Snap Shot
- The question: Once a harmful outcome is known, can witnesses, jurors, judges, and reviewing experts fairly judge what could have been foreseen beforehand, and does knowing the outcome change memory of what came before?
- What the research shows: Outcome knowledge makes an outcome seem more likely, more inevitable, and more foreseeable than it seemed in advance, and it shifts memory of one’s own earlier judgments toward the known result. The effect appears in children and older adults, in physicians, mock jurors, and judges, and in judgments of negligence, care, and culpability.
- What it does not show: The effect is generally small to moderate in size, and it does not show that any particular judgment of foreseeability is wrong, that a particular witness’s memory has changed, or that a person who failed to prevent harm acted reasonably. Some risks were in fact foreseeable.
- The law: Use-of-force and ineffective-assistance doctrine expressly direct courts to judge conduct from the actor’s position at the time (Graham v. Connor; Barnes v. Felix; Strickland v. Washington), and Arizona and military self-defense standards turn on what a reasonable person would have perceived. Expert testimony is governed by Rule 702, Houser, and Chapple, and may not address a particular witness’s credibility (Lindsey).
- Issues for each side: Prosecutors should build foreseeability from what the actor knew and saw before the outcome, without assuming that harm proves the risk was obvious; defense counsel can explain why outcome knowledge inflates foreseeability, while recognizing that the same bias can reshape an accused’s own memory of a threat.
Hindsight bias is the tendency, after learning how an event turned out, to regard the outcome as more predictable than it really was. People who know the result tend to believe they would have expected it, to see it as having been more or less bound to happen, and to remember their own earlier views as closer to the result than they were. The tendency is sometimes called the “knew-it-all-along” effect. It is not dishonesty; it arises from the ordinary way new knowledge is absorbed into what a person already knows.
The phenomenon reaches court for two reasons. First, much of the law asks a factfinder who knows the outcome to evaluate a decision made by someone who did not: whether a risk was foreseeable in a negligence or negligent-homicide case, whether an officer’s or accused’s use of force was reasonable, whether a parent, caregiver, or professional should have recognized danger to a child, and whether defense counsel’s choices were reasonable when made. Second, hindsight bias is partly a memory effect. Witnesses learn how events ended long before they testify, and that knowledge can reshape what they remember about the warning signs, their own expectations, and the sequence that led to the outcome.
This paper describes the research on hindsight bias, its effect on memory, studies of legal decision makers, and attempts to reduce it, followed by the limits of the evidence, the governing law, and recurring issues for each side.
What Hindsight Bias Is
Fischhoff (1975) first demonstrated the phenomenon experimentally. Participants read accounts of historical or clinical events and estimated the likelihood of several possible outcomes; some were told which outcome had occurred and asked to judge as if they did not know. Those given an outcome rated it as more likely than participants given no outcome, and they did not appear to recognize that the information had influenced them. In a companion study, people who had predicted the outcomes of President Nixon’s 1972 trips to China and the Soviet Union were later asked to recall their predictions; they tended to remember having assigned higher probabilities to events that occurred than they actually had (Fischhoff & Beyth, 1975).
A widely cited review describes hindsight bias as three related components that can occur separately (Roese & Vohs, 2012). The first component is memory distortion, the misremembering of one’s own earlier judgment. The second is a sense of inevitability, the belief that the outcome had to happen. The third is foreseeability, the belief that one could have predicted it. The review attributes the bias to the way people make sense of outcomes: once a result is known, people construct an explanation for it, and the facts that fit the explanation become more salient while facts that pointed elsewhere recede.
How Robust Is the Effect?
Two meta-analyses have found hindsight bias to be a reliable effect across many studies, materials, and populations (Christensen-Szalanski & Willham, 1991; Guilbault et al., 2004). Both also found that its size varies with the task and the material, and the average effect in this literature is generally described as small to moderate. That distinction matters in court: a reliable effect is not necessarily a large one.
The bias appears across the lifespan. In a study of 194 participants aged 3 to 95, every age group showed hindsight bias on each of three tasks; preschoolers and older adults showed more than older children and younger adults, for different reasons (Bernstein et al., 2011). Hindsight bias also occurs in perception: people who have seen a clear image of an object overestimate how early they, or others, would identify it from a degraded version (Harley et al., 2004).
Hindsight Bias and Memory
The memory component of hindsight bias is the part most directly relevant to witness testimony. When people try to recall an earlier judgment or impression after learning the outcome, they either retrieve the original or reconstruct it, and reconstruction tends to move toward what they now know (Bernstein et al., 2011). Reviewers have noted that this process resembles the misinformation effect and other forms of memory distortion in which later information is blended into memory for earlier events (Mazzoni & Vannucci, 2007).
The forensic implication should be stated carefully. A witness who now knows that an event ended in harm may sincerely remember the earlier warning signs as clearer, or his or her own concern as greater, than they were at the time. A witness who knows that a person was later charged may recall that person’s earlier conduct as more suspicious. The research shows that this can happen, not that it happened in a particular case, and most of it concerns memory for one’s own estimates rather than for observed events. Many warning signs are real and were recognized at the time. In a study in which 526 undergraduates rated vignettes describing an adult’s behavior toward a child, participants told that the adult was later found to be a child molester overestimated how likely they would have been to predict it, yet participants were also able to recognize sexual grooming behaviors in the vignettes (Winters & Jeglic, 2016). Contemporaneous records such as messages, reports, notes, and earlier statements are the best available check on whether a recollection of an earlier impression has shifted.
Hindsight in Legal Judgment
Negligence and Professional Care
The law of negligence asks whether a risk was foreseeable and whether precautions were reasonable, judged as of the time of the decision. Experimental research has generally found that outcome knowledge pushes those judgments toward liability. In a study by Kamin and Rachlinski (1995), participants in a foresight condition, reviewing evidence as if at an administrative hearing, decided whether a city should take precautions against flooding; about three quarters concluded that a flood was too unlikely to justify them. Participants in a hindsight condition, told that a flood had occurred and asked whether the city had been negligent, more often found that it had, and gave higher estimates of the probability of the flood. In a study of 297 community residents asked to evaluate a therapist’s handling of a potentially dangerous patient, those told that the patient later became violent rated the violence as more foreseeable and the therapist’s actions as less reasonable, and more often judged the therapist negligent (LaBine & LaBine, 1996). A review of this literature concluded that hindsight bias affects many kinds of legal decisions that turn on foreseeability (Harley, 2007).
Professionals are not immune. Physicians given a clinical case and told which diagnosis was correct assigned it a higher probability than physicians who were not told (Arkes et al., 1981). When 112 practicing anesthesiologists rated the appropriateness of care in cases that were identical except for the severity of the outcome, the proportion of ratings of appropriate care fell by 31 percentage points when the outcome was changed from temporary to permanent injury (Caplan et al., 1991). A reviewer who knows the outcome, including one reviewing the care of an injured child, evaluates the earlier decision with information the clinician did not have. Judges are not immune either: in a study of 167 federal magistrate judges, each of five cognitive illusions tested, hindsight bias among them, significantly affected responses to hypothetical problems (Guthrie et al., 2001).
Criminal Culpability, Child Protection, and Police Response
Hindsight affects judgments of blame as well as probability. In ten preregistered experiments (N = 2,043), harmful outcomes, compared with neutral ones, increased participants’ estimates of the probability of harm and, through that increase, their attributions of negligence and culpability (Kneer & Skoczeń, 2023). The authors argued that the distortion is a systematic cognitive error and that it is a particular concern in criminal law, where mental-state findings such as recklessness and negligence depend on what the actor should have perceived.
The same pattern appears in settings that frequently reach court after a death or serious injury. Inquiries into child deaths are necessarily conducted with knowledge of the outcome; a content analysis of 45 British child abuse inquiry reports found that the errors of the professionals involved, such as reliance on vivid or readily available information and slowness to revise an initial view, were predictable products of ordinary human reasoning rather than random lapses (Munro, 1999). In an experiment with 200 university students, participants who were told that a man assessed by police later killed his intimate partner rated the homicide as more likely and foreseeable, and gave the police lower performance ratings, than participants given no outcome; asking participants to take the officers’ perspective did not remove the effect (Haynes & Kebbell, 2025). Neither finding means that the professionals or officers in a given case acted reasonably; outcome knowledge is a source of error whether or not the earlier decision was sound.
Can Hindsight Bias Be Reduced?
Telling people about the bias, or asking them to disregard what they know, has generally had little effect. Participants who were warned and tested repeatedly continued to show the bias (Pohl & Hell, 1996), and a debiasing procedure in the Kamin and Rachlinski (1995) study did not significantly change hindsight judgments. In a study of 355 jury-eligible adults who watched a reenacted automobile negligence trial, repeated admonitions from the judge were generally ineffective, deliberation did not correct the error, and bifurcation, which kept outcome evidence from the jurors when they decided the conduct question, was relatively more effective (Smith & Greene, 2005).
The best-supported technique asks decision makers to consider how the outcome could have turned out differently. Neuropsychologists who were asked to give a reason why each alternative diagnosis might have been correct showed reduced hindsight bias (Arkes et al., 1988). The technique has limits: when participants were asked to list ten ways an event might have turned out otherwise, the difficulty of the task increased hindsight bias, while listing two produced no significant bias (Sanna et al., 2002). Some debiasing strategies in recent preregistered experiments also showed promise, while others did not (Kneer & Skoczeń, 2023). Considering a few concrete alternatives appears more useful than general warnings, but the research has not identified a method that reliably eliminates the bias.
Limits of the Evidence
Most hindsight research uses students or community participants making judgments about written vignettes, almanac questions, or hypothetical cases, without the evidence, argument, instructions, deliberation, and stakes of a trial. The effect is reliable but generally modest, and it varies with the material and the measure. The memory research largely concerns recall of one’s own earlier answers and estimates, not recall of observed events, and its extension to eyewitness memory is an inference supported by related research on post-event information rather than a direct demonstration. Group findings cannot establish whether a particular witness’s memory has shifted or whether a particular risk was foreseeable; those are questions for the evidence of what was known at the time.
Questions to Ask About This Research
When Kamin and Rachlinski (1995) or LaBine and LaBine (1996) is cited to show that jurors will judge conduct too harshly, counsel can ask who was studied and under what conditions: these were students or community residents reading case materials, without live evidence, cross-examination, or deliberation. When Caplan et al. (1991) is cited, counsel can ask whether it applies to this reviewer, including whether the reviewing expert knew the outcome before forming an opinion. When Fischhoff and Beyth (1975) or Bernstein et al. (2011) is offered to argue that later knowledge reshaped a witness’s recollection of warning signs, counsel can ask what exactly was measured, because those studies measured memory for one’s own earlier predictions and answers, and whether contemporaneous records show what the witness said before the outcome was known. When the consider-the-alternatives remedy (Arkes et al., 1988) is offered, counsel can ask how much evidence supports it in legal settings, given that asking for too many alternatives backfired in Sanna et al. (2002). These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.
Hindsight Bias and the Law
Federal and Military Standards
The law has expressly recognized hindsight as a source of error in several doctrines. Excessive-force claims arising from an arrest are judged under the Fourth Amendment’s objective reasonableness standard, from the perspective of a reasonable officer on the scene “rather than with the 20/20 vision of hindsight,” allowing for split-second judgments in tense, rapidly evolving circumstances (Graham v. Connor, 1989). The inquiry considers the totality of the circumstances, not only the final moment of threat (Barnes v. Felix, 2025); the unanimous Court rejected a moment-of-threat rule and left open whether an officer’s own creation of danger bears on reasonableness. In reviewing claims of ineffective assistance, courts must make every effort to “eliminate the distorting effects of hindsight” and assess the conduct as it looked to counsel when it occurred (Strickland v. Washington, 1984).
Expert testimony on hindsight bias is governed by Federal Rule of Evidence 702, as amended December 1, 2023, under which the proponent must show that it is more likely than not that the rule’s requirements are met; Daubert v. Merrell Dow Pharmaceuticals, Inc. (1993) assigned the gatekeeping role to the trial judge, and Kumho Tire Co. v. Carmichael (1999) extended it to all expert testimony, including testimony based on technical or other specialized knowledge. In courts-martial, Military Rule of Evidence 702 tracks the federal rule, the 2023 federal amendment took effect in the military rule in 2025 through Military Rule of Evidence 1102 (Joint Service Committee on Military Justice, 2025), and practitioners should confirm current text; admissibility is assessed under the six factors of United States v. Houser (1993). Military self-defense in cases involving deadly force requires that the accused apprehended, on reasonable grounds, that death or grievous bodily harm was about to be inflicted wrongfully, the first element of which is an objective test (R.C.M. 916(e)), and child endangerment by culpable negligence under Article 119b turns on disregard for consequences that were foreseeable (10 U.S.C. § 919b).
Arizona
Arizona Rule of Evidence 702 adopted the federal formulation and the Daubert standard effective January 1, 2012, and was amended effective January 1, 2024, to conform to the 2023 federal amendment (Ariz. Sup. Ct. Order R-23-0004). State v. Chapple (1983) is Arizona’s leading case on expert testimony about memory and eyewitness reliability, holding exclusion an abuse of discretion on its facts; admission otherwise remains within the trial court’s discretion, and an expert may be limited to general factors (State v. Nordstrom, 2001). An expert may not tell the jury whether a particular witness is truthful (State v. Lindsey, 1986, 149 Ariz. at 475). Arizona’s self-defense instruction measures the defendant’s belief against what a reasonable person in the situation would have believed and states that actual danger is not necessary, which directs jurors to the apparent circumstances at the time (RAJI (Criminal) Standard 4.04). Criminal negligence, the mental state for offenses such as negligent homicide, requires a failure to perceive a substantial and unjustifiable risk that is a gross deviation from what a reasonable person would observe in the situation (A.R.S. § 13-105(10)(d)).
Special Issues in Legal Practice
The principles described above apply with equal force to every party in a legal proceeding. The science operates the same way regardless of who is asking the questions, and the research summarized here does not by itself favor either side. What differs between the parties is not the science but the vantage point — the particular places where these issues tend to surface, and the particular errors each party is best positioned to guard against. The two summaries below identify those recurring issues. They are offered for orientation, not as litigation strategy; in every case the controlling question is what the evidence in that case will, and will not, support, and the expert’s role is to inform the trier of fact rather than to advocate for either side.
Special Issues for Prosecutors
The research supports several propositions prosecutors can legitimately present. Hindsight bias is a tendency, not proof that a risk was unforeseeable; many risks were foreseeable, and research shows that people can recognize warning behaviors such as grooming. The bias also affects an accused’s own account: a person who learns after a confrontation that the other person was unarmed, or that someone was badly hurt, may remember the events in light of that knowledge, and a self-defense claimant who later learns of a weapon may remember the threat as clearer than it appeared. Prosecutors are well positioned to ground foreseeability in what the actor knew and observed before the outcome, through contemporaneous messages, reports, and statements, and to have expert reviewers document what they knew when they formed an opinion.
The research also counsels caution. Arguments that treat the severity of the harm as proof that the danger was obvious ask the factfinder to reason in the way the research describes as biased, and the law of use of force, self-defense, and negligence directs the inquiry to the circumstances as they appeared at the time. Witnesses, including professionals, who recall having seen clear warning signs may sincerely be recalling them through later knowledge, and their earlier statements and records are the better measure. Expert reviews of an officer’s, caregiver’s, or clinician’s conduct that were prepared with knowledge of the outcome are open to the criticism that the reviewer could not set that knowledge aside.
Special Issues for Defense
The research supports several propositions defense counsel can legitimately present. Outcome knowledge makes a result look more foreseeable, and more nearly inevitable, than it was; it affects jurors, professionals, and judges, and admonitions to disregard it are generally ineffective. Defense counsel can ask that conduct be judged by the information available at the time, request the instructions the law provides, seek procedures that separate conduct from outcome where the rules permit, offer general expert testimony where admissible, and ask whether a reviewing expert or witness formed an opinion before or after learning the outcome.
The research also counsels caution about the limits that apply to the defense. Hindsight bias does not establish that a particular risk was unforeseeable or that particular conduct was reasonable, and its average effect is modest. An expert cannot tell the factfinder that a particular witness’s memory has been distorted by outcome knowledge, which is a credibility judgment. The same bias may have shaped the accused’s own recollection, and favorable outcomes produce hindsight too, as when a risky decision that happened not to cause harm is later described as having been safe.
This paper is for educational purposes only and is not legal advice. The author is a psychologist, not an attorney; summaries of law are general and must be verified against current authority or with a licensed attorney.
References
Arkes, H. R., Faust, D., Guilmette, T. J., & Hart, K. (1988). Eliminating the hindsight bias. Journal of Applied Psychology, 73(2), 305–307. https://doi.org/10.1037/0021-9010.73.2.305
Arkes, H. R., Wortmann, R. L., Saville, P. D., & Harkness, A. R. (1981). Hindsight bias among physicians weighing the likelihood of diagnoses. Journal of Applied Psychology, 66(2), 252–254. https://doi.org/10.1037/0021-9010.66.2.252
Bernstein, D. M., Erdfelder, E., Meltzoff, A. N., Peria, W., & Loftus, G. R. (2011). Hindsight bias from 3 to 95 years of age. Journal of Experimental Psychology: Learning, Memory, and Cognition, 37(2), 378–391. https://doi.org/10.1037/a0021971
Caplan, R. A., Posner, K. L., & Cheney, F. W. (1991). Effect of outcome on physician judgments of appropriateness of care. JAMA, 265(15), 1957–1960. https://doi.org/10.1001/jama.1991.03460150061024
Christensen-Szalanski, J. J. J., & Willham, C. F. (1991). The hindsight bias: A meta-analysis. Organizational Behavior and Human Decision Processes, 48(1), 147–168. https://doi.org/10.1016/0749-5978(91)90010-Q
Fischhoff, B. (1975). Hindsight ≠ foresight: The effect of outcome knowledge on judgment under uncertainty. Journal of Experimental Psychology: Human Perception and Performance, 1(3), 288–299. https://doi.org/10.1037/0096-1523.1.3.288
Fischhoff, B., & Beyth, R. (1975). “I knew it would happen”: Remembered probabilities of once-future things. Organizational Behavior and Human Performance, 13(1), 1–16. https://doi.org/10.1016/0030-5073(75)90002-1
Guilbault, R. L., Bryant, F. B., Brockway, J. H., & Posavac, E. J. (2004). A meta-analysis of research on hindsight bias. Basic and Applied Social Psychology, 26(2–3), 103–117. https://doi.org/10.1080/01973533.2004.9646399
Guthrie, C., Rachlinski, J. J., & Wistrich, A. J. (2001). Inside the judicial mind. Cornell Law Review, 86(4), 777–830.
Harley, E. M. (2007). Hindsight bias in legal decision making. Social Cognition, 25(1), 48–63. https://doi.org/10.1521/soco.2007.25.1.48
Harley, E. M., Carlsen, K. A., & Loftus, G. R. (2004). The “saw-it-all-along” effect: Demonstrations of visual hindsight bias. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30(5), 960–968. https://doi.org/10.1037/0278-7393.30.5.960
Haynes, C., & Kebbell, M. R. (2025). Intimate partner homicide: Hindsight bias, outcome bias and perspective-taking influences on perceptions of police responses. Journal of Criminal Psychology, 15(7), 1–14. https://doi.org/10.1108/JCP-08-2024-0083
Joint Service Committee on Military Justice. (2025). Amendments to the Military Rules of Evidence since 1984 (G. E. Maggs, Ed.; 2025 ed.). https://jsc.defense.gov/Portals/99/MRE%20Amendments%202025.pdf
Kamin, K. A., & Rachlinski, J. J. (1995). Ex post ≠ ex ante: Determining liability in hindsight. Law and Human Behavior, 19(1), 89–104. https://doi.org/10.1007/BF01499075
Kneer, M., & Skoczeń, I. (2023). Outcome effects, moral luck and the hindsight bias. Cognition, 232, Article 105258. https://doi.org/10.1016/j.cognition.2022.105258
LaBine, S. J., & LaBine, G. (1996). Determinations of negligence and the hindsight bias. Law and Human Behavior, 20(5), 501–516. https://doi.org/10.1007/BF01499038
Mazzoni, G., & Vannucci, M. (2007). Hindsight bias, the misinformation effect, and false autobiographical memories. Social Cognition, 25(1), 203–220. https://doi.org/10.1521/soco.2007.25.1.203
Munro, E. (1999). Common errors of reasoning in child protection work. Child Abuse & Neglect, 23(8), 745–758. https://doi.org/10.1016/S0145-2134(99)00053-8
Pohl, R. F., & Hell, W. (1996). No reduction in hindsight bias after complete information and repeated testing. Organizational Behavior and Human Decision Processes, 67(1), 49–58. https://doi.org/10.1006/obhd.1996.0064
Roese, N. J., & Vohs, K. D. (2012). Hindsight bias. Perspectives on Psychological Science, 7(5), 411–426. https://doi.org/10.1177/1745691612454303
Sanna, L. J., Schwarz, N., & Stocker, S. L. (2002). When debiasing backfires: Accessible content and accessibility experiences in debiasing hindsight. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28(3), 497–502. https://doi.org/10.1037/0278-7393.28.3.497
Smith, A. C., & Greene, E. (2005). Conduct and its consequences: Attempts at debiasing jury judgments. Law and Human Behavior, 29(5), 505–526. https://doi.org/10.1007/s10979-005-5692-5
Winters, G. M., & Jeglic, E. L. (2016). I knew it all along: The sexual grooming behaviors of child molesters and the hindsight bias. Journal of Child Sexual Abuse, 25(1), 20–36. https://doi.org/10.1080/10538712.2015.1108945
Legal Authorities
Barnes v. Felix, 605 U.S. 73 (2025).
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).
Graham v. Connor, 490 U.S. 386 (1989).
Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).
State v. Chapple, 135 Ariz. 281, 660 P.2d 1208 (1983).
State v. Lindsey, 149 Ariz. 472, 720 P.2d 73 (1986).
State v. Nordstrom, 200 Ariz. 229, 25 P.3d 717 (2001).
Strickland v. Washington, 466 U.S. 668 (1984).
United States v. Houser, 36 M.J. 392 (C.M.A. 1993).
10 U.S.C. § 919b (Article 119b, Uniform Code of Military Justice).
Ariz. Rev. Stat. Ann. § 13-105(10)(d).
Fed. R. Evid. 702.
Mil. R. Evid. 702.
Mil. R. Evid. 1102.
Rule for Courts-Martial 916(e).
Ariz. R. Evid. 702.
Ariz. Sup. Ct. Order R-23-0004 (amending Ariz. R. Evid. 702, effective Jan. 1, 2024).
Revised Arizona Jury Instructions (Criminal), Standard 4.04 (Justification for Self-Defense) (6th ed. 2022).
Related topic papers: Reconstruction and Recall · The Misinformation Effect · Consistency Bias · Confirmation Bias · Egocentric Bias · Remembering Changes Memory
Related topics from Arizona Forensic Services: Courts-Martial · Child Abuse Allegations