Many people enter debates expecting that if they just present the facts clearly, the other side will come around. Usually it does not happen. This is not because people are stupid or facts do not matter. It is because beliefs are tied to much more than evidence.

Motivated reasoning

Psychologists use the term motivated reasoning for the tendency to evaluate evidence in ways that favor conclusions we already want to reach. We scrutinize unwelcome evidence closely and accept welcome evidence easily. A related habit, confirmation bias, leads us to seek out and remember information that fits what we already believe. Everyone does this, including highly educated people; some studies suggest that people with more knowledge can be even better at finding reasons to dismiss inconvenient facts.

Identity

Beliefs often signal membership in a group. If your friends, family, church or community share a view, changing it can feel like betraying them. The legal scholar Dan Kahan and colleagues call this identity-protective cognition: people tend to process evidence in ways that protect their standing in groups they care about. From this angle, rejecting a fact can be rational for the individual, even when it is bad for the community. A useful lesson follows: a message that makes accepting a fact feel like switching teams will usually fail.

Different moral foundations

The psychologist Jonathan Haidt and colleagues developed moral foundations theory, presented for general readers in The Righteous Mind (2012). In paraphrase, it proposes that human moral intuitions rest on several distinct foundations, which people weigh differently. The main ones are usually listed as:

• Care versus harm: protecting the vulnerable.
• Fairness versus cheating: justice and reciprocity.
• Loyalty versus betrayal: standing by one's group.
• Authority versus subversion: respect for legitimate order and tradition.
• Sanctity versus degradation: treating some things as sacred or pure.
• Liberty versus oppression: resisting domination (added later).

Haidt also argues that moral judgments usually come first as intuitions, with reasoning following to justify them, rather than the other way round. Critics question whether the list is complete, whether the foundations are innate, and how well the survey measures work. Still, the core practical lesson is widely accepted: people who disagree may not be ignoring morality; they may be weighing different moral concerns. An argument framed only in terms you care about may not register with someone who weighs other values more.

The backfire effect: weaker than once thought

Around 2010, a well-known study by political scientists Brendan Nyhan and Jason Reifler reported cases where correcting a false belief seemed to make some people hold it more strongly. This backfire effect became famous, and many concluded that corrections are pointless.

Later research changed the picture. Larger studies, including work by Thomas Wood and Ethan Porter testing many corrections on thousands of people, found backfire to be rare. Corrections usually move beliefs at least somewhat toward accuracy, even among people whose side the correction is uncomfortable for. Nyhan himself has written that backfire appears to be uncommon. The more realistic worry is different: corrections often have modest effects that fade, and they may change belief about a specific fact without changing the attitudes or votes built on it.

What follows

None of this means facts are useless. It means facts work best when:

• they come from a source the listener trusts;
• accepting them does not require abandoning one's identity or group;
• they connect to values the listener already holds;
• the conversation feels like shared inquiry rather than a contest.

The next level looks at methods built on these lessons.

Worked example: the roundabout argument at dinner

At a family dinner, Kim's uncle says he opposes the state's plan to replace the old four-way stop on Main Street with a roundabout. Kim launches into statistics showing that roundabouts usually reduce serious crashes. Her uncle digs in. What went wrong?

Identity. Her uncle's friends at the hardware store all oppose the roundabout, and to him it is partly about not letting distant planners remake the town. Accepting Kim's statistics would feel like siding with outsiders against his own circle.

Moral foundations. Kim framed everything around care (fewer injuries). Her uncle is weighing authority and tradition (respect for the town's way of doing things and its own elected officials) and perhaps liberty (resentment at being told how to drive). They were not arguing about the same values.

Backfire, or not? Kim worries that correcting him will just make him believe more strongly. The research suggests that is unlikely. Accurate information usually nudges beliefs somewhat toward accuracy. The more realistic problem is that a single correction will have a small effect and may not change his overall position.

A better approach. Kim might first ask what concerns him most and listen. She might acknowledge his concern honestly: “I get not wanting decisions made for the town by people who don't live here.” She could share evidence from a source he trusts, such as a neighboring town's fire chief. And she could make it clear that accepting a fact about crashes does not require him to agree with her about anything else.

Her uncle may not change his mind tonight. But he is far more likely to keep thinking about it if he did not feel defeated at dinner.