Before defending a view, decide what could count against it. The important test is whether you apply the same standard when the result is inconvenient.
Agreement can look like investigation
You search for an explanation of a disagreement and click the result that sounds closest to your view. Its examples feel persuasive, so you search for more of the same. By the end, you have gathered information. You may also have made your preferred conclusion harder to question without testing it very much.
Confirmation bias is a broad label for ways people can favor information consistent with what they already believe or expect. It is not a useful insult to throw at an opponent. The most productive place to begin is an interpretation you have some reason to want to be true.
Make the claim small enough to test
“Remote work is better” contains many possible claims about different people, tasks, and outcomes. “For this team, two quiet mornings at home would reduce interruptions during report writing” is narrower. You can inspect what happens during those mornings and what tradeoffs appear elsewhere.
Write the outcome you care about before collecting examples. If your criteria change whenever an unfavorable observation appears, the test has become difficult to fail. A useful definition should survive the arrival of evidence you did not expect.
Consider an alternative without pretending it is true
Lord and colleagues’ experiments on considering the opposite examined how an explicit alternative can affect particular social judgments. The research does not establish a universal procedure that removes bias. It does suggest a more concrete activity than telling yourself to “be objective.”
For your claim, write the strongest reasonable alternative explanation. Perhaps the quiet morning helped because the deadline was clearer, not because of the location. Identify one comparison that would help separate those explanations. An alternative is useful because it can be investigated, not because every argument needs an equal opposite.
Audit the standards you use
Take two pieces of evidence: one that supports your view and one that challenges it. What questions do you ask of each source? Do you inspect methods, dates, relevant expertise, and possible limitations in both cases? Or does one side receive a generous reading while the other must meet an impossible standard?
The sources need not be equally good. A careful audit may justify giving one more weight. The point is to make the difference rest on something more defensible than which conclusion you prefer.
_-_DPLA_-_9f3555e673ab102033cbd1eb647f4636_(page_108).jpg){kind=link}
Keep a condition for revision
Finish a short note: “I would become less confident if…” Then name an observation you could realistically recognize. Avoid a condition so extreme that nothing could satisfy it. Also record what would strengthen the view. Revision is not a ritual of doubt; it is a relationship between a claim and the evidence that bears on it.
Try the exercise with an ordinary opinion before a highly charged one. It is easier to learn the movement when your identity is not immediately at stake. The useful habit is not perpetual uncertainty. It is knowing why your confidence deserves the level you have given it.
Sources and context
Considering the opposite: a corrective strategy for social judgment (Lord et al., 1984)
Published by CreativePulse. Research, writing and production assisted by AI. The practical exercises are editorial suggestions; research limits are stated in the article. Editorial standards.
Continue reading
How to Think for Yourself Without Ignoring Everyone Else
Ten Things Worth Thinking Through Yourself Before Asking AI
Try a short Thinking Gym exercise. Your response stays in your browser.