Snap Shot
- The question: When a defendant’s mental state at the time of an offense is disputed, what can memory science say about the evidence of that mental state, and what may a psychologist tell the jury about it under Clark v. Arizona and Arizona’s Mott rule?
- What the research shows: Evidence of a past mental state is reconstructed from memory: the defendant’s account, witnesses’ recollections of what the defendant said and did, and records. Reports of one’s own past reasons and feelings are shaped by later knowledge and present appraisals, although reports of what a person noticed at the time can be informative. A review estimates that about 20% to 30% of people who have committed violent crimes claim some amnesia for the offense; some claims reflect genuine memory failure, such as alcohol blackout or brain injury, and some are feigned.
- What it does not show: Memory loss after an offense does not show that the person lacked the required mental state when acting; a person in an alcohol blackout can act purposefully. No test can establish that a particular amnesia claim is genuine or feigned with certainty, and group research cannot say what a particular defendant thought or now remembers.
- The law: Clark (2006) upheld Arizona’s rule, from State v. Mott (1997), confining mental-disease and capacity evidence to the guilty-except-insane defense while leaving observation evidence available on mens rea; State v. Malone (2019) extended the rule to brain damage. Federal Rule 704(b) bars expert opinion on whether the defendant had the mental state, though not testimony about what most members of a group know (Diaz, 2024). In courts-martial, mental-condition evidence short of lack of mental responsibility may be offered on mens rea, most clearly specific intent (R.C.M. 916(k)(2); Ellis v. Jacob, 1988), and R.C.M. 706 and Mil. R. Evid. 302 govern sanity boards.
- Issues for each side: Prosecutors can show that amnesia claims are common, sometimes feigned, and legally distinct from lack of intent, while recognizing that genuine amnesia occurs and tests cannot rule it out. Defense counsel can present observation evidence and the limits of retrospective reports, while recognizing that in Arizona diagnosis and capacity opinions belong to the insanity defense and that a defendant’s own account is open to the same reconstruction.
Most serious crimes require proof that the defendant acted with a culpable mental state, such as intent, knowledge, or premeditation. Mental state can rarely be observed directly. Factfinders infer it from what the defendant did and said, from the defendant’s later account, and from people who saw the defendant before, during, and after the events. Nearly all of that evidence is memory evidence, and much of it is gathered months later, after the outcome is known.
When the defendant has a mental illness, the scope of a clinician’s testimony on mens rea is a legal question as well as a scientific one. Clark v. Arizona (2006) divided such evidence into three kinds and upheld Arizona’s limits on two of them when the issue is mens rea. In Arizona, a psychologist’s testimony about a defendant’s perception, memory, and thinking at the time of an offense must fit those categories; federal courts and courts-martial apply different rules.
This paper reviews research on memory for one’s own past thoughts and actions, on amnesia claims and their assessment, and on the reliability of retrospective evaluations, then the limits of that research and the governing federal, military, and Arizona law.
Mental State as a Reconstruction
An evaluation of mental state at the time of an offense is retrospective by design. The evaluator usually sees the defendant weeks or months later and works from the defendant’s account, witnesses, police reports, recordings, and records; standard forensic texts advise drawing on multiple independent sources and testing the defendant’s account against them (Melton et al., 2018). Each source has its own memory history. People who know an outcome tend to judge it as having been more foreseeable than it was and underestimate how much that knowledge shaped their judgment (Fischhoff, 1975). A witness who later learns of a killing may recall the defendant’s earlier behavior as more disturbed, or more calculated, than it seemed then.
The Clark categories map onto these sources. Observation evidence is testimony about what the defendant did and said, including expert testimony about the defendant’s characteristic ways of thinking and behaving. Mental-disease evidence is a professional diagnosis. Capacity evidence is an opinion about whether the defendant could form the required mental state (Clark v. Arizona, 2006). The memory research summarized here bears most directly on observation evidence, which consists largely of recollections.
Remembering One’s Own Thoughts and Actions
People are sometimes poor reporters of why they did something, particularly when asked to explain or interpret rather than to recall what they noticed. Nisbett and Wilson (1977) reviewed studies in which participants confidently gave reasons for their choices that the experiments showed had not influenced them, reasons that matched what outside observers guessed; the authors argued that such reports often rest on plausible theories rather than direct access to causes. That conclusion has limits. Ericsson and Simon (1980) showed that verbal reports are more accurate when they concern information the person was attending to at the time, are given concurrently or soon afterward, and do not require the person to explain or interpret. “I thought he had a gun” reports remembered content; “I did it because I was confused” is an explanation, more open to later theorizing.
Memory for past feelings is also shaped by the present. Supporters of a presidential candidate who withdrew from the 1992 race recalled their feelings at the time of the withdrawal in ways biased toward how they appraised him months later (Levine, 1997). A defendant’s memory of how angry or frightened he or she felt, and a witness’s memory of how the defendant seemed, can drift the same way. Reconstruction is not lying and does not make a report false; it means the report should be compared with what was said or recorded closer to the time.
Memory for one’s own violent acts is often good. Among 105 young offenders convicted of serious violence, about 19% reported partial amnesia for the offense and one reported complete amnesia; most reported no memory loss, and amnesia was associated with heavy drinking, an emotional relationship with the victim, and the offender’s thinking during the assault (Evans et al., 2009).
Claims of Amnesia for an Offense
Claims of amnesia are common in serious cases. Studies reviewed by Jelicic (2018) found claims in 26% to 31% of homicide samples and in 29% of 207 life-sentence prisoners (Pyszora et al., 2003); the review estimated that roughly 20% to 30% of violent offenders make such claims. Researchers distinguish three sources. Organic amnesia arises from brain dysfunction, such as head injury, hypoglycemia, or intoxication; alcohol can block new long-term memories while the person keeps conversing and acting in complex ways, so a person in a blackout may act purposefully and later remember nothing of it (White, 2003). Dissociative amnesia is attributed to psychological processes associated with extreme stress. Feigned amnesia is a claim of memory loss by a person who remembers.
Researchers disagree about how often each occurs. Christianson and Merckelbach (2004) argued that many crime-related amnesia claims are best understood as deception, noting that emotional events are usually well remembered and that the claims cluster where they offer an advantage. Jelicic (2018) regarded dissociative amnesia for offenses as rare but emphasized that organic amnesia is plausible in many cases and that a claim should be judged not credible only on converging evidence. A systematic review of 128 published case studies of dissociative amnesia found that most did not rule out ordinary forgetting or malingering (Mangiulli, Otgaar, et al., 2022). Dissociative amnesia nonetheless remains a recognized diagnosis (American Psychiatric Association, 2022), and some researchers argue that critics have overstated the case against forgetting of traumatic experiences (Brewin, 2021). Genuine and feigned amnesia both occur, and their base rates are not well established.
Feigning can itself change memory. In laboratory studies, participants who watched a mock crime and simulated amnesia later recalled fewer crime details, after they stopped simulating, than participants who were truthful from the start (Mangiulli, Riesthuis, & Otgaar, 2022). The studies used students and mock crimes, and the effect has not been tested in actual offenders.
For mens rea, the central point is timing. An amnesia claim concerns what a person can retrieve now; mens rea concerns what the person intended or knew when acting. Memory loss after the fact, even when genuine, does not by itself show that the required mental state was absent, although the conditions associated with genuine amnesia, such as heavy intoxication, may bear on it where the law permits.
Assessing Amnesia Claims and Retrospective Evaluations
Symptom validity testing presents a series of two-alternative forced-choice questions about details of the offense. A person with no memory should score near chance, and a score significantly below chance indicates that the person recognized the correct answers and avoided them, though it does not show the reason (van Oorsouw & Merckelbach, 2010). Self-report measures of improbable symptoms and memory-effort tests are also used, and researchers recommend combining methods (Giger et al., 2010; Jelicic, 2018). Each method has limits. Forced-choice testing needs many items about details known only to someone who was present, and details learned through questioning, discovery, or media undermine it; a score at or above chance does not establish genuine amnesia; and general effort tests say little about a claim limited to an intoxicated episode (Jelicic, 2018). A 2023 review concluded that most validity tests have had little research specific to crime-related amnesia and that estimates of the base rate of feigning vary so widely that ruling out feigning may not be possible even with the best available tools (Rubenzer, 2023).
Retrospective evaluations also vary among evaluators. In 161 Hawaii cases with three independent sanity evaluations, all three agreed in 55%; agreement was higher when evaluators diagnosed a psychotic disorder or the defendant had been recently hospitalized, and lower when the defendant had been abusing substances or was charged with a violent felony (Gowensmith et al., 2013). The study concerned insanity in one state, but it bears on Clark‘s concern that capacity opinions involve judgments on which qualified professionals disagree.
Limits of the Evidence
Most research on self-report and reconstruction uses students and everyday judgments, not defendants describing a serious offense. Prevalence figures for amnesia claims come from convicted offenders and from studies whose definitions of partial and complete amnesia vary. Simulated-amnesia studies use mock crimes with no real consequences. The base rate of feigned amnesia in actual cases is unknown, which limits what any test result can say about the probability that a particular claim is genuine. None of this research can establish what a particular defendant thought, intended, or now remembers; that remains a question for the factfinder on all the evidence.
Questions to Ask About This Research
When an expert cites the estimate that 20% to 30% of violent offenders claim amnesia (Jelicic, 2018), counsel should ask what the figure counts: how often the claim is made, not how often it is true, in convicted samples. When forced-choice testing is offered, counsel should ask how many items were used, how they were built, and whether the details were available to the defendant through discovery, interviews, or media (van Oorsouw & Merckelbach, 2010; Rubenzer, 2023).
When an expert relies on the defendant’s account, counsel can ask whether it reports what the defendant noticed at the time or an explanation built later, when it was first given, and whether it fits statements and behavior closer to the event (Ericsson & Simon, 1980; Nisbett & Wilson, 1977). When lay witnesses describe the defendant’s behavior, counsel can ask when the witness first described it and what the witness had learned in between (Fischhoff, 1975). When a retrospective opinion is offered with confidence, a court may ask what independent sources support it and how often qualified evaluators disagree on similar questions (Gowensmith et al., 2013). These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.
Observation Evidence and the Law
Federal and Military Standards
Expert testimony on mental state must satisfy Federal Rule of Evidence 702, as amended December 1, 2023, under which the proponent must show it is more likely than not that the rule’s helpfulness and reliability requirements are met (Daubert v. Merrell Dow Pharmaceuticals, Inc., 1993); Kumho Tire Co. v. Carmichael (1999) extended the gatekeeping obligation to all expert testimony, including non-scientific specialized knowledge. Federal Rule 704(b) bars an expert in a criminal case from opining on whether the defendant had a mental state or condition that is an element of the crime or a defense; Diaz v. United States (2024) held that the rule reaches only opinions about the defendant, so an expert’s testimony that most members of a group (there, drug couriers) know what they carry does not violate it, while testimony that all members have the mental state would. The federal insanity defense requires a severe mental disease or defect that left the defendant unable to appreciate the nature and quality or wrongfulness of the acts, proved by clear and convincing evidence (18 U.S.C. § 17), and federal courts have generally allowed psychiatric evidence offered to negate specific intent (United States v. Pohlot, 1987). States have wide latitude; Kahler v. Kansas (2020) upheld a scheme that admits mental-illness evidence on mens rea but has no moral-incapacity insanity defense.
In courts-martial, Article 50a, UCMJ, and R.C.M. 916(k)(1) define lack of mental responsibility in the federal statute’s terms, and the accused must prove it by clear and convincing evidence (R.C.M. 916(k)(3)(A)). R.C.M. 916(k)(2) provides that a lesser mental condition is not an affirmative defense; its Discussion states that such evidence may be admissible on whether the accused had a required state of mind, with pretrial notice of expert testimony (R.C.M. 701(b)(2)). The Court of Military Appeals held that an earlier version of the rule could not bar evidence offered to negate specific intent (Ellis v. Jacob, 1988). The military cases in this line concern evidence offered to negate specific intent, and the government keeps the burden of proving the required mental state beyond a reasonable doubt. Under R.C.M. 706, a sanity board of physicians or clinical psychologists makes separate findings on mental responsibility, diagnosis, and competence; the full report goes to the defense, and others generally receive only the ultimate conclusions. Mil. R. Evid. 302 bars the accused’s statements at the examination, and derivative evidence, unless the accused first offers them; when the defense offers expert mental-condition testimony, the military judge must on motion release the report, other than the accused’s statements, to the prosecution. Mil. R. Evid. 704 has no counterpart to Federal Rule 704(b). Expert testimony is assessed under Mil. R. Evid. 702 and United States v. Houser (1993); the 2023 amendment to Federal Rule 702 took effect in the military rule in 2025 through Mil. R. Evid. 1102, and practitioners should confirm current text.
Arizona
Arizona’s “guilty except insane” defense requires proof by clear and convincing evidence that the defendant had a mental disease or defect so severe that the defendant did not know the criminal act was wrong; it excludes, among other things, disorders resulting from acute voluntary intoxication, character defects, and impulse control disorders (A.R.S. § 13-502). Voluntary intoxication is not a defense to any criminal act or required mental state (A.R.S. § 13-503). In State v. Mott (1997), the Arizona Supreme Court held that, because the legislature had not adopted a diminished-capacity defense, expert evidence of a mental disorder short of insanity is not admissible to negate mens rea, and it affirmed exclusion of battered-woman and low-intelligence testimony offered by a mother charged in her child’s death. The court distinguished State v. Christensen (1981), which allowed expert testimony that a defendant had a character trait of acting reflexively under stress.
In Clark v. Arizona (2006), the Court, in an opinion by Justice Souter, held that neither Arizona’s moral-incapacity insanity test nor the Mott rule, as applied to mental-disease and capacity evidence, violated due process. It reasoned that Arizona may presume sanity and that channeling diagnosis and capacity evidence to the insanity issue reduces the risk of misleading jurors, given professional disagreement over diagnoses, the gap between clinical and legal categories, and the risk that capacity opinions carry an “apparent authority that psychologists and psychiatrists do not claim to have.” The Court described observation evidence as unrestricted by Mott and declined to decide whether Clark’s observation evidence had been improperly limited, because that claim had not been presented to the Arizona courts. Justice Breyer joined all of the opinion except Parts III-B and III-C and the disposition; he accepted the majority’s reading of the record but was concerned that the distinction among the kinds of evidence would be unclear in some cases and would have remanded for the Arizona courts to decide whether Arizona law and its application in the trial court were consistent with the Court’s distinction. Justice Kennedy, joined by Justices Stevens and Ginsburg, dissented: in his view the categories blur in practice, a jury cannot make sense of evidence such as Clark’s belief that townspeople were aliens unless it also hears about the illness behind the belief, and ordinary evidentiary rules could screen unreliable or confusing testimony case by case.
In State v. Malone (2019), the Arizona Supreme Court held that Mott covers brain damage and that a defendant offering expert evidence of an impulsive behavioral tendency to rebut premeditation may not corroborate it with brain-damage evidence; it called the Christensen category “behavioral-tendency evidence” and left open whether corroboration is permitted if the state disputes the trait. Expert testimony must also satisfy Arizona Rule of Evidence 702, amended effective January 1, 2012, to adopt the Daubert standard and effective January 1, 2024, to track the 2023 federal amendment (Ariz. Sup. Ct. Order R-23-0004). In practice, a psychologist’s testimony on mens rea may describe what the defendant did and said and a Christensen-type behavioral tendency, but not a diagnosis (including a dissociative explanation of amnesia) or a capacity opinion, which belong to the insanity defense. This paper did not locate an Arizona appellate decision deciding where general memory testimony, such as the point that later amnesia does not show an absent mental state, falls among the Clark categories.
Special Issues in Legal Practice
The principles described above apply with equal force to every party in a legal proceeding. The science operates the same way regardless of who is asking the questions, and the research summarized here does not by itself favor either side. What differs between the parties is not the science but the vantage point — the particular places where these issues tend to surface, and the particular errors each party is best positioned to guard against. The two summaries below identify those recurring issues. They are offered for orientation, not as litigation strategy; in every case the controlling question is what the evidence in that case will, and will not, support, and the expert’s role is to inform the trier of fact rather than to advocate for either side.
Special Issues for Prosecutors
The research supports several propositions prosecutors can legitimately present. Amnesia claims are common in serious cases, some are feigned, and most offenders studied report remembering their offenses. Genuine inability to remember now does not show that the defendant lacked intent or knowledge when acting, and a person in a blackout can act purposefully. A defendant’s later explanation of his or her reasons may be a sincere reconstruction rather than a memory, and statements and behavior close to the event provide a check on it. A score significantly below chance on a properly built forced-choice test indicates that the person recognized the correct answers and avoided them; it does not show the reason. In Arizona, Clark, Mott, and Malone support objection when diagnosis or capacity opinions are offered on mens rea, and in federal court Rule 704(b) bars an expert’s opinion on whether the defendant had the mental state.
The research also counsels caution. Genuine amnesia occurs, particularly with heavy intoxication and brain injury, and no test establishes with certainty that a particular claim is feigned; an above-chance score is not evidence of malingering. Rule 704(b) binds prosecution experts too: after Diaz, group-level testimony is permitted, but not an opinion that the defendant, or every member of a group, had the mental state. Forced-choice testing loses value when the items concern details the defendant could have learned from investigators or discovery, which the prosecution is well placed to document. The reconstruction that affects a defendant’s account affects the state’s lay witnesses as well. An objection that sweeps in observation evidence or behavioral-tendency evidence risks error, because Clark and Mott do not restrict that category.
Special Issues for Defense
The research supports several propositions defense counsel can legitimately present. Evidence of a past mental state is a reconstruction from imperfect sources, and lay recollections of a defendant’s behavior can be shaped by outcome knowledge. Genuine amnesia for an offense occurs, and validity tests have limited research specific to crime-related amnesia and cannot rule a claim in or out with certainty. Retrospective evaluations involve judgments on which qualified evaluators disagree, a point that cuts against overconfident opinions from either side. In Arizona, observation and behavioral-tendency evidence remain available on mens rea, and the state keeps the burden of proving the culpable mental state beyond a reasonable doubt; in courts-martial, mental-condition evidence short of lack of mental responsibility may be offered on whether the accused had the required state of mind, most clearly for specific-intent offenses.
The research also counsels caution. Amnesia, even when genuine, does not establish that the defendant lacked intent or knowledge at the time, and in Arizona voluntary intoxication cannot be offered to negate a mental state. Researchers regard feigned amnesia as a real possibility in serious cases, a score significantly below chance on a properly built forced-choice test indicates that the person recognized the correct answers and avoided them, though not why, and simulated amnesia may itself erode memory. In Arizona, a diagnosis or capacity opinion belongs to the guilty-except-insane defense, and Malone bars brain-damage evidence offered to corroborate a behavioral tendency. In courts-martial, offering expert mental-condition testimony opens the R.C.M. 706 report to the prosecution under Mil. R. Evid. 302. The defendant’s own account of a past mental state is subject to the same reconstruction as any other witness’s.
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
American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.). https://doi.org/10.1176/appi.books.9780890425787
Brewin, C. R. (2021). Tilting at windmills: Why attacks on repression are misguided. Perspectives on Psychological Science, 16(2), 443–453. https://doi.org/10.1177/1745691620927674
Christianson, S. Å., & Merckelbach, H. (2004). Crime-related amnesia as a form of deception. In P. A. Granhag & L. A. Strömwall (Eds.), The detection of deception in forensic contexts (pp. 195–226). Cambridge University Press. https://doi.org/10.1017/CBO9780511490071.009
Ericsson, K. A., & Simon, H. A. (1980). Verbal reports as data. Psychological Review, 87(3), 215–251. https://doi.org/10.1037/0033-295X.87.3.215
Evans, C., Mezey, G., & Ehlers, A. (2009). Amnesia for violent crime among young offenders. Journal of Forensic Psychiatry & Psychology, 20(1), 85–106. https://doi.org/10.1080/14789940802234471
Fischhoff, B. (1975). Hindsight is not equal to foresight: The effect of outcome knowledge on judgment under uncertainty. Journal of Experimental Psychology: Human Perception and Performance, 1(3), 288–299. https://doi.org/10.1037/0096-1523.1.3.288
Giger, P., Merten, T., Merckelbach, H., & Oswald, M. (2010). Detection of feigned crime-related amnesia: A multi-method approach. Journal of Forensic Psychology Practice, 10(5), 440–463. https://doi.org/10.1080/15228932.2010.489875
Gowensmith, W. N., Murrie, D. C., & Boccaccini, M. T. (2013). How reliable are forensic evaluations of legal sanity? Law and Human Behavior, 37(2), 98–106. https://doi.org/10.1037/lhb0000001
Jelicic, M. (2018). Testing claims of crime-related amnesia. Frontiers in Psychiatry, 9, Article 617. https://doi.org/10.3389/fpsyt.2018.00617
Levine, L. J. (1997). Reconstructing memory for emotions. Journal of Experimental Psychology: General, 126(2), 165–177. https://doi.org/10.1037/0096-3445.126.2.165
Mangiulli, I., Otgaar, H., Jelicic, M., & Merckelbach, H. (2022). A critical review of case studies on dissociative amnesia. Clinical Psychological Science, 10(2), 191–211. https://doi.org/10.1177/21677026211018194
Mangiulli, I., Riesthuis, P., & Otgaar, H. (2022). The memory-undermining effect of simulated crime-related amnesia and its legal implications: A review. Psychological Injury and Law, 15(2), 213–226. https://doi.org/10.1007/s12207-021-09441-x
Melton, G. B., Petrila, J., Poythress, N. G., Slobogin, C., Otto, R. K., Mossman, D., & Condie, L. O. (2018). Psychological evaluations for the courts: A handbook for mental health professionals and lawyers (4th ed.). Guilford Press.
Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84(3), 231–259. https://doi.org/10.1037/0033-295X.84.3.231
Pyszora, N. M., Barker, A. F., & Kopelman, M. D. (2003). Amnesia for criminal offences: A study of life sentence prisoners. Journal of Forensic Psychiatry & Psychology, 14(3), 475–490. https://doi.org/10.1080/14789940310001599785
Rubenzer, S. (2023). Ruling out feigned crime-related amnesia? A response to Acklin (2022). Behavioral Sciences & the Law, 41(4), 207–223. https://doi.org/10.1002/bsl.2623
van Oorsouw, K., & Merckelbach, H. (2010). Detecting malingered memory problems in the civil and criminal arena. Legal and Criminological Psychology, 15(1), 97–114. https://doi.org/10.1348/135532509X451304
White, A. M. (2003). What happened? Alcohol, memory blackouts, and the brain. Alcohol Research & Health, 27(2), 186–196.
Legal Authorities
Clark v. Arizona, 548 U.S. 735 (2006).
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).
Diaz v. United States, 602 U.S. 526 (2024).
Ellis v. Jacob, 26 M.J. 90, 93 (C.M.A. 1988).
Kahler v. Kansas, 589 U.S. 271 (2020).
Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).
State v. Christensen, 129 Ariz. 32, 628 P.2d 580 (1981).
State v. Malone, 247 Ariz. 29, 444 P.3d 733 (2019).
State v. Mott, 187 Ariz. 536, 931 P.2d 1046 (1997).
United States v. Houser, 36 M.J. 392 (C.M.A. 1993).
United States v. Pohlot, 827 F.2d 889 (3d Cir. 1987).
10 U.S.C. § 850a (Article 50a, Uniform Code of Military Justice).
18 U.S.C. § 17.
Ariz. Rev. Stat. Ann. § 13-502.
Ariz. Rev. Stat. Ann. § 13-503.
Fed. R. Evid. 702.
Fed. R. Evid. 704.
Ariz. R. Evid. 702.
Ariz. Sup. Ct. Order R-23-0004 (2023).
Mil. R. Evid. 302.
Mil. R. Evid. 702.
Mil. R. Evid. 704.
Mil. R. Evid. 1102.
Rule for Courts-Martial 701(b)(2).
Rule for Courts-Martial 706.
Rule for Courts-Martial 916(k).
Related topic papers: Memory Expert Testimony · Reconstruction and Recall · Intoxication and Memory · Dissociation and Memory · Hindsight Bias · Detecting Deception
Related topics from Arizona Forensic Services: Alcohol Blackouts: Alcohol Blackout / Amnesia · Courts-Martial