by Zoe Farkas*

Fed­er­al Rule of Evi­dence 702 pro­vides that a wit­ness may only offer opin­ion tes­ti­mo­ny if they are an expert and sets out the stan­dard for expert tes­ti­mo­ny, requir­ing that:

(a) the expert’s sci­en­tif­ic, tech­ni­cal, or oth­er spe­cial­ized knowl­edge will help the tri­er of fact to under­stand the evi­dence or to deter­mine a fact in issue; (b) the tes­ti­mo­ny is based on suf­fi­cient facts or data; (c) the tes­ti­mo­ny is the prod­uct of reli­able prin­ci­ples and meth­ods; and (d) the expert has reli­ably applied the prin­ci­ples and meth­ods to the facts of the case.1

The Supreme Court expand­ed upon this rule in Daubert v. Mer­rell Dow Phar­ma­ceu­ti­cals, where it estab­lished a two-prong test for dis­trict courts to apply when faced with a pro­posed expert wit­ness: “whether the expert is plan­ning to tes­ti­fy to (1) sci­en­tif­ic knowl­edge that (2) will assist the tri­er of fact to under­stand or deter­mine the facts in issue.”2 The Daubert Court made clear that the objec­tive of this test was to ensure reli­a­bil­i­ty and rel­e­vance of the pro­posed expert tes­ti­mo­ny, as such tes­ti­mo­ny can deter­mine the out­come of a case.3

A cir­cuit split has emerged in recent years over whether the tes­ti­mo­ny of eye­wit­ness iden­ti­fi­ca­tion experts is admis­si­ble.4 Experts with spe­cial­ized knowl­edge about the cred­i­bil­i­ty of wit­ness mem­o­ry are often enlist­ed by the defense in cas­es where an eye­wit­ness has iden­ti­fied the defen­dant. His­tor­i­cal­ly, courts have been hes­i­tant to allow these experts to tes­ti­fy.5 How­ev­er, courts are grad­u­al­ly becom­ing more amenable to their tes­ti­mo­ny. As one court explained: “It may be that a door once large­ly shut is now some­what ajar.”6 This Con­tri­bu­tion will demon­strate why tri­al courts should open that door wider, there­by allow­ing juries to ful­fill their role as factfinders.

Under the first prong of Rule 702, a tri­al court must deter­mine whether an expert’s tes­ti­mo­ny con­sti­tutes sci­en­tif­ic or oth­er expe­ri­ence-based knowl­edge.7 In decid­ing whether the evi­dence the expert intends to rely upon is sci­en­tif­ic, the tri­al court can con­sid­er: (1) whether the the­o­ry or tech­nique can be test­ed, (2) whether the the­o­ry or tech­nique has been sub­ject to peer review or pub­li­ca­tion, (3) the known or poten­tial rate of error, and (4) the par­tic­u­lar degree of accep­tance with­in the sci­en­tif­ic com­mu­ni­ty.8 Eye­wit­ness reli­a­bil­i­ty experts come from social sci­ence back­grounds, and there­fore deci­sions regard­ing these types of wit­ness­es vary wide­ly because courts will often dif­fer in how they apply the Daubert fac­tors. Judges who choose to apply the Daubert fac­tors strict­ly under the ratio­nale that psy­chol­o­gy is a type of sci­ence may reject an expert because, for exam­ple, they can­not test their tech­nique on an indi­vid­ual basis with the eye­wit­ness whose reli­a­bil­i­ty is at issue or there isn’t a cal­cu­la­ble rate of error on an indi­vid­ual lev­el. How­ev­er, in Kumho Tire Co. v. Carmichael, just six years after Daubert, the Court clar­i­fied that these were not strict ele­ments, but rather fac­tors to be weighed.9 The Kumho Court empha­sized that the fac­tors used in assess­ing reli­a­bil­i­ty can be flex­i­ble based on the con­tent and nature of the issue.10 There­fore, if cer­tain Daubert fac­tors are not rel­e­vant to the exper­tise being con­sid­ered by the court, they should not be applied.11 The fac­tors serve as guide­posts and should not be treat­ed as a one-size fits all stan­dard, which courts reject­ing these experts have failed to rec­og­nize.12

While it is true that the evi­dence relied upon by eye­wit­ness reli­a­bil­i­ty experts can­not be val­i­dat­ed in the same way as tra­di­tion­al “hard sci­ences” can, the body of research in the area is now robust enough to sup­port a gen­er­al con­sen­sus in the sci­en­tif­ic com­mu­ni­ty about the lim­its of eye­wit­ness tes­ti­mo­ny and exac­er­bat­ing fac­tors that wors­en reli­a­bil­i­ty. The Sec­ond Cir­cuit rec­og­nized this real­i­ty in Young v. Con­way, where it held that such research “has been reviewed, repli­cat­ed, and retest­ed, and is gen­er­al­ly accept­ed in the sci­en­tif­ic com­mu­ni­ty.”13 The court in Con­way relied upon research pre­sent­ed by the Inno­cence Project that con­clud­ed that the pres­ence of a weapon, the stress of the sit­u­a­tion, an eye­wit­ness being sup­plied with post-inci­dent infor­ma­tion that could solid­i­fy false beliefs, the iden­ti­fi­ca­tion being cross-racial, the witness’s expo­sure to defen­dant through mul­ti­ple iden­ti­fi­ca­tion pro­ce­dures, and the pas­sage of time between obser­va­tion and iden­ti­fi­ca­tion, among oth­er fac­tors, can all impair a witness’s abil­i­ty to accu­rate­ly iden­ti­fy some­one.14 The evi­dence demon­strates that spe­cif­ic yet com­mon cir­cum­stances may wors­en an eyewitness’s already unre­li­able mem­o­ry; there­fore, even under a stricter Daubert analy­sis, this sci­ence should still be admissible.

After decades of mass incar­cer­a­tion, it is more vital than ever for courts to take a hard look at the past fail­ures of the crim­i­nal jus­tice sys­tem and do what they can to ensure juries come to the ver­dict jus­tice requires. Although the caus­es of mass incar­cer­a­tion are abun­dant and inter­con­nect­ed, wrong­ful con­vic­tions con­tribute sig­nif­i­cant­ly to this prob­lem.15 The dis­turb­ing real­i­ty that as many as one in twen­ty peo­ple incar­cer­at­ed in the Unit­ed States are inno­cent16 is dri­ven in large part by mis­tak­en iden­ti­fi­ca­tions.17 Many courts have rec­og­nized this injus­tice and have become more recep­tive to the sci­ence behind expert tes­ti­mo­ny on wit­ness iden­ti­fi­ca­tion.18

The sec­ond prong of Rule 702 requires that the expert tes­ti­mo­ny be help­ful to the jury.19 The emerg­ing research in the field estab­lish­es that juries absolute­ly need this infor­ma­tion to reach an informed ver­dict, and when they are deprived of it, the con­se­quences are a stain on the jus­tice sys­tem. Despite this real­i­ty, judges con­tin­ue to reject these experts as improp­er or unhelp­ful to juries in dis­charg­ing their duties.

One rea­son judges fre­quent­ly cite for reject­ing these experts is that deter­min­ing a witness’s cred­i­bil­i­ty is the respon­si­bil­i­ty of the jury alone. How­ev­er, it is becom­ing increas­ing­ly clear that, with­out this infor­ma­tion, jurors are unable to make cred­i­bil­i­ty assess­ments of eye­wit­ness­es. Over 50 years ago, the Supreme Court acknowl­edged the endur­ing prob­lems with eye­wit­ness tes­ti­mo­ny in Unit­ed States v. Wade, stat­ing that “[t]he vagaries of eye­wit­ness iden­ti­fi­ca­tion are well-known; the annals of crim­i­nal law are rife with instances of mis­tak­en iden­ti­fi­ca­tion.”20 The data pro­duced since Wade has only pro­vid­ed fur­ther sup­port for this concern.

Whether the sub­stance of a pro­posed expert’s tes­ti­mo­ny is con­sid­ered com­mon sense is often a deter­mi­na­tive fac­tor for judges in their Rule 702 analy­ses.21 While the gen­er­al notion of unre­li­able mem­o­ry may be well known, the spe­cif­ic mechan­ics of degen­er­a­tive mem­o­ry, the con­tribut­ing fac­tors, and the extent to which cer­tain fac­tors can impact mem­o­ry, are not. Fac­tors that are known to impact the form­ing of mem­o­ries are often present in crim­i­nal tri­als, for exam­ple cross-racial iden­ti­fi­ca­tions, extreme stress, and weapons focus. The aver­age juror would not be versed in how these fac­tors con­tribute to an eyewitness’s unre­li­a­bil­i­ty and may even believe the oppo­site.22

Sim­i­lar­ly, lay jurors are not gen­er­al­ly aware, nor is it obvi­ous, that there is a very weak cor­re­la­tion between con­fi­dence and accu­ra­cy of iden­ti­fi­ca­tions. The com­mon notion of unre­li­able mem­o­ry evokes an image of some­one strug­gling to remem­ber what they saw, not some­one con­fi­dent­ly assert­ing their false beliefs under oath. For this rea­son, cross-exam­i­na­tion, which is often employed to pro­vide evi­dence of a witness’s lack of cred­i­bil­i­ty to the jury, is insuf­fi­cient.23 Assum­ing lay jurors know not to empha­size con­fi­dence in deter­min­ing the cred­i­bil­i­ty of an eye­wit­ness is dan­ger­ous and leads to severe mis­car­riages of jus­tice. Accord­ing to the Inno­cence Project, eye­wit­ness misiden­ti­fi­ca­tions were present in an over­whelm­ing 71% of the cas­es in which sub­se­quent DNA test­ing estab­lished a wrong­ful­ly con­vict­ed defendant’s fac­tu­al inno­cence,24 and those are only cas­es where exon­er­a­tion was possible.

Courts that exclude experts often rea­son that cross-exam­i­na­tion and jury instruc­tions can rem­e­dy the issue. But those less rig­or­ous mech­a­nisms are sim­ply inef­fec­tive. Cross-exam­i­na­tion will rarely be help­ful because a mis­tak­en eye­wit­ness is wrong, but not lying. And jury instruc­tions are sim­ply a judge, who is a layper­son in this con­text, telling oth­er laypeo­ple about com­pli­cat­ed psy­chol­o­gy with­out the ben­e­fit of expand­ed ques­tion­ing and clar­i­fi­ca­tion on cross-exam­i­na­tion. In these cir­cum­stances, there is a lot of room for mis­takes, mis­un­der­stand­ings, mis­com­mu­ni­ca­tions, and gaps of knowl­edge to wreak hav­oc. Hav­ing an empir­i­cal­ly qual­i­fied expert on the stand to elu­ci­date the infor­ma­tion and con­cede weak­ness­es in con­clu­sions through cross-exam­i­na­tion from an adver­sary is much more effec­tive at achiev­ing an equi­table outcome.

Anoth­er con­cern often cit­ed by oppo­nents of the admis­si­bil­i­ty of eye­wit­ness iden­ti­fi­ca­tion experts is the “aura of spe­cial reli­a­bil­i­ty and trust­wor­thi­ness” sur­round­ing an expert wit­ness that has the poten­tial to lend them unearned cred­i­bil­i­ty with a jury.25 While there is a valid con­cern that juries will sim­ply lis­ten to an expert because they seem author­i­ta­tive, the same could be said for eye­wit­ness tes­ti­mo­ny that is already giv­en much more cred­i­bil­i­ty than it is due. In fact, “[a]ll the evi­dence points rather strik­ing­ly to the con­clu­sion that there is almost noth­ing more con­vinc­ing than a live human being who takes the stand, points a fin­ger at the defen­dant, and says ‘That’s the one!’”26 It is clear that sole­ly rely­ing on cross-exam­i­na­tion and jury instruc­tions against iden­ti­fi­ca­tion wit­ness tes­ti­mo­ny has failed. There­fore, per­haps it is time we instead rely on those mech­a­nisms to counter the qual­i­fied experts instead, either with anoth­er expert or through thor­ough cross-examination.

Of course, all issues of admis­si­bil­i­ty must always be gov­erned by the cir­cum­stances of the case at issue to deter­mine whether the evi­dence is sub­stan­tial­ly more prej­u­di­cial than pro­ba­tive, as Rule 403 requires.27 An expert’s tes­ti­mo­ny may, for exam­ple, not pre­vail in a Rule 403 analy­sis if they are only speak­ing to gen­er­al knowl­edge but can­not point to any spe­cif­ic fac­tor influ­enc­ing the con­test­ed witness’s mem­o­ry.28 Rule 403 is a rule of inclu­sion, not exclu­sion, and when the min­i­mal poten­tial prej­u­dice is weighed against the pro­ba­tive val­ue of this tes­ti­mo­ny, admis­si­bil­i­ty should pre­vail. Accord­ing­ly, tri­al courts should gen­er­al­ly find that the Daubert test per­mits qual­i­fied, reli­able eye­wit­ness iden­ti­fi­ca­tion experts who will tes­ti­fy about the impli­ca­tions of per­ti­nent fac­tors present in the case by default, and only exclude such experts if the cir­cum­stances make their tes­ti­mo­ny sub­stan­tial­ly more prej­u­di­cial than probative.

The con­se­quences of reject­ing such an appli­ca­tion of Daubert are par­tic­u­lar­ly grave because once a tri­al judge makes a rul­ing regard­ing these experts, there is lit­tle recourse for a con­vict­ed defen­dant to appeal this deter­mi­na­tion. Tri­al judges are meant to serve a gate­keep­ing role in apply­ing Rule 702 and appel­late courts defer to tri­al court find­ings unless it ris­es to an abuse of dis­cre­tion.29 This high bar requires that the tri­al judge’s use of dis­cre­tion be “man­i­fest­ly erro­neous” in order to be over­ruled,30 and, nat­u­ral­ly, few cas­es have met this high bar for reversal.

The bur­den is always on the gov­ern­ment to con­vince a jury of a defendant’s guilt beyond a rea­son­able doubt. Grow­ing evi­dence now defin­i­tive­ly shows that this can­not be achieved with­out allow­ing juries to have access to the infor­ma­tion only an expert can suf­fi­cient­ly pro­vide. Eye­wit­ness iden­ti­fi­ca­tion experts are a vital tool for defen­dants to get a fair tri­al and should be pre­sumed essen­tial, not prejudicial.


* Zoe Farkas is a J.D. Can­di­date (2023) at New York Uni­ver­si­ty School of Law. This Con­tri­bu­tion is a com­men­tary on the prob­lem at the 2022 Prince Evi­dence Com­pe­ti­tion host­ed by Brook­lyn Law School. The ques­tion pre­sent­ed was whether expert tes­ti­mo­ny on the reli­a­bil­i­ty of eye­wit­ness iden­ti­fi­ca­tions is admis­si­ble under Fed­er­al Rules of Evi­dence 702 and 403. This con­tri­bu­tion presents a dis­til­la­tion of the side of the argu­ment assigned to the author in the Prince Evi­dence Com­pe­ti­tion, and the views expressed here­in do not nec­es­sar­i­ly reflect the views of the author.

1. Fed. R. Evid. 702.

2. 509 U.S. 579, 592 (1993); see also Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999).

3. Kumho Tire Co., 526 U.S. at 152.

4. Com­pare Unit­ed States v. Lump­kin, 192 F.3d 280, 289 (2d Cir. 1999) (affirm­ing the exclu­sion of expert tes­ti­mo­ny on the unre­li­a­bil­i­ty of eye­wit­ness iden­ti­fi­ca­tion because it would usurp the jury’s role in deter­min­ing wit­ness cred­i­bil­i­ty), with Unit­ed States v. Stevens, 935 F.2d 1380, 1383 (3d Cir. 1991) (uphold­ing the inclu­sion of an expert on eye­wit­ness tes­ti­mo­ny and find­ing it to be nei­ther high­ly sug­ges­tive nor misleading).

5. E.g., Unit­ed States v. Fos­h­er, 590 F.2d 381, 382 (1st Cir. 1979); Unit­ed States v. Purham, 725 F.2d 450, 454 (8th Cir. 1984); Unit­ed States v. Lan­gan, 263 F.3d 613, 624 (6th Cir. 2001).

6. Unit­ed States v. Brien, 59 F.3d 274, 277 (1st Cir. 1995) (decid­ing not to “adopt a blan­ket rule that qual­i­fied expert tes­ti­mo­ny on eye­wit­ness iden­ti­fi­ca­tion must rou­tine­ly be admit­ted or excluded”).

7. Daubert, 509 U.S. at 592; see also Kumho Tire Co., 526 U.S. at 151 (hold­ing that expert knowl­edge can be based on expe­ri­ence as long as it is reliable).

8. Daubert, 509 U.S. at 593–94.

9. 526 U.S. at 150.

10. Id. at 141 (“[A]s the Court stat­ed in Daubert, the test of reli­a­bil­i­ty is ‘flex­i­ble,’ and Daubert’s list of spe­cif­ic fac­tors nei­ther nec­es­sar­i­ly nor exclu­sive­ly applies to all experts or in every case.”).

11. Id.

12. See, e.g., Unit­ed States v. Hud­son, 884 F.2d 1016, 1024 (7th Cir. 1989) (hold­ing eye­wit­ness expert tes­ti­mo­ny will not aid the jury because it address­es issues of which the jury is already gen­er­al­ly aware, despite the pres­ence of stress on the iden­ti­fy­ing wit­ness and a cross-racial identification).

13. 698 F.3d 69, 79 (2d Cir. 2012) (quot­ing State v. Hen­der­son, 27 A.3d 872, 916 (2011) (refer­ring to a com­pre­hen­sive 86-page report based on sev­en experts’ tes­ti­mo­ny and hun­dreds of sci­en­tif­ic stud­ies, includ­ing many of those cit­ed by the Inno­cence Project, as the gold stan­dard in apply­ing social sci­ence research to the law)).

14. Id. at 78­–79.

15. Claire Gilbert, Beneath the Sta­tis­tics: The Struc­tur­al and Sys­temic Caus­es of Our Wrong­ful Con­vic­tion Prob­lem, Geor­gia Inno­cence Project (Feb. 1, 2022), https://www.georgiainnocenceproject.org/2022/02/01/beneath-the-statistics-the-structural-and-systemic-causes-of-our-wrongful-conviction-problem/ (find­ing as high as 6% of peo­ple incar­cer­at­ed in the Unit­ed States pris­ons are actu­al­ly innocent).

16. Id.

17. Eye­wit­ness Iden­ti­fi­ca­tion Reform, The Inno­cence Project, https://innocenceproject.org/eyewitness-identification-reform/ (find­ing mis­tak­en iden­ti­fi­ca­tions to be a lead­ing cause of wrong­ful con­vic­tions); Richard A. Wise et al., An Exam­i­na­tion of the Caus­es and Solu­tions to Eye­wit­ness Error, Fron­tiers in Psy­chi­a­try, Aug. 2014, at 1 (esti­mat­ing that as many as one in three eye­wit­ness­es make an erro­neous identification).

18. E.g., Brien, 59 F.3d at 277 (“There is more expert lit­er­a­ture on the sub­ject [of wit­ness iden­ti­fi­ca­tion], more experts press­ing to tes­ti­fy, and pos­si­bly more skep­ti­cism about the reli­a­bil­i­ty of eye­wit­ness­es.”); Unit­ed States v. Stevens, 935 F.2d at 1384 (con­clud­ing that expert tes­ti­mo­ny about the weak rela­tion­ship between con­fi­dence and accu­ra­cy in eye­wit­ness iden­ti­fi­ca­tions was admis­si­ble under cir­cuit prece­dent and would have been help­ful to the jury).

19. Daubert, 509 U.S. at 592.

20. 388 U.S. 218, 228 (1967).

21. See e.g., Young, 698 F.3d at 79 (observ­ing that “many of these fac­tors are coun­ter­in­tu­itive and not coter­mi­nous with com­mon sense”); Unit­ed States v. Nolan, 956 F.3d 71, 81–82 (2d Cir. 2020) (“[I]t appears to us unlike­ly that [cross-racial iden­ti­fi­ca­tions being less accu­rate than intra-racial iden­ti­fi­ca­tions] is com­mon knowl­edge among lay jurors.”).

22. Inno­cence Staff, How Eye­wit­ness Iden­ti­fi­ca­tion Can Send Inno­cence Peo­ple to Prison, (Apr. 15, 2020), https://innocenceproject.org/how-eyewitness-misidentification-can-send-innocent-people-to-prison/ (“[M]ost peo­ple believe that high­ly stress­ful events are par­tic­u­lar­ly mem­o­rable . . . . How­ev­er, the high lev­els of stress that most of us would nat­u­ral­ly expe­ri­ence dur­ing a crime actu­al­ly impede reli­able mem­o­ry for­ma­tion.”); Young, 698 F.3d at 79 (stat­ing these fac­tors “are coun­ter­in­tu­itive and there­fore, not coter­mi­nous with ‘com­mon sense’”).

23. Con­tra Fos­h­er, 590 F.2d at 382 (1st Cir. 1979) (hold­ing that issues con­cern­ing eye­wit­ness cred­i­bil­i­ty can be expli­cat­ed more effi­cient­ly and less prej­u­di­cial­ly defense’s thor­ough cross exam­i­na­tion of the eye­wit­ness and the court’s com­pre­hen­sive jury instruction).

24. Nolan, 956 F.3d at 75 (cit­ing Eye­wit­ness Iden­ti­fi­ca­tion Reform, The Inno­cence Project, https://innocenceproject.org/eyewitness-identification-reform/).

25. Fos­h­er, 590 F.2d at 383.

26. Watkins v. Sow­ders, 449 U.S. 341, 352 (1981) (Brey­er, J., dis­sent­ing) (inter­nal cita­tion omitted).

27. Fed. R. Evid. 403.

28. See, e.g., Fos­h­er, F.2d at 382 (reject­ing an expert on mem­o­ry unre­li­a­bil­i­ty in part because the writ­ten offer of proof pro­claimed that the expert would not com­ment on the tes­ti­mo­ny of any named wit­ness in the tri­al); cf. Brien, 59 F.3d at 276 (bar­ring an expert wit­ness from tes­ti­fy­ing about mem­o­ry because he intend­ed to tes­ti­fy gen­er­al­ly with few details about the case and dis­claimed any abil­i­ty to deter­mine if a par­tic­u­lar wit­ness is cred­i­ble or accu­rate in mak­ing an identification).

29. Gen. Elec. Co. v. Join­er, 522 U.S. 136, 139 (1997) (hold­ing abuse of dis­cre­tion is appro­pri­ate stan­dard of review for tri­al court deci­sions to admit or exclude expert tes­ti­mo­ny under Daubert).

30. Id. at 142.