Imagine accepting a new job after checking the responsibilities, speaking with future colleagues, and considering what the move would mean for your family. Months later, an unexpected closure leaves you unemployed. Someone else accepts a job with little investigation and happens to thrive. Looking backward, it becomes tempting to call your decision foolish and theirs perceptive.

The ending gives each story a persuasive shape. Yet it leaves an important question unanswered: how reasonable was the choice when it was made? Understanding outcome bias helps us examine that question without dismissing either the consequences we face or the uncertainty we once faced.

What Is Outcome Bias

We usually care about decisions because of what they might bring. Wanting a favorable result is sensible. This form of biased judgment arises when knowing the result changes our evaluation of reasoning that has already taken place. A success can make an inadequately considered choice appear insightful, while a failure can make a careful choice appear incompetent (Baron & Hershey, 1988).

Decision quality concerns how a choice was reached: the evidence considered, the alternatives available, the foreseeable risks, and the goals at stake. Outcome quality concerns what actually followed. The two are related, but circumstances beyond the decision maker’s control can separate them. Even a strong probability of success leaves room for failure.

In the job example, checking the role and speaking with colleagues may have been reasonable steps. Their value depends on whether they addressed the uncertainties that mattered, including information about the employer’s stability that could realistically have been obtained. A careful-looking process can still overlook something important. Equally, an unexpected closure does not retroactively make every earlier inquiry inadequate.

Outcome Bias Versus Hindsight Bias

Outcome bias and hindsight bias can occur together, but they concern different judgments. Outcome bias changes how good a decision seems. Hindsight bias changes how predictable the result seems after we know what happened. Baron and Hershey explicitly distinguished the effect of outcomes on decision evaluation from their effect on perceived probabilities (Baron & Hershey, 1988).

“The project failed, so approving it was foolish” illustrates outcome bias when the conclusion rests on failure alone. “Everyone could have seen the failure coming” illustrates hindsight bias when the result was less predictable beforehand. A person can also accept that failure was unlikely and still downgrade the decision because it failed. Outcome bias therefore does not require a claim that the ending was obvious.

Research on Outcome Bias

Jonathan Baron and John Hershey examined outcome bias in five experiments involving undergraduate participants. The decisions concerned medical situations or monetary gambles. Participants evaluated such things as the quality of the decision maker’s thinking, competence, or suitability to make decisions on their behalf (Baron & Hershey, 1988).

Baron and Hershey gave participants the information needed to understand a choice as it appeared beforehand. They then examined whether knowing the result would change the evaluation, even when it supplied no relevant new information about the earlier reasoning. Favorable outcomes generally led to more favorable judgments. Some participants were influenced even though they said outcomes should not affect their ratings (Baron & Hershey, 1988).

Aiyer and colleagues revisited the first experiment in a preregistered replication published in 2023. The analyzed online sample included 692 participants. Each participant saw a version of one medical decision scenario, with either a successful or unsuccessful outcome, rather than comparing both outcomes. Successful outcomes again led to higher ratings of decision quality (Aiyer et al., 2023).

This newer study strengthens confidence in the basic phenomenon, while leaving questions about its reach. It used a single medical scenario, and the authors caution against generalizing its findings to other contexts without further research. Neither study tells us that every judgment after a disappointing result is biased, or how much outcome bias shapes any particular person’s life (Aiyer et al., 2023).

Why Outcomes Can Distort Our Judgment

Once we know how a choice ended, some reasons can become easier to notice. After a failure, warnings may seem especially compelling; after success, the same warnings may look overly cautious. Baron and Hershey investigated whether outcomes changed the attention given to arguments favoring or opposing a decision. Their findings offered partial support for this explanation, but it did not fully account for the bias (Baron & Hershey, 1988).

This possibility matters whenever we reconstruct a decision afterward. A convincing explanation can reflect what has become prominent in the finished story. To evaluate it fairly, we need to ask whether the same reason deserved that much weight before the ending was known.

Outcome Bias in Everyday Life

Now imagine two versions of the career decision. In the first, the employer’s closure follows an event that reasonable checks would not have revealed. In the second, credible reports of missed payroll were available, but the applicant dismissed them without an inquiry that was feasible. Both versions end with job loss. The second gives us an additional reason to question the decision: evidence was neglected before the outcome was known.

The result does not tell us which version occurred. That requires evidence about the earlier circumstances, including the applicant’s options and constraints. Even accepting a known risk can be reasonable when the alternatives are worse. The point is to investigate how the choice was made before turning the ending into a verdict about the person.

At work, the reverse problem can arise when a rushed project succeeds despite ignored warnings. The team may treat that success as permission to repeat the same omissions, allowing its learning from experience to rest on a result it has not examined. This hypothetical example shows why favorable outcomes deserve scrutiny too. A process can carry avoidable risks even when those risks have not yet produced visible harm.

In a relationship, someone may raise a concern respectfully and receive an angry response. The unpleasant ending alone does not establish that speaking was wrong. The content, timing, foreseeable risks, and other person’s behavior all matter. Where intimidation or retaliation is present, protecting safety takes priority; the language of bias should never be used to dismiss someone’s experience of harm.

How Success and Failure Can Mislead Learning

When success becomes sufficient proof of sound judgment, weaknesses can escape examination. After failure, a useful approach may be abandoned alongside an assumption that actually needs correction. These are possible consequences of evaluating the process through the result. Learning requires a more specific account of what should be retained, investigated, or changed.

Failure also creates a separate obstacle. In five studies involving 1,674 participants, Eskreis-Winkler and Fishbach found that people learned less from feedback about their own incorrect answers than from feedback about correct answers. The tasks had two answer choices, so either kind of feedback revealed the correct response. Participants learned more from other people’s failures than from their own, consistent with the authors’ account that personal failure can threaten the self and interrupt attention (Eskreis-Winkler & Fishbach, 2019).

How we give and receive feedback matters too. Kluger and DeNisi brought together findings from many studies. Feedback improved performance on average, but more than one-third of the effects they examined showed worse performance. Feedback tended to be less helpful when it drew attention away from the task and toward the person’s sense of self. A judgment such as “You are incompetent” offers less direction than identifying a specific assumption to check. The research concerns feedback generally; it does not show that outcome bias caused these difficulties (Kluger & DeNisi, 1996).

Learning from Outcomes Without Unfair Blame

The career example invites three different judgments. Was accepting the job reasonable at the time? What does the closure teach us about future choices? What support or response does the resulting loss now require? Keeping these questions separate gives us a practical way to examine the experience. This distinction is an organizing approach for the article, rather than a tested psychological model.

Evaluating the Original Decision

In the first version, the closure adds little reason to condemn the earlier choice if reasonable inquiry could not have revealed the threat. In the second, the available payroll warning changes the evaluation because of what could have been considered beforehand. A favorable ending would not erase that omission. In either version, we still need to examine the alternatives and constraints before deciding how serious the weakness was.

Learning from the Result

Outcomes can legitimately teach us something new. When we know little about the quality of a process, results may provide a reason to investigate it. Across comparable situations, they can also help us revise estimates of what is likely. Baron and Hershey distinguished these uses of outcome information from giving a result undue weight when evaluating an already documented decision (Baron & Hershey, 1988).

For the job applicant, a useful lesson might be to investigate a particular sign of instability next time, if it can realistically be checked. It would take more evidence to conclude that changing jobs is always a mistake. Kahneman and Klein argue that reliable intuitive expertise depends on learnable patterns and adequate opportunities to learn them. Repeated outcomes are most informative when we understand how the situations compare and what the feedback actually reveals (Kahneman & Klein, 2009).

Responding to Harm and Judging Responsibility

Losing work brings practical consequences whether or not accepting the job was a mistake. A need for support does not depend on first proving that the earlier choice was flawless. Similarly, in decisions that affect other people, acknowledging harm and deciding what response is needed remain important even when the original reasoning was defensible.

A bad result also leaves open the question of blame. Malle and colleagues distinguish judging an event as harmful from judging a person as blameworthy. Their account asks whether the person caused the event and acted intentionally. If the action was intentional, the reasons matter. If the harm was unintended, the questions become whether the person had a duty to prevent it and a realistic ability to do so. Wanting a good outcome does not answer those questions (Malle et al., 2014).

This gives us room to recognize bad luck without excusing neglected duties or foreseeable risks. It also gives us room to acknowledge pain without assuming that its severity proves someone’s incompetence. Where intimidation, abuse, or retaliation is present, safety and support take priority over debating whether the harmed person’s account reflects a cognitive bias.

How to Evaluate Decisions More Fairly

A fair review gives the earlier decision room to be understood before the ending takes over. The following practices bring the article’s distinctions into a usable sequence. They are evidence-informed suggestions, not a tested program guaranteed to remove outcome bias.

Reconstruct the Decision and Examine the Reasoning

Begin with the information available before the choice. What evidence was accessible? What remained uncertain? Which alternatives were realistic? Contemporary notes, messages, and stated expectations can help distinguish what someone believed then from what seems clear now. Also ask what relevant information could reasonably have been obtained before acting.

Review how the evidence was used. Were important risks considered? Did the choice fit the goals and responsibilities involved? Were inconvenient facts dismissed? This is where the broader practice of wise decisions becomes relevant. Fairness allows us to identify a weak process even when it succeeds and a reasonable process even when it disappoints.

Consider a Different Ending

Imagining a different ending can help us question our evaluation. Would the same reasoning look better or worse if chance had produced a different outcome? In two experiments, Lord and colleagues found that considering opposite possibilities reduced certain biases in social judgment more effectively than instructions simply to be fair and unbiased. Applying that question here is a reasoned extension; those experiments did not test this complete outcome-bias review process (Lord et al., 1984).

In the payroll-warning version, imagine that the employer recovered and the new job worked out well. Would dismissing the warning now seem sensible? The changed ending should not erase the question about neglected evidence. Conversely, imagining success after an unforeseeable closure can help reveal whether disappointment alone is carrying the criticism.

Use the Review to Improve Future Decisions

End with a specific adjustment that the evidence supports. Perhaps an important assumption needs checking, a safeguard needs changing, or more comparable cases are needed before reaching a conclusion. In the job example, identify the inquiry that would have been both useful and feasible. Avoid replacing one sweeping verdict with another, such as “I should never take a risk again.”

For consequential choices, a brief decision record may be useful. Write down the options, your reasons, the major uncertainties, and what you expect could happen. Later, compare the outcome with that account. The purpose is to preserve something concrete to examine, including assumptions that deserve revision. A record should make learning easier rather than become a defense against all criticism.

A Few Words from Psychology Fanatic

The person who lost a new job still has to face the loss. A fair account of the earlier decision will not restore an income or remove the uncertainty of starting again. It may, however, make the next question more precise: was there a warning to investigate, an assumption to revise, or a limit to what anyone could reasonably have known?

We can examine an omission without turning it into a judgment of someone’s worth. We can recognize a reasonable choice while acknowledging that its consequences still need attention. Leaving room for both allows the past to inform our next decision without requiring its ending to explain everything.

Associated Concepts

• Thinking errors: Outcome bias belongs to the broader study of how judgment can depart from the evidence relevant to a question.
• Wise decisions: Evaluating a choice requires attention to its reasoning and uncertainty as well as its eventual consequences.
• Cognitive heuristics: Using an outcome as a shortcut can be informative in some settings and misleading in others.
• Receiving advice: A recommendation can be assessed by the evidence and reasoning available before its consequences are known.
• Hindsight Bias: The tendency to see an event as more predictable after it has occurred. While outcome bias distorts judgments of decision quality, hindsight bias distorts judgments of what could have been anticipated.
• Counterfactual Thinking: Imagining how events might have unfolded differently. Considering alternative outcomes can help separate the quality of a decision from the particular result it produced.

References

Aiyer, Sriraj; Kam, Hoi Ching; Ng, Ka Yuk; Young, Nathaniel A.; Shi, Jiaxin; Feldman, Gilad (2023). Outcomes Affect Evaluations of Decision Quality: Replication and Extensions of Baron and Hershey’s (1988) Outcome Bias Experiment 1. International Review of Social Psychology, 36(1), Article 12, 1–16. DOI: 10.5334/irsp.751.
(Return to Main Text)

Baron, Jonathan; Hershey, John C. (1988). Outcome Bias in Decision Evaluation. Journal of Personality and Social Psychology, 54(4), 569–579. DOI: 10.1037/0022-3514.54.4.569.
(Return to Main Text)

Eskreis-Winkler, Lauren; Fishbach, Ayelet (2019). Not Learning from Failure—the Greatest Failure of All. Psychological Science, 30(12), 1733–1744. DOI: 10.1177/0956797619881133.
(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)

Kluger, Avraham N.; DeNisi, Angelo (1996). The Effects of Feedback Interventions on Performance: A Historical Review, a Meta-Analysis, and a Preliminary Feedback Intervention Theory. Psychological Bulletin, 119(2), 254–284. DOI: 10.1037/0033-2909.119.2.254.
(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)

Malle, Bertram F.; Guglielmo, Steve; Monroe, Andrew E. (2014). A Theory of Blame. Psychological Inquiry, 25(2), 147–186. DOI: 10.1080/1047840X.2014.877340.
(Return to Main Text)