Skip to content
Search resources Open menu

False Confessions

Pretext Phone Calls and Recorded Confrontations

Published

Snap Shot

  • The question: When investigators arrange for a complainant, relative, or friend to call, message, or meet a suspect while the conversation is recorded, how should the resulting admissions, apologies, and denials be weighed, and what limits does the law place on the technique?
  • What the research shows: Pretext calls are an investigative tool used in sexual offense investigations, although one case-file sample found them underused (St. George, 2026), and they can capture statements in the suspect’s own words, with a complete record of what was said, including statements later confirmed by independent evidence. Review chapters describe callers using tactics similar to those of interrogators, such as presenting the evidence as settled, minimizing the act, and appealing to the relationship, and identify a motive formal interrogation lacks: a suspect may concede or apologize to calm or keep someone he cares about.
  • What it does not show: No published study measures how often pretext-call admissions are true or false, and exoneration registries do not count statements made to private persons as confessions. The only quantitative work consists of one study of police use, one descriptive program report on children’s calls, and unpublished mock-juror studies.
  • The law: A consenting party may record the call (18 U.S.C. § 2511(2)(c); A.R.S. § 13-3012(9)). Miranda does not apply when the suspect does not know he is speaking to the police (Illinois v. Perkins), but a police-directed call after charges are filed or preferred can violate the right to counsel (Maine v. Moulton; Mil. R. Evid. 305(c)(3)). In courts-martial, Article 31(b) can reach a questioner acting in an official capacity (Gilbreath), and failure to deny is not an admission once a service member is under investigation (Mil. R. Evid. 304(a)(2)).
  • Issues for each side: For prosecutors, the research bears on the value of a complete recording and of independent evidence that confirms what was admitted; for defense counsel, on which party introduced each fact and what pressure the caller applied. Neither an apology nor a denial or angry response shows guilt or innocence by itself.

In a pretext call, an investigator arranges for someone the suspect knows, usually the complainant or a family member, to telephone or message him while the conversation is recorded, in the hope that he will discuss the alleged offense. A recorded confrontation is the in-person version: the complainant or a relative meets the suspect while carrying a recording device. The technique is used in sexual offense investigations, where offenses often occur in private between people who know each other.

Pretext communications differ from police interrogation in ways that cut in both directions. The suspect is usually not in custody, can end the conversation, and speaks in his own words with a complete recording of both sides. But he does not know the police are involved, his relationship with the caller is at stake, and the caller has fewer legal limits than an officer would. The resulting statements can be among the most persuasive evidence at trial, and both sides ask the factfinder to read them in their favor. Placing a call can also be distressing for a complainant, particularly a child, and practice guidance puts the complainant’s safety first (Eisenga, 2004).

This paper describes how these communications are used, the research on what they produce and how it is perceived, and the governing federal, military, and Arizona law. It treats the call as covert questioning; the interpretation of apologies, non-denials, and text messages is addressed in “Pretext Calls, Apologies, and Post-Incident Messages” in the Counterintuitive Testimony resource.

How Pretext Calls and Recorded Confrontations Are Used

Police practice guidance, such as an International Association of Chiefs of Police training key, treats the call as a planned conversation, stresses the lawfulness of the recording, and advises against threats (Eisenga, 2004). The guidance is aimed at obtaining statements and offers no data on their accuracy. In one Arizona case, detectives told the caller that the goal was the suspect’s account and warned her not to make promises or threats (State v. Davitt, 2023).

How often police use the technique has been measured in one located study: in case files from the Los Angeles Sexual Assault Study (n = 616), pretext phone calls, witness interviews, and suspect interviews were all underused (St. George, 2026). The only published program evaluation concerns children: a review of four years of New York State Police control calls placed by children aged 3 to 18 after a forensic interview described the calls as highly effective, reported that many suspects pleaded guilty, sparing children from testifying, and cautioned about adverse effects on the child (Canavan et al., 2021). That review was descriptive, had no comparison group, and did not assess whether the statements were accurate. Pretext calls are often followed by a formal interview in which what the suspect said on the call is presented to him as evidence (Davis et al., 2021).

The Caller as Questioner

Review chapters by researchers who study interrogation describe pretext callers as surrogate interrogators (Davis et al., 2021; Livingston et al., 2019). Like an interrogator, a caller may present the evidence of guilt as settled, offer face-saving explanations that minimize the act, suggest that acknowledging it will lead to a better outcome for the suspect or the family, and press for details. Unlike an interrogator, the caller is a private person whose words are not bound by the rules on promises, threats, and custody, and whose relationship with the suspect gives the conversation emotional force.

In laboratory studies with students accused of a minor infraction, police-style minimization and explicit offers of leniency increased both true and false confessions (Russano et al., 2005; 296 students analyzed), and false incriminating evidence led innocent participants to sign a confession to damaging a computer (Kassin & Kiechel, 1996; 79 students, 75 analyzed); the 2025 scientific review of police-induced confessions identifies minimization that implies leniency and false-evidence presentations as situational risk factors (Kassin et al., 2025). Applying these findings to pretext calls is reasoning by analogy: no study has tested whether the same tactics, used by a relative on the telephone, change the accuracy of what a suspect says, and a pretext call usually lacks custody, a feature the research associates with risk.

Relationship Pressure and Placating Statements

The feature most specific to pretext calls is the relationship. A suspect speaking with a spouse, child, or parent may apologize, agree, or concede a point to calm a distressed caller, to keep the family together, or to prevent the caller from going to the police (Davis et al., 2021). In an Arizona memorandum decision, a suspect on a two-hour call with the complainants’ mother repeatedly denied sexual contact, said he would admit to anything to bring the family back together, and retracted each concession immediately (State v. Davitt, 2023). Such statements show why the whole conversation matters: a concession made and withdrawn in the same breath, or an apology for causing distress, admits something different from an account of what occurred. They do not show that apologies or concessions on pretext calls are usually false, and some pretext-call admissions describe specific acts, places, and times.

Disclosure of Details by the Caller

A complainant necessarily knows the details of the allegation, and in the course of a call may state them: where something happened, what was said, what the suspect did. A detail the suspect later repeats on the call, or in a subsequent interview, is then less informative about his own knowledge than a detail he supplied first. Research on contaminated confessions, which concerns police interviews rather than pretext calls, shows that false confessions in exoneration cases often contained accurate details that investigators were believed to have disclosed (Garrett, 2010). The same reasoning applies with equal force in the other direction: when the recording shows that the suspect volunteered an accurate detail the caller did not mention, and the detail is later confirmed, the admission carries considerably more weight. Because the call is usually recorded from its first word, it is often possible to tell which party introduced each fact.

What Pretext Calls Produce and How It Is Perceived

Pretext calls produce full admissions, partial admissions that concede contact but assert consent or accident, apologies, evasions, angry denials, and outright denials. The Counterintuitive Testimony paper reviews the research on apologies and on denial-and-counterattack responses, which shows that apologies arise from sympathy and habit as well as blame and that a denial combined with an attack on the complainant can lower observers’ belief in her (Blackman & Stubbs, 2001; Harsey & Freyd, 2023). Neither line of research distinguishes the responses of people who committed an offense from those who did not.

Research on how observers weigh pretext-call statements is preliminary. In an unpublished conference poster, mock jurors’ guilt ratings fell as their perception of caller pressure rose, particularly for threats to the relationship (Livingston et al., 2018). Davis and colleagues (2021) also describe unpublished mock-juror studies in which participants rated motives other than guilt, such as soothing the caller or preserving the relationship, as important reasons for a suspect’s admissions, and in which caller tactics affected perceived pressure, which in turn affected perceived guilt. None of this work has been peer-reviewed, and it does not show that real juries discount such admissions. Research on confessions to police, which finds that jurors discount pressured confessions only partly, is reviewed in “How Jurors Respond to Confession Evidence” in this resource (Mindthoff et al., 2024).

Pretext Calls and Exoneration Data

Exoneration data say little about pretext calls, for a definitional reason: the National Registry of Exonerations counts a statement as a confession only if it was made to law enforcement (as quoted in Denne et al., 2026), so an admission to a complainant or relative on a pretext call would not be counted even if it were false.

In an analysis of the 339 child sexual abuse exonerations in the Registry as of December 2025, false confession was a contributing factor in 29 (8.6%), and 87.9% of the exonerees were men (Denne et al., 2026). These are shares of known wrongful convictions, not rates among all sexual abuse prosecutions or all pretext calls; they say nothing about the accuracy of complaints in general, and the study did not code pretext calls.

Limits of the Evidence

The pretext-call literature is thin. No study has compared the statements of suspects who committed the act with those who did not, measured how often pretext calls yield false or misleading admissions, or tested which call designs produce more accurate information. The suspects in the cases described here were men, as were most exonerees in the child sexual abuse data (Denne et al., 2026), and no located study examined whether the sex of the caller or the suspect affects what is said.

Group research cannot determine whether a particular admission on a particular call is true. That depends on the words used, which party introduced each fact, the pressure applied, the suspect’s circumstances, and above all whether independent evidence corroborates what was said.

Questions to Ask About This Research

When the review chapters (Davis et al., 2021; Livingston et al., 2019) are cited to show that a pretext call produced a misleading admission, ask how much evidence the claim rests on: they describe how calls can produce such statements, not how often. When interrogation experiments (Russano et al., 2005; Kassin & Kiechel, 1996) are applied to a call, ask whether students accused of trivial infractions by a police-like experimenter resemble a suspect on the telephone with a relative. When the New York program review (Canavan et al., 2021) is cited to show that pretext calls work, ask what exactly was counted: pleas and avoided testimony say nothing about accuracy, and the callers were children. When a low false-confession share in exoneration data (Denne et al., 2026) is cited to show that pretext-call admissions are reliable, ask whether the conclusion follows, since the Registry does not count statements to private persons. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.

Pretext Calls and the Law

Federal and Military Standards

Federal law permits interception by a person acting under color of law with one party’s consent (18 U.S.C. § 2511(2)(c)), and a plurality of the Supreme Court has held that the Fourth Amendment does not protect against a confidant who records a conversation for the police (United States v. White, 1971). Miranda warnings are not required when the suspect does not know he is speaking with an agent of the police, because Miranda forbids coercion, not strategic deception (Illinois v. Perkins, 1990). Due-process involuntariness requires coercive police activity (Colorado v. Connelly, 1986), so a private caller’s pressure raises a constitutional voluntariness question only when the caller acts as an agent of the state.

The right to counsel imposes a firmer limit after formal charges. Once that right has attached, the government may not deliberately elicit statements about the charged offense through a recorded conversation with a codefendant (Maine v. Moulton, 1985) or an informant who initiates conversation for that purpose (Massiah v. United States, 1964; United States v. Henry, 1980), although a passive listener who only reports does not violate it (Kuhlmann v. Wilson, 1986). A police-directed pretext call about a charged offense raises that question.

In courts-martial, Article 31(b), UCMJ, and Military Rule of Evidence 305(c)(1) require warnings before a person subject to the code interrogates or requests a statement from a suspect, and Rule 305(b)(1) extends the requirement to “a knowing agent” of such a person or of a military unit. The Court of Appeals for the Armed Forces has required warnings when an active-duty Marine sergeant had junior Marines telephone a reservist and then took the phone and questioned him about a missing pistol; the statement was inadmissible and the findings were set aside (United States v. Gilbreath, 2014). The test asks whether the questioner was acting, or could reasonably be considered to be acting, in an official law-enforcement or disciplinary capacity (United States v. Jones, 2014). After charges have been preferred, Rule 305(c)(3) bars use of statements obtained by anyone acting in a law-enforcement capacity, “or the agent of such a person,” about the preferred charges when the accused has counsel or has requested counsel and counsel was not present.

Case Example. In United States v. Rios (1998), investigators asked the accused’s 14-year-old stepdaughter to place a recorded telephone call to him about her allegations, and his commander had told him to call her. The Court of Appeals for the Armed Forces held that no Article 31(b) warning was required, because he believed he was speaking with his stepdaughter, there was no coercion based on rank or duty, and the commander’s directive did not shape the conversation; two judges dissented. Rios applied an earlier form of the test that asked how the accused perceived the questioner; Jones (2014) later dropped that subjective element, so the case now illustrates the facts courts weigh rather than the current test.

Military Rule of Evidence 304(a)(2) provides that failure to deny an accusation is not an admission if the person was under investigation or in confinement, arrest, or custody for the alleged wrongdoing. The rule applied to silence in response to a friend’s question once an investigation had begun (United States v. Cook, 1998), and where the defense did not object to evidence of an accused’s failure to deny during a pretext call and text exchange, or to argument that he “never denies anything,” the court held the Rule 304(a)(2) claim waived (United States v. Ahern, 2017). An admission, including a self-incriminating statement meant to be exculpatory, requires independent evidence tending to establish its trustworthiness (Mil. R. Evid. 304(a)(1)(C), (c)), and the defense may introduce the rest of a partly introduced statement (Mil. R. Evid. 304(h)). In civilian court, the accused’s statements are admissions of a party-opponent (Fed. R. Evid. 801(d)(2)), and Federal Rule of Evidence 106, as amended December 1, 2023, permits completion of a partly introduced written or oral statement.

Arizona

Arizona permits interception with the consent of a party or a person present (A.R.S. § 13-3012(9)), and the Court of Appeals held that a police-arranged confrontation call between a minor and her stepfather comported with the state statute and the state constitution (State v. Allgood, 1992). The Arizona Supreme Court, assuming that a victim who called the suspect at police suggestion acted as a police agent, held that her demands for return of her property, including a threat to call the police, did not render his statements involuntary (State v. Keller, 1977). In memorandum decisions, which may be cited only as Arizona Supreme Court Rule 111(c) permits, the Court of Appeals assumed without deciding that a complainant placing a police-arranged call was a state agent and held the statements voluntary because the suspect could end the call and was not promised leniency, and held that Miranda and the right to counsel did not apply because he was neither in custody nor charged (State v. Deng, 2017; State v. Davitt, 2023).

A.R.S. § 13-3988 requires the judge to decide voluntariness before a confession is admitted and permits the jury to hear evidence on voluntariness, but subsection (C) provides that the statute does not bar a voluntary confession made “to any other person without interrogation by anyone,” or when the person was not under arrest or detention. The standard instruction tells jurors not to consider statements “to a law enforcement officer” unless they find beyond a reasonable doubt that the statements were voluntary, and treats a statement as involuntary if it resulted from violence, coercion, threats, or “any direct or implied promise, however slight” (RAJI (Criminal) Standard 19, 2022). Whether the instruction reaches a statement to a private caller acting at police direction is an agency question the instruction does not answer and no published Arizona decision located for this paper has resolved. Arizona applies a corpus delicti rule to confessions (State v. Carlson, 2015), addressed in “Corroboration and Assessing the Reliability of a Confession” in this resource.

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

A pretext call recorded from first word to last gives the factfinder the suspect’s own words in a setting without custody, in which he could end the conversation, and shows which party introduced each fact. Specific admissions that include details the caller did not supply are informative when independent evidence later confirms those details: confirmed, volunteered details are the kind of independent evidence the research treats as most informative about reliability, and no research permits an expert to say that a particular admission is true or false. A call can also resolve a case without requiring a child complainant to testify (Canavan et al., 2021). The weight such evidence carries depends on the completeness of the record, including the instructions given to the caller and every communication on every platform, and on independent corroboration of what was admitted.

The research and law also counsel caution. An apology, a concession withdrawn in the same breath, or a statement offered to keep the family together may admit less than its words suggest (Davis et al., 2021). In courts-martial, Military Rule of Evidence 304(a)(2) limits treating a failure to deny as an admission once the person is under investigation (Cook); questioning by someone acting in an official capacity can require Article 31(b) warnings (Gilbreath); and after charges are filed or preferred, the Sixth Amendment and Rule 305(c)(3) limit police-directed questioning about those charges. Details the caller supplied first carry little weight when the suspect repeats them, and exoneration data cannot show that pretext-call admissions are reliable, because statements to private persons are not counted.

Special Issues for Defense

The research supports close attention to how the call was designed and conducted. Review chapters describe callers using minimization, appeals to the relationship, and claims that the evidence is settled, with fewer limits than an interrogator (Davis et al., 2021; Livingston et al., 2019), and laboratory research associates police-style minimization and false-evidence tactics with more false confessions as well as more true confessions (Russano et al., 2005; Kassin et al., 2025). A complete record of the call, including any planning materials, instructions to the caller, and unrecorded contacts, is what allows a factfinder to see which facts the caller introduced and what pressure was applied, and the law permits the circumstances of a statement to be presented to the factfinder (Mil. R. Evid. 304(g); A.R.S. § 13-3988(A)). Whether the caller acted as an agent of the police can matter under Rule 305(b)(1), Gilbreath, the right to counsel, and the Arizona voluntariness instruction.

The research and law also counsel caution. The analogy from laboratory interrogation to a telephone call with a relative is untested, the observer studies are unpublished, and no study shows that pretext-call admissions are false at any measured rate. Courts have repeatedly found pretext-call statements voluntary where the suspect was not in custody and could end the call (Keller; Deng; Davitt), and Miranda does not apply when the suspect does not know he is speaking to the police (Perkins). Where a recorded admission includes details the caller did not supply and independent evidence confirms them, that is the kind of evidence the research treats as most informative about reliability, no research permits an expert to say that a particular admission is true or false, and a denial or angry response no more shows innocence than an apology shows guilt.

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

Blackman, M. C., & Stubbs, E. C. (2001). Apologies: Genuine admissions of blameworthiness or scripted, sympathetic responses? Psychological Reports, 88(1), 45–50. https://doi.org/10.2466/pr0.2001.88.1.45

Canavan, J. W., Borowski, C., Essex, S., & Perkowski, S. (2021). Multidisciplinary responses to the sexual victimization of children: Use of control phone calls. Journal of Interpersonal Violence, 36(1–2), NP384–NP401. https://doi.org/10.1177/0886260517730565

Davis, D., Leo, R. A., Livingston, T. N., & Rerick, P. O. (2021). On the synergy between pretext caller and police interrogator. In N. Deslauriers-Varin & C. Bennell (Eds.), Criminal investigations of sexual offenses (pp. 115–130). Springer. https://doi.org/10.1007/978-3-030-79968-7_9

Denne, E., Sullivan, C. E., & Neal, T. M. S. (2026). Characteristics of exonerated cases of child sexual abuse. Frontiers in Psychology, 17, Article 1794898. https://doi.org/10.3389/fpsyg.2026.1794898

Eisenga, H. (2004). Pretext phone calls in sexual assault investigations (Training Key No. 574). International Association of Chiefs of Police. https://evawintl.org/wp-content/uploads/IACP-574PretextPhoneCalls.pdf

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

Harsey, S. J., & Freyd, J. J. (2023). The influence of deny, attack, reverse victim and offender and insincere apologies on perceptions of sexual assault. Journal of Interpersonal Violence, 38(17–18), 9985–10008. https://doi.org/10.1177/08862605231169751

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., & Kiechel, K. L. (1996). The social psychology of false confessions: Compliance, internalization, and confabulation. Psychological Science, 7(3), 125–128. https://doi.org/10.1111/j.1467-9280.1996.tb00344.x

Livingston, T., Rerick, P., & Davis, D. (2018, May). The effect of caller threats on guilt ratings in a “pretext” phone call [Poster presentation]. Association for Psychological Science 30th Annual Convention, San Francisco, CA, United States. https://www.psychologicalscience.org/conventions/archive/2018annual/paper/11320/

Livingston, T. N., Rerick, P. O., Villalobos, J. G., & Davis, D. (2019). Deception induced confession: Strategies of police interrogators and their lay collaborators. In T. Docan-Morgan (Ed.), The Palgrave handbook of deceptive communication (pp. 747–767). Palgrave Macmillan. https://doi.org/10.1007/978-3-319-96334-1_39

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

Russano, M. B., Meissner, C. A., Narchet, F. M., & Kassin, S. M. (2005). Investigating true and false confessions within a novel experimental paradigm. Psychological Science, 16(6), 481–486. https://doi.org/10.1111/j.0956-7976.2005.01560.x

St. George, S. (2026). Before arrest: The effects of rape myths on evidence collection decisions in sexual assault investigations. Behavioral Sciences & the Law, 44(2), 265–280. https://doi.org/10.1002/bsl.70037

Legal Authorities

Colorado v. Connelly, 479 U.S. 157 (1986).

Illinois v. Perkins, 496 U.S. 292 (1990).

Kuhlmann v. Wilson, 477 U.S. 436 (1986).

Maine v. Moulton, 474 U.S. 159 (1985).

Massiah v. United States, 377 U.S. 201 (1964).

United States v. Henry, 447 U.S. 264 (1980).

United States v. White, 401 U.S. 745 (1971).

State v. Allgood, 171 Ariz. 522, 831 P.2d 1290 (App. 1992).

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

State v. Davitt, No. 1 CA-CR 22-0553 (Ariz. Ct. App. Dec. 19, 2023) (mem. decision) (not for publication; citable for persuasive value only under Ariz. R. Sup. Ct. 111(c)).

State v. Deng, No. 1 CA-CR 15-0638 (Ariz. Ct. App. Feb. 9, 2017) (mem. decision).

State v. Keller, 114 Ariz. 572, 562 P.2d 1070 (1977).

United States v. Ahern, 76 M.J. 194 (C.A.A.F. 2017).

United States v. Cook, 48 M.J. 236 (C.A.A.F. 1998).

United States v. Gilbreath, 74 M.J. 11 (C.A.A.F. 2014).

United States v. Jones, 73 M.J. 357 (C.A.A.F. 2014).

United States v. Rios, 48 M.J. 261 (C.A.A.F. 1998).

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

Uniform Code of Military Justice art. 31, 10 U.S.C. § 831.

18 U.S.C. § 2511(2)(c)–(d).

Ariz. Rev. Stat. Ann. § 13-3012(9).

Ariz. Rev. Stat. Ann. § 13-3988.

Fed. R. Evid. 106, 801(d)(2).

Manual for Courts-Martial, United States, Mil. R. Evid. 304, 305 (2024 ed.).

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

Ariz. R. Sup. Ct. 111(c).

Related topic papers: Minimization, Maximization, and Implied Promises of Leniency · False Evidence Ploys and Police Deception · Contamination: Non-Public Details and Post-Admission Narratives · Corroboration and Assessing the Reliability of a Confession · Confessions in Courts-Martial: Article 31(b), Mil. R. Evid. 304 and 305, and Expert Assistance · How Jurors Respond to Confession Evidence

Related topics from Arizona Forensic Services: Counterintuitive Testimony · Child Abuse Allegations · Courts-Martial