: why did the underdog win? Why is this stranger so willing to help me? The relationship between contradiction and persuasion may hinge on this same concept. If someone changes their mind about education policy, friends may attribute the change to careful consideration and an informed revision of opinion. Whether or not this is accurate, the friend’s opinion suddenly carries more authority.

On the second question—when is contradiction persuasive—Reich and Tormala ran four experiments (including one in the field) to tease out some of the boundaries. They found three important conditions: first, as described above, the strength of an argument is essential. With weak arguments, contradiction is not persuasive. Second, the effect holds only if a source is trustworthy. Third, the contradiction must come from a single source. Divergent opinions between two different people don’t make for a persuasive case.

The implications of these findings touch many domains, but especially topical is the realm of COVID-19 communications. There is no reason to doubt that public health recommendations will continue to change in step with advancing scientific understanding of the disease –  the news of tomorrow contradicting the news of today. Perhaps this should be embraced. If handled properly—with strong arguments in defense of the change and with information coming from the same, trusted source—the presence  of uncertainty could keep us safe.