Skip to content
Search resources Open menu

False Confessions

Confessions at Trial and After Conviction

Published

Snap Shot

  • The question: Once a confession has been admitted, how is its reliability presented and contested at trial, how is it reviewed on appeal, and what routes exist after conviction when its truth is disputed?
  • What the research shows: Confessions strongly influence jurors, who discount coerced confessions only partly; in a sample of DNA exonerations, false confessions were often followed by other errors in forensic and witness evidence, a pattern the authors interpreted as the confession influencing later evidence; and most proven false confessors who went to trial in the studied samples were convicted. Complete recordings and independent corroboration allow factfinders to evaluate a confession on its merits.
  • What it does not show: These findings come mostly from mock jurors and from selected samples of proven false confessions; they do not provide a rate of wrongful conviction among confession cases, and they cannot show that a particular confession or verdict is wrong.
  • The law: A defendant may present the circumstances of the interrogation to the jury (Crane v. Kentucky, 1986), and federal, military, and Arizona law require the factfinder to give a confession the weight it deserves; erroneous admission of a coerced confession is reviewed for harmless error (Arizona v. Fulminante, 1991). After conviction, relief depends on newly discovered evidence, DNA testing statutes, and actual-innocence standards (Fed. R. Crim. P. 33; 18 U.S.C. § 3600; Article 73, UCMJ; Ariz. R. Crim. P. 32.1(e), (h); A.R.S. § 13-4240).
  • Issues for each side: For prosecutors, the complete recording and independent corroboration let factfinders evaluate a confession on its merits; for defense counsel, the law protects evidence about the interrogation, and post-conviction relief requires genuinely new evidence rather than reargument.

A ruling that a confession is admissible ends the suppression stage but not the dispute. At trial, the jury decides what weight the confession deserves, and in Arizona it decides voluntariness again. On appeal, a court may find that the confession should not have been admitted and then ask whether the error affected the verdict. After the appeal, new evidence, such as DNA results, an alternative suspect, or proof that the confession’s details came from the investigation, may support a motion for a new trial or a post-conviction petition.

This paper describes how confession evidence is presented and contested at trial, what research shows about its effect on the rest of the case and on factfinders, and the appellate and post-conviction routes in federal, military, and Arizona courts. The admissibility stage is treated in Voluntariness and Suppression of Confessions, juror research in detail in How Jurors Respond to Confession Evidence, and expert testimony in Expert Testimony on Interrogations and False Confessions.

How a Confession Shapes the Rest of the Case

A confession often arrives early and changes what follows. In an analysis of DNA exoneration cases, false-confession cases were more likely than eyewitness cases to contain additional errors, most often invalid or improper forensic science, then mistaken identifications and informant testimony, and where several errors were present the confession usually came first (Kassin et al., 2012). The authors interpreted this pattern as evidence that a confession can influence the analysts, witnesses, and informants who learn of it. The study examined only cases later shown to be wrongful convictions; in cases where the confession was true, later evidence that agrees with it is expected, so agreement alone does not show influence. What matters is whether the later evidence was developed independently of knowledge of the confession, a question discussed in Investigator Bias, Tunnel Vision, and Forensic Confirmation Bias.

A confession also affects whether a case goes to trial. Most criminal cases are resolved by plea, and a defendant whose confession has been ruled admissible faces strong pressure to plead. In an analysis of the exonerations catalogued by the National Registry, wrongful convictions by guilty plea were about four times more likely than wrongful convictions at trial to involve a false confession (Redlich et al., 2025). Pleas are discussed in False Guilty Pleas.

Presenting and Contesting Reliability at Trial

The Constitution protects the defendant’s ability to challenge a confession’s reliability before the jury. In Crane v. Kentucky (1986), the Supreme Court held that excluding testimony about the length of the interrogation and the manner in which it was conducted violated the defendant’s right to present a defense, because the physical and psychological environment that produced a confession bears on its credibility even after the judge has found it voluntary. Federal law requires that the jury hear relevant evidence on voluntariness and be instructed to give the confession the weight it deserves (18 U.S.C. § 3501(a)); Military Rule of Evidence 304(g) and A.R.S. § 13-3988(A) do the same.

The rules of completeness also matter. Under Federal Rule of Evidence 106, as amended in 2023, when a party introduces part of a statement, including an oral statement, the opposing party may require introduction of any other part that in fairness ought to be considered at the same time, over a hearsay objection. Military Rule of Evidence 304(h) allows the defense to introduce the remainder of a partly introduced confession. These rules address a common problem: a short excerpt, or an officer’s summary of a long interview, may omit the questions, disclosures, and assurances that preceded the admission. A complete recording allows both sides and the factfinder to see the sequence, and it protects investigators against inaccurate claims as well as suspects (Kassin et al., 2025; see Electronic Recording of Interrogations).

Corroboration rules operate at trial as a threshold. Federal law requires independent evidence tending to establish the trustworthiness of the statement (Opper v. United States, 1954; Smith v. United States, 1954); Mil. R. Evid. 304(c) requires independent evidence tending to establish trustworthiness; and Arizona requires corroborating evidence from which jurors could reasonably infer that the crime occurred (State v. Carlson, 2015). The research explains why the source of the details matters more than their number: in known false confessions, the confessions often contained accurate, apparently non-public details later traced to the investigation (Garrett, 2010), while undisclosed details that were independently verified, or new information that led to evidence, are strong signs of reliability. These issues are discussed in Corroboration and Assessing the Reliability of a Confession.

Instructions vary. Federal and military instructions tell jurors to give the statement the weight it deserves; the military model instruction adds that members should consider the specific evidence, their common sense and knowledge of human nature, and the corroborating evidence (Department of the Army, 2025, Instruction 4-1). Arizona’s instruction requires jurors to find voluntariness beyond a reasonable doubt and treats any direct or implied promise, however slight, as making a statement involuntary (RAJI (Criminal) Standard 19, 2022). Some states give special instructions when an interrogation was not recorded (Commonwealth v. DiGiambattista, 2004).

What Factfinders Do With Confessions

Juror research is summarized in the jurors paper; three findings bear most directly on trial. First, confessions are powerful. A meta-analysis of 83 samples found that confessions increased mock jurors’ guilty verdicts, and that coerced confessions were partly discounted compared with uncoerced ones but still increased convictions compared with no confession (Mindthoff et al., 2024). Second, jurors may be influenced by a confession they believe should not count: in one experiment, a high-pressure confession increased guilty verdicts even when mock jurors judged it involuntary and said it did not affect them (Kassin & Sukel, 1997). Third, in samples of proven false confessions, most defendants who went to trial were convicted: 73% in one sample and 81% (30 of 37) in another (Leo & Ofshe, 1998; Drizin & Leo, 2004). Those samples were selected because the confessions were later shown to be false, so the figures describe the outcomes of known false confessors, not the accuracy of juries in confession cases generally.

The research also identifies conditions under which factfinders evaluate confessions more carefully. In a 2026 experiment with a contaminated confession, 38% of mock jurors who received neither questions directing attention to contamination nor an expert’s explanation voted guilty, compared with 17% to 22% of those who received either or both (Vornhagen & Hart, 2026). In a study of 132 judges, the judges recognized a high-pressure confession as coerced and set it aside when required to perform a harmless-error analysis, although the improper confession still increased their conviction rate when they judged guilt directly (Wallace & Kassin, 2012). These findings suggest that structured attention to specific features of a confession may help; they come from single studies and do not show how any particular jury or judge will respond.

Appeal and Harmless Error

An appellate court reviews a suppression ruling with deference to the trial court’s findings of fact. If it concludes that an involuntary confession was admitted, it asks whether the government has shown beyond a reasonable doubt that the error did not contribute to the verdict (Arizona v. Fulminante, 1991). Fulminante held that this harmless-error rule applies to coerced confessions, while stating that a confession “is like no other evidence” and directing reviewing courts to exercise extreme caution before finding its admission harmless (499 U.S. at 296); on the facts of that Arizona case, the Court held that the error was not harmless. The juror and judge research relates to that judgment, because it suggests that a confession can influence decision-makers even when they believe they have set it aside (Kassin & Sukel, 1997; Wallace & Kassin, 2012); those were a mock-juror experiment and a study of judges, and the harmless-error determination is a legal one for the court. In courts-martial, the C.A.A.F. applies a similar analysis, asking whether a nonconstitutional error had a substantial influence on the findings or, for constitutional error, whether it was harmless beyond a reasonable doubt (United States v. Kohlbek, 2019).

After Conviction

Post-conviction routes require new evidence or a constitutional violation, and the standards are demanding by design, reflecting the value the law places on the finality of a verdict reached at a fair trial.

In federal court, a motion for a new trial based on newly discovered evidence must be filed within three years after the verdict (Fed. R. Crim. P. 33(b)(1)). A federal prisoner who asserts actual innocence under penalty of perjury may obtain DNA testing if the statutory conditions are met, including that testing may produce new material evidence raising a reasonable probability that the applicant did not commit the offense (18 U.S.C. § 3600). The Supreme Court has held that there is no freestanding constitutional right to post-conviction DNA testing, and that a state’s procedures satisfy due process unless they are fundamentally inadequate (District Attorney’s Office v. Osborne, 2009). A claim of actual innocence alone does not entitle a state prisoner to federal habeas relief absent an independent constitutional violation (Herrera v. Collins, 1993), but a petitioner who shows that it is more likely than not that no reasonable juror would have convicted in light of new evidence may pass through a procedural bar to have constitutional claims heard (Schlup v. Delo, 1995).

In Arizona, Rule 32.1(e) permits relief for newly discovered material facts that probably would have changed the verdict, if the defendant was diligent and the facts are not merely cumulative or impeaching; Rule 32.1(h) permits relief on clear and convincing evidence that no reasonable fact-finder would have found the defendant guilty beyond a reasonable doubt; and A.R.S. § 13-4240 requires DNA testing when there is a reasonable probability that exculpatory results would have prevented prosecution or conviction and the other conditions are met. In the military, a petition for a new trial based on newly discovered evidence or fraud on the court-martial may be filed with the Judge Advocate General within three years after entry of judgment, and newly discovered evidence must probably produce a substantially more favorable result; such a petition is not available on newly discovered evidence where the conviction rested on a guilty plea (Article 73, UCMJ; R.C.M. 1210).

A confession can be a particular obstacle after conviction. One account describes cases in which a confession was treated as a reason to deny post-conviction DNA testing or to discount exculpatory results (Kassin, 2022). Exonerations involving false confessions remain a minority of known exonerations, about 13% in the national registry (Kassin et al., 2025; National Registry of Exonerations, 2025), and that figure is a share of known wrongful convictions, not a rate among confession cases.

Case Example. In People v. Rivera (2011), an Illinois defendant had been convicted twice of a 1992 murder after signing a confession. DNA testing in 2005 excluded him as the source of the sperm recovered from the victim, his conviction was vacated, and a jury convicted him a third time after the State argued that the sample had been contaminated or had another source unrelated to the crime. The Illinois Appellate Court reversed outright. It held that the State’s theories were highly improbable on the record, that no physical evidence linked the defendant to the crime, and that the evidence was insufficient to prove guilt beyond a reasonable doubt. The court also stated that DNA evidence does not automatically outweigh all other evidence. The case illustrates how a confession can sustain a prosecution after exculpatory scientific evidence, and that the outcome turned on the whole record.

Hypothetical Example. Consider a hypothetical defendant convicted of a burglary-homicide on a recorded confession that described where a tool was discarded, a detail the interviewers had not disclosed and that led them to the tool. Years later, the defendant petitions for DNA testing of material found under the victim’s fingernails, asserting innocence. The testing produces a profile matching the defendant. The example shows that post-conviction testing can confirm a confession as well as undermine one, and that a confession corroborated by undisclosed details and independent evidence is the kind most likely to withstand later review.

Limits of the Evidence

The juror findings come mostly from mock jurors reading or watching summaries, often without deliberation, and few studies involve judges. The field data on trial outcomes and on the corruptive effect of confessions come from proven false confessions and DNA exonerations, a selected sample in which every confession was false; they do not show how often confessions corrupt other evidence or how often juries err in confession cases generally. Exoneration figures reflect cases that were reinvestigated, which depends on the availability of DNA evidence, legal resources, and the seriousness of the charge. A 2026 analysis estimated that the probability that lawful interrogation tactics produce a false-confession wrongful conviction is low, clustering near 1% (Mourtgos & Adams, 2026a); because known exonerations are not a complete count of wrongful convictions, neither that estimate nor the exoneration figures can be checked directly. Most known false confessors are men; the juror studies cited here mostly did not analyze juror sex, and the meta-analysis did not report a sex moderator (Mindthoff et al., 2024).

Questions to Ask About This Research

When conviction rates for proven false confessors are cited, ask how the sample was selected and whether the figure describes juries in general or only cases later shown to be wrong. When the corruptive-effect study is cited, ask when each piece of later evidence was developed, who knew of the confession, and whether the evidence was obtained independently. When mock-juror studies are cited, ask whether participants deliberated, what they saw, and whether any instruction resembled the one in this case. When post-conviction relief is sought, ask what evidence is new, whether it was available with diligence, and how it relates to the specific details of the confession. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.

Confessions at Trial, After Conviction, and the Law

Federal and Military Standards

Jackson v. Denno (1964) requires a reliable determination of voluntariness by the judge, and Lego v. Twomey (1972) permits that determination by a preponderance. Crane v. Kentucky (1986) protects the presentation of the interrogation’s circumstances at trial, and 18 U.S.C. § 3501(a) requires a weight instruction. Fed. R. Evid. 106 governs completeness, and Fed. R. Evid. 702 as amended December 1, 2023, with Kumho Tire extending gatekeeping to all expert testimony, governs expert testimony on confessions. Opper and Smith govern corroboration. Fulminante governs harmless error. Post-conviction relief is governed by Fed. R. Crim. P. 33, 18 U.S.C. § 3600, Herrera, Schlup, and Osborne. In courts-martial, Mil. R. Evid. 304(c), (g), and (h) govern corroboration, weight, and completeness; expert testimony is governed by Mil. R. Evid. 702, to which the 2023 federal amendment applies from June 1, 2025 under Mil. R. Evid. 1102(a) (Joint Service Committee on Military Justice, 2025); and new-trial petitions by Article 73 and R.C.M. 1210. Military law is described in Confessions in Courts-Martial: Article 31(b), Mil. R. Evid. 304 and 305, and Expert Assistance.

Arizona

A.R.S. § 13-3988(A) requires that the jury hear evidence on voluntariness and receive a weight instruction, and RAJI (Criminal) Standard 19 requires jurors to find voluntariness beyond a reasonable doubt. Corpus delicti requires a reasonable inference that the crime occurred (Carlson, 2015). Expert testimony is governed by Ariz. R. Evid. 702 as amended effective January 1, 2024 (Order R-23-0004). Post-conviction relief is available under Ariz. R. Crim. P. 32.1(e) and (h), and DNA testing under A.R.S. § 13-4240. Fulminante itself arose from an Arizona prosecution. Arizona law is described 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 presenting a confession through its complete recording and its independent corroboration. Accurate details that were not disclosed to the suspect, new information that led to evidence, and forensic or witness evidence developed without knowledge of the confession give jurors a principled basis to rely on it, and the post-conviction standards, including Herrera, Schlup, Rule 33’s time limit, and Rule 32.1’s diligence and materiality requirements, protect verdicts reached at fair trials from reargument. Most confessions are presumed to be true; the proportion that are false is unknown, and no study measures it directly. The conviction-rate and corruption findings come from selected samples of wrongful convictions.

The research also counsels caution. Confessions influence jurors strongly, coerced confessions are discounted only partly, and Fulminante calls for extreme caution before treating a confession’s erroneous admission as harmless, so the strength of the independent evidence matters on appeal. An excerpt or a summary can leave unclear what was disclosed before the admission, and Rule 106 and Mil. R. Evid. 304(h) allow the remainder to be introduced. After conviction, new exculpatory evidence has to be weighed against the confession on the whole record; Rivera illustrates a case in which the theories offered to reconcile a confession with excluding DNA results did not survive review.

Special Issues for Defense

The law and the research support a full presentation of the interrogation at trial. Crane, § 3501(a), Mil. R. Evid. 304(g), and § 13-3988 protect evidence of the interrogation’s circumstances; Fed. R. Evid. 106 and Mil. R. Evid. 304(h) address excerpts; and the research on contamination and on the corruptive effect of confessions (Garrett, 2010; Kassin et al., 2012) explains why the source of the details and the independence of the other evidence matter. After conviction, DNA statutes, Rule 33, Rule 32.1(e) and (h), and Article 73 provide routes when evidence is genuinely new.

The research also counsels caution. In a meta-analysis of mock-juror studies, expert testimony reduced the likelihood of conviction whether or not the confession was coerced (described as a skepticism effect), while jury instructions reduced convictions for coerced but not uncoerced confessions (Mindthoff et al., 2024); the authors conclude that safeguards are needed but that more research is needed to identify the most effective instructions and expert testimony. Whether presentations connecting specific features of the confession, such as what was disclosed, what was wrong, and what was not corroborated, to the record improve discrimination has not been established, although one 2026 experiment with a contaminated confession found fewer guilty votes when attention was directed to such features (Vornhagen & Hart, 2026). The selected-sample figures on false confessors do not show that a particular confession is false. Post-conviction standards are demanding, require diligence and materiality, and are not a forum for reargument, and new testing may confirm rather than undermine a confession. Post-conviction petitions rest on newly discovered evidence, so evidence specific to the confession’s details is more likely to satisfy those standards than general research on interrogation methods.

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

Department of the Army. (2025). Military judges’ benchbook (DA Pam 27-9). Headquarters, Department of the Army.

Drizin, S. A., & Leo, R. A. (2004). The problem of false confessions in the post-DNA world. North Carolina Law Review, 82(3), 891–1007.

Garrett, B. L. (2010). The substance of false confessions. Stanford Law Review, 62(4), 1051–1118.

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. (2022). Duped: Why innocent people confess—and why we believe their confessions. Prometheus Books.

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

Kassin, S. M., Cleary, H. M. D., Gudjonsson, G. H., Leo, R. A., Meissner, C. A., Redlich, A. D., & Scherr, K. C. (2025). Police-induced confessions, 2.0: Risk factors and recommendations. Law and Human Behavior, 49(1), 7–53. https://doi.org/10.1037/lhb0000593

Kassin, S. M., & Sukel, H. (1997). Coerced confessions and the jury: An experimental test of the “harmless error” rule. Law and Human Behavior, 21(1), 27–46. https://doi.org/10.1023/A:1024814009769

Leo, R. A., & Ofshe, R. J. (1998). The consequences of false confessions: Deprivations of liberty and miscarriages of justice in the age of psychological interrogation. Journal of Criminal Law and Criminology, 88(2), 429–496. https://doi.org/10.2307/1144288

Mindthoff, A., Ferreira, P. A., & Meissner, C. A. (2024). The effect of confession evidence on jurors’ verdict decisions: A systematic review and meta-analysis. Law and Human Behavior, 48(3), 163–181. https://doi.org/10.1037/lhb0000563

Mourtgos, S. M., & Adams, I. T. (2026a). Recalibrating the risk of false confession wrongful convictions: Interrogation tactics and inverse probability. Journal of Criminal Justice, 103, Article 102600. https://doi.org/10.1016/j.jcrimjus.2026.102600

National Registry of Exonerations. (2025). 2024 annual report. https://exonerationregistry.org/sites/exonerationregistry.org/files/documents/2024_Annual_Report.pdf

Redlich, A. D., Wilford, M. M., DiPano, M., & Berger, N. (2025). Commonalities in false guilty plea cases. Psychology, Crime & Law, 31(1), 64–82. https://doi.org/10.1080/1068316X.2023.2213381

Vornhagen, J., & Hart, K. J. (2026). How evidence of confession contamination affects perceptions and judgments in a mock case. Psychology, Crime & Law. Advance online publication. https://doi.org/10.1080/1068316X.2026.2647334

Wallace, D. B., & Kassin, S. M. (2012). Harmless error analysis: How do judges respond to confession errors? Law and Human Behavior, 36(2), 151–157. https://doi.org/10.1037/h0093975

Legal Authorities

Arizona v. Fulminante, 499 U.S. 279 (1991).

Crane v. Kentucky, 476 U.S. 683 (1986).

District Attorney’s Office for the Third Judicial District v. Osborne, 557 U.S. 52 (2009).

Herrera v. Collins, 506 U.S. 390 (1993).

Jackson v. Denno, 378 U.S. 368 (1964).

Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).

Lego v. Twomey, 404 U.S. 477 (1972).

Opper v. United States, 348 U.S. 84 (1954).

Schlup v. Delo, 513 U.S. 298 (1995).

Smith v. United States, 348 U.S. 147 (1954).

United States v. Kohlbek, 78 M.J. 326 (C.A.A.F. 2019).

Commonwealth v. DiGiambattista, 442 Mass. 423, 813 N.E.2d 516 (2004).

People v. Rivera, 2011 IL App (2d) 091060.

State v. Carlson, 237 Ariz. 381, 351 P.3d 1079 (2015).

U.S. Const. amends. V, VI, XIV.

18 U.S.C. §§ 3501, 3600.

Uniform Code of Military Justice art. 73, 10 U.S.C. § 873.

Ariz. Rev. Stat. Ann. §§ 13-3988, 13-4240.

Fed. R. Crim. P. 33.

Fed. R. Evid. 106, 702 (as amended Dec. 1, 2023).

Ariz. R. Crim. P. 32.1.

Ariz. R. Evid. 702 (amended effective Jan. 1, 2024, Ariz. Sup. Ct. Order R-23-0004).

Manual for Courts-Martial, United States, Mil. R. Evid. 304; R.C.M. 1210 (2024 ed.).

Revised Arizona Jury Instructions (Criminal), Standard Criminal 19 (State Bar of Arizona 2022).

Related topic papers: How Jurors Respond to Confession Evidence · Voluntariness and Suppression of Confessions · Corroboration and Assessing the Reliability of a Confession · Contamination: Non-Public Details and Post-Admission Narratives · False Guilty Pleas · Expert Testimony on Interrogations and False Confessions

Related topics from Arizona Forensic Services: Jurors and Jury Decision-Making · Courts-Martial