Snap Shot
- The question: Why do truthful alibis so often turn out to be vague, wrong, or changed, and how do investigators, prosecutors, and jurors interpret those problems?
- What the research shows: In experiments, many people who give true alibis for a past day make errors and change details on retelling, rely on what they usually do rather than what they did, and have little independent evidence to offer; strangers who could corroborate them often fail to recognize them. Evaluators tend to treat a weak or changed alibi as a sign of guilt, and in one laboratory study, learning that the suspect had confessed made students less likely to maintain their corroboration.
- What it does not show: The research does not show that a particular weak or changed alibi is true. Guilty suspects also give false alibis, and observers, including police, distinguish true from false alibis only modestly better than chance. Most studies use students or community volunteers and low-stakes mock accusations.
- The law: Federal, military, and Arizona rules require the defense to give notice of an alibi, and the prosecution keeps the burden of proving beyond a reasonable doubt that the accused committed the offense (Williams v. Florida; R.C.M. 701(b)(2); Ariz. R. Crim. P. 15.2). Expert testimony on alibi research is governed by Rule 702.
- Issues for each side: For prosecutors, the research favors alibi checks that are prompt, documented, and open to confirming the alibi; for defense counsel, early preservation of records and care not to over-read research that describes group tendencies rather than a particular account.
An alibi is a claim that the accused was somewhere else when the crime occurred. It is among the most direct answers an innocent person can give to an accusation, and investigators routinely ask for one. Yet researchers who study alibis have found that innocent people often give accounts that are incomplete, mistaken, or later revised, and that these problems are easily read as lies. In some wrongful-conviction cases, an alibi that was disbelieved, abandoned, or never investigated has been reported alongside a false confession or a mistaken identification; no rate is available.
Alibis reach court in several ways. The defense must give notice of an alibi and the witnesses who will support it. The prosecution may argue that an alibi that changed, or was shown to be wrong, reflects consciousness of guilt. During an interrogation, an investigator may confront a suspect with gaps or inconsistencies in the alibi, and the suspect’s response may become part of the case. Jurors then decide whether the alibi raises a reasonable doubt.
This paper reviews the research on how truthful alibis are generated, why they fail, how investigators and jurors evaluate them, and how a confession can affect alibi witnesses. It then summarizes the procedural law on alibis in federal, military, and Arizona courts. The memory processes involved are discussed more generally in the Memory in the Courtroom papers Forgetting, Schema and Script Memory, and The Telescoping Effect.
What Makes an Alibi Believable
Olson and Wells (2004) proposed that an alibi is supported by two kinds of evidence: physical evidence, such as receipts, surveillance video, or phone records, and person evidence, such as a witness who was with the suspect. In their study, 252 participants playing the role of detectives rated alibis that combined different levels of each. Physical evidence carried more weight than expected and tended to overwhelm person evidence. Witnesses who were friends or relatives of the suspect added little, and evaluators did not seem to consider that a stranger who corroborated an alibi might be mistaken about who he or she had seen. A related mock-juror study found that jurors acquitted more often when an alibi witness with no relationship to the defendant testified, and that they discounted a witness who had a relationship with the defendant even when the witness was not a family member (Culhane & Hosch, 2004).
The difficulty for innocent suspects is that ordinary days leave little of the evidence evaluators value most. In a study of 841 Dutch laypeople asked to generate a true alibi after being falsely accused of a mock robbery, nearly all could name where they had been and most could point to some supporting evidence, but the type of evidence depended heavily on the day of the week and the time of day; the authors argued that judging an alibi only by the strength of its support is not a fair measure without knowing what innocent people typically can produce (Nieuwkamp et al., 2016). In an American study in which innocent participants were asked to provide alibis and supporting evidence, many could offer only weak corroboration, usually from people close to them (Olson & Charman, 2012).
Strangers may not help as much as investigators expect. When university employees who had briefly interacted with a student were shown that student’s photograph a day later, most did not recognize the student, although false recognitions were rare, and the students were overconfident that they would be remembered (Charman et al., 2017). In that study, failure to recognize a student was common, so a clerk’s failure to recall a customer may carry little weight on its own; a clear, specific recollection is a different matter.
Memory Errors and Changing Alibis
Many alibis are requested days or weeks after the event, for a time that held no special meaning for the person when it happened. Memory for routine events is reconstructed from fragments and from knowledge of what one usually does. In two experiments, participants who were later asked where they had been at particular times relied heavily on schemas of their typical week, especially when cued with a time rather than a place; accuracy was lower when their actual whereabouts were out of the ordinary, and confidence was high regardless of accuracy (Leins & Charman, 2016). An innocent person may therefore give a confident, sincere, and wrong alibi by describing a usual Tuesday rather than the Tuesday in question.
Alibis also change. Strange, Dysart, and Loftus (2014) asked participants to give a truthful alibi for an afternoon three weeks earlier, to search for evidence of where they had actually been, and to retell the alibi a week later. Participants were largely inconsistent across every aspect of the account, including what they were doing, with whom, where, and for how long. Those who changed their accounts were less confident and remembered the period less vividly. The authors concluded that inconsistency is a normal byproduct of memory rather than a reliable sign of deception. Crozier, Strange, and Loftus (2017) described how an early error can cascade: investigators read the change as a lie, prosecutors weigh it in charging, and jurors may treat it as evidence of guilt.
The same pattern can arise in the interrogation room. Gudjonsson (2018) describes how an innocent suspect who cannot remember a period clearly may be confronted with that vagueness until he or she loses confidence in the alibi, and notes that people who distrust their own memories are more open to suggestion. Leo (2008), drawing on observation of American interrogations, describes attacking the suspect’s alibi as inconsistent or implausible as a common accusatorial tactic. These descriptions come from researchers who are generally critical of accusatorial methods, and they describe tendencies rather than the conduct of any particular interviewer.
How Investigators and Factfinders Evaluate Alibis
Police and laypeople commonly believe that a true alibi should remain consistent over time, a belief documented in a survey of law enforcement officers (Dysart & Strange, 2012, as summarized in Strange et al., 2014). That belief is not unreasonable on its face: a guilty suspect who invents an alibi may later abandon it when it is checked. The research problem is that innocent people also revise their accounts, so a change by itself does not separate the two.
Observers are not very good at telling true from false alibis by their content. In a study with remand prisoners who generated true or false alibis, experienced police officers and students who later evaluated the written alibis did not exceed about 60% accuracy, and the two groups attended to different features (Nieuwkamp et al., 2018). Earlier work with student participants reached a similar conclusion about the difficulty of detection (Culhane et al., 2013). These findings fit the broader research on deception detection, summarized in the Memory hub’s Detecting Deception.
Changes in a witness’s account also matter. In three experiments, listeners heard an interview in which an alibi witness either changed or maintained the account; a change lowered ratings of the witness’s credibility and of the suspect’s innocence, and the effect was somewhat smaller when the interviewer explained how memory works than when the interviewer made no attempt to influence the witness or used an implicit threat (Eastwood et al., 2026). The study suggests that how an alibi witness is questioned can shape how the witness’s later account is perceived.
The investigator’s perspective deserves equal weight. Checking alibis is a standard investigative step: an Army law-enforcement manual lists whether a suspect’s alibi is supported among the questions in developing a suspect and directs investigators to interview the people who can confirm or refute it (Department of the Army, 2005). Guilty suspects do offer false alibis, and friends or relatives sometimes support them. In a laboratory study, witnesses were willing to give false corroboration for a suspect who seemed innocent, whether or not they liked the suspect (Marion & Burke, 2013). A changed alibi can be the product of memory, of a lie, or of a guilty suspect correcting a lie once it fails, and the research does not tell an investigator which applies in a given case.
Confessions, Alibi Witnesses, and the Order of Investigation
A confession can affect the witnesses who might support an alibi. In a laboratory study, participants who had spent time with a confederate initially corroborated her alibi that she had never left the room. When told she had denied a theft, 95% maintained their support; when told she had confessed, only 45% did, and only 20% did when the experimenter suggested that supporting her alibi might imply their own involvement (Marion et al., 2016). The authors suggested that a confession can strip an innocent person of exculpatory evidence. The study used students in a brief, low-stakes setting, and the pressure applied was mild compared with a real investigation, so its numbers should not be carried directly into the field.
Archival research on wrongful convictions points in the same direction. Analyses of DNA exoneration cases found that false-confession cases often contained additional errors, such as mistaken identifications and informant testimony, and that the confession was usually obtained first (Kassin et al., 2012). That pattern is discussed further in “Investigator Bias, Tunnel Vision, and Forensic Confirmation Bias.” For alibis, one practical recommendation concerns sequence. Trainum (2016), a former homicide detective, recommends that investigators verify a suspect’s alibi before an accusatory interrogation rather than after, and cautions against questioning alibi witnesses with the aim of discrediting them. An alibi checked promptly and neutrally either clears an innocent suspect early or, if it fails, gives the prosecution stronger evidence than an alibi challenged only after a confession.
Hypothetical Example. Consider a hypothetical burglary investigation in which a suspect, interviewed nine days after the offense, says he spent the evening at a weekly class. Records later show the class was cancelled that night. Confronted with the records, he becomes unsure and then says he may have been at home. A detective treats the change as evidence of a fabricated alibi. Further inquiry finds that his phone’s location data place him at a relative’s house across town during the relevant hours. The example illustrates the research on schema reliance: a person asked about a routine evening may report what he usually does, with confidence, and be wrong. It also shows why independent records, sought promptly, are more informative than the consistency of the suspect’s account.
Hypothetical Example. Consider a hypothetical assault investigation in which a suspect says she was at work during the incident and a coworker confirms it. Badge records and the employer’s surveillance system show that she left the building forty minutes before the assault and returned an hour later, and the coworker, when interviewed separately with the records in hand, says she assumed the suspect was present because her car was in the lot. The example illustrates the other side of the research: alibi corroborators can be mistaken or willing to help, and physical records can expose a false alibi as readily as they can confirm a true one.
Limits of the Evidence
Most alibi research uses students or community volunteers who are asked to imagine being suspected, or who are falsely accused of a mock crime with no real consequence. The Dutch studies with large community samples and remand prisoners are important steps toward realism, but no study has followed real suspects’ alibis through investigation and trial with known ground truth. Field data on how often innocent suspects’ alibis fail, or how often guilty suspects’ false alibis are exposed, are not available. Effect sizes vary, and many findings rest on a small number of laboratories.
The confession and alibi-witness study (Marion et al., 2016) is a single paradigm with student participants, and the archival exoneration data show associations within a selected set of known errors, not how often confessions cause alibi evidence to collapse. Sex and gender were not a focus of the studies reviewed here: the abstracts and samples did not report sex differences in alibi accuracy, consistency, or evaluation, and no conclusion about women or men as alibi providers can be drawn from them. Group research can show that a weak or changed alibi is common among innocent people; it cannot show whether a particular alibi is true.
Questions to Ask About This Research
When alibi studies are cited, ask who generated the alibis (students, community volunteers, or prisoners), how long after the target period they were asked, whether the accusation had any consequence, and whether the study measured accuracy against records or only consistency between tellings. Ask whether the alibi in the case was checked before or after the suspect confessed or was identified, whether alibi witnesses were interviewed separately and neutrally, and whether physical records were sought while they still existed. When detection studies are cited, ask what accuracy observers achieved and whether that figure was better than chance. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.
Alibis and the Law
Federal and Military Standards
An alibi is not an affirmative defense in the sense of shifting any burden to the accused; the prosecution must prove every element of the offense, including that the accused committed it, beyond a reasonable doubt (In re Winship, 1970). Procedural rules nonetheless require notice. Under Federal Rule of Criminal Procedure 12.1, the government may make a written demand stating the time, date, and place of the offense, and the defense must then disclose each place the defendant claims to have been and the witnesses on whom the defense will rely; the government must in turn disclose its rebuttal witnesses, and the court may exclude the testimony of an undisclosed witness, though not the defendant’s own testimony. The Supreme Court upheld notice-of-alibi requirements in Williams v. Florida (1970) and held in Wardius v. Oregon (1973) that due process requires reciprocal discovery when the defense must give notice.
In courts-martial, R.C.M. 701(b)(2) requires the defense to notify trial counsel in writing, before the beginning of trial on the merits, of its intent to offer an alibi, including the place or places the accused claims to have been and the names and addresses of supporting witnesses (Manual for Courts-Martial, 2024 ed.). Under R.C.M. 701(g)(3), the military judge may order discovery, grant a continuance, exclude undisclosed evidence, or enter another just order for noncompliance, without limiting the accused’s right to testify. The Military Judges’ Benchbook alibi instruction tells members that the burden remains on the government to prove that the accused was present and committed the offense (Department of the Army, 2025, Instruction 5-3). Expert testimony about alibi research, where offered, is governed by Fed. R. Evid. 702 as amended Dec. 1, 2023, and by Mil. R. Evid. 702; the 2023 federal amendment to Rule 702 applies to the military rule from June 1, 2025, 18 months after its December 1, 2023 effective date, under Mil. R. Evid. 1102(a) unless the President acts otherwise (Joint Service Committee on Military Justice, 2025). Kumho Tire Co. v. Carmichael (1999) extends the gatekeeping obligation to all expert testimony.
Arizona
Arizona’s disclosure rules require the defense to give written notice of the defenses it intends to raise at trial, including alibi, and to disclose the names of witnesses it intends to call (Ariz. R. Crim. P. 15.2(b)). The State bears the burden of proving beyond a reasonable doubt that the defendant committed the offense, and an alibi need only raise a reasonable doubt. Expert testimony on alibi research would be governed by Ariz. R. Evid. 702, which Arizona conformed to the federal Daubert standard in 2012 and amended effective Jan. 1, 2024 (Order R-23-0004). Under State v. Lindsey (1986) and State v. Moran (1986), an expert may describe general research on memory and the evaluation of accounts but may not offer an opinion on whether a particular witness or defendant is telling the truth. Arizona’s rules on confessions and the jury’s consideration of voluntariness are covered in “Arizona Confession Law.”
Special Issues in Legal Practice
The research described above is the same whichever party relies on it. 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 the prosecution’s reliance on alibi evidence that has been checked against independent records. Physical evidence such as phone, badge, transaction, or video records is weighted heavily by evaluators (Olson & Wells, 2004), and when such records contradict an alibi, they are more probative than the suspect’s inconsistency alone. Witnesses sometimes give false corroboration (Marion & Burke, 2013), and observers cannot reliably detect a false alibi from its content (Nieuwkamp et al., 2018), which is a reason to rely on records rather than impressions. An alibi that was checked promptly, documented, and investigated with the possibility of confirming it in mind is less open to the concerns the research raises.
The research also counsels caution. A changed or vague alibi was common among innocent participants in the studies reviewed (Strange et al., 2014; Leins & Charman, 2016), so the research cannot support the assumption that inconsistency alone distinguishes a false alibi from a true one; whether a change reflects memory, error, or deception is for the factfinder to decide on all the evidence. If alibi witnesses withdrew their support only after learning of a confession, the defense may point to one laboratory experiment with students in which learning of a confession reduced corroboration (Marion et al., 2016), which cannot show that it did so for these witnesses. Delays in checking an alibi, or interviews of alibi witnesses, by any party, that were confrontational or not recorded, leave gaps in the record that the research makes relevant.
Special Issues for Defense
The research supports explaining to the factfinder why a truthful alibi may be weak, mistaken, or revised. Innocent people often lack physical corroboration for ordinary days (Nieuwkamp et al., 2016; Olson & Charman, 2012), rely on their usual routines when reconstructing a particular day (Leins & Charman, 2016), and change details on retelling (Strange et al., 2014). Strangers who saw the accused may not recognize him or her (Charman et al., 2017). Defense counsel are often best positioned to gather records quickly, before routine retention periods expire, and to ask when and how the alibi was checked and whether alibi witnesses were told of a confession first.
The research also counsels caution. Guilty suspects give false alibis, and friends and relatives do sometimes support them; jurors tend to discount alibi witnesses with close relationships to the accused (Culhane & Hosch, 2004). The memory studies show that inconsistency is common among truthful people, not that a particular inconsistent alibi is true, and an alibi contradicted by reliable records is not rescued by group research. An alibi disclosed late, or one that shifts to fit the prosecution’s evidence, warrants scrutiny, and notice rules carry consequences for noncompliance.
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
Charman, S. D., Reyes, A., Villalba, D. K., & Evans, J. R. (2017). The (un)reliability of alibi corroborators: Failure to recognize faces of briefly encountered strangers puts innocent suspects at risk. Behavioral Sciences & the Law, 35(1), 18–36. https://doi.org/10.1002/bsl.2264
Crozier, W. E., Strange, D., & Loftus, E. F. (2017). Memory errors in alibi generation: How an alibi can turn against us. Behavioral Sciences & the Law, 35(1), 6–17. https://doi.org/10.1002/bsl.2273
Culhane, S. E., & Hosch, H. M. (2004). An alibi witness’ influence on mock jurors’ verdicts. Journal of Applied Social Psychology, 34(8), 1604–1616. https://doi.org/10.1111/j.1559-1816.2004.tb02789.x
Culhane, S. E., Kehn, A., Horgan, A. J., Meissner, C. A., Hosch, H. M., & Wodahl, E. J. (2013). Generation and detection of true and false alibi statements. Psychiatry, Psychology and Law, 20(4), 619–638. https://doi.org/10.1080/13218719.2012.729018
Department of the Army. (2005). Law enforcement investigations (Field Manual 3-19.13).
Department of the Army. (2025). Military judges’ benchbook (DA Pamphlet 27-9).
Dysart, J. E., & Strange, D. (2012). Beliefs about alibis and alibi investigations: A survey of law enforcement. Psychology, Crime & Law, 18(1), 11–25. https://doi.org/10.1080/1068316X.2011.562867
Eastwood, J., Snow, M., Crough, Q., Han, T., Snook, B., Gregory, M., Fallon, L., & Lively, C. (2026). Your alibi better not be a-changin’: The effect of alibi change and interview strategy on perceptions of alibi witnesses. Psychology, Crime & Law, 32(6), 898–922. https://doi.org/10.1080/1068316X.2024.2441822
Gudjonsson, G. H. (2018). The psychology of false confessions: Forty years of science and practice. Wiley.
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
Kassin, S. M., Bogart, D., & Kerner, J. (2012). Confessions that corrupt: Evidence from the DNA exoneration case files. Psychological Science, 23(1), 41–45. https://doi.org/10.1177/0956797611422918
Leins, D. A., & Charman, S. D. (2016). Schema reliance and innocent alibi generation. Legal and Criminological Psychology, 21(1), 111–126. https://doi.org/10.1111/lcrp.12035
Leo, R. A. (2008). Police interrogation and American justice. Harvard University Press.
Marion, S. B., & Burke, T. M. (2013). False alibi corroboration: Witnesses lie for suspects who seem innocent, whether they like them or not. Law and Human Behavior, 37(2), 136–143. https://doi.org/10.1037/lhb0000021
Marion, S. B., Kukucka, J., Collins, C., Kassin, S. M., & Burke, T. M. (2016). Lost proof of innocence: The impact of confessions on alibi witnesses. Law and Human Behavior, 40(1), 65–71. https://doi.org/10.1037/lhb0000156
Nieuwkamp, R., Horselenberg, R., & van Koppen, P. J. (2016). The illusion of the perfect alibi: Establishing the base rate of non-offenders’ alibis. Journal of Investigative Psychology and Offender Profiling, 14(1), 23–42. https://doi.org/10.1002/jip.1470
Nieuwkamp, R., Horselenberg, R., & van Koppen, P. J. (2018). True and false alibis among prisoners and their detection by police detectives. Psychiatry, Psychology and Law, 25(6), 902–921. https://doi.org/10.1080/13218719.2018.1482570
Olson, E. A., & Charman, S. D. (2012). ‘But can you prove it?’ – Examining the quality of innocent suspects’ alibis. Psychology, Crime & Law, 18(5), 453–471. https://doi.org/10.1080/1068316X.2010.505567
Olson, E. A., & Wells, G. L. (2004). What makes a good alibi? A proposed taxonomy. Law and Human Behavior, 28(2), 157–176. https://doi.org/10.1023/B:LAHU.0000022320.47112.d3
Strange, D., Dysart, J., & Loftus, E. F. (2014). Why errors in alibis are not necessarily evidence of guilt. Zeitschrift für Psychologie, 222(2), 82–89. https://doi.org/10.1027/2151-2604/a000169
Trainum, J. L. (2016). How the police generate false confessions: An inside look at the interrogation room. Rowman & Littlefield.
Legal Authorities
In re Winship, 397 U.S. 358 (1970).
Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).
State v. Lindsey, 149 Ariz. 472, 720 P.2d 73 (1986).
State v. Moran, 151 Ariz. 378, 728 P.2d 248 (1986).
Wardius v. Oregon, 412 U.S. 470 (1973).
Williams v. Florida, 399 U.S. 78 (1970).
Fed. R. Crim. P. 12.1.
Fed. R. Evid. 702 (as amended Dec. 1, 2023).
Ariz. R. Crim. P. 15.2.
Ariz. R. Evid. 702 (as amended effective Jan. 1, 2024, Order R-23-0004).
Manual for Courts-Martial, United States, R.C.M. 701(b)(2), 701(g)(3) (2024 ed.).
Manual for Courts-Martial, United States, Mil. R. Evid. 702, 1102 (2024 ed.; Mil. R. Evid. 702 as amended 2025).
Related topic papers: Investigator Bias, Tunnel Vision, and Forensic Confirmation Bias · Internalized False Confessions and False Memories of Committing a Crime · Interrogative Suggestibility and Compliance · Corroboration and Assessing the Reliability of a Confession · Risk Factors for False Confessions: An Overview
Related topics from Arizona Forensic Services: Memory in the Courtroom · Jurors and Jury Decision-Making