Snap Shot
- The question: Do men and women differ as eyewitnesses, and can a witness’s gender tell a court anything about the accuracy of that witness’s account?
- What the research shows: Across hundreds of studies, women show a small average advantage in episodic memory, especially for verbal material and faces, and men a small advantage for spatial material. In eyewitness studies, neither sex consistently reports more accurate information overall; women tend to recall more person details and men somewhat more about the setting, and people of both sexes often remember people of their own gender better. Several studies found that men expressed more confidence than women without greater accuracy; the studies are few, and the findings describe groups and say nothing about any individual witness.
- What it does not show: The differences are small and the distributions overlap heavily, so a witness’s gender says almost nothing about whether that witness’s account is accurate. Most eyewitness studies are laboratory studies of university students, and their findings on suggestibility conflict.
- The law: Every person is competent to testify (Rule 601); expert testimony is governed by Rule 702 and Houser, must fit the case, and may not address a particular witness’s accuracy or credibility (Lindsey, Nordstrom). Chapple permits general testimony on factors affecting eyewitness reliability in an appropriate case.
- Issues for each side: Both parties can use the research to answer arguments that rest on gender stereotypes; neither should present a group difference as evidence about the accuracy of an individual witness or treat a witness’s expressed certainty as a gauge of accuracy.
Whether men or women make better witnesses is one of the oldest questions in the psychology of testimony. In an early study, William Stern reported that women’s testimony was less accurate and less resistant to misleading information than men’s, but his groups were not comparable in age, and later studies with comparable groups did not replicate the finding (Butts et al., 1995). The question still surfaces in court, usually implicitly, when counsel or a factfinder assumes that a witness of one gender is more emotional, more observant, more suggestible, or more objective.
The research now available permits a more precise answer than either the old claim or a flat denial of any difference. There are average differences between men and women in some kinds of memory, and some of them are reliable. They are small, they depend on what is being remembered, and they describe groups rather than individuals. The forensic significance lies mainly in what the research rules out: inferences about a particular witness’s accuracy drawn from that witness’s gender.
This paper summarizes the meta-analytic evidence on sex differences in episodic memory, research on what male and female eyewitnesses tend to remember, the own-gender bias in face recognition, and findings on confidence and suggestibility, and it reviews the law that governs expert testimony on the subject. Most studies compare participants who identified as male or female; “gender” is used here as it is in the research titles and in the hub’s paper list. Related questions about memory for people are addressed in Face Recognition and Memory for Persons and Faces, and differences among other groups of witnesses are addressed in Memory in Special Populations.
Episodic Memory: A Small Average Difference
The largest analysis of the question combined 617 studies, conducted from 1973 to 2013 with more than 1.2 million participants, and found an overall female advantage in episodic memory of g = 0.19, a small effect (Asperholm et al., 2019). The direction depended on the material. Women did better on verbal tasks such as words, sentences, and prose (g = 0.28), on faces (g = 0.26), and on smells, tastes, and colors; men did better on spatial tasks such as remembering abstract images and routes (about g = -0.20 to -0.24). The differences were stable across the four decades studied, varied by region, and were smaller in childhood and old age. A companion analysis found that differences in the spread of scores between men and women were also mostly small and may partly reflect tests that were too easy for high scorers (Asperholm et al., 2020).
An effect of this size implies extensive overlap. The meta-analysis does not express it this way, but by standard statistical conversions a difference of g = 0.19 corresponds to distributions that overlap by roughly 90%, and to a probability of about 55% that a randomly chosen woman would outscore a randomly chosen man on a typical episodic memory task. That pattern is consistent with the broader conclusion that most psychological differences between men and women are small (Hyde, 2005). Recent work has shown that many small differences can, in combination, predict group membership or group-level outcomes (Herlitz et al., 2026); that is a finding about aggregates of many measures, and it does not make any single difference informative about the memory of one person.
What Male and Female Eyewitnesses Remember
Eyewitness studies ask a narrower question: whether men and women differ in recalling a staged or filmed event. A 2024 scoping review identified 22 studies from seven countries over more than four decades and concluded that neither sex consistently reported more accurate information in total; the more common finding was a difference in the kinds of details reported accurately (Russell et al., 2024). Women were more often accurate about person-related details, such as clothing, hair, and other features of the people involved, and men were in some studies slightly more accurate about the surroundings. Studies that counted the number of details reported generally found no difference in quantity.
Individual studies, as summarized in the scoping review (Russell et al., 2024), illustrate both the pattern and its variability. In the first of a classic pair of experiments, women were more accurate about details such as a woman’s clothing and actions, and men about the male thief’s appearance and the surrounding scene, with no overall difference in accuracy (Powers et al., 1979). In a study of 280 Slovenian university students who watched a video of a violent robbery with a female victim and a male perpetrator, women were more accurate overall, reported fewer false details, and were more accurate about the people involved, especially the victim, while men had a small advantage in describing the incident itself; apart from person details, most differences were small or absent, and the amount recalled did not differ (Areh, 2011). Other studies, including one with 20 men and 20 women, found no significant difference in recall accuracy or resistance to false information (Butts et al., 1995). Situational factors operate on both sexes: in an early study, recall was less accurate for a violent version of a filmed event than for a nonviolent version; women were less accurate than men in the violent version, although overall accuracy did not differ by sex across conditions (Clifford & Scott, 1978; Russell et al., 2024).
The usual explanation is attention: people attend more closely to what interests them or seems relevant to them, and attention at encoding shapes what is later remembered. The scoping review noted, however, that most explanations were proposed after the fact and that almost none of the studies tested them directly (Russell et al., 2024). The practical point is modest. Any witness, male or female, notices some parts of a scene more than others, and what caught a particular witness’s attention is a question about that witness and that event.
The Own-Gender Bias in Face Recognition
The most consistent gender-linked finding in this literature concerns faces. A meta-analytic review examined sex differences and the own-gender bias in face recognition (Herlitz & Lovén, 2013), and later research describes the established finding as a female own-gender bias, in which women recognize female faces better than male faces (Palmer et al., 2013). The episodic memory meta-analysis found a small female advantage for faces generally (Asperholm et al., 2019). In eyewitness studies, the scoping review found consistent evidence that participants were more accurate about people of their own gender, in several studies for both men and women (Russell et al., 2024).
Attention appears to contribute. When participants’ attention was divided while they viewed faces, women’s recognition of female faces declined more than their recognition of male faces, in both a face-recognition task (113 participants) and an eyewitness identification task (502 participants), reducing the female own-gender bias (Palmer et al., 2013). For men, the pattern differed between the two studies, and evidence of an own-gender bias in men is less consistent. The own-gender bias resembles the cross-race effect discussed in The Cross-Race Effect in that people tend to recognize members of their own group better, but it is generally described as smaller and less consistent, and it has received far less study in identification settings.
Confidence and Suggestibility
Several studies have found that men express more confidence in their memory reports than women without being more accurate. In a field study in which witnesses recalled a young woman who had spoken with them briefly, men and women performed similarly on most characteristics, but men were more confident in their answers (Yarmey & Yarmey, 1997). Men also reported more confidence in their memories in the Slovenian robbery study (Areh, 2011; Russell et al., 2024), and the scoping review found no study in which men’s greater confidence was accompanied by greater overall accuracy (Russell et al., 2024). Because factfinders often treat confidence as a sign of accuracy, this is the gender-linked finding with the most direct forensic relevance. It concerns how certainty is expressed, not the quality of memory, and its general lesson, that confidence expressed at trial, or after feedback, is a weak indicator of accuracy for witnesses of any gender while confidence recorded at a fair first test can be informative, is addressed in Confidence and Accuracy.
The evidence on suggestibility conflicts. In one experiment women were more suggestible overall, while in a second experiment each sex was more resistant to misleading suggestions about details relevant to its own gender (Powers et al., 1979). A study of nearly 2,000 visitors to a science museum found men more susceptible to misinformation (Loftus et al., 1992), and another study found no difference (Butts et al., 1995); the scoping review summarizes all three (Russell et al., 2024). The research therefore does not support a general claim that witnesses of either gender are more suggestible.
Limits of the Evidence
Most eyewitness studies on this question used university students, filmed or staged events, and short delays, and many were conducted decades ago; the scoping review noted the variety of methods, the predominance of student samples from a few developed countries, and the scarcity of research testing proposed explanations (Russell et al., 2024). Some samples were small, so studies reporting no difference may have lacked the power to detect a small one. The meta-analytic effects are precise but small, and they come mainly from standard memory tests rather than from witnesses to crimes. Most important, a group difference of this size provides almost no information about an individual. Research on gender cannot show that a particular witness’s account is accurate or inaccurate, and the factors that matter more, such as viewing conditions, stress, delay, and the way the witness was questioned, apply to everyone.
Questions to Ask About This Research
When the Asperholm et al. (2019) meta-analysis is cited for a female memory advantage, counsel should ask how large the effect is and what it means for one person: g = 0.19 overall, which corresponds to distributions that overlap by roughly 90%, measured mostly on laboratory tests of words, pictures, and faces. When Powers et al. (1979) or Areh (2011) is used to suggest that a witness of one gender was unlikely to notice a particular detail, counsel should ask what was counted, how large the difference was in absolute terms, whether the detail at issue resembles the details tested, and whether other studies found the same pattern. When the own-gender bias in face recognition (Herlitz & Lovén, 2013; Palmer et al., 2013) is raised, counsel should ask whether it was observed in men as well as women in the studies relied on and whether any study has measured its size in real identifications. When gender is offered as a reason to doubt or credit a witness’s susceptibility to suggestion, counsel should ask how the conflicting findings (Butts et al., 1995; Loftus et al., 1992; Powers et al., 1979) are reconciled. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.
Gender Differences and the Law
Federal and Military Standards
Federal Rule of Evidence 601 and Military Rule of Evidence 601 make every person competent to testify unless the rules provide otherwise, and nothing in the law of evidence makes a witness’s gender a measure of competence or reliability. Expert testimony about gender and memory is governed by Federal Rule of Evidence 702, as amended December 1, 2023, which requires the proponent to show that it is more likely than not that the testimony will help the trier of fact, rests on sufficient facts or data, and reflects a reliable application of reliable methods. Daubert v. Merrell Dow Pharmaceuticals, Inc. (1993) and Kumho Tire Co. v. Carmichael (1999) extend that gatekeeping obligation to all expert testimony, including non-scientific specialized knowledge. Because the differences are small and group-level, testimony that applied them to a particular witness would raise questions of fit and reliable application. In courts-martial, Military Rule of Evidence 702 and United States v. Houser (1993) govern; the 2023 federal amendment took effect in the military rule in 2025 through Military Rule of Evidence 1102 (Joint Service Committee on Military Justice, 2025), and practitioners should confirm current text. A witness’s level of certainty is one of the factors courts weigh in assessing an identification under Neil v. Biggers (1972) and Manson v. Brathwaite (1977); the confidence findings summarized above are one reason that factor is evaluated with care.
Arizona
Arizona Rule of Evidence 601 and A.R.S. § 13-4061 make every person competent to be a witness in a criminal trial. Arizona Rule of Evidence 702 adopted the Daubert standard effective January 1, 2012, and was amended effective January 1, 2024, to track the 2023 federal amendment (Ariz. Sup. Ct. Order R-23-0004). State v. Chapple (1983), Arizona’s leading case on eyewitness expert testimony, held exclusion of such testimony an abuse of discretion on its facts, and State v. Nordstrom (2001) approved limiting an eyewitness expert to the general factors affecting reliability, without an opinion on the accuracy of a particular witness. Under State v. Lindsey (1986) and State v. Moran (1986), an expert may not opine on the accuracy, reliability, or credibility of a particular witness, so testimony about gender differences, if admitted, must remain general and educational.
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. Neither men nor women are generally more accurate witnesses, the research does not support the view that women are generally more suggestible, and an argument that a complainant or witness is less reliable because of gender has no basis in the memory literature. Prosecutors are well positioned to keep attention on the factors that matter more, such as the conditions of observation, the delay, and how the account was obtained, and to document an early account for witnesses of any gender.
The research also counsels caution. The group differences are small and do not show that a particular witness noticed or remembered a particular detail, so they cannot corroborate an account. Confidence expressed later or after feedback does not show that a witness is accurate, although confidence recorded at a fair first test can be informative, and several studies found that men expressed more confidence than women without greater accuracy, a group-level finding that says nothing about any individual witness; confidence concerns apply to prosecution witnesses as well as defense witnesses. A tentative witness, whatever the witness’s gender, is not for that reason inaccurate.
Special Issues for Defense
The research supports several propositions defense counsel can legitimately present. Confidence expressed later or after feedback is not a reliable measure of accuracy, and some studies found that men expressed more confidence than women without greater accuracy. The own-gender bias in face recognition and the differences in attention to person and setting details are real at the group level, and expert testimony describing them, together with the factors that affect any witness’s memory, can help the factfinder understand why two witnesses to the same event may remember different things.
The research also counsels caution about the limits that apply to the defense. The differences are small, overlapping, and inconsistent across studies, and they do not show that a particular witness failed to notice or misremembered a detail. An expert may not testify that a witness’s account is inaccurate, and arguments that rely on a witness’s gender rather than on the circumstances of the observation risk inviting the stereotypes the research does not support. The same limits apply when the accused’s own recollection is at issue.
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
Areh, I. (2011). Gender-related differences in eyewitness testimony. Personality and Individual Differences, 50(5), 559–563. https://doi.org/10.1016/j.paid.2010.11.027
Asperholm, M., Högman, N., Rafi, J., & Herlitz, A. (2019). What did you do yesterday? A meta-analysis of sex differences in episodic memory. Psychological Bulletin, 145(8), 785–821. https://doi.org/10.1037/bul0000197
Asperholm, M., van Leuven, L., & Herlitz, A. (2020). Sex differences in episodic memory variance. Frontiers in Psychology, 11, Article 613. https://doi.org/10.3389/fpsyg.2020.00613
Butts, S. J., Mixon, K. D., Mulekar, M. S., & Bringmann, W. G. (1995). Gender differences in eyewitness testimony. Perceptual and Motor Skills, 80(1), 59–63. https://doi.org/10.2466/pms.1995.80.1.59
Clifford, B. R., & Scott, J. (1978). Individual and situational factors in eyewitness testimony. Journal of Applied Psychology, 63(3), 352–359. https://doi.org/10.1037/0021-9010.63.3.352
Herlitz, A., Frostegård, J. K. E., Asperholm, M., Bränström, R., Guest, E., Martinsen, J., Sonnegård, H., Sorjonen, K., Thorell, L. B., & Persson, B. N. (2026). The power of many small sex differences in cognition, personality, and interests. Scientific Reports, 16, Article 16572. https://doi.org/10.1038/s41598-026-53824-6
Herlitz, A., & Lovén, J. (2013). Sex differences and the own-gender bias in face recognition: A meta-analytic review. Visual Cognition, 21(9–10), 1306–1336. https://doi.org/10.1080/13506285.2013.823140
Hyde, J. S. (2005). The gender similarities hypothesis. American Psychologist, 60(6), 581–592. https://doi.org/10.1037/0003-066X.60.6.581
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
Loftus, E. F., Levidow, B., & Duensing, S. (1992). Who remembers best? Individual differences in memory for events that occurred in a science museum. Applied Cognitive Psychology, 6(2), 93–107. https://doi.org/10.1002/acp.2350060202
Palmer, M. A., Brewer, N., & Horry, R. (2013). Understanding gender bias in face recognition: Effects of divided attention at encoding. Acta Psychologica, 142(3), 362–369. https://doi.org/10.1016/j.actpsy.2013.01.009
Powers, P. A., Andriks, J. L., & Loftus, E. F. (1979). Eyewitness accounts of females and males. Journal of Applied Psychology, 64(3), 339–347. https://doi.org/10.1037/0021-9010.64.3.339
Russell, E. M., Longstaff, M. G., & Winskel, H. (2024). Sex differences in eyewitness memory: A scoping review. Psychonomic Bulletin & Review, 31(3), 985–999. https://doi.org/10.3758/s13423-023-02407-x
Yarmey, A. D., & Yarmey, M. J. (1997). Eyewitness recall and duration estimates in field settings. Journal of Applied Social Psychology, 27(4), 330–344. https://doi.org/10.1111/j.1559-1816.1997.tb00635.x
Legal Authorities
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).
Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).
Manson v. Brathwaite, 432 U.S. 98 (1977).
Neil v. Biggers, 409 U.S. 188 (1972).
State v. Chapple, 135 Ariz. 281, 660 P.2d 1208 (1983).
State v. Lindsey, 149 Ariz. 472, 720 P.2d 73 (1986).
State v. Moran, 151 Ariz. 378, 728 P.2d 248 (1986).
State v. Nordstrom, 200 Ariz. 229, 25 P.3d 717 (2001).
United States v. Houser, 36 M.J. 392 (C.M.A. 1993).
Ariz. Rev. Stat. Ann. § 13-4061.
Fed. R. Evid. 601.
Fed. R. Evid. 702.
Ariz. R. Evid. 601.
Ariz. R. Evid. 702.
Ariz. Sup. Ct. Order R-23-0004 (amending Ariz. R. Evid. 702, effective Jan. 1, 2024).
Mil. R. Evid. 601.
Mil. R. Evid. 702.
Mil. R. Evid. 1102.
Related topic papers: Memory in Special Populations · Face Recognition · Memory for Persons and Faces · The Cross-Race Effect · Confidence and Accuracy · Eyewitness Memory
Related topics from Arizona Forensic Services: Courts-Martial