Snap Shot
- The question: How does military law govern the questioning of service members and the use of their statements, and when may the defense obtain, or the members hear, an expert on interrogations and confessions?
- What the research shows: The general findings may be relevant to service members, although almost no research examines military interrogations: in civilian research, long sessions, implied leniency, false claims about evidence, and personal vulnerabilities are associated with a higher risk of unreliable statements, while recorded, well-documented interviews protect both the accused and the investigator. Rank and the duty to answer superiors may add pressures that Article 31(b) addresses; no study has measured their effect.
- What it does not show: Almost no research studies military interrogations specifically, and no study, test, or expert can establish whether a particular service member’s confession is true or false.
- The law: Article 31(b) requires warnings when a person subject to the UCMJ, acting in an official law-enforcement or disciplinary capacity, questions a suspect; Mil. R. Evid. 305 adds counsel and waiver rules and 304 governs voluntariness, corroboration, and weight; expert assistance turns on R.C.M. 703(d) and the necessity test of United States v. Gonzalez (1994), and expert testimony on Mil. R. Evid. 702, which took on the 2023 federal amendment in 2025.
- Issues for each side: Trial counsel benefit from complete recordings, clear advisements, and an accurate record of any polygraph session; defense counsel, from a specific showing of why expert help is needed, what it would do, and why counsel cannot do it.
Statements by the accused are central evidence in many courts-martial. Military law regulates them through constitutional rules, a statute older than Miranda, and detailed evidentiary rules. Article 31 of the Uniform Code of Military Justice (UCMJ) was enacted in 1950, sixteen years before Miranda v. Arizona, because Congress recognized that service members are trained to answer the questions of superiors and may feel obliged to answer even when they have a right not to.
Confession issues arise at several points: whether warnings were required, a motion to suppress before plea, corroboration, a defense request for an expert, the members’ assessment of weight, and appeal. This paper describes each, with the research that bears on it. The constitutional voluntariness test is treated in Voluntariness and Suppression of Confessions, Miranda research in Miranda Warnings, Waiver, and Comprehension, and the admissibility of expert testimony in civilian courts in Expert Testimony on Interrogations and False Confessions.
Article 31(b): Who Must Warn, and When
Article 31(b) provides that no person subject to the UCMJ may interrogate or request any statement from an accused or a suspect without first informing the person of the nature of the accusation, advising that the person need not make any statement about the offense, and advising that any statement may be used as evidence in a court-martial. The warnings do not include a right to counsel; counsel rights come from the Fifth and Sixth Amendments and are set out in Mil. R. Evid. 305(c)(2)–(3). Unlike Miranda, Article 31(b) is not limited to custodial interrogation. Since United States v. Tempia (1967), military courts have also applied Miranda‘s counsel warnings to custodial interrogation (see Mil. R. Evid. 305(b)(3)).
The key question in many cases is whether the questioner was acting in an official capacity. The Court of Appeals for the Armed Forces (C.A.A.F.) asks whether, on all the facts at the time, the questioner was acting, or could reasonably be considered to be acting, in an official law-enforcement or disciplinary capacity, judged from the position of a reasonable person in the suspect’s place (United States v. Jones, 2014). In Jones, the court expressly rejected the older second prong of United States v. Duga (1981), which had asked whether the suspect subjectively perceived the inquiry as more than casual conversation. Persons not assigned to investigate offenses ordinarily do not trigger Article 31(b) (United States v. Cohen, 2006), and undercover agents and informants do not, because a reasonable person would not feel compelled to answer them (Jones, 2014). Mixed-purpose questioning is decided on the totality of the circumstances (United States v. Ramos, 2017).
The two cases cut in different directions. In Jones, the court upheld admission of a statement made to an infantryman who was serving as a military police augmentee and was involved in the burglary investigation, because he had a personal reason for the conversation, had limited law-enforcement authority, and held no disciplinary role over the accused. In Ramos, the court held that the military judge abused his discretion in admitting statements to agents who had come to discuss threats to the accused and his wife: the agents suspected him of a UCMJ offense, chose not to advise him in order to keep him talking, and their questions were part of a law-enforcement investigation rather than an operational exigency.
Army investigative doctrine reflects the same structure. The military police field manual distinguishes an interview, an open-ended conversation used when guilt is unknown, from an interrogation, a planned and structured session used when the investigator has reason to believe the person committed the offense; it requires Article 31(b) warnings before questioning a suspect, prefers written sworn statements, and recommends video recording of interrogations so that claims of coercion can be answered (Department of the Army, 2005, ch. 4). Most confessions obtained this way are presumed to be true; the proportion that are false is unknown, no study measures it directly, and the research addresses the conditions that raise the risk when a confession is false.
Research supports the concern behind the statute. Compliance, the tendency to go along with requests from a person in authority despite private disagreement, is one of the personal factors associated with false confessions in field and laboratory research, although it is also present among people who confess truthfully and is measured by self-report instruments (Gudjonsson, 2018; Kassin et al., 2025). A setting in which the questioner outranks the suspect, or in which the suspect expects career consequences for refusing to cooperate, adds a source of pressure that civilian studies do not directly measure. That is a reason for care in applying the research, not evidence that statements made to superiors are unreliable; the law addresses the pressure through warnings.
Waiver, Counsel, and Civilian Questioning Under Mil. R. Evid. 305
A statement obtained in violation of Rule 305 is treated as involuntary under Rule 304 (Mil. R. Evid. 305(a)). A suspect may waive rights only freely, knowingly, and intelligently; no written waiver is required, but the suspect must affirmatively acknowledge understanding the rights, affirmatively decline counsel, and affirmatively consent to make a statement (Mil. R. Evid. 305(e)(1)). After a request for counsel in custodial interrogation, questioning must stop, and a later waiver is valid only if the prosecution shows that the suspect initiated the communication or was not continuously restrained in the interval (Mil. R. Evid. 305(c)(4), (e)(3)); after preferral, interrogation about the preferred charges requires counsel if the accused requests or has counsel (Mil. R. Evid. 305(c)(3)).
When civilian federal, state, or local officials question a person subject to the UCMJ, Rule 305(f)(1) applies the warning and waiver principles used in federal district courts, not Article 31(b). Statements obtained by foreign officials abroad do not require Article 31 or Miranda warnings unless U.S. military personnel or their agents conducted, instigated, or participated in the interrogation, but such statements remain inadmissible if obtained through coercion, unlawful influence, or unlawful inducement (Mil. R. Evid. 305(f)(2)).
The research on Miranda comprehension applies to rights advisements generally: most suspects waive, and comprehension is lower on average among people with intellectual disability and those who are intoxicated or distressed (Kassin et al., 2025), group-level findings that do not show a particular service member failed to understand a properly given advisement.
Voluntariness, Corroboration, and Weight Under Mil. R. Evid. 304
Article 31(d) bars statements obtained in violation of Article 31 or through coercion, unlawful influence, or unlawful inducement, and Mil. R. Evid. 304(a) applies that bar, and the Fifth Amendment, to the accused’s statements and to evidence derived from them. The defense must move before plea, and the prosecution must establish voluntariness by a preponderance (Mil. R. Evid. 304(f)). The courts apply a totality-of-the-circumstances test that considers both the characteristics of the accused and the details of the interrogation (United States v. Griffin, 1999). In United States v. Bresnahan (2005), a father’s statement that he may have shaken his infant son was held voluntary, even though a detective told him the doctors needed to know what happened in order to treat the child: the statement was accurate, she was not confrontational, and he was an experienced soldier who was not detained, isolated, or questioned at length. The court added that the detective’s appeal supplied no motive for a false admission.
Corroboration is a separate requirement. An admission or confession may be considered on guilt only if independent evidence has been admitted that would tend to establish its trustworthiness; the independent evidence need raise only an inference of truth, need not prove every element or every fact in the statement, and may not itself be another uncorroborated confession (Mil. R. Evid. 304(c)). The military judge alone decides when corroboration is sufficient, but the corroborating evidence should go to the members so that they can weigh the statement (Department of the Army, 2025, Instruction 4-1, Note 5). The research on contamination and corroboration explains why the source of a confession’s details matters: in known false confessions, accurate details often traced to what investigators disclosed, while accurate, non-public details that were not disclosed, or new information that led to evidence, are among the strongest indicators of reliability (Garrett, 2010; see Corroboration and Assessing the Reliability of a Confession).
If a statement is admitted, the defense may present evidence on voluntariness, and the members are instructed to give the statement the weight it deserves under all the circumstances (Mil. R. Evid. 304(g)). The model instruction tells members to consider the specific evidence, their common sense and knowledge of human nature, and the corroborating evidence (Department of the Army, 2025, Instruction 4-1, Note 3). The juror research summarized in How Jurors Respond to Confession Evidence bears on how well that instruction is likely to work; no study has examined court-martial members specifically.
Polygraph Examinations and Post-Test Interviews
Polygraph examinations are used in some military investigations, and an admission may follow a post-test interview in which the examiner discusses the results. Mil. R. Evid. 707 bars the results, the examiner’s opinion, and any reference to an offer to take, failure to take, or taking of a polygraph, but does not bar an otherwise admissible statement made during the examination. Courts have upheld confessions obtained after polygraph sessions; in United States v. Freeman (2008), a confession was held voluntary although the questioning lasted nearly ten hours, with breaks, and included two polygraph examinations and false claims that witnesses had seen the accused and that his fingerprints had been found.
Case Example. In United States v. Kohlbek (2019), a soldier waived his Article 31 rights, took a four-hour polygraph examination, was told his answers indicated deception, and then wrote a detailed admission. The military judge read Rule 707 to bar his evidence that the admission followed the polygraph, his discomfort during it, and the report of deception. The C.A.A.F. held that Rule 707 does not exclude evidence of the facts and circumstances of a polygraph procedure offered to explain why a person confessed, so military judges have discretion to admit it. It held the error harmless because the complainant’s detailed testimony and prompt reports made the government’s case strong. The case illustrates both that the circumstances of a confession may be presented and that other evidence can decide the outcome.
Telling a suspect that a test showed deception is a confrontation with evidence; it is lawful, and when the claim is overstated the research on false-evidence ploys suggests it can increase the risk of compliance or of self-doubt in vulnerable people (Kassin et al., 2025; see False Evidence Ploys and Police Deception). The model instruction now limits the members’ use of polygraph-related evidence to the weight of the statement (Department of the Army, 2025, Instruction 4-1, Note 4); the polygraph itself is discussed in The Pre-Interrogation Interview: Behavior Analysis, Investigator Judgment, and the Polygraph.
Expert Assistance
The government provides defense experts on a showing of need. Defense requests are made under service regulations, and after referral the defense may raise a request before the military judge, ex parte for a consultant. The judge decides whether an expert witness’s testimony is relevant and necessary and whether a consultant’s assistance is necessary for an adequate defense; if the motion is granted, the expert or an adequate substitute must be provided (R.C.M. 703(d)(2)). These provisions were revised by Executive Order 14103 (2023). Its Annex 1, effective July 28, 2023, required the services to issue funding regulations; its Annex 2, which took effect with the special trial counsel reforms on December 27, 2023, expanded the defense requests that may be raised before the judge (previously only a denied request) and added ex parte consultant motions. Earlier commentary, written when requests went first to the convening authority, describes the same standard and the adequate-substitute option (Stein & Younggren, 2019).
The C.A.A.F. requires more than the mere possibility of help. The accused must show a reasonable probability both that an expert would assist the defense and that denial would result in a fundamentally unfair trial, and the first part is assessed under three questions from United States v. Gonzalez (1994): why the assistance is needed, what it would accomplish for the accused, and why defense counsel cannot gather and present the evidence themselves (Bresnahan, 2005; Freeman, 2008). The right rests on the Sixth Amendment, Article 46, UCMJ, and R.C.M. 703(d) (United States v. Kreutzer, 2005), and parallels the principle of Ake v. Oklahoma (1985) that an indigent defendant is entitled to expert psychiatric assistance when mental condition is a significant factor at trial.
Case Example. In United States v. Bresnahan (2005), the defense sought a consultant on interrogation techniques to examine whether a detective’s methods at a hospital and a police station made the accused’s confession unreliable. The military judge denied it, finding no evidence that the confession was false or that the accused had a mental or emotional condition or a submissive personality making false admissions likely. The C.A.A.F. accepted that the expert knew more than counsel but found no abuse of discretion. It called the ruling a close call and said it would also have upheld a decision to grant the request. The case shows how much the outcome depends on the specificity of the showing and on the deference given to the military judge.
In Freeman (2008), the court likewise upheld denial of a consultant in police interrogations because counsel did not explain why they could not develop the knowledge and cross-examine the investigators themselves. Read together, the cases suggest that showings tied to concrete features of the record, such as the length of the interview, specific statements by the interviewer, or the accused’s documented condition, fit the Gonzalez questions better than general descriptions of the research; they do not establish that such requests will be granted.
Expert Testimony at Trial
An expert who testifies must meet Mil. R. Evid. 702 and the factors of United States v. Houser (1993): the expert’s qualifications, the subject matter, the basis for the opinion, legal relevance, reliability, and probative value under Rule 403. In Griffin (1999), the C.A.A.F. upheld exclusion of a psychologist’s testimony offered to show that a confession was involuntary, because the expert could not say the confession was false or coerced, only that it was consistent with a coerced-compliant type; relied in part on the accused’s account of the interrogation, which the judge found not credible; expressed reservations about the norms used; and shed little light on whether the confession was coerced. Commentary on military practice describes rulings on such experts as mixed and notes that an expert may not opine that a confession was false (Stein & Younggren, 2019). Service-court decisions exist but are often unpublished and should be read in full before reliance.
Limits of the Evidence
Almost all interrogation research comes from civilian settings. Laboratory studies use mostly college students in low-stakes tasks, and a 2026 meta-analysis argued that absolute laboratory confession rates cannot be generalized to real interrogations (Mourtgos & Adams, 2026b); a 2025 integrative review made a similar argument about ecological validity (Welner et al., 2025). The 2025 scientific review rests on converging case, field, and experimental evidence (Kassin et al., 2025). Exoneration data are a selected sample and do not provide a false-confession rate. No published study has measured how rank, the duty to obey, or career consequences affect true and false confession rates in military interrogations, and none has examined court-martial members. The key studies cited here did not report sex or gender differences relevant to military suspects; most exonerated false confessors are men.
Questions to Ask About This Research
When research on compliance or authority is cited, ask whether the questioner in this case held authority over the accused, what was actually said about rank, duty, or consequences, and whether the study involved anything comparable. When the research on false-evidence ploys is applied to a polygraph post-test interview, ask what the examiner actually told the accused and whether it was accurate. When an expert’s help is requested, ask what specific facts in this record the expert would address and what counsel could not do without the expert. When laboratory rates are cited, ask what the participants were accused of, what they stood to lose, and whether the critique of laboratory paradigms applies. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.
Courts-Martial Confessions and the Law
Federal and Military Standards
The Fifth Amendment’s privilege and Due Process Clause apply in courts-martial; the Fourteenth Amendment does not. Beyond the rules described above, a statement obtained in violation of Article 31 or Rule 305 may be used only to impeach by contradiction or in a later perjury or false-statement prosecution (Mil. R. Evid. 304(e)), and a guilty plea waives Rule 304 objections to that offense except as R.C.M. 910(a)(2) allows (Mil. R. Evid. 304(f)(8)). Expert testimony is governed by Mil. R. Evid. 702, Houser, and Daubert, and Kumho Tire extends gatekeeping to all expert testimony. The 2024 Manual prints the earlier text of Rule 702. Under Mil. R. Evid. 1102(a), amendments to the Federal Rules of Evidence amend the military rules by operation of law 18 months after they take effect unless the President acts otherwise, so the December 1, 2023 federal amendment, which requires the proponent to show that each requirement is more likely than not met, applies to Mil. R. Evid. 702 from June 1, 2025 (Joint Service Committee on Military Justice, 2025). Mil. R. Evid. 704 allows opinions on an ultimate issue, but the members are instructed that witnesses may not give opinions on credibility or guilt (Department of the Army, 2025, Instruction 7-26). A petition for a new trial on newly discovered evidence or fraud on the court-martial may be filed within three years after entry of judgment (Article 73, UCMJ; R.C.M. 1210); see Confessions at Trial and After Conviction.
Arizona
Service members in Arizona may be questioned by civilian police and prosecuted in either system. When a state or local officer questions a service member, a court-martial applies the federal district court standards to warnings and waiver (Mil. R. Evid. 305(f)(1)), while an Arizona prosecution applies Arizona law, including the State’s burden to prove voluntariness, A.R.S. § 13-3988, and the jury instruction that a statement induced by “any direct or implied promise, however slight” is involuntary (RAJI (Criminal) Standard 19, 2022); Arizona case law requires that the statement have been induced by the promise (State v. Blakley, 2003). 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 law and the research support many military confessions. Most confessions are presumed to be true; the proportion that are false is unknown, and no study measures it directly. Article 31(b) and Rule 305 give service members broader warnings than civilians receive; a clear, recorded advisement followed by an affirmative waiver gives the military judge a documented basis for deciding warning issues. Bresnahan, Freeman, and Jones show that courts have upheld statements obtained through accurate confrontation, lawful persuasion, and conversations outside a law-enforcement role, and Griffin, Bresnahan, and Freeman show that the C.A.A.F. defers to military judges who decline to admit general testimony or to fund experts when the necessity showing is not made. Accurate, non-public details and corroboration that meets Rule 304(c) give the members a principled basis to rely on a confession, and Kohlbek shows that strong independent evidence can make disputes about a confession’s circumstances harmless.
The cases also counsel caution. Ramos shows that statements may be excluded when agents who suspect an offense delay warnings to keep a member talking. After Kohlbek, the circumstances of a polygraph session may come before the members, so the examiner’s statements about results, the length of the session, and the accused’s condition should be documented and should be accurate. An unrecorded or partly recorded post-test interview makes it difficult to show what details were disclosed and what the accused supplied. Where the accused is young, impaired, or has a documented cognitive or psychiatric condition, a request for expert assistance is more likely to meet the Gonzalez test, and the C.A.A.F. described Bresnahan itself as a close call.
Special Issues for Defense
The research and the rules give the defense several bases to examine a statement. Article 31(b) applies without custody, the official-capacity test is objective, and Rule 305(e)(1) requires affirmative acknowledgment and consent. Rule 304(g) and the model instruction allow the members to hear the circumstances of the interrogation, and Kohlbek allows the circumstances of a polygraph session to be offered to explain why the accused confessed. The research on compliance, false-evidence ploys, long sessions, and personal vulnerability (Kassin et al., 2025; Gudjonsson, 2018) gives context to those facts, and Rule 304(c) requires independent evidence of trustworthiness before a confession can be considered at all.
The research also counsels caution. Gonzalez, Bresnahan, and Freeman require a concrete explanation of why expert help is needed, what it would accomplish, and why counsel cannot do it, and requests that described the literature in general terms were denied in Bresnahan and Freeman; service-court decisions are often unpublished and should be read in full. In Griffin, testimony that rested partly on the accused’s disputed account, and that could not address whether this confession was coerced, was excluded, and no expert may opine that a confession is false. The laboratory research is contested (Mourtgos & Adams, 2026b; Welner et al., 2025), no study addresses military interrogations directly, and the presence of a risk factor does not show that a statement is false. Requests and proffers tied to specific facts in the record or to documented characteristics of the accused are more likely to satisfy the Gonzalez questions and the Houser factors than those that describe the literature in general terms, but decisions vary by judge, and published decisions are a selected sample.
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. (2005). Law enforcement investigations (Field Manual No. 3-19.13). Headquarters, Department of the Army.
Department of the Army. (2025). Military judges’ benchbook (DA Pam 27-9). Headquarters, Department of the Army.
Garrett, B. L. (2010). The substance of false confessions. Stanford Law Review, 62(4), 1051–1118.
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., 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
Mourtgos, S. M., & Adams, I. T. (2026b). What do laboratory false confession paradigms measure? A calibration meta-analysis. Journal of Quantitative Criminology. Advance online publication. https://doi.org/10.1007/s10940-026-09687-1
Stein, C. T., & Younggren, J. N. (Eds.). (2019). Forensic psychology in military courts. American Psychological Association. https://doi.org/10.1037/0000141-000
Welner, M., DeLisi, M., & Janusewski, T. (2025). False confessions: An integrative review of the phenomenon. Behavioral Sciences & the Law, 43(2), 185–202. https://doi.org/10.1002/bsl.2707
Legal Authorities
Ake v. Oklahoma, 470 U.S. 68 (1985).
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).
General Electric Co. v. Joiner, 522 U.S. 136 (1997).
Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).
Miranda v. Arizona, 384 U.S. 436 (1966).
State v. Blakley, 204 Ariz. 429, 65 P.3d 77 (2003).
United States v. Bresnahan, 62 M.J. 137 (C.A.A.F. 2005).
United States v. Cohen, 63 M.J. 45 (C.A.A.F. 2006).
United States v. Duga, 10 M.J. 206 (C.M.A. 1981).
United States v. Freeman, 65 M.J. 451 (C.A.A.F. 2008).
United States v. Gonzalez, 39 M.J. 459 (C.M.A. 1994).
United States v. Griffin, 50 M.J. 278 (C.A.A.F. 1999).
United States v. Houser, 36 M.J. 392 (C.M.A. 1993).
United States v. Jones, 73 M.J. 357 (C.A.A.F. 2014).
United States v. Kohlbek, 78 M.J. 326 (C.A.A.F. 2019).
United States v. Kreutzer, 61 M.J. 293 (C.A.A.F. 2005).
United States v. Ramos, 76 M.J. 372 (C.A.A.F. 2017).
United States v. Tempia, 16 C.M.A. 629, 37 C.M.R. 249 (1967).
U.S. Const. amends. V, VI.
Uniform Code of Military Justice arts. 31, 46, 73, 10 U.S.C. §§ 831, 846, 873.
Ariz. Rev. Stat. Ann. § 13-3988.
Exec. Order No. 14103, 88 Fed. Reg. 50535 (Aug. 2, 2023).
Manual for Courts-Martial, United States, R.C.M. 703(d), 1210 (2024 ed.).
Manual for Courts-Martial, United States, Mil. R. Evid. 304, 305, 702, 704, 707, 1102 (2024 ed.).
Revised Arizona Jury Instructions (Criminal), Standard Criminal 19 (State Bar of Arizona 2022).
Related topic papers: Voluntariness and Suppression of Confessions · Miranda Warnings, Waiver, and Comprehension · Expert Testimony on Interrogations and False Confessions · The Pre-Interrogation Interview: Behavior Analysis, Investigator Judgment, and the Polygraph · Corroboration and Assessing the Reliability of a Confession · Arizona Confession Law
Related topics from Arizona Forensic Services: Courts-Martial