A friend sends us a health claim with a confident explanation. A professional recommends a course of action. A widely shared post warns that an ordinary event signals a hidden danger. In each case, we face the same practical problem. We cannot personally verify everything we are told, yet accepting every message would leave us vulnerable to error and manipulation.

Human knowledge grows through other minds. We learn from parents, teachers, neighbors, journalists, researchers, clinicians, and institutions. This dependence makes culture and accumulated knowledge possible. It also exposes us to honest mistakes, limited expertise, persuasion, and deception. Good judgment requires openness and protection to work together.

Epistemic vigilance names the processes involved in evaluating communicated information. Dan Sperber and his colleagues proposed that people assess both the source of a message and its content, drawing on judgments about competence, honesty, plausibility, coherence, and reasons. Their account describes a family of cognitive and social mechanisms rather than a single skill or fixed personality trait (Sperber et al., 2010).

The goal is calibrated trust. We want to remain open enough to learn while becoming more careful about what deserves confidence. A useful way to organize this task is to examine four things: the source, the claim, our own response, and the stakes of being wrong. This source, claim, self, and stakes framework is a Psychology Fanatic synthesis of the research discussed below. It is not an established scale or a substitute for professional judgment.

Learning Through Other Minds

Most people know that the earth orbits the sun, that certain medicines affect the body, and that distant places exist without verifying each claim through direct observation. Testimony carries knowledge beyond individual experience. Trust in communication is therefore not an intellectual weakness. It is one of the conditions that makes complex learning possible.

Research on children’s learning makes this dependence especially visible. Paul Harris and his colleagues reviewed evidence that infants and children seek information from other people and become increasingly attentive to who has been accurate, informed, or well positioned to know. Children do not evaluate only whether a speaker was right or wrong. As their judgment develops, they also consider why an error occurred and what the speaker could reasonably have known (Harris et al., 2018).

The need for testimony does not disappear in adulthood. Specialized knowledge makes dependence more extensive. A thoughtful person cannot reproduce every clinical trial, inspect every bridge, or audit every economic report before acting. Mature judgment often involves deciding when deference is warranted and how much confidence to place in conclusions that rest on other people’s work.

This tension cannot be solved by choosing trust or doubt once and for all. Trust that is too open invites exploitation. Doubt that is too rigid prevents learning and cooperation. Epistemic vigilance helps us decide which messages deserve acceptance, which require further examination, and which should remain unsettled.

How Epistemic Vigilance Evaluates Sources and Claims

Judging a Source in Context

When we assess a source, two questions matter especially: Is this person or institution competent about the subject, and do they have reason to communicate honestly in this situation? Neither question can be answered by a general impression alone. Expertise is usually bounded. A skilled engineer may know little about nutrition, and a caring friend may confidently repeat a rumor.

Reliable expertise also depends on the conditions in which it developed. Daniel Kahneman and Gary Klein concluded that skilled intuition is most defensible when an environment contains sufficiently valid cues and the person has had adequate opportunity to learn them through experience and feedback. Subjective confidence alone is not evidence of accuracy (Kahneman & Klein, 2009).

Sperber and his colleagues emphasize that trust is often specific to a topic and circumstance. We may rely on someone’s knowledge of local history while remaining cautious about their medical advice. Money, status, fear, or divided loyalties can increase the need for independent support, but a possible motive does not by itself refute a claim (Sperber et al., 2010).

Examining the Claim

Vigilance also turns toward content. Some messages contain contradictions or conflict with well-established knowledge. Others become doubtful because the evidence offered does not match the certainty of the conclusion. A dramatic possibility may be described as a common outcome, or one person’s experience may be presented as proof of a general rule.

Existing knowledge provides a starting point for evaluation, not an infallible court of appeal. A new claim that conflicts with what we believe deserves closer examination. Yet our background beliefs can also be incomplete, outdated, or mistaken. Healthy vigilance allows the new claim and the old belief to place pressure on each other.

Following the Reasons

This kind of critical thinking lets us evaluate a message without resting entirely on the speaker’s authority. We can ask what evidence supports the conclusion, which assumptions connect the evidence to it, and whether another explanation fits the same facts. A persuasive message may still rely on a missing premise, a misleading comparison, or an emotional appeal that does not establish the point at issue.

An argument can sound convincing and still be weak. Walton’s work suggests checking the reasoning rather than treating confidence or status as proof. When a claim rests on expertise, ask four practical questions: Is the person qualified in the relevant field? Does the claim fall within that expertise? What evidence supports it? Do other qualified sources broadly agree? Expert opinion can be reasonable evidence without becoming unquestionable evidence (Walton, 2007; Walton et al., 2008).

Epistemic Vigilance Between Trust and Suspicion

Epistemic vigilance is sometimes mistaken for cultivated distrust. That interpretation reverses its purpose. The proposed mechanisms help communication remain useful despite the risk of error or manipulation. They protect trust by making it more selective, not by making it disappear.

Credulity accepts too readily. Cynicism rejects too readily. Calibrated vigilance is conditional. It asks how a source knows, what the source has been reliable about, whether the claim is independently supported, and what an error would cost. It can produce trust, doubt, or a temporary suspension of judgment.

The term epistemic trust has a more specific meaning in research associated with Peter Fonagy. It describes openness to knowledge from other people when that knowledge feels relevant and usable beyond the immediate exchange. Epistemic trust and epistemic vigilance overlap, but they are not simple opposites. One concerns openness to learning; the other concerns how communicated information is evaluated (Fonagy & Allison, 2023).

How Bias Shapes Our Judgment

Evaluating a message may feel like an objective inspection of facts. In practice, the evaluator brings hopes, loyalties, fears, identities, and prior commitments to the task. Those motives do not make careful reasoning impossible, but they can shape which evidence receives attention and how demanding our standards become.

Accuracy and Directional Goals

Ziva Kunda’s account of motivated reasoning distinguishes reasoning guided by an accuracy goal from reasoning guided toward a desired conclusion. When accuracy matters, people may invest more effort, attend more carefully to relevant information, and use more complex strategies. When a particular conclusion is desired, they may search memory and evidence for material that makes that conclusion easier to justify (Kunda, 1990).

Directional reasoning is not necessarily deliberate dishonesty. A person may sincerely experience the preferred conclusion as the most reasonable one. The bias can appear in the search process and the standards of evaluation. Welcome evidence passes quickly, while unwelcome evidence faces a longer interrogation. Knowledge and reasoning ability do not by themselves guarantee evenhanded evaluation.

Confirmation Bias and Asymmetric Standards

Confirmation bias includes seeking or interpreting evidence in ways that favor an existing belief, expectation, or working hypothesis. Raymond Nickerson’s review emphasizes that this selectivity can occur without conscious intent and can arise with or without a strong desire to reach a particular conclusion (Nickerson, 1998).

Epistemic vigilance can therefore be deployed selectively. We may scrutinize an opposing claim for every weakness while accepting a congenial claim on the strength of a headline. The relevant question is not only whether we can identify flaws in another person’s reasoning. It is whether comparable evidence receives comparable scrutiny when it supports our own position.

Intellectual Humility and Belief Revision

Intellectual humility offers a useful corrective. Mark Leary and his colleagues define it as recognizing that a particular personal belief may be fallible and attending appropriately to limitations in the evidence supporting it. Across four studies, people higher in intellectual humility showed greater openness in several domains and, in one study, greater sensitivity to the strength of persuasive arguments (Leary et al., 2017).

Humility does not require vague self-doubt or automatic deference. It asks us to hold confidence in proportion to the available support. A person can have good reasons for a firm conclusion and still acknowledge what evidence would count against it. This stance makes revision possible without treating every belief as equally uncertain.

Epistemic Vigilance in Digital Media

Digital media did not create the problem of deciding whom to believe. It changed the speed, scale, and social setting in which that decision occurs. Claims arrive beside popularity signals, familiar slogans, professional design, and recommendations selected by systems whose operations are often invisible to the reader.

Familiarity Is Not Evidence

Repetition can make a statement easier to process, and that feeling of fluency can be misread as evidence of truth. Nadia Brashier and Elizabeth Marsh review research on this illusory truth effect, including a meta-analysis in which repeated claims were judged truer than new claims. Knowledge can help, but it does not always prevent familiarity from influencing judgment (Brashier & Marsh, 2020).

This does not mean that repeated claims are false. It means repetition and truth are different kinds of information. A familiar claim still requires attention to its source, evidence, and fit with better-established knowledge.

Why Corrections Do Not Erase Misinformation

False information can continue to shape later reasoning after it has been corrected. Ullrich Ecker and his colleagues review explanations for this continued influence, including failures to integrate a correction into an existing mental model and later retrieval of familiar misinformation without its corrective context (Ecker et al., 2022).

Corrections are often more useful when they explain why the original claim was wrong and provide a credible alternative account. The same review finds support for both reactive debunking and pre-emptive prebunking, while emphasizing that effectiveness depends on the message, context, and outcome being measured. Correction helps, but it does not guarantee that the earlier information leaves no trace.

When Individual Checking Is Not Enough

Misinformation is not solely a failure inside an individual mind. Information systems, including algorithmic feeds and echo chambers, influence which claims appear, how often they appear, and which social identities surround them. Lewandowsky, Ecker, and Cook argue that responses must therefore consider technological, political, and social conditions as well as individual cognition (Lewandowsky et al., 2017).

No reader can personally verify every specialized claim. Responsible reliance often depends on social practices such as transparent sourcing, independent replication, disclosed uncertainty, correction procedures, and meaningful expert disagreement. Institutions remain fallible, but the presence or absence of these practices gives us more to evaluate than confidence, reputation, or institutional branding alone.

Practicing Calibrated Vigilance

No short checklist can guarantee freedom from error. Still, directing attention toward accuracy can matter. Gordon Pennycook and David Rand combined 20 experiments conducted by their group, involving 26,863 participants. Accuracy prompts increased sharing discernment, mainly by reducing intentions to share false headlines. Because this was an internal meta-analysis of one research program and measured sharing intentions in particular settings, it should support a modest conclusion: a timely reminder about accuracy can improve some sharing judgments (Pennycook & Rand, 2022).

The questions below make the grounds for confidence more explicit. Together they apply the article’s source, claim, self, and stakes framework. They are practical prompts, not a validated epistemic vigilance test.

Clarify What Is Being Claimed

Separate observation from interpretation. A report that two events occurred together differs from a claim that one caused the other. A prediction differs from a description, and a recommendation includes values and tradeoffs beyond the facts used to support it. Clear claims are easier to test and harder to defend by quietly changing their meaning.

Match Expertise to the Question

Ask whether the source has relevant knowledge for the claim at hand. Credentials, reputation, and experience can matter, but their relevance is specific. Also ask whether the field provides valid cues, meaningful feedback, and opportunities for learning. Confidence is less informative where outcomes are highly irregular or feedback is delayed and ambiguous (Kahneman & Klein, 2009).

Consider Motives and Constraints

Consider what the communicator knows, what access they had to the evidence, what they might gain or fear, and what circumstances could explain an error. These questions should change the amount of corroboration we seek. They should not serve as an automatic verdict on truthfulness.

Use Lateral Reading to Check a Source

Look for the connection between evidence and conclusion. Are important assumptions left unstated? Could another explanation fit the same observations? Do several reports offer genuinely independent confirmation, or do they all trace back to one original source?

Online evaluation often requires leaving the page under review. In a think-aloud study of 10 historians, 10 professional fact-checkers, and 25 university students, Sam Wineburg and Sarah McGrew found that fact-checkers commonly investigated unfamiliar sites by opening other sources and comparing what they found. The authors called this lateral reading. The study was small and descriptive, but it identifies a concrete expert practice: investigate who is behind a source and what independent sources say before being persuaded by the source’s own presentation (Wineburg & McGrew, 2019).

Audit Your Own Standards

Ask whether the same evidence would feel persuasive if it supported the opposite conclusion. Charles Lord, Mark Lepper, and Elizabeth Preston tested this idea in studies of biased assimilation and hypothesis testing. Participants instructed to consider opposing possibilities showed less judgmental bias than participants who were merely told to be unbiased (Lord et al., 1984).

Also ask what evidence would change your mind. If no possible evidence counts, the belief has been protected from evaluation. This question should be applied with care: refusing a weak objection is not closed-mindedness, and changing a well-supported conclusion requires evidence strong enough to matter.

Match Scrutiny and Confidence to the Stakes

Not every statement deserves a personal investigation. The reasonable level of scrutiny depends on the cost of being wrong, the reversibility of the decision, the urgency of action, and access to qualified help. A casual historical detail and a decision about medical treatment do not require the same effort.

Some questions do not have one obvious method or a fully settled answer. Evidence may conflict, experts may disagree, or the standards for deciding may themselves be disputed. Karen Kitchener used the term epistemic cognition for reflection on what can be known, how certain a conclusion can be, and which standards fit the problem. In these situations, good judgment may mean stating a provisional level of confidence, waiting for better evidence, or seeking qualified guidance rather than forcing a yes-or-no answer (Kitchener, 1983).

A Few Words by Psychology Fanatic

It is comforting to imagine that falsehood belongs to gullible people while good judgment belongs to those who are harder to fool. Human reasoning is less tidy. We all depend on trust, and we all bring motives into the evaluation of what we hear. The practical question is whether we can become more honest about the conditions under which confidence is earned.

The source, claim, self, and stakes framework returns responsibility to the whole judgment. Sometimes responsibility means leaving an attractive page to investigate its source. Sometimes it means listening to an expert whose conclusion unsettles us. Sometimes it means waiting because the evidence does not support certainty. These responses share a willingness to let reasons influence what we believe.

We will still be wrong at times. Vigilance offers no immunity from error. Its value lies in helping us notice when trust needs support, when doubt has become indiscriminate, and when confidence should change. The aim is a mind that can be persuaded, but not cheaply, and corrected without losing its capacity to trust.

Associated Concepts

• Epistemic trust: Openness to socially communicated knowledge. One influential developmental-clinical tradition emphasizes whether information is experienced as relevant and generalizable.
• Motivated reasoning: The influence of goals and preferred conclusions on how people search for and evaluate information.
• Epistemic cognition: Reflection on what can be known, how certain a conclusion can be, and which standards of evidence are appropriate for the question.
• Critical thinking: The deliberate evaluation of claims, evidence, assumptions, and alternative explanations before reaching or revising a conclusion.
• Intellectual humility: Recognition that a belief may be mistaken or inadequately supported, paired with a willingness to consider stronger reasons.
• Confirmation bias: Selective searching for or interpretation of information that supports an existing belief, expectation, or hypothesis.
• Misinformation: Inaccurate or misleading information, whether the error is accidental or deliberately produced.

References

Brashier, Nadia M.; Marsh, Elizabeth J. (2020). Judging Truth. Annual Review of Psychology, 71, 499–515. DOI: 10.1146/annurev-psych-010419-050807.
(Return to Main Text)

Ecker, Ullrich K. H.; Lewandowsky, Stephan; Cook, John; Schmid, Philipp; Fazio, Lisa K.; Brashier, Nadia; Kendeou, Panayiota; Vraga, Emily K.; Amazeen, Michelle A. (2022). The Psychological Drivers of Misinformation Belief and Its Resistance to Correction. Nature Reviews Psychology, 1, 13–29. DOI: 10.1038/s44159-021-00006-y.
(Return to Main Text)

Fonagy, Peter; Allison, Elizabeth (2023). Beyond Mentalizing: Epistemic Trust and the Transmission of Culture. The Psychoanalytic Quarterly, 92(4), 599–640. DOI: 10.1080/00332828.2023.2290023.
(Return to Main Text)

Harris, Paul L.; Koenig, Melissa A.; Corriveau, Kathleen H.; Jaswal, Vikram K. (2018). Cognitive Foundations of Learning from Testimony. Annual Review of Psychology, 69, 251–273. DOI: 10.1146/annurev-psych-122216-011710.
(Return to Main Text)

Kahneman, Daniel; Klein, Gary (2009). Conditions for Intuitive Expertise: A Failure to Disagree. American Psychologist, 64(6), 515–526. DOI: 10.1037/a0016755.
(Return to Main Text)

Kitchener, Karen Strohm (1983). Cognition, Metacognition, and Epistemic Cognition: A Three-Level Model of Cognitive Processing. Human Development, 26(4), 222–232. DOI: 10.1159/000272885.
(Return to Main Text)

Kunda, Ziva (1990). The Case for Motivated Reasoning. Psychological Bulletin, 108(3), 480–498. DOI: 10.1037/0033-2909.108.3.480.
(Return to Main Text)

Leary, Mark R.; Diebels, Kate J.; Davisson, Erin K.; Jongman-Sereno, Katrina P.; Isherwood, Jennifer C.; Raimi, Kaitlin T.; Deffler, Samantha A.; Hoyle, Rick H. (2017). Cognitive and Interpersonal Features of Intellectual Humility. Personality and Social Psychology Bulletin, 43(6), 793–813. DOI: 10.1177/0146167217697695.
(Return to Main Text)

Lewandowsky, Stephan; Ecker, Ullrich K. H.; Cook, John (2017). Beyond Misinformation: Understanding and Coping with the Post-Truth Era. Journal of Applied Research in Memory and Cognition, 6(4), 353–369. DOI: 10.1016/j.jarmac.2017.07.008.
(Return to Main Text)

Lord, Charles G.; Lepper, Mark R.; Preston, Elizabeth (1984). Considering the Opposite: A Corrective Strategy for Social Judgment. Journal of Personality and Social Psychology, 47(6), 1231–1243. DOI: 10.1037/0022-3514.47.6.1231.
(Return to Main Text)

Nickerson, Raymond S. (1998). Confirmation Bias: A Ubiquitous Phenomenon in Many Guises. Review of General Psychology, 2(2), 175–220. DOI: 10.1037/1089-2680.2.2.175.
(Return to Main Text)

Pennycook, Gordon; Rand, David G. (2022). Accuracy Prompts Are a Replicable and Generalizable Approach for Reducing the Spread of Misinformation. Nature Communications, 13, Article 2333. DOI: 10.1038/s41467-022-30073-5.
(Return to Main Text)

Sperber, Dan; Clément, Fabrice; Heintz, Christophe; Mascaro, Olivier; Mercier, Hugo; Origgi, Gloria; Wilson, Deirdre (2010). Epistemic Vigilance. Mind & Language, 25(4), 359–393. DOI: 10.1111/j.1468-0017.2010.01394.x.
(Return to Main Text)

Walton, Douglas N. (2007). Media Argumentation: Dialectic, Persuasion, and Rhetoric. Cambridge University Press. DOI: 10.1017/CBO9780511619311. ISBN: 978-0-521-87690-2; 978-0-521-70030-6.
(Return to Main Text)

Walton, Douglas N.; Reed, Christopher; Macagno, Fabrizio (2008). Argumentation Schemes. Cambridge University Press. DOI: 10.1017/CBO9780511802034. ISBN: 978-0-521-89790-7; 978-0-521-72374-9.
(Return to Main Text)

Wineburg, Sam; McGrew, Sarah (2019). Lateral Reading and the Nature of Expertise: Reading Less and Learning More When Evaluating Digital Information. Teachers College Record, 121(11), 1–40. DOI: 10.1177/016146811912101102.
(Return to Main Text)

Last Updated: October 9, 2026