People v. Williams

14 Misc. 3d 571

Opinion

OPINION OF THE COURT

Joseph Kevin McKay, J.

Introduction

This opinion is an expanded version of this court’s June 20, 2006 oral ruling and short form decision and order granting defendant’s pretrial in limine application to call an expert witness at trial on the issue of eyewitness identifications. My ruling was, of course, heavily influenced by the two most recent Court of Appeals decisions on this issue (see People v Drake, 7 NY3d 28 [2006]; People v Young, 7 NY3d 40 [2006]), and was based upon the Frye hearing (see Frye v United States, 293 F 1013 [DC Cir 1923]) I ordered and conducted. For reasons discussed hereafter, significant focus was placed by this court during the Frye hearing on the decision in People v LeGrand (196 Misc 2d 179 [Sup Ct, NY County 2002], affd 28 AD3d 318 [1st Dept 2006]).¹

Defendant, Jeffrey Williams, has been indicted for robbery in the first degree (Penal Law § 160.15 [4]) and related charges arising out of an alleged gunpoint robbery of a laundromat managed by the complaining witness. By notice of motion dated May 19, 2006, defendant sought the court’s leave to offer expert testimony at trial regarding scientific studies of factors affecting the perceptual ability and memory of eyewitnesses to make identifications. The People orally opposed the motion in court on June 19, 2006, but conceded the need for a Frye hearing. That hearing was held and completed the next day, followed immediately by the court’s findings and conclusions and a written short form decision and order, dated June 20, 2006,² granting defendant’s motion.

Factual Background

On November 9, 2005 at approximately 5:40 p.m., complainant Maribel Camacho was working as the manager in a laundromat at 1090 Sutter Avenue, Brooklyn, when a person later identified as defendant³ and another unapprehended male entered. At the time, Camacho’s teenage son, Gustavo Dominguez, was inside visiting his mother. Perpetrator No. 1 had a gun in his hand and pointed it at the son’s rib cage, seeking to know who was in charge and demanding money. Camacho went with perpetrator No. 1 into an area where the cash register was located while the other male remained as a lookout at the front door. Camacho was so nervous that she was unable to open the register so perpetrator No. 1 attempted to do so by banging it with the gun. Upon observing this, Camacho again tried to open it and succeeded. Perpetrator No. 1 removed the money, put it in his jacket and left.

At the point when perpetrator No. 1 went behind the counter, Dominguez was able to escape past the male guarding the door but could hear the sounds of the gun banging against the register. He went to a bodega for help and, when he returned to the laundromat with the police, perpetrator No. 1 and the other male were gone. Camacho and Dominguez participated in an immediate canvass of the area with the police with negative results.

The son viewed a computer-generated photo array that same day at the 75th Precinct and selected defendant’s photograph as perpetrator No. I.⁴ Defendant was subsequently arrested on an unrelated marijuana charge on November 27, 2005 and based on the earlier robbery photo “hit” he was taken from central booking to the 75th Precinct and placed in a lineup that same day. Camacho and Dominguez separately viewed the lineup and each identified defendant as perpetrator No. 1. The People concede that there is no independent evidence corroborating the identification testimony of the two witnesses.

Frye Hearing

The sole witness to testify at the Frye hearing was Dr. Margaret Bull Kovera, called by the defense, and subject to extensive cross-examination by the People.⁵1 found Dr. Kovera’s testimony to be scholarly and balanced. She presented her research and her opinions with thoughtful circumspection.

Dr. Kovera is a social psychologist and the director of the Forensic Psychology Doctorad Program at John Jay College, City University of New York. At the time of her testimony Dr. Kovera had just been named the president-elect of the American Psychology-Law Society, a post currently held by Dr. Gary Wells, whom she regards as the leading expert in the field of eyewitness identification.⁶ Dr. Kovera received her B.A. in psychology from Northwestern University and her Ph.D. in social psychology from the University of Minnesota. She specializes in the application of social psychological theory to the study of jury decision making and eyewitness identifications.⁷ She has done extensive funded research and has published in both areas. Dr. Kovera’s expert testimony had been admitted at a Frye hearing in a Florida court, but she had never before testified in a New York court. Dr. Kovera was questioned about those six areas relating to eyewitness identification previously enumerated (see n 2, supra), which defendant Williams asserted were all applicable to his case: (1) cross-racial identification, (2) weapon focus, (3) exposure duration, (4) confidence malleability, (5) mug shot exposure, and (6) double-blind lineups and the National Institute of Justice standards for lineups. Before discussing these topics, Dr. Kovera described the analytic aspect of research in this field, explaining that the majority of researchers of eyewitness identification rely upon the now well-accepted technique called “meta-analysis.”⁸ It is a statistical method utilized to synthesize or pool large numbers of separate but related studies to determine if there is a general'effect that can be reliably ascertained.

Dr. Kovera also testified that Dr. Saul Kassin conducted an empirical survey in 1989⁹ of leading experts studying eyewitness identification to see whether they believed that certain enumerated eyewitness effects or phenomena were considered reliable enough to be the subject of expert testimony in court. An updated Kassin survey was conducted in 2001.¹⁰ It is chiefly this updated 2001 survey which was the subject of Dr. Kovera’s testimony and upon which defendant relies heavily in order to meet his Frye burden.¹¹

On her direct testimony, as well as cross-examination, Dr. Kovera was asked to address concerns raised by the prosecutor and the court during oral argument on the previous day regarding Dr. Ebbe Ebbesen’s testimony in the Frye hearing held in People v LeGrand (196 Misc 2d 179 [Sup Ct, NY County 2002], affd 28 AD3d 318 [1st Dept 2006], lv granted 7 NY3d 758 [2006]). Dr. Ebbesen is one of the major critics of the use of meta-analysis as a statistical tool for research in the field of eyewitness identification.

Dr. Kovera explained and to a large extent refuted Dr. Ebbesen’s criticisms. She does not consider him one of the leading experts on eyewitness identification because to her knowledge he had not done substantial research in this area until “somewhat recently.” She believes he has been inaccurate and simplistic in some of his evaluations and characterizations of research methodologies. For example, he dismisses meta-analysis by arguing that averaging a number of studies might show an effect that does not fully explain variables between studies. Dr. Kovera testified, however, that meta-analysis looks for “moderating variables” present in the included studies and codes and weights them proportionately in drawing conclusions. Further, researchers are also expected to weight the different studies to be synthesized in a meta-analysis by the number of participants in each study. It follows from this more sophisticated approach that the fewer people in the study, the less the statistical effect in the overall conclusions drawn in the meta-analysis. Dr. Kovera knows of no published criticism of meta-analysis as a research methodology¹² and she believes only a small number of people espouse Dr. Ebbesen’s views.¹³ Dr. Ebbesen is also a vocal critic of the Kassin surveys.¹⁴ He faulted the 2001 survey for having less than a 50% reply rate from those solicited to participate.¹⁵ Dr. Kovera countered that the return should still be considered high for experts because it is likely many opted out based on their self-assessments that they were not sufficiently qualified as experts in that specialized field. Dr. Kovera testified that one of the criteria for participation in the more recent survey was scholarly publication on eyewitness identification within the previous 10 years of the survey. She noted that the responders were among the most widely published in the field, a fact which in her view added to the prominence of the survey.¹⁶ Dr. Kovera was a participant. As for Dr. Ebbesen’s criticism that the survey field of witness identification experts was too small and should have been expanded to all researchers in human memory, apparently to include himself,¹⁷ Dr. Kovera strongly disagreed, predicting that such expansion would in all likelihood have reduced the reply rate to 5%. She pointed out that memory is a huge field and most researchers within that broad spectrum do not stay current with the much more specialized applied studies in eyewitness identification.

(1) Cross-Racial Identification

Dr. Kovera testified that a meta-analysis study conducted by Drs. Meissner and Brigham¹⁸ on cross-racial identification concluded that people are better at identifying members of their own race than those from other races.¹⁹ Included in that study was a field study by Drs. Platz and Hosch involving the interaction of African-Americans, Hispanics of Mexican descent and Caucasians in a convenience store.²⁰ Defendant Williams is an African-American male and the complaining witness and her son have been described as Hispanic (they are actually of Dominican heritage). Dr. Kovera stated that the less exposure time there is to the face of a person of another race, the greater the effect.²¹

Dr. Kovera addressed “encoding,” which refers to the storage of information in a person’s memory at the time of exposure to a stimulus, which information cannot be changed. She testified that scientists are beginning to see that the effect of cross-racial identification occurs at the encoding stage, that is, at the time one first sees the face of a subject. She conceded that there is debate within the scientific community as to the exact cause of cross-racial identification bias, and there is ongoing study to discover the underlying psychological mechanisms at work. Nevertheless, Dr. Kovera insisted that the phenomenon itself has been studied by social scientists for over 25 years and it is generally accepted in the scientific community. Dr. Kovera testified that the 2001 Kassin survey revealed that 97% of respondents said the effect of cross-racial identification was research based and 90% felt it was a reliable area about which to testify. Although the question was not presented to Dr. Kovera, the 2001 survey indicates that 72% also stated that they would be willing to personally testify about this effect in court.²²

(2) Weapon Focus

Dr. Kovera testified that Dr. Nancy Steblay²³ and others have done meta-analysis studies in the field of “weapon focus.” According to their findings, when a witness is exposed to a weapon during the course of a crime, she has a tendency to focus her attention on that weapon and not the perpetrator’s face, which impairs the ability of that witness to make a subsequent identification of the perpetrator.

Dr. Ebbesen had belittled the research, observing that of 19 included studies only 6 showed a positive effect of a weapon’s presence while 13 showed no effect. Dr. Kovera again explained, however, that Dr. Ebbesen’s criticism was too simplistic in that it merely counted the studies numerically and compared their results, overlooking the way meta-analysis weights different factors to account for variables and ignoring a statistical tool called the “fail-safe N.”²⁴ Using that tool, an additional 1,500 studies showing no weapon focus effect would have to be performed in order to negate the conclusion drawn regarding “the significant effect of weapons,” as Dr. Kovera characterized it, found by the meta-analysis of the combined 19 studies.

Dr. Ebbesen has also argued that the failure to brandish actual weapons in laboratory settings diminishes the value of these weapon focus studies. Dr. Kovera explained that, although ethical considerations prevent the use of actual weapons, researchers attempt to simulate the same effect by showing participants videotapes during the experiments with weapons present and absent. Dr. Kovera opined that the weapon focus effect in the lab would naturally be expected to be less than for a real life crime.²⁵

Dr. Kovera testified that the Kassin weapon focus survey found that 97% of the responders reported there was a research basis underlying this particular effect. Adso, 87% thought the effect was reliable enough to testify about in court, and over 73%²⁶ would personally testify about the effect. These numbers represent a growth in consensus from 1989 to 2001, as measured by the Kassin surveys.

(3) Exposure Duration

The longer a person is exposed to a face, the more likely he is to make a correct identification later. Dr. Kovera testified that although this sounds commonsensical, it still helps to have an expert explain it to a jury. Dr. Ebbesen criticized those studies as only measuring short periods of time, that is, less than a minute. Dr. Kovera agreed there are not many facial recognition paradigm studies of longer observation periods. She pointed out, however, that longer exposure times raise confidence/accuracy relationship issues.²⁷ Dr Kovera testified that, according to the 2001 Kassin survey, 93% of the responders believed there was a research basis for the “exposure duration” findings, 81% thought the conclusions were reliable enough to be the subject of expert testimony and 68% reported that they would personally testify about the results.

(4) Confidence Malleability

Research studies have shown that, when a witness makes an identification and receives confirming feedback from the police, whether by an overt cheer or a more subtle nod or smile, the witness’s confidence about her identification is likely to be enhanced, even when she has selected the wrong person. This increased confidence may in turn distort her original memory of the crime and encourage the witness to believe she had a better and longer view of the perpetrator than she actually did. This effect could also lead the witness, though not deliberately, to exaggerate these details in court testimony.

As will also be noted later in the section on “double-blind lineups,” this phenomenon of confidence malleability is the reason scientists prefer the administrator of a lineup to be unaware of the identity of the suspect. Failing that, scientists urge at least that the witness’s confidence level be recorded immediately after the identification. Neither of these precautionary steps was taken in this case.²⁸

Because this is a relatively new area of research, there is currently no published meta-analysis on the subject of confidence malleability. The Kassin survey also found the confidence malleability effect very reliable: 95% of the experts responding thought there was a strong research basis behind this data, 95% also viewed the effect reliable enough to be the subject of testimony in court, and 79% were willing to testify about it themselves.²⁹

(5) Mug Shot Exposure

The research on mug shot exposure is based on the theory of “unconscious transference.” Memory is fallible and a person can see someone in a situation and misremember where she saw him. In other words, if a person sees a mug shot or photo array containing a suspect, such person is more likely to identify the suspect later in the lineup, whether or not such suspect is the actual perpetrator. Moreover, studies show that the effect is exacerbated when one has actually chosen the mug shot previously and committed herself to it. The relevance for the jury in the trial of this indictment lies in the risk that a witness to a crime, while having a memory of the perpetrator, may, after extensive exposure to mug shots of potential suspects, choose somebody who looks somewhat similar, but who may not be the actual perpetrator. That choice in and of itself will increase the likelihood that the witness will pick the same person out of the lineup, in some instances to the exclusion of the actual perpetrator who is present (as shown in studies using controlled experiments).

According to Dr. Kovera, Dr. Ebbesen devalues this research because the experiments in mug shot exposure are often performed in a laboratory with college students, who are too select a group and who may seek to please the instructor or researcher. Dr. Kovera’s answer is that a good researcher would hide the underlying hypothesis from the participants, just as a pharmaceutical researcher would do in drug-testing trials, thereby making it very difficult for a participant to skew the results by trying to please the researcher. Also, Dr. Kovera believes that real-life witnesses to a crime viewing mugbooks or lineups may have as great or greater a tendency to try to please the police as a student would to please an instructor. Dr. Kovera further asserted that she is unaware of any evidence that the cognitive process of college students regarding encoding and recall is different from the population at large for remembering faces.

Dr. Kovera testified that, according to the 2001 Kassin survey, mug shot exposure effect (referred to in the survey as “Mugshot-induced-bias’’; see n 11, supra, and 56 Am Psychologist [No. 5] at 412) is one of the effects deemed most reliable by the experts: 97% of responders thought that it was research based, 95% thought it was a reliable phenomenon (that is, reliable enough to testify about in court), and 77% were personally willing to testify about it. These impressive results predate the newest meta-analysis.³⁰

(6) Double-Blind Lineups and the National Institute of Justice Standards for Lineups³¹

In 1998 then Attorney General Janet Reno directed the National Institute of Justice (NIJ) (the research, development and funding branch of the United States Department of Justice), then headed by Jeremy Travis, now president of John Jay College, to study and report on certain aspects of the criminal justice system, in the wake of startling revelations of wrongful convictions, which were often the product of mistaken identifications.³² NIJ formed a panel comprised of prosecutors, defense counsel, police officers, psychologists and social scientists to explore what the criminal justice system could do to minimize mistaken identifications and wrongful convictions. Neither Dr. Ebbesen nor Dr. Kovera were invited to participate. Some of the conclusions of the panel bear on the trial of this indictment because the lead detective in this case did not implement them. One was that a witness viewing a lineup should be instructed that the perpetrator may or may not be in that lineup. Another recommendation was that fillers should be chosen to match the description given by the witness, if adequate, rather than to match the suspect in the lineup.³³

Other relevant recommendations urged by the scientists did not make it into the final guidelines because of objections by police and prosecutors. One such recommendation was that lineups be conducted sequentially, that is, one filler is viewed at a time rather than together with the entire array. Another recommendation not adopted was the use of double-blind lineups, where neither the witness nor the administrator of the lineup knows who the suspect is. The scientists also recommended, in the absence of a double-blind lineup, that the witness’s confidence in the identification be recorded in her own words immediately after the identification, before any feedback from the administrator might have a chance to affect that confidence level.³⁴

Discussion

The law in New York on expert testimony relating to eyewitness identification, like the underlying social-psychological research field itself, has evolved significantly in recent years. In People v Mooney (76 NY2d 827 [1990]), the Court held that the trial court properly exercised its discretion in excluding expert testimony concerning factors that may influence a witness’s perception and memory and affect the reliability of identifica- ' tion testimony, but the Court did not decide whether such testimony could properly be admitted at trial “as a matter of law.” (Id. at 828.)³⁵ In People v Lee (96 NY2d 157 [2001]), the Court revisited the issue of the admissibility of such expert testimony and held it “is not inadmissible per se” and the decision whether to admit it rests within the sound discretion of the trial court (id. at 160), even when “to some degree, it invades the jury’s province.” (Id. at 162.)

As noted previously, on May 9, 2006, the Court of Appeals handed down two separate opinions involving expert eyewitness identification testimony which (1) further indicate the Court’s growing acceptance of such testimony and (2) reminded trial courts that the “sound discretion” to exclude expert testimony is by no means unfettered. In People v Drake (supra), an assault in the first degree case, identification of the assailant was hotly contested. The trial judge granted the defendant’s application to admit the testimony of Dr. Elizabeth Loftus, a psychologist, as an expert on the reliability of eyewitness identifications,³⁶ but, in a controversial portion of the court’s final charge, seemed to curtail severely the jury’s use of that testimony.³⁷

In People v Young (supra), another hotly contested identification case, the defendant was convicted at a second trial,³⁸ at which the People introduced for the first time corroboration testimony in the form of the victims’ stolen property. Defendant sought to introduce the testimony of Dr. John Brigham to testify about factors affecting the accuracy of eyewitness identifications.³⁹ The trial court ruled that the evidence would be excluded in the exercise of its discretion. The Court of Appeals held that the trial judge’s ruling was not an abuse of discretion in light of two factors: (1) the extent to which the research findings discussed by Dr. Brigham were relevant (or not) to the victim wife’s identification of the defendant and (2) the extent to which that identification was corroborated by other evidence. (Id. at 44-45.) Interestingly, however, the Court also stated that it would have been proper to admit the expert testimony as well, and cautioned: “But if this case turned entirely on an uncorroborated eyewitness identification, it might well have been an abuse of discretion to deny the jury the benefit of Brigham’s opinions.” (Id. at 45.)⁴⁰

In light of those very recent opinions, it seemed questionable whether a Frye hearing was even necessary in Williams, but the Appellate Division’s LeGrand affirmance (People v LeGrand, 28 AD3d 318 [1st Dept 2006], lv granted 7 NY3d 758 [2006]) caused me to hold one.

Defendant Williams’s In Limine Application/Frye

The parties presented this court with two related issues: First, the “traditional” Frye aspect of the hearing — whether the six social-psychological theories and the underlying research which defendant seeks to introduce through an expert witness are based on scientific principles and research generally accepted by the relevant scientific community. (See Parker v Mobil Oil Corp., 7 NY3d 434 [2006]; People v Wernick, 89 NY2d 111, 115-116 [1996]; People v Wesley, 83 NY2d 417, 423-429 [1994].) Second, if the first is answered in the affirmative, whether this court should exercise its discretion to permit this expert testimony at trial in order to assist the jury in reaching its verdict. (See People v Young, supra; People v Lee, supra.)41

Before addressing the core Frye issue, I will deal briefly with the traditional admissibility tests. Based on her impressive professional credentials and knowledge of eyewitness identification as a subspecialty of the social-psychological field of human memory, I was satisfied that Dr. Kovera had the scientific background and knowledge to testify as an expert in the field of eyewitness identification. (See Matott v Ward, 48 NY2d 455, 459 [1979].) Her competence was manifest and even conceded by the People.

Expert testimony, like all proffered testimony, must be relevant in order to be admissible. “It is well settled that evidence is relevant if it has any ‘tendency in reason to prove any material fact’ (People v Alvino, 71 NY2d 233, 241 [1987]).” (People v Mateo, 2 NY3d 383, 424 [2004],)⁴²1 find that the proffered expert testimony of Dr. Kovera is clearly relevant since it relates directly to the central fact of this case, the identity of the perpetrator of the laundromat robbery, and also to the facts and circumstances surrounding those identifications. Moreover, the prosecution concedes there is no corroborating evidence independent of identification testimony, nor have the People ever argued that they will be prejudiced by the admission of this expert evidence or that it will mislead the jury. The prosecution’s only argument has always been that the proffered expert testimony has not met general acceptance within the relevant scientific community, the core Frye issue.

Whether Dr. Kovera’s expert testimony is within the ken of ordinary jurors has been answered to a large extent by the Court of Appeals in People v Lee (at 162): “Despite the fact that jurors may be familiar from their own experience with factors relevant to the reliability of eyewitness observation and identification, it cannot be said that psychological studies regarding the accuracy of an identification are within the ken of the typical juror.” As phrased in People v Young (7 NY3d at 45, quoting People v Cronin, 60 NY2d 430, 433 [1983]), “[A] court’s exercise of discretion . . . [should] depend[ ] in large part on whether the ‘specialized knowledge’ of the expert can give jurors more perspective than they get from ‘their day-to-day experience, their common observation and their knowledge’.”

In Williams, I conclude that the proffered expert testimony on double-blind lineups, mug shot exposure, and confidence malleability depends upon underlying scientific principles and research beyond the ken of the average juror and can provide jurors a better perspective on these effects or phenomena than they normally enjoy from their ordinary knowledge and experience. I believe this is also true, although to a lesser extent, for weapon focus, cross-racial identification and the more commonsensical “exposure duration” phenomena. I am satisfied that, although the average juror may have some awareness of these effects, there is sufficient underlying scientific research involved to permit Dr. Kovera to tell the jurors something significant they may not already know about these factors. Indeed, a balanced presentation of “exposure duration” (as well as other effects) could easily support arguments favoring the prosecution, depending on the relative strength or weakness of the eyewitness testimony itself, as it is presented at trial.

Frye

The court now reaches the core Frye issue: whether the expert testimony being offered is based on scientific principles, research and procedures which have gained general acceptance of reliability within the relevant scientific community. (Frye v United States, supra; see also People v Wesley, 83 NY2d at 422.)⁴³

The burden of proving general acceptance in the relevant scientific community rests upon the proponent of the disputed testimony. (See Zito v Zabarsky, 28 AD3d 42 [2d Dept 2006]; People v Kanani, 272 AD2d 186 [1st Dept 2000], lv denied 95 NY2d 935 [2000].)⁴⁴ The Frye decision itself recognized that it is not always easy to define precisely when a scientific principle or discovery crosses the line from the experimental to the demonstrable stage, referring to this grey area as “this twilight zone.” (Frye v United States, 293 F at 1014.) It is important to appreciate, however, that it is not the responsibility of a trial judge under Frye to determine whether or not a novel scientific theory, principle or procedure is in fact reliable but only whether it is generally accepted as such by the relevant scientific community. We are to count the number and percentage of scientists favoring the principle, not measure its soundness. (Parker v Mobil Oil Corp., supra.)

There remains the critical task of identifying the relevant scientific community.⁴⁵ The key to resolving this “general acceptance” issue is the latest Kassin survey. The criticism is misplaced that it

“represented only a ‘very small sample of the thousands of researchers who do research on human memory/ thereby excluding a ‘highly published minority’ of researchers in the eyewitness identification discipline from participating in the survey (e.g., Cynthia Reinick, Vladimir Konecni, John Wixted, Rogers Elliott, Michael McCloskey and JohnYuille).” (People v LeGrand, 196 Misc 2d at 191.)⁴⁶

I agree with Dr. Kovera that to include the entire community of psychologists who conduct research in the vast area of memory is far too broad and impractical. Most researchers within this broad spectrum do not stay current with the more specialized applied research in the much narrower field of eyewitness identification.

Similarly, devaluing the 2001 Kassin survey because of the seemingly low response rate (of the 197 prospective respondents, 64 answered the survey, yielding a 34.4% response rate) is unpersuasive. Dr. Kovera and the Kassin survey itself addressed this issue, making it clear that the return rate should still be regarded as substantial. Many prospective respondents would have been expected to have opted out from participating based on their own assessments that they did not qualify as experts in the specialized field covered by the survey. According to the Kassin survey, those who did respond were a “highly prolific subgroup” and “could be described as a blue ribbon group of leading researchers.” (See n 16, supra.) Nor does the disparity between the percentages of experts endorsing the research and those willing to testify about it raise any red flags. Dr. Kovera’s knowledgeable explanation, that some experts are unwilling to testify in a courtroom no matter what the issue is or what the case involves, is a satisfactory answer.

In addition, any concern that the Kassin survey was designed expressly for litigation is easily answered by the 2001 authors themselves: “ ‘[B]y providing empirical evidence of the consensus within the community of experts, this survey has proved useful to judges ruling on the admissibility of expert witnesses.’ ” (People v LeGrand, 196 Misc 2d at 192.) The original 1989 Kassin survey makes it clear that its intent was to address the Frye test, just as the updated 2001 survey acknowledged that it was addressing the newer Daubert standard. The purpose of the surveys, however, was not advocacy but education: that is, to fill a vacuum with empirical guidance to judges, defense counsel, prosecutors and the experts themselves when addressing certain eyewitness phenomena in the courtroom.⁴⁷

I conclude that defendant has established that Dr. Kovera’s proposed expert testimony has been generally accepted by the relevant scientific community, based on her own testimony as a respected, active member of that community, and more significantly as evidenced by the Kassin surveys, notwithstanding that the acceptance is not unanimous. (See People v Middleton, 54 NY2d 42, 49 [1981].)

Conclusion

Dr. Margaret Bull Kovera, who holds a Ph.D. in social psychology, is an expert eminently qualified in the subspecialty of eyewitness identification. The testimony she presented on behalf of the defense at the Frye hearing on the six categories explained in this opinion is relevant to the facts and circumstances of the two pertinent identifications at issue in this case. These identifications are uncorroborated by any additional prosecution evidence. Dr. Kovera’s opinions are grounded in scientific research generally accepted by the relevant scientific community identified in this opinion, and they are likely to benefit jurors by providing them with a better perspective and significant information beyond their common knowledge and experience. Under these circumstances, and, based upon this record, it is my opinion that this testimony fits squarely within recent Court of Appeals precedent cited herein. For these reasons I have exercised my discretion to admit such testimony at the trial of this indictment.⁴⁸

1 . Two days after I concluded the hearing and rendered my ruling, the Court of Appeals granted leave in LeGrand (7 NY3d 758 [2006, Kaye, Ch. J.]), setting the stage for the Court to address once again expert testimony on eyewitness identification and the Frye test.

2 . That decision and order reads in part:

“Without testifying about the ultimate issues before the jury, Dr. Kovera may explain to the jury the problems associated with the following six factors, which relate to the identifications in this case in varying degrees: (1) cross-racial identification, (2) weapon focus, (3) exposure duration, (4) confidence malleability, (5) mug shot exposure, and (6) lack of double-blind lineups.”

3 . Hereafter this person will be referred to as perpetrator No. 1.

4 . There was conflicting testimony at the jury trial of this case subsequently conducted, commencing on July 24, 2006, concerning whether Camacho viewed photographs and whether she was shown the photo picked out by her son. A mistrial was declared on August 3, 2006 on the single count of robbery in the first degree submitted to the jury, based on the jury’s failure to reach a unanimous verdict after five deadlock notes over the course of three days.

5 . I am aware of the challenges courts face in conducting an adversarial Frye hearing to explore scientific research, even for the limited purpose of determining the admissibility of this proffered expert testimony. A principal constraint is the record made reflecting counsels’ arguments and choices regarding witnesses. For example, in this case the prosecutor chose not to call a competing expert.

6 . That society is a division of the highly regarded American Psychological Association. (See <http://www.apa.org>.)

7 . Dr. Kovera described herself as best known and recognized in the field of eyewitness identifications for her research on double-blind lineups. Her 19-page curriculum vitae listing her numerous publications was admitted into evidence.

8 . In fact, Dr. Kovera stated that it is very difficult to get “regulatory” research published unless meta-analysis is a component of that research.

9 . See Saul Kassin, Phoebe Ellsworth and Vicki Smith, The “General Acceptance” of Psychological Research on Eyewitness Testimony: A Survey of the Experts, 44 Am Psychologist (No. 8) 1089 (1989) (1989 Kassin survey).

10 . See Saul Kassin, Anne Tubb, Harmon Hosch and Amina Memon, On the “General Acceptance” of Eyewitness Testimony Research: A New Survey of the Experts, 56 Am Psychologist (No. 5) 405 (2001) (2001 Kassin survey).

11 . In light of subsequent research undertaken in the 1990’s, 13 new phenomena were added to those in the 1989 survey, including “confidence malleability” and “mugshot exposure.” (Id. at 407-408.)

12 . But see generally Robert L. Ferrer, Graphical Methods for Detecting Bias in Meta-analysis, 30 Fam Med (No. 8) 579-583 (1998) (containing citations to published criticisms in footnotes).

13 . Dr. Kovera also noted that Dr. Ebbesen’s curriculum vitae consisted of an inordinate number of references to his work with consulting attorneys in addition to his scholarly publications.

14 . The survey addressed in LeGrand is the updated 2001 Kassin study.

15 . In 1989, the overall response rate was 53%. (See 1989 Kassin survey, supra at 1090.) For the 2001 survey, questionnaires were mailed to 197 prospective respondents, six were returned because the recipients either indicated they had no time to participate or no longer considered themselves to be experts, five were undeliverable. Of the remaining 186 prospective participants, 64 returned data in usable form, yielding a 34% response rate. (See 2001 Kassin survey, supra at 407.)

16 . Although the 2001 survey respondents were not identified, the survey reveals that 62 of the 64 held a Ph.D. in psychology (four had also earned a J.D.); one held only a D.S. and another an M.A. (Id.) Based on the number of eyewitness publications self-reported by respondents compared with the total publications of all prospective participants, as derived from PsycINFO, the Kassin survey concluded that respondents constituted “a highly prolific subgroup of the total population” and “could be described as a blue ribbon group of leading researchers.” (Id.) In addition, 92% of the respondents had published one or more books, chapters, or articles on the psychology of eyewitness identification. (Id. at 409.)

17 . Dr. Kovera assumed that Dr. Ebbesen’s exclusion from the second Kassin survey was based on his failure to meet the criterion regarding publication in this field within the preceding 10 years. (But see, e.g., Ebbe B. Ebbesen and Cynthia B. Rienick, Retention Interval and Eyewitness Memory for Events and Personal Identifying Attributes, 83 J Applied Psychol [No. 5] 745-762 [1998]; Ebbe B. Ebbesen and Vladimir J. Konecni, Eyewitness Memory Research: Probative v. Prejudicial Value, 5 Intl Dig of Hum Behav, Sci & L 2-28 [1996].)

18 . See Christian A. Meissner and John C. Brigham, Thirty Years of Investigating the Own-Race Bias in Memory for Faces: A Meta-analytic Review, 7 Psychol, Pub Pol’y & L 3 (2001).

19 . As with her other opinions, Dr. Kovera was careful not to quantify or estimate the percentages of misidentifications which may result from this cross-racial effect. She recognized that there is too much variation in individual cases for such percentages to be reliable and useful for a jury.

20 . Stephanie J. Platz and Harmon M. Hosch, Crossracial/Ethnic Eyewitness Identification-. A Field Study, 18 J Applied Soc Psychol 972 (1988). One finding in that study was that Hispanic clerks were more likely to recognize Hispanic customers (53.5%) than Caucasian (35.7%) or African-American (25%) customers. (Id. at 978.) Dr. Kovera never intended to make any reference to those percentages before the jury.

21 . The name of this phenomenon in the 2001 survey, “cross-race bias” (see 2001 Kassin survey, supra at 412), can be misleading. Racial prejudice only tangentially affects cross-racial identification in that racial bias may impact the amount and quality of contact a person of one race has with persons of another race. Even if there is increased contact with a person of another race, if little or no “attention” is paid to that person, then this will not moderate the cross-racial identification effect.

22 . Id.

23 . See Nancy M. Steblay, A Meta-analytic Review of the Weapon Focus Effect, 16 L & Hum Behav 413 (1992).

24 . One criticism of meta-analysis is “publication bias,” which is a tendency for journals to publish studies regarding statistically significant findings and not those studies with statistically nonsignificant results, which some critics argue threaten the external validity of meta-analysis. (See S. Iyengar and J. Greenhouse, Selection Models and the File Drawer Problem, 3 Statistical Sci [No. 1] 133-135 [1988].) Dr. Robert Rosenthal in 1991 styled this as the “file drawer” problem, which represents those unpublished nonsignificant result reports residing in the file drawers of researchers. Rosenthal developed a statistical procedure to calculate the “fail-safe N,” that is, the number of unreported studies that would have to be sitting in file drawers in order to negate the significant effect of a meta-analysis study. (See Robert Rosenthal, The “File Drawer Problem” and Tolerance for Null Results, 86 Psychol Bull 638-641 [1979].)

25 . While this point seems logical, it is hard to dismiss the competing commonsense notion that real life victims of weapon crimes may have an incentive and an accompanying heightened degree of focus to see and remember perpetrators unmatched by college students or other volunteers in a laboratory setting. This point of view, however, can easily be brought out at trial and argued to a jury, and in my opinion is not a sufficient reason to discard the research discussed by Dr. Kovera, or to preclude this part of her expert testimony at trial. In this connection, Judge G.B. Smith’s observation in People v Young (at 49 [dissenting op]) is pertinent:

“[T]he trial judge seemed inappropriately concerned by Dr. Brigham’s studies on stress and reliability, which did not include persons who were personally robbed, but people who witnessed staged robberies (efi People v Aphaylath, 68 NY2d 945 [1986] [holding that an expert need not have knowledge of a defendant’s particular characteristics for testimony to be admitted]).”

26 . This appears to be a slightly understated figure. The table 4 reference actually lists the figure of 77%. (See 2001 Kassin survey, supra.)

27 . A person who has seen a perpetrator for an extended period of time may have inflated confidence in his ability to make an accurate identification, which studies show is disproportionate in some cases. This can cause the witness unconsciously to exaggerate the clarity of his original observations and the certainty with which he testifies in court. This confidence/accuracy effect was not raised at trial.

28 . Dr. Kovera was careful to point out that confidence malleability did not affect the identification itself, but only the confidence in that identification, whether or not the actual perpetrator was identified.

29 . In response to the court’s question why so many responders were unwilling to testify, most of whom supported the research and its conclusions, Dr. Kovera offered the explanation that some experts simply prefer never to testify as experts in courts of law.

30 . At the time of Dr. Kovera’s testimony this new meta-analysis in the field of mug shot exposure was only available to on-line subscribers (but see Kenneth A. Deffenbacher, Brian H. Bornstein and Steven D. Penrod, Mugshot Exposure Effects: Retroactive Interference, Mugshot Commitment, Source Confusion, and Unconscious Transference, 30 Law and Hum Behav 287-307 [2006]).

31 . This factor has not been the subject of the Kassin surveys, but arguably gains acceptance from the diverse panel of experts discussed in this section of the opinion.

32 . These wrongful convictions were exposed in large measure by a legal clinic and resource center at Benjamin N. Cardozo School of Law in New York City called the Innocence Project, using postconviction DNA to exonerate the innocent. (See <http://www.innocenceproject.org>.)

33 . Neither side made an issue of this “description” point at the trial.

34 . It is noteworthy in this regard that the Supreme Court of New Jersey very recently announced, in an opinion affirming a criminal conviction in State v Delgado (188 NJ 48, 902 A2d 888 [2006]), a new rule “to require that, as a condition to the admissibility of an out-of-court identification, law enforcement officers make a written record detailing the out-of-court identification procedure, including the place where the procedure was conducted, the dialogue between the witness and the interlocutor, and the results.” (188 NJ at 63, 902 A2d at 897 [emphasis added].) The court directed the Criminal Practice Committee to prepare a rule to this effect for the court’s consideration and adoption.

35 . In a strongly worded and frequently cited dissenting opinion, then Associate Judge Kaye criticized the majority for its “cursory treatment” (76 NY2d at 828) of the defendant’s claim. The late Judge Fritz Alexander concurred in the dissent.

36 . Dr. Loftus testified, inter alia, about weapon focus, cross-racial identification, accuracy-confidence and that memory fades over time and can become susceptible to postevent information or become distorted by new information from outside sources, such as media reports. (7 NY3d at 35 [dissenting op].)

37 . The trial court instructed the jury that the testimony of Dr. Loftus “may not be used to discredit or accredit the reliability of eyewitness testimony in general, or in this case.” (7 NY3d at 32.) The Court of Appeals, while encouraging courts to admit such expert testimony “in appropriate cases,” also opined that this portion of the charge was “ill advised” and improper. However, the Court affirmed the.conviction on the ground that the charge, when read as a whole, conveyed to the jury the correct standard for the proper use of the expert’s testimony. (Id.) Judge G.B. Smith, the lone dissenter, wrote that the charge “effectively negated” the expert’s testimony. (Id. at 37-38.)

38 . Defendant’s first trial resulted in a conviction which was reversed because certain evidence, including the victim wife’s lineup identification, was held to have been fruits of an illegal arrest (see People v Young, 255 AD2d 905 [4th Dept 1998]). Prior to the retrial, eight years after the crime, the trial court found the wife had an independent source to make an in-court identification.

39 . Dr. Brigham was a former president of the American Psychology-Law Society, of which Dr. Kovera is now the president-elect. (See People v Young, 7 NY3d at 48 [dissenting op].) His proposed testimony included many factors in common with Dr. Kovera’s in Williams.

40 . Judge G.B. Smith wrote in his dissent in Young: “The majority’s ruling misses the opportunity to hold that here, as a matter of law, where eyewitness identification is attenuated and possibly tainted, and corroborating evidence is weak, courts should allow expert testimony concerning eyewitness identification.” (7 NY3d at 50.)

41 . Some trial courts, including the court in People v LeGrand (196 Misc 2d at 186) have described a “four prong” Frye test, in effect combining the classic Frye issue regarding general acceptance in the relevant scientific community with the other traditional tests for admissibility of expert testimony. (See, e.g., People v Smith, 2 Misc 3d 1007[A], 2004 NY Slip Op 50172[U] [Sup Ct, NY County 2004].) However, I will treat these related but distinct admissibility issues separately.

42 . A trial court may, however, exercise its discretion and preclude “technically relevant” evidence “if its probative value is substantially outweighed by the danger that it will unfairly prejudice the other side or mislead the jury.” (People v Scarola, 71 NY2d 769, 777 [1988] [emphasis added].)

43 . The Wesley court stated that the Frye standard and not the judicial gatekeeping standard set forth in Daubert v Merrell Dow Pharmaceuticals, Inc. (509 US 579 [1993] [based on Fed Rules Evid rule 702]) is the test to be applied in New York State courts. (See People v Wesley, 83 NY2d at 423 n 2.) Frye was most recently confirmed as the New York test in Parker v Mobil Oil Corp. (supra).

44 . General acceptance of reliability may not be demonstrated through discordant judicial opinions alone. (See People v Jeter, 80 NY2d 818 [1992].) Several diverse trial court decisions have been published in New York on the admissibility of expert eyewitness identification testimony, most notably, People v LeGrand (supra). Other trial court decisions denying admission are: People v Carrieri (4 Misc 3d 307 [Sup Ct, Queens County 2004]); People v Smith (2 Misc 3d 1007[A], 2004 NY Slip Op 50172[U] [Sup Ct, NY County 2004]); People v Santiago (2 Misc 3d 652 [Sup Ct, NY County 2003]). Examples of decisions admitting the expert evidence are: People v Radcliffe (196 Misc 2d 381 [Sup Ct, Bronx County 2003]); People v Smith (191 Misc 2d 765 [Sup Ct, NY County 2002]).

45 . In People v Wesley (supra), Chief Judge Kaye in a concurring opinion (joined by Judge Ciparick) stated:

“In defining the relevant scientific field, the court must seek to comply with the Frye objective of containing a consensus of the scientific community. If the field is too narrowly defined, the judgment of the scientific community will devolve into the opinion of a few experts. The field must still include scientists who would be expected to be familiar with the particular use of the evidence at issue, however, whether through actual or theoretical research (Gianelli, The Admissibility of Novel Scientific Evidence: Frye v United States, a Half-Century Later, 80 Colum L Rev 1197, 1209-1210).” (83 NY2d at 438.)

It should also be noted that while the Chief Judge counseled “counting scientists” for Frye purposes, the same concurring opinion stated, “Where controversy rages, a court may conclude that no consensus has been reached.” (Id. at 439.)

46 . It is LeGrand’s definition of the relevant scientific community, along with its criticism of the methodology used by some of the experts, with which I disagree. (See People v LeGrand, 196 Misc 2d at 189-192.) Supreme Court’s discussion of the effect of various eyewitness phenomena appears to seek a consensus well beyond that which is required by Frye. For example, the fact that there is serious disagreement among experts about how and why memory works in certain ways (id. at 207) does not nullify the results of the extensive research described at Williams’s Frye hearing, if conducted according to proper scientific methodology. Moreover, criticism in LeGrand of the methodologies used, such as meta-analysis, was adequately answered by Dr. Kovera, as discussed in the “Frye Hearing” portion of this opinion.

47 . Other criticism of the research endorsed by Dr. Kovera, such as disparagement of meta-analysis, have been adequately answered by Dr. Kovera, as explained in the “Frye Hearing” portion of this opinion.

48 . It is ironic and somewhat unsettling that, after such extensive analysis and discussion in academia and courts of law about this scholarly research and the relevant scientific community, it remains very difficult to measure how much, if any, real guidance this type of testimony actually provides to juries. For example, during frequent and lengthy read-backs and re-charges requested by the deliberating jury in the Williams trial, none of those requests or questions involved the expert’s testimony. Counsel’s discussion with the jurors after the declaration of the mistrial seemed to confirm that they were not particularly enlightened or persuaded by the eyewitness identification expert. (There was some suggestion that the expert may have addressed too many factors and used too much professional jargon.) This experience does not cause me to abandon the use of such testimony in appropriate cases, but it does caution me to allow it only with careful circumspection. It should also cause the experts to continue to explore and refine this field of study, particularly as it relates to courts and juries.