James M. Doyle, is of counsel to Carney & Bassil, Boston, MA. He is the co-author of Eyewitness Testimony:
Civil and Criminal (3d ed. 1997) with Elizabeth Loftus. His practice is focused on trial and appellate litigation in civil
and criminal cases. A former professor at Georgetown Law Center, Washington, DC, he was also Senior Trial Counsel
for the Roxbury Defenders (MA) and Deputy Chief Counsel (Head of Public Defender Division) for the statewide
Committee for Public Counsel Services.
©1997 James M. Doyle. All rights reserved.
One thing about eyewitness trials: any experienced lawyer can look pretty good while losing them. Among the first things the trial advocacy faculty teaches and a new defender masters is the Mr. Potatohead cross-examination:
Q: You didn't describe his eyes?
Q: You didn't describe his ears?
Q: You didn't describe his nose?
Q: You didn't even describe his little mustache?
Q: Nothing further! [Stalk to counsel table, drop pad, sit down. During closing argument, produce Mr. Potato-head and remove his features, one by one.]
Unfortunately, in real life this gratifying sequence is usually followed by a redirect examination seldom reproduced at the advocacy programs:
Q: How do you remember the robber?
A: I just remember he had a head sort of like a potato.
Q: How certain are you this is the man?
A: Absolutely. I'll never forget that face.
The eyewitness may be wrong, and the eyewitness may be impeached, but the eyewitness will be (and look) confident. That confidence has been the subject of extensive psychological study, and all of the news from the psychologists is bad for the defense. Wrong and impeached, the confident eyewitness will probably be believed,1 and hundreds of innocent defendants will be convicted as a result.2
The psychological experiments indicate that in most criminal cases the eyewitness' confidence has little or no correlation with the eyewitness' accuracy.3 But while psychologists are convinced that they have shown that confidence is at best a weak indicator of accuracy, and is usually no indicator at all, jurors rely more on the eyewitness' confidence than they do on genuinely influential factors such as disguise, weapon focus, violence and retention intervals in evaluating eyewitness testimony.4 Jurors seem to believe confident incorrect identification testimony just as frequently as confident correct testimony.5 Eyewitness confidence is extremely malleable, and is easily enhanced when an identification is confirmed by another witness, the police, or some innocent circumstance.6 Even worse, experienced lawyers don't seem to be any better at improving this situation than do third-year law students.7
Can Anything Be Done?
There are three traditional remedies for this problem: suppression motions, offers of expert psychological testimony, and cautionary instructions. None of these works very often. Since Manson v. Braithwaite8 only a tiny portion of eyewitness identifications are suppressed. In fact, the Manson opinion holds that witness confidence is a perfectly good indicator of reliability.9 Besides, it's important to remember that the suppression of one suggestive identification is not much help. Unless all identifications are suppressed the defense will need the first, suggestive, identification to undercut subsequent "independent" identifications.10
It's possible to discern a weak trend towards the admissibility of expert testimony on eyewitness identification in the aftermath of Downing v. United States 11 and Daubert v. Merrill-Dow Pharmaceuticals,12 but judicial hostility to expert testimony on eyewitness identification remains strong,13 judicial ingenuity in avoiding its admission can be very impressive,14 and it will be a long time before psychologists testify routinely for the defense in criminal eyewitness cases.
The First Circuit's observation on the admissibility of expert testimony seems entirely accurate: "A door once largely shut is now somewhat ajar."15 On the issue of eyewitness confidence, the most frequently discussed cautionary instruction -- one derived from the District of Columbia Circuit's opinion in United States v. Telfaire 16 -- actually pushes jurors in the wrong direction by encouraging them to employ confidence in evaluating identifications.
This article outlines a strategy for attacking the confident eyewitness that includes filing a suppression motion and offering an expert witness but recognizes -- as most realistic lawyers will -- that those efforts can't be counted on to succeed. This approach is no silver bullet -- it's more like a war of attrition -- but a consistent, determined campaign aimed at the issue of eyewitness confidence can show significant results.
This campaign has three principal thrusts:
- First, fight any effort to elicit testimony about, or make arguments based on, eyewitness confidence at every conceivable opportunity.
- Second, recognize that even if you succeed some elements of the eyewitness' confidence will be communicated to the jury, if only through the eyewitness' demeanor. Try to mute these effects by separating the eyewitness from the identification testimony, outlining what an ideal, reliable identification process would have been, and then contrasting that ideal process with the shoddy, suggestive process actually used.
- Finally, by pointing out the gap between reliable procedures and the procedures to which the eyewitness was subjected, encourage the jurors to rely on their own judgment, rather than the eyewitness'.
One lesson from the psychological findings is unmistakable: investigate now. In a typical case, the eyewitness' confidence grows as the trial approaches. The earlier in the case an interview is conducted, the lower the ceiling that can be imposed on the eyewitness' expressions of certainty. Frequently the witness who might say "75 percent" during the first week after the crime has arrived at "99 percent" by end of the direct examination. Obviously, the defense will want to have the "75 percent" figure available. Even if the eyewitness' early estimate is depressingly high it is worth having: a very small movement in the eyewitness' level of confidence will still prove that confidence can be changed. That qualitative point can be as important to the defense as a quantitative one: it makes it possible to point out that the witness could not have become more accurate as the trial approached, only more confident. It can show, in other words, that forces outside the witnessing situation are at work on the eyewitness' confidence.
In some cases, an early investigation may help not only to impose a ceiling on confidence estimates but actually to lower the estimates. Eyewitnesses readily revise their confidence estimates upwards, and downgrade them less easily.17 Even so, telling an eyewitness that another witness has failed to identify the defendant, or that the defendant has passed a polygraph, or has an alibi can sometimes shake (even if it never eliminates) the eyewitness' sense of certainty.18
The importance of this effort cannot be overestimated. Recent research indicates that as eyewitness' levels of confidence improve, they improve their accounts of the details of the event (e.g., the lighting, their opportunity to observe, the features they remember) accordingly.19 Interrupting that process is crucial.
In investigating eyewitness cases it is important to remember that the results of the identification process are not the best focus; the prosecutor will cheerfully tell you those. The real subject of an eyewitness investigation is the process itself. The answers the witness gave are less important here than the means by which the answers were elicited, and the ways in which the answers were received. Specifically, how did the process help to build the eyewitness' confidence?
Here, as in all areas of eyewitness cases, it is important for lawyers to make a one-time investment in reading at least a summary of the psychological literature on how the police could do it right.20 For example, psychologists believe that photo-arrays should be administered blind: that is, by an investigator who does not know who the suspect is and therefore cannot give suggestions or confirm feedback.
Similarly, psychologists believe that eyewitnesses should be asked to state their level of confidence immediately after making an identification, before any confirming influences can intervene.21 Knowledge of the extensive psychological findings on these issues (which are easily available in a variety of forms) fuels an accurate investigation of how the police -- sometimes unknowingly -- augmented the witness's progress towards confidence. 22
Confidence in Suppression Litigation
When the Supreme Court decided Manson v. Braithwaite it guaranteed that the defense will hardly ever win a suppression motion. Luckily, the Manson standard of "reliability" is so amorphous that the Manson opinion simultaneously created a situation in which the defense can almost always file a suppression motion. This opportunity should never be overlooked.
Because the Manson opinion cites confidence as a significant reliability factor, confidence is always relevant in suppression litigation. Attack it. Misconceptions about confidence can be challenged in a variety of ways during pretrial litigation. Even hopeless suppression litigation presents a vehicle for educating the judge on the distorting impact of eyewitness confidence. Judges who believe experts should not be allowed to testify at trial because they will overwhelm the credulous jurors are unlikely to admit that they are in danger of being overwhelmed themselves.23 Even if an expert does not testify in suppression proceedings, the memorandum in support of the motion to suppress can serve notice that the defense plans to contest the folk wisdom of the accuracy/confidence relationship. The amount of published material evaluating the confidence/accuracy relationship is enormous. 24 The memorandum can be either extensive or succinct, but it should present the basic challenge to the legitimacy of the inference that the eyewitness' confidence confirms his or her reliability.
Questioning during the hearing on the motion to suppress can also influence the witness' progression towards confidence and the events that drove the progression. Partly of course, this is important because you might win the motion, but it serves trial preparation and record-making purposes too. Because the trial testimony of an eyewitness has such devastating potential, the pretrial questioning has to be done in detail, so that counsel are armed with tight, useful impeachment for the trial cross-examination.25 Besides, the testimony at the suppression hearing can also form an important element of the campaign to seek introduction of expert testimony.
Offer of Expert Evidence About Confidence
In one survey 87 percent of experimental psychologists queried agreed that they would testify that confidence is not an indicator of accuracy. 26 Among psychologists who have been permitted to testify, the confi-dence/accuracy relationship was the second most frequent topic.27 Moreover, experiments with mock jurors indicate that expert testimony is the most effective way to modulate the jurors' reliance on eyewitness confidence. It tends to direct jurors attention away from inappropriate and misleading confidence estimates and towards witnessing and identification conditions (such as weapons focus) that are genuinely related to reliability. In short, the admission of expert testimony is worth a try.
The case for admitting expert psychological testimony has been ably made many times,28 and the best strategic and tactical approaches for accomplishing its admission are outside the scope of this article.29 For now, it will be enough to make two points: one about how to approach the issue, a second about why the effort will be useful even when it fails.
Although there are still courts that reject expert psychological testimony as "bad science,"30 that position will become more and more difficult to sustain as lawyers accustom themselves to the post-Daubert world and learn to make appropriate records. The ultimate test of admissibility set out in Daubert is very promising for proponents of expert testimony, but Daubert also institutes a formal "gatekeeper" role for the trial judge. This requires an adjustment: lawyers must now provide the judge in the offer of proof with enough of the underlying background of the scientific testimony to permit the judge to make a threshold decision.31 But even after the "bad science" hurdle has been surmounted, the problem of "fit" will always be present: does the expert's testimony, however valid, address a significant issue in this case? 32 An effective offer of proof in an eyewitness case will tell the trial judge why the confidence/accuracy relationship is crucial.
The judge must be shown two things: both that confidence does not predict accuracy and that jurors wrongly believe that it does. The offer of proof is incomplete unless it shows both of these facts. Usually, the success of the effort will depend on whether counsel has exploited the opportunity that investigation and suppression hearings presented to illuminate the witness' progression towards confidence and the other factors affecting reliability that the jurors' fascination with the confidence/accuracy connection threatens to obscure.
The offer of expert testimony should include information concerning the scientific method of the inquiries that led the expert to his or her conclusions, and a careful statement of the scientific conclusions on the confidence/accuracy relationship and its impact on jurors, but it must also include close analysis of the facts of the case. Part of this analysis should stress that experimental findings indicate that many of the factors traditionally used to buttress identifications (e.g., opportunity to observe) are entirely self-reported by the eyewitnesses, and are themselves suspect products of the confidence-enhancing procedures.33
Even when this approach does not compel an exercise of discretion admitting expert testimony on confidence it will frame the issue as one of two competing scientific propositions, one asserting, one denying, the existence of a predictive relationship between confidence and accuracy. That approach will prepare the ground for later battles.
Motion In Limine To Exclude Evidence of Confidence
If the judge cannot be persuaded to admit expert psychological testimony regarding confidence and accuracy there is another step available: move to exclude all prosecution reference to confidence at trial. This option will not be available unless counsel has taken advantage of earlier opportunities to challenge the confidence/accuracy relationship. Counsel who don't use those opportunities and object to testimony and argument about certainty for the first time mid-trial will be looked at as if they are Martians. But if earlier opportunities have been utilized and the judge sensitized to the issue it becomes possible to argue that since the defense cannot prove its scientific proposition, "confidence does not show accuracy," then the prosecutor cannot exploit the competing, inculpatory version, "confidence does show accuracy." Now the defense can ask, "If my evidence about confidence doesn't "fit," why does theirs?" On the surface this approach is novel, but in fact a motion in limine regarding eyewitness confidence draws on a number of black letter legal doctrines.
As unfamiliar as this attack is, judges on whom it has been tried have found its logic hard to evade. The motion in limine may not win an absolute ban on questions about confidence, but it will probably result in the judge's exerting some control over the prosecutor's presentation. It will certainly improve the appellate record, and it will increase the pressure on the trial judge to address the confidence/accuracy problem during instructions.
Trial Presentation: Addressing Confidence in the Opening Statement
Even if this pretrial skirmishing is successful, the witness will still testify, and although confidence is never mentioned in so many words, the witness will exude it. The impact of eyewitness confidence will have to be muted in every eyewitness trial. Where can you start?
Jurors don't wait until all the evidence is in and then evaluate the case. From a trial's first moments jurors are trying to fit each piece of evidence into a story or stories that will explain the facts and help them to resolve the dispute. Litigators have always known that opening statements are important, psychologists have explained why: openings provide the stories that jurors use to frame the evidence. 34
The prosecutor's story is simplicity itself: he did it; they saw it; they're certain; that's it. In the prosecutor's story the eyewitness' confidence is the conclusion, and in it, confidence amounts to the same thing as correctness.
The defense must change this story. In the defense version it is conceded that the witness is confident, but so what? The defense story is about how the mistaken witness became confident: Mrs. Smith will take the stand. She will point at my client. She tell you that in her opinion, that's the man who robbed her. She will be sincere: she will have no interest in a wrongful conviction. She will seem certain. The evidence will show you that she is wrong, and that you cannot accept her opinion beyond a reasonable doubt. How can that be? The evidence will show that too.
This approach has the virtue of stealing the prosecutor's thunder,35 but it should do more than that. It can begin the project of separating the sympathetic, sincere witness from the identification and freeing the jurors to reject the eyewitness' judgment. How to do this during opening statement is a topic for a whole article, 36 but two suggestions may be helpful here.
The first suggestion is on the level of structure. Jurors tend to use the video tape metaphor in thinking about eyewitness. The opening can begin to subvert that misconception if it explicitly breaks the process down into its component tasks of perception, storage and recall. For example, the opening can be organized into paragraphs, each one beginning with something like: "The evidence will show you just how difficult it would have been for Mr. Carney to see (remember, identify, etc.) the robber...." Within those paragraphs, the ideal viewing (or storage, or retrieval) conditions can be introduced and contrasted with the conditions governing this identification. So, the opening might continue: "The evidence will show that Mr. Carney was taken to the police station and told to view a photo-spread; not one photograph at a time so that he could make a true/false choice, but in a group of five, prompting a 'most like' decision. . . ."
The second suggestion is on the level of language, one phrase at a time. In the defense story the sympathetic, well-meaning eyewitness is the victim of the shoddy processes used by the authorities. To make this point, eyewitnesses appear in the passive voice. The don't go to the police station; they are brought. They don't look at an array; they are shown an array.
Cross-Examination and Confidence
An important benefit of separating the eyewitness from the identification is that it provides a new point of attack. Because eyewitnesses are almost always sympathetic, unbiased figures with whom the jurors will identify, lawyers are compelled to treat them gently. Besides, eyewitnesses are sincere, and prolonged attack on eyewitnesses will simply enhance the impact of their sincere (and confident) demeanor on the jurors.37 By conceptualizing the eyewitness as the victim of the authorities' defective procedures the authorities administering the process are made available as targets who can be examined with no holds barred.
In essence then, cross-examination about confidence is two cross-examinations, one gentle, one aggressive. These examinations share a common background. Psychologists have uncovered many of the elements of a reliable identification process. For example: a witness should be instructed that there may not be a suspect in the photo-array; shown a "blank" array (with no suspect) first; shown the photographs sequentially, not all at once (to promote absolute, rather than relative judgment); be shown photographs that selected to match their description of the culprit (rather than resemble the suspect); be shown photographs by "blind" investigators who do not know who is the suspect, and be shielded from the reactions of investigators and other witnesses to their selections. 38 The goal of cross-examination must be to show that these procedures could have been followed but were not. Many commentators have provided advice about cross-examining on these issues. 39 For now, it is probably enough to provide a brief example of each version.
First, consider cross-examining the witness about the construction of confidence through feed-back:
Q: The police called you to the station house?
Q: And said they wanted you to look at some pictures?
Q: You thought about why they were doing that?
Q: You didn't think they were just wasting your time?
Q: You thought they had a reason for bringing you down?
Q: You thought they had a suspect?
Q: Now, they didn't force you to go there, you wanted to?
Q: To catch the robber?
Q: Before he robbed again?
Q: And they showed you a group of pictures?
Q: And had you compare these pictures to each other?
Q You never saw them one at a time?
A: No, they were in a group.
Q: And you chose the one who looked most like the robber?
Q: And Detective Bassil stopped right there?
Q: She didn't show you any more pictures?
Q: But she let you study the one you'd picked?
Q: Now after that you were curious about how you'd done?
Q: Then she asked you to come to a line-up?
Q: And in the line-up you saw the man in the picture you'd chosen?
Q: And here he is again?
For the detective, a more aggressive approach is in order:
Q: You have access to hundreds of pictures?
Q: But you only showed six?
Q: And you never showed them one picture at a time?
Q: So the witness had to say yes or no?
Q: You showed them in a group?
Q: So the witness could say "most like?"
A: I don't know what you mean.
Q: You knew who the suspect was?
Q: And you administered the array yourself?
Q: You were there while the witness was looking?
Q: And you stayed there after she made her choice?
Q: And you knew she had chosen your man?
Q: You never told her that there might be no suspect in the group?
The research on the optimal ways to conduct identification procedures is so compelling that it would be easy to make a policy argument that following those optimal procedures should be the required foundation for evidence of eyewitness confidence. The day that those policies are instituted, however, is a long way off, and in the meantime, cross-examination, while it will rarely if ever shake the eyewitness' confidence, can at least begin to explain it, by contrasting reliable identification procedures with those actually employed.
Confidence in Instructions and Argument
Many defense lawyers spent the 1980s beating their brains out arguing for a Telfaire cautionary instruction on eyewitness identification. There is now reason to believe that in some circumstances the Telfaire instruction actually leads to a higher rate of conviction.40 One reason for this is the Telfaire instruction's direction that the jurors consider the "strength" of the identification. This can only aggravate the jurors' pre-existing reliance on witness confidence. The least a lawyer can do in an eyewitness case is to object to that language and ask that it be excised. At least where the trial judge has been bombarded (and the record saturated) with psychological material disputing the relationship, that request has a reasonable chance of success. If the case is at all close, the judge, having refused to suppress, having excluded expert testimony, and allowed argument on confidence, is now feeling a little nervous.41
A more ambitious approach is to seek an instruction directly addressing confidence. For example: Identification testimony is an expression of opinion or impression by the witness. In deciding whether or not to believe a witness who identifies the defendant, remember that you must consider not only whether the witness is telling the truth or is lying. You must also consider the possibility of "good faith mistake." In weighing the testimony of an eyewitness, you should remember that there is no proven relationship between a witness' confidence and a witness' accuracy. An eyewitness' certainty about an identification may have many sources, and does not necessarily bear on the correctness of the identification. Indeed, some believe that an eyewitness' confidence is among the least significant factors in predicting the accuracy of an identification. The accuracy of an identification is a question for you the jury to decide beyond a reasonable doubt.
Remember that the point of these instructions is not simply to lower the jurors reliance on the eyewitness but to enhance the jurors' reliance on themselves. In their brilliant article on closing argument Randy Hertz and Anthony Amsterdam point out that a potential story for the defense closing argument is not the story of the crime, but the story of the trial.42 In that story, the jury is the hero, and it is embarked on a heroic quest. As Amsterdam and Hertz describe the argument:
1. The Hero, the Jury, is set on stage and reminded of its oath;
2. The jury's virtues (dedication, independence, etc.) are extolled;
3. The temptations that might tempt the Hero (e.g., to defer to the eyewitness) are described.
4. The jurors are called to their task (e.g., to answer whether the eyewitness is correct or mistaken);
5. The jurors struggle with the evidence;
6. The jurors find the answer to the Riddle, by keeping faith with their oath, respecting the burden of proof and employing their own judgment.
In the Amsterdam/Hertz analysis jurors, "faithful to their oath, acquitted the defendant although . . . sorely tempted . . . do otherwise."43
The jurors, reminded that they have had ideal conditions to decide (the witnesses, the lawyers, each other, the instructions, a clean well-lighted place, and all the time in the world) must be encouraged to substitute their vote for that of the eyewitness who labored under conditions (a quick, frightening crime on a dark night, a suggestive identification procedure and a badly flawed pretrial process). The traditional reasonable doubt and burden of proof instructions can be integrated into the attack on the identification procedures to stress for the jurors that as a matter of law as well as a matter of fact it is their confidence that counts.
A Good Place To Start
Trial lawyers have an endemic tendency to treat psychologists who study eyewitness identifications as a quick fix for a bad case: get them on the stand and they'll solve the problem. A better approach is to absorb the psychologists' knowledge and integrate it into a unified presentation. The issue of the eyewitness' confidence -- arguably the most crucial of all the jurors' misconceptions about eyewitness -- is a good place to start.
1. G. Wells & D. Murray, Eyewitness Confidence, in G.L. Wells & Elizabeth Loftus, Eyewitness Testimony: Psychological Perspectives 155 (1984); Penrod & Cutler, Witness Confidence and Witness Accuracy: Assessing Their Forensic Relation, 1 Psych. Pub. Pol. & Law 817 (1995).
2. U.S. Department of Justice, Convicted By Juries, Exonerated By Science: Case Studies in the Use of DNA Evidence to Establish Innocence After Trial, 16-17 (1996)(majority of wrongful convictions cited depended on eyewitness testimony at trial.)
3. Penrod & Cutler, supra, note 1, at 825, marshall the numerous studies and conclude that the most appropriate generalization is that "under the conditions that typically prevail in short criminal encounters . . . witness confidence is largely unrelated to accuracy, and confidence in having made a correct identification is, at best, only modestly associated with identification accuracy."
4. Cutler, Penrod & Stuve, Juror Decision-Making in Eyewitness Identification Cases, 12 Law & Human Behavior 41 (1988).
5. Luus & Wells, The Malleability of Eyewitness Confidence: Co-Witness and Perseverance Effects, 79 J. Applied Psychology 714 (1994).
6. Lindsay, Wells & O'Connor, Mock Juror Belief of Accurate and Inaccurate Eyewitnesses: A Replication and Extension, 13 Law & Human Behavior 333 (1989).
8. Manson v. Braithwaite, 432 U.S. 98 (1977).
10. Elizabeth Loftus & James Doyle, Eyewitness Testimony: Civil and Criminal, (LEXIS Law Publishing 3d ed. 1997).
11. Downing v. United States, 753 F.2d 1224 (3d Cir. 19__ ).
12. Daubert v. Merrill Dow Pharmaceuticals, ___ U.S.___ , 113 S. Ct. 2786.
13. See, e.g., United States v. Labanstat, 94 F.3d 527 (9th Cir. 1996); United States v. Kime, 99 F.3d 870 (8th Cir. 1996).
14. United States v. Brien, 59 F.3d 274 (First Cir. 1995).
15. Id. at 281.
16. 469 F.2d 552, 558-59 (D.C.Cir. 1979).
17. Luus & Wells, supra, note 5.
19. Cohen, I Could Swear It Was Him, Officer, New Scientist, January 18, 1997, at 11, reporting forthcoming research by Iowa State psychologist Gary Wells and his student Amy Bradfield, forthcoming in ___J. Applied Psych___.
20. Wells & Seelau, Eyewitness Identification: Psychological and Legal Policy on Lineups, 1 Psych.Pub.Pol. & Law 765 (1995); Nettles, Nettles & Wells, Eyewitness Identification: "I Noticed You Paused At Number Three," The Champion, Nov. 1996, at 11.
22. A comprehensive collection of experimental findings can be found at Brian Cutler & Stephen Penrod, Mistaken Identification: The Eyewitness, Psychology and the Law (1995).
23. A similar point is made in regard to the admission of polygraph evidence in non-jury proceedings in United States v. Posado, 57 F.3d 428 (5th Cir. 1995).
24. See, Cutler & Penrod, supra, note 22.
25. See generally, Loftus & Doyle, supra, note 10, at 148-159; 246-249.
26. Kassin, Ellsworth & Smith, The General Acceptance of Psychological Research On Eyewitness Testimony: A Survey of the Experts, 44 Am.Psychologist 1089 (1989).
28. An excellent version, integrating the legal and psychological materials is M. Leippe, The Case for Expert Testimony About Eyewitness Memory, 1 Psych.Pub.Pol.& Law 909 (1995).
29. See generally, E. Loftus & J. Doyle, supra, note 10, at 273-312.
30. See, e.g., United States v. Labanstat, 94 F.3d 527 (9th Cir. 1996). 31. The First Circuit lays out the "gatekeeper" requirements in United States v. Brien, 59 F.3d 274 (First Cir. 1995). For cases acknowledging the Brien gatekeeper concerns and finding that they were met by the defense offer of proof, see, United States v. Jordan, 924 F. Supp. 443 (W.D.N.Y. 1996); United States v. Norwood, 939 F. Supp. 1132 (D.N.J. 1996). Effective presentations of the science of eyewitness identification that address "gatekeeper" concerns are found in Cutler & Penrod, supra, note 22, and Leippe, The Case for Expert Testimony About Eyewitness Memory, 1 Psych.Pub.Pol. & L. 909 (1995).
32. See, generally, United States v. Downing, 753 F.2d 1224 (3d Cir. 1985).
33. See note 18, supra.
34. Pennington & Hastie, Explaining the Evidence: Tests of the Story Model for Juror Decision Making, 62 J. Pers. & Soc.Psych. 189 (1992).
35. Williams, Bourgeois & Croyle, The Effects of Stealing Thunder in Criminal Trials, 17 L. & Hum. Behav. 597 (1993).
36. Loftus & Doyle, supra, note 10, 181-197.
37. Leippe, supra, note 28, at 923, notes that even when a witness' confidence can be shaken, that does not improve the ability to detect inaccuracies: (". . .tactics (in cross and direct examination) are used that almost inevitably lead to altering and making salient exactly what can most mislead jurors: eyewitness confidence. Instead of focusing jurors on a reasonable assessment of the physical and social conditions of witnessing and identification, cross-examination is likely to direct attention to less diagnostic features involving demeanor. . . .")
38. See, generally, Nettles, Nettles & Wells, supra, note 20.
39. Id. See also, Loftus & Doyle, supra, note 10 at 229-259.
40. Greene, Judge's Instructions on Eyewitness Testimony: Evaluation and Revision, 18 J. Applied Soc. Psych. 252 (1988).
41. For cases holding that instructions should not encourage the consideration of witness confidence see, Commonwealth v. Jones, 423 Mass. 99 (1996); Commonwealth v. Santoli, 424 Mass. 837 (Sup. Judicial Court 5/8/97). In Santoli the Supreme Judicial Court noted that: "It is probably true that the challenged [confidence] instruction has merit in so far as it deals with the testimony of a witness who expressed doubt about the accuracy of her identification, whether that identification was made during her testimony , or at a "showup" or lineup. Where, however, the witness has expressed great confidence in her identification of the defendant, the challenged instruction may pose a problem because...there is significant doubt about whether there is any correlation between a witness's confidence in her identification and the accuracy of her recollection." The Santoli opinion was the culmination of a campaign along the lines of that suggested in this article, by public defenders Cathy Bryne, Ben Keehn, Paul McManus and Larry Tipton, among others.
42. Amsterdam & Hertz, An Analysis of Closing Argument to a Jury, 37 N.Y.L. Sch. L. Rev. 55 (1992).
43. Id. at 73.
1025 Connecticut Ave. NW, Ste. 901, Washington DC 20036