United States v. Nguyen

793 F. Supp. 497 | D.N.J. | 1992

793 F. Supp. 497 (1992)

UNITED STATES of America Prosecution,
v.
Tom NGUYEN, a/k/a "Johnny" Defendant.

Cr. A. No. 91-526(AJL).

United States District Court, D. New Jersey.

May 22, 1992.

Jane Meyers, Mark Rufolo, Asst. U.S. Attys., U.S. Atty.'s Office, Newark, N.J., for Government.

James C. Patton, Livingston, N.J., for defendant.

OPINION

LECHNER, District Judge

In this case defendant Tom Nguyen, a/k/a "Johnny" ("Nguyen"), was indicted on three counts (the "Indictment").^[1] Count One alleged that during the time period beginning January 1991 and ending July 1991 Nguyen conspired to engage in the business of dealing in firearms without a federal license in violation of 18 U.S.C. § 922(a)(1)(A) and (2). Count Two alleged Nguyen knowingly and willfully engaged in the business of dealing in firearms without a federal license and sold a weapon in violation of 18 U.S.C. § 922(a)(1)(A) and (2). Count Three charged Nguyen with possession of a weapon while a convicted felon in violation of 18 U.S.C. § 922(g)(1). On 19 May 1992, after a three-day trial, Nguyen was found not guilty on all three counts of the Indictment.

Prior to trial, Nguyen moved for a bifurcated trial on the count charging him with possession of a weapon while a convicted felon, for an order precluding in-court and out-of-court identification testimony of Nguyen and for the introduction of expert testimony regarding eyewitness identification.^[2]

At a pretrial hearing, for the reasons set forth below, the motion to bifurcate the trial on the issue that Nguyen possessed a weapon while a convicted felon was denied and the motion to exclude in-court and out-of-court identification testimony was denied. Prior to trial, Nguyen indicated that he was uncertain whether he would call the expert,^[3] therefore, no ruling on the motion to introduce expert testimony concerning eyewitness identification was made prior to trial. At trial, Nguyen did not call the expert to testify. If Nguyen had called the expert during trial, the motion would have been denied for the reasons set forth below.

Factual Allegations

The Indictment charged Nguyen neither applied for nor obtained a license to deal in firearms, as required by federal law. Indictment at 2. It alleged that on or about 2 February 1991 Nguyen placed a telephone call to an individual to discuss the sale of firearms and ammunition. Id. at 3. The Indictment charged, on that same date, Nguyen willfully and knowingly sold a .25 caliber semi-automatic pistol, Raven Arms model MP-25, bearing serial number 1733273 (the "Weapon"), to an individual in Jersey City, New Jersey. Id. at 3, 5. The individual was Investigator Mark Peterkin ("Peterkin") of the Hudson County Prosecutor's Office, who was working undercover. The sale of the Weapon occurred around dusk in Lincoln Park (the "Park") in Jersey City, New Jersey. 2 April 1992 Letter Brief at 3; 8 May Hearing Tr. at 5. The Indictment charged that at the time of such sale, Nguyen had previously been convicted of forcible theft with a firearm which is a felony offense.^[4] Indictment at 10.

On 2 February 1991, an hour after Peterkin arrived at the Park, a white Camaro, in which four individuals were riding, drove up next to the Park.^[5] 8 May Hearing Tr. at 19. Peterkin stated a man got out of the car and swaggered over to him. Id. at 6. He described the man as a five foot, three inches to five foot, five inches, Asian, wearing a long black, full-length coat, a scarf, black patent leather shoes, white socks and a white shirt, completely buttoned to the collar. Id.

Peterkin stated he had a three to five minute conversation during which the man sold Peterkin the Weapon and discussed future sales. Id. Peterkin testified that during this time, he stood face to face with this person, only an arm's length from him. Id. at 7. Peterkin said that although it was dark, he could clearly see this man because of the artificial lights from the street lamps overhead. Id. He testified the man pulled out the Weapon, put a clip in it, but did not put a round in the chamber, and handed it to Peterkin, initially with the barrel pointing at him but then by holding the barrel and flipping the butt of the Weapon over to him. Id. at 7, 24-25.

Following the sale of the Weapon, Peterkin returned to the Hudson County Prosecutor's Office where he looked through a book of approximately sixty photographs of Asian men (street gang members)^[6] between the ages of twenty and twenty-five.^[7]Id. at 8. Peterkin identified a photograph of Nguyen as the man in the Park who sold him the Weapon. Id.

On 15 April 1991 Nguyen was arrested in New York City and charged with first degree robbery. Moving Brief at 1. Subsequently, Nguyen was incarcerated at Rikers Island New York State Prison ("Rikers Island") on the robbery charge. Id. In August 1991, during Peterkin's investigation, he discovered that Nguyen, the man he identified on 2 February 1991, may have then been at Rikers Island. 8 May Hearing Tr. at 11. To verify this information, Peterkin again identified a photograph of Nguyen as the man in the Park who sold him the Weapon; Peterkin received the photograph by facsimile transmission. Id. at 12.

Because the facsimile was dark, however, Peterkin went to Rikers Island for a lineup. Id. Peterkin was accompanied by Officer Wisniewski.^[8] 13 May Hearing Tr. at 185. After checking in, Peterkin and Officer Wisniewski went to an office. After about ten to fifteen minutes they left the office and went into the corridor. Id. at 186. Peterkin immediately recognized Nguyen, who was walking in a corridor about twenty to twenty-five feet from Peterkin, as the man in the Park. Id. at 186-89; 8 May Hearing Tr. at 12. Nguyen was being escorted by two prison guards at the time Peterkin identified him; however, there were other prisoners and guards in the corridor.^[9] 13 May Hearing Tr. at 186-88.

On 3 December 1991 Nguyen was arraigned on the Indictment and entered a plea of not guilty. Opp. Brief at 1. On that date, Nguyen was transferred from Rikers Island to Union County Jail to await trial in this action. Id.

Discussion

A. Bifurcation of Count Three

Prior to trial, Nguyen moved for a bifurcated trial with respect to elements of the crime charged in Count Three of the Indictment. 25 March 1992 Letter Brief at 2. Specifically, Nguyen requested "that the jury be asked initially to determine whether ... Nguyen possessed a firearm. If the jury finds that he did, ... Nguyen [would] stipulate that prior to February 2, 1991, the date in question, he had been convicted of a felony. The jury would then be asked to retire to determine whether ... Nguyen was guilty of possessing a firearm while a felon." Id. Nguyen argued bifurcation was necessary because the prior felony conviction was unfairly prejudicial; he argued "a juror [would] use the evidence of ... Nguyen's prior conviction to conclude that ... Nguyen [was] the man in the [P]ark who sold [the Weapon] to [Peterkin]...."^[10]Id. at 2.

The Government argued the felony alleged in the Indictment is admissible as an element of Count Three which asserts a charge for possessing a firearm while being a convicted felon. Opp.Brief at 7.

Count Three alleged a violation of 18 U.S.C. § 922(g)(1). Section 922(g)(1) makes it unlawful for a person who has been convicted of a crime punishable by a term exceeding one year to possess a firearm. 18 U.S.C. § 922(g)(1); see also United States v. Paolello, 951 F.2d 537, 541 n. 3 (3d Cir.1991); United States v. Schoolcraft, 879 F.2d 64, 69 (3d Cir.), cert. denied, 493 U.S. 995, 110 S. Ct. 546, 107 L. Ed. 2d 543 (1989). "In order to convict someone under this statute, the government must prove: (1) that the defendant had a previous felony conviction, (2) that the defendant possessed a firearm, and (3) that the firearm had travelled in or affected interstate commerce." United States v. Petitjean, 883 F.2d 1341, 1346 (7th Cir.1989).

The Government is entitled to offer proof on an element of the crime. See, e.g., United States v. Williams, 612 F.2d 735, 740 (3d Cir.1979), cert. denied, 445 U.S. 934, 100 S. Ct. 1328, 63 L. Ed. 2d 770 (1980) (government does not have to accept stipulation of prior felony conviction which is element of charged crime). The argument that the prior felony conviction would unfairly prejudice Nguyen because it would show a propensity for crime or a bad character was rejected. The nature of the prior felony conviction, armed robbery, was not admitted into evidence, but only the fact that Nguyen had a prior felony conviction was introduced into evidence. Moreover, strong limiting instructions were given during trial, at the time the evidence was introduced and again when the jury was charged, with respect to the introduction of a prior criminal conviction. Accordingly, the motion for bifurcation was denied.

B. Identification Testimony

Nguyen moved to suppress the in-court and out-of-court identifications on the ground that they were the product of impermissively suggestive procedures which violated his right to due process under the Fifth and Fourteenth Amendments. Moving Brief at 11. Nguyen argued the out-of-court photographic identifications by Peterkin were impermissively suggestive because he made the identifications from a book of mug shots of Asian men. Nguyen argued because the photobook was specifically compiled for an ongoing investigation of Asian street gangs and Peterkin was aware the photographs in the book included members of the particular Asian street gang to which Peterkin believed the man in the Park belonged, Peterkin believed the man in the Park would be in the book. 14 May Hearing Tr.

Nguyen also argued testimony of the identification at Rikers Island should be excluded because it was the result of a suggestive procedure. Specifically, he argued because Nguyen was escorted by two prison guards there was a suggestion to Peterkin that Nguyen was the man in the Park. Id. Nguyen argued the in-court identification cannot be permitted because it was the product of impermissively suggestive photographic identifications and a "show up" type identification rather than the result of independent knowledge.

The Government argued that prior to prohibiting out-of-court and in-court identification testimony, the undercover agent's opportunity to view the suspect, the degree of attention paid to the suspect, the accuracy of any prior description, the level of certainty of the identification and the length of time the undercover agent had to view the suspect must be considered. Opp. Brief at 18-19.

The Due Process Clause prohibits the use of identification procedures which are "unnecessarily suggestive and conducive to irreparable mistaken identification." Stovall v. Denno, 388 U.S. 293, 302, 87 S. Ct. 1967, 1972, 18 L. Ed. 2d 1199 (1967); see also Neil v. Biggers, 409 U.S. 188, 196, 93 S. Ct. 375, 380-81, 34 L. Ed. 2d 401 (1972); Simmons v. United States, 390 U.S. 377, 384, 88 S. Ct. 967, 971, 19 L. Ed. 2d 1247 (1968). Where identification procedures have been unnecessarily suggestive and conducive to irreparable mistaken identification, the identification testimony cannot be admitted in trial. Manson v. Brathwaite, 432 U.S. 98, 117, 97 S. Ct. 2243, 2254, 53 L. Ed. 2d 140 (1977).

"[R]eliability is the linchpin in determining the admissibility of identification testimony...." Id. at 114, 97 S.Ct. at 2253. Accordingly, "[t]he admission of testimony concerning a suggestive and unnecessary identification procedure does not violate due process so long as the identification possesses sufficient aspects of reliability." Id. at 106, 97 S.Ct. at 2249. The asserted suggestiveness of identification procedures must be considered in the totality of circumstances, considering the following five factors:

[T]he opportunity of the witness to view the criminal at the time of the crime, the witness' degree of attention, the accuracy of the witness' prior description of the criminal, the level of certainty demonstrated by the witness at the confrontation, and the length of time between the crime charged and the confrontation.

Biggers, 409 U.S. at 199-200, 93 S.Ct. at 382; see also Reese v. Fulcomer, 946 F.2d 247, 258 (3d Cir.1991), cert. denied, ___ U.S. ___, 112 S. Ct. 1679, 118 L. Ed. 2d 396 (1992).

On 8 May 1992, an identification hearing was held to determine the issue of suggestiveness. United States v. Stevens, 935 F.2d 1380, 1386 n. 3 (3d Cir.1991) (citing United States v. Wade, 388 U.S. 218, 87 S. Ct. 1926, 18 L. Ed. 2d 1149 (1967)) (A pretrial hearing will be held where the defendant raises the issue that an identification procedure has possibly violated a constitutional right.).

The identification process in this case possessed sufficient aspects of reliability. At the time of the crime, Peterkin viewed the man face to face, at arm's length for three to five minutes. Peterkin was able to describe the clothing the man was wearing. He was working undercover in a major investigation. Peterkin arranged the meeting in the Park for the sole purpose of purchasing the Weapon and knew he would have to make an identification of the seller of the Weapon after the transaction was completed. Peterkin focused on the man's characteristics and dress. Peterkin directed his attention at only the man who sold him the Weapon; he could not see the individuals in the Camaro; he was not distracted.

The same night he purchased the Weapon, Peterkin made an identification of the seller of the Weapon. He looked through a book of approximately sixty photographs of Asian men between the ages of twenty to twenty-five and selected a photograph of Nguyen.

After considering the totality of the circumstances, Peterkin's initial identification of Nguyen in the photobook is reliable; the motion to exclude the initial out-of-court identification testimony was denied.

The subsequent identification at Rikers Island is also reliable. Because the initial identification was not made under suggestive circumstances, the identification at Rikers Island was not tainted by the initial identification. Although Peterkin was specifically going to Rikers Island to identify the man in the Park and also knew the man in the photograph he selected from the photobook was held at Rikers Island, he intended to view a lineup. Prior to the lineup, however, Peterkin, apparently by accident, recognized Nguyen in the hallway as the man in the Park.

Peterkin's identification in the hallway was spontaneous and unexpected. Although spontaneous one-on-one identifications are not as favorable as line-up identifications, they are generally reliable because of the unexpected nature of the identification. Stevens, 935 F.2d at 1390-91 (spontaneous wanted board identification admissible). The fact that Nguyen was being escorted in the corridor by two prison guards is of no moment. Peterkin testified he thought he was going to observe a lineup; it appears he was never told that Nguyen would be brought to him and he was not expecting to see Nguyen in the corridor. The motion to exclude the identification at Rikers Island was denied.

Because the out-of-court identifications were not the products of impermissively suggestive procedures, the in-court identification testimony was admissible. See Stevens, 935 F.2d at 1392-93 n. 16. The motion to exclude the in-court identification was denied.

C. Expert Testimony on Identification

Nguyen moved to introduce expert testimony of Steven Penrod ("Penrod")^[11] pursuant to Rule 702 of the Federal Rules of Evidence on the issue of the reliability of eyewitness identifications.^[12] Specifically, Nguyen seeks to have Penrod testify in six areas as follows: (1) that there is a lack of correlation between the confidence a witness has in the identification and the accuracy of that identification; (2) that cross-racial identifications, such as the one in this case, are less accurate than same-race identifications; (3) that observations made during stressful situations tend to be less accurate than observations made under less stressful situations; (4) that when a weapon is visible during the perpetration of a crime, the observer focuses on the weapon rather than the face of the perpetrator, therefore, the identification will be less accurate; (5) that a witness who is asked to make a subsequent identification will tend to relate back to the initial identification rather than to the crime itself; and (6) that a witness who is a police officer is subject to the same limitations as any other witness. 2 April 1992 Letter Brief at 1-2; 8 May Hearing Tr. at 44.

Of the six areas, Penrod has conducted studies in only one area, the correlation between confidence a witness has in an identification and accuracy of that identification. Id. at 86. With respect to the other five areas, Penrod claimed to have expertise from reviewing and reading studies conducted by other researchers. Id. Nguyen did not propose to have Penrod testify as to Peterkin's ability to accurately identify the man in the Park, but rather he proposed to have Penrod introduce results of studies on the accuracy of eyewitness identifications.

Penrod testified summarily that the effect of expert testimony on eyewitness identification makes jurors "act upon knowledge they claim to possess, ... acquire and act upon knowledge they didn't have prior to the expert testimony and ... give less weight ... to confidence expressed by a witness and begin to give consideration to these other factors."^[13] 13 May Hearing Tr. at 295.

Penrod stated expert testimony concerning eyewitness identification is helpful to jurors because it "sensitizes jurors to evaluate evidence in a more appropriate way." 8 May Hearing Tr. at 69. He stated: "In the absence of expert testimony, a number of studies indicate that jurors essentially go on the basis of the confidence expressed by the witness ... and essentially ignore ... the underlying circumstances that give rise to the identification." Id.

Rule 702 of the Federal Rules of Evidence "authorizes the admission of expert testimony so long as it is rendered by a qualified expert and is helpful to the trier of fact."^[14]DeLuca v. Merrell Dow Pharmaceuticals, Inc., 911 F.2d 941, 954 (3rd Cir.1990); accord Stevens, 935 F.2d at 1397. In United States v. Downing, 753 F.2d 1224 (3d Cir.1985), the Circuit held that an expert may be allowed to testify as to the reliability of eyewitness identifications. Id. at 1226.

The Circuit held that prior to the admission of such expert testimony, the trial court must conduct an in limine hearing to determine the helpfulness of the expert testimony. Id. Downing set forth three factors which must be considered to determine relevancy under Rule 702. The court must balance two of the factors:

(1) the reliability of the scientific principles upon which the expert testimony rests, hence the potential of the testimony to aid the jury in reaching an accurate resolution of a disputed issue; and (2) the likelihood that introduction of the testimony may in some way overwhelm or mislead the jury.

Id.; see also Stevens, 935 F.2d at 1397; Hines v. Consolidated Rail Corp., 926 F.2d 262, 273 (3d Cir.1991); United States v. Sebetich, 776 F.2d 412, 419 (3rd Cir. 1985).

In addition, the court must consider "whether expert testimony proffered in the case is sufficiently tied to the facts of the case that it will aid the jury in resolving a factual dispute." Downing, 753 F.2d at 1242. The admission of expert testimony "depends upon the `fit,' i.e., upon a specific proffer showing that scientific research has established that particular features of the eyewitness identifications involved may have impaired the accuracy of those identifications." Id. at 1226; see also Stevens, 935 F.2d at 1397; Hines, 926 F.2d at 273.

With respect to the first factor, the reliability of expert testimony, Downing addressed the foundational requirements for the admission of scientific testimony on novel issues of scientific expertise. The Circuit recognized two lines of case law: (1) the predominant view is the standard set forth in Frye v. United States, 293 F. 1013 (D.C.Cir.1923); under Frye, the reliability of scientific testimony is determined by whether the scientific principle is generally accepted in its field; and (2) the minority view is a less stringent approach; under this approach, the relevance of scientific testimony is determined by applying a Rule 403 balancing test under which the novelty of the scientific principle bears on the weight of the evidence. Downing, 753 F.2d at 1233.

After a review of the advantages and disadvantages of Frye, the Circuit rejected the general acceptance standard. Id. at 1233-1237. The Downing court reasoned:

The [Frye] test has proved to be too malleable to provide the method for orderly and uniform decision-making envisioned by some of its proponents. Moreover, in its pristine form the general acceptance standard reflects a conservative approach to the admissibility of scientific evidence that is at odds with the spirit, if not the precise language, of the Federal Rules of Evidence.

Id. at 1237. The Court held "a particular degree of acceptance of scientific technique within the scientific community is neither a necessary nor a sufficient condition for admissibility; it is, however, one factor that a district court normally should consider in deciding whether to admit evidence based upon the technique." Id.

The Downing court explained, the "reliability inquiry that we envision is flexible and may turn on a number of considerations, in contrast to the process of scientific `nose-counting' that would appear to be compelled by a careful reading of Frye." Id. at 1238. Under the flexible approach, trial courts should consider, in addition to the degree of acceptance in the scientific community, a variety of factors, including the relationship of the new technique to more established methods of scientific analysis, the "existence of specialized literature" on the new technique, the "qualifications and professional stature" of the expert witness, the "non-judicial uses" to which the new scientific technique is put, the "frequency with which [the new] technique leads to erroneous results" and the type of error generated by the new technique. Id. at 1238-39.

If the court determines that scientific testimony is reliable, it must also assess whether the testimony will confuse or mislead the jury. Id. at 1239. The Downing court cautioned against a situation "where the jury is not presented with the data on which the expert relies, but must instead accept the expert's assertions as to the accuracy of his conclusions." Id.

Rule 702 requires the court to balance the reliability of the expert testimony against the dangers that the evidence might, nonetheless, confuse or mislead the jury. Id. at 1240. The Circuit explained in balancing these factors, "there is the `presumption of helpfulness' accorded expert testimony generally under Fed.R.Evid. 702." Id. at 1241 (quoting In re Japanese Electronic Prods. Antitrust, 723 F.2d 238, 279 (3d Cir.1983), rev'd on other grounds, 475 U.S. 574, 106 S. Ct. 1348, 89 L. Ed. 2d 538 (1986)). It nevertheless stated that courts should exercise some caution, "especially in the criminal context, whenever proffered novel scientific evidence, if unreliable or likely to mislead, will increase the likelihood of an erroneous verdict." Id.

With respect to the fit requirement, the Downing court stated:

[A] defendant who seeks the admission of expert testimony must make an on-the-record detailed proffer to the court, including an explanation of precisely how the expert's testimony is relevant to the eyewitness identifications under consideration. The offer of proof should establish the presence of factors (e.g., stress, or differences in race or age as between the eyewitness and the defendant) which have been found by researchers to impair the accuracy of eyewitness identifications.

Id. at 1242 (emphasis added); accord Stevens, 935 F.2d at 1397.

While Downing and its progeny permit the introduction of expert testimony under Rule 702, such testimony is nonetheless subject to a Rule 403 analysis. The Circuit explained the analysis outlined above "incorporates ... the danger of unfair prejudice, enumerated in Fed.R.Evid. 403." Downing, 753 F.2d at 1242. Rule 403, however, may be invoked to "exclude the [expert testimony] evidence if the court finds its probative value to be substantially outweighed by other dangers, e.g., confusion of the issue or waste of time." Id. at 1243; see also, Stevens, 935 F.2d at 1399. The Downing court offered as examples of when Rule 403 may be invoked situations when "there is evidence of a defendant's guilt other than eyewitness evidence, e.g., fingerprints, or other physical evidence." 753 F.2d at 1243.

The Government contested the reliability of the proposed testimony of Penrod. The Government also contended the proffered testimony would confuse and mislead the jury. The Government further argued the proffered Penrod testimony did not fit with the facts of this case. 8 April 1992 Letter Opp. at 2-5; 14 May Hearing Tr. The Government argued, therefore, the proposed expert testimony would be inadmissible under Rule 702. Id.

1. Lack of Correlation between Confidence and Accuracy

Nguyen proposed to offer testimony that studies have shown there is a lack of correlation between eyewitness confidence of an identification and the accuracy of the identification. 2 April 1992 Letter Brief at 1.

a. Penrod's Testimony

At the Downing hearing, Penrod testified he is familiar with around fifty studies which have been conducted directly examining the relationship between confidence and accuracy.^[15] 8 May Hearing Tr. at 44; 13 May Hearing Tr. at 285. He stated he has conducted eight to ten studies examining the correlation between confidence and accuracy of an eyewitness identification. 8 May Hearing Tr. at 45, 110.

By way of an example, Penrod testified he conducted a study involving one hundred and fifty students who viewed an approximately one hundred second videotape of a re-enacted crime.^[16]Id. at 116-17. After playing the videotape, the students viewed an array of photographs and were asked to identify the perpetrator of the crime. The students were also asked to specify how confident they were as to the accuracy of their identification and how confident they were they could avoid a mistaken identification if the perpetrator were not in the array of photographs. Id. at 117. The students were also asked whether they would be willing to sign a sworn statement that the person identified was or was not the person they saw commit the crime and about the accuracy of the identification. Id. He stated that this question was not geared toward ensuring the students used the penalty of perjury standard which is used by a witness testifying at trial.^[17]Id. at 125.

Penrod explained that there was a correlation of .27 between the level of confidence to the accuracy of the students. Id. at 117. He testified that similar studies, all either laboratory studies or field studies,^[18] showed similar results of a correlation of .25. Id. at 47, 49, 137. Penrod testified that a correlation coefficient "doesn't mean very much ... unless you're a statistician." Id. at 47. When asked to explain what "correlation of .27" meant, Penrod could not explain it verbally, but rather went to a drawing board to give a lengthy and confusing illustration which was, as he termed it, a "crude way of explaining what a correlation of .27 is."^[19]Id. at 118-123.

Penrod testified he could explain to the jury that the relationship between the level of expressed confidence and the accuracy of the identification is weak; accordingly to Penrod, the expression of confidence is not a reliable source to determine accuracy of the identification. Id. at 48. He stated these principles are counter-intuitive, and therefore, these principles are an educational benefit which is helpful to the jury. Id.; 13 May Hearing Tr. at 294. He testified the principles revealed from the studies are "determined to a reliable degree of scientific certainty based upon the statistical analysis of the studies." Id. at 289; see also 8 May Hearing Tr. at 49.

Penrod testified he could not transfer the results in the studies to "understand exactly how a person in a real life situation is going to react...." Id. at 90. He further agreed that there was no way to predict with any degree of certainty whether the group of people in a study would react the same way in real life as they did in the studies. Id.

Penrod concluded, nonetheless, he did not believe the studies were unreliable even though they did not involve real crime situations because the pattern of results in laboratory studies are similar to the pattern of results in field studies. Id. at 51-53. In addition, Penrod stated a person's cognitive abilities^[20] do not change from a laboratory or a field setting to real life; the fundamental limits on a person's ability to store and retrieve information are the same regardless of the setting. Id.

Penrod testified that the cognitive process is dependent upon numerous variables such as the degree of attention, the proximity of the subject to the witness, the lighting conditions, the duration of exposure to the subject, the amount of stimuli, the seriousness of the offense, an awareness that the witness will be called upon to make an identification,^[21] the length of time between the event and the identification, the individual makeup of the witness, motivation and, under certain categories, training. Id. at 98-103. Penrod testified that when conducting studies he manipulates one or two of these variables while keeping everything else the same. Id. at 103. He stated he can keep other variables, such as motivation, under control by randomly assigning the subjects of the study. Id. at 105-108.

Penrod admitted that if several of these variables were positive, he would expect a higher correlation between accuracy and confidence. Id. at 127-28, 130-31. For instance, he stated, if a witness had more experience with a target group than the average subject studied, there would be an increase in the correlation between confidence and accuracy. Id. at 128. Similarly, Penrod stated a witness with a great opportunity to view, in a close proximity to the suspect and who knew he was going to have to make an identification would also have a higher correlation between confidence and accuracy. Id. at 128-29.

He stated putting all of these factors — experience with a target group, opportunity to view, close proximity and awareness of need to make identification — together, he would expect the correlation between confidence and accuracy to be .45, almost double from the original level of .27. Id. at 133. Penrod stated he had not conducted studies where positive variables existed in such a manner because it would be too easy for everyone to get correct answers. Id. at 133. When asked to translate the meaning of .45, Penrod stated he could not put it in terms of a percentage and could only translate it by giving an illustration as he did earlier. Id. at 133-34.

Lastly, Penrod admitted he was aware of a study by Professor Dethenbocker of the University of Nebraska, conducted in 1980, which reviewed forty-three different studies. Id. at 137-38. He acknowledged this review concluded that twenty-two of the forty-three studies showed a positive correlation of confidence to accuracy. Id. at 138. He also stated he had no reason to dispute a study conducted by Professor Brigham of the University of Florida. Id. at 139. This study showed that eighty-five percent of those witnesses who stated they were confident in their identifications were correct.^[22]Id.

Penrod offered two possible explanations for the difference in the results of the studies.

One is when you run a statistical test, which is the sort of the deciding criteria of whether you observed an effect. The likelihood that you will observe a significant result is in a very substantial degree dependent on the sample size that you're using.

. . . . .

The other reason people fail to observe a difference is as a result of sampling error.

13 May Hearing Tr. at 287-88. He stated the benefit of performing a Meta-analysis^[23] of individual studies is that it studies larger number of samples and observations. Id. at 288.

With respect to this case, Penrod conceded the circumstances under which Peterkin viewed Nguyen — a three to five minute opportunity to view in close proximity, awareness of the need to identify and no distracting external stimuli — would promote a higher correlation between confidence and accuracy. 8 May Hearing Tr. at 140-41.

b. Admissibility — Rule 702

In this case, Nguyen argued Peterkin, if called, would have testified that he is confident in the accuracy of the photograph identifications, the person to person identification at Rikers Island and the in-court identification. As mentioned, Nguyen proposed to offer Penrod's testimony regarding the lack of a correlation between confidence and accuracy in eyewitness identification to rebut the natural assumption that Peterkin's expressed confidence in the accuracy of his identifications means his identifications are reliable. 2 April 1992 Letter Brief at 3-4.

Penrod's educational and independent research experience on the issue of the correlation between confidence and accuracy qualify him to testify as an expert on this issue. Although there do not appear to be any non-judicial uses for the confidence/accuracy studies, a significant number of studies have been conducted reaching the same results,^[24] which appear to be reliable. Despite Penrod's first hand knowledge of the confidence/accuracy studies, there is nonetheless, a danger that he will not be able to effectively communicate the studies to the jury.

As mentioned, during the Downing hearing, when Penrod was asked to explain the meaning of "a correlation of .27," his explanation all but clarified the issue. Aside from providing a long and confusing crude illustration on the drawing board, Penrod was only able to summarize the findings of the studies as showing that there is a "weak" relationship between confidence and accuracy. To describe the conclusion as weak without being able to clearly articulate how the conclusion was derived, requires the jury to accept Penrod's assertions as to the accuracy of his conclusions without the benefit of having the data upon which he relies. As indicated in Downing, this is one of the dangers to be avoided when introducing expert testimony. 753 F.2d at 1239.

If Nguyen had sought to call Penrod at trial, there would have been a great danger the proffered Penrod testimony on the correlation of confidence to accuracy would have confused or misled the jury. This danger would have outweighed the reliability and helpfulness of the proffered Penrod testimony on confidence/accuracy. The motion to introduce this testimony would, therefore, have been denied.^[25]

2. Cross-Racial Identifications

Nguyen also proposed to introduce expert testimony regarding studies showing that cross-racial identifications are less accurate than same-race identifications. 2 April 1992 Letter Brief at 1, 4.

a. Penrod's Testimony

At the Downing hearing Penrod testified he had not conducted studies on cross-racial identifications. 8 May Hearing Tr. at 54, 86, 141. He has reviewed studies conducted by other researchers and is familiar with the pattern of results of around eighteen studies.^[26]Id. at 141, 148. He stated he could talk about the specifics of about ten studies. Id. at 143-44, 148-49. Penrod testified that the general principles regarding cross-racial identifications have a reasonable degree of scientific certainty. 13 May Hearing Tr. at 297; 8 May Hearing Tr. at 55-56. Penrod did not testify, however, that the types of studies used to reach such conclusions were the type reasonably relied on by experts in the field of human perception. See Fed.R.Evid. 703.

Penrod testified the majority of the studies conducted involved the ability of caucasians and African-Americans to identify members of the other group. Id. at 55. He stated the studies compared caucasians identifying caucasians, caucasians identifying African-Americans, African-Americans identifying African-Americans and African-Americans identifying caucasians. Id. He testified the pattern of the results was that caucasians could more accurately identify caucasians than caucasians could identify African-Americans. Similarly, African-Americans could identify African-Americans more accurately than African-Americans could identify caucasians. Peterkin could not quantify what he meant by saying it's "more difficult" to make accurate cross-racial identifications. Id. at 156.

Penrod stated the studies showed, however, that African-Americans could more accurately identify caucasians than caucasians could identify African-Americans. Id. at 152-53. When asked to quantify the differential between the ability of African-Americans to identify caucasians, he stated African-Americans are eighty percent as bad at identifying caucasians as caucasians are at identifying African-Americans.^[27]Id. at 155. Penrod stated, from his examination of the studies conducted, he believed the results were scientifically reliable. Id. at 155-56. While Penrod testified this was the general pattern of results in the studies he has reviewed, he acknowledged that a review by Professors Lindsey and Wells and a study by Professor Brigham both concluded there was no cross-racial impairment. Id. at 144-45, 147. Penrod also acknowledged Professor Loftus concluded there was a problem with the reliability of cross-racial studies because they used photographs.^[28]Id. at 146.

Penrod also testified as to one study conducted by Professors Chance and Goldstein (the "Chance and Goldstein Study") which involved cross-racial identifications among African-Americans, Asians and caucasians. Id. at 156. He stated this study concluded that African-Americans were less accurate at identifying Asians than they were at identifying caucasians, which was less accurate than their ability to identify African-Americans. Id. at 157.

Although Penrod agreed that the reliability of conclusions is judged, in part, by the number of studies conducted reaching the same conclusion, he stated he would feel confident in relying on the one study conducted in the area of cross-racial impairment of African-Americans and Asians. Without explanation, Penrod stated he would rely on the Chance and Goldstein Study over the review of studies, conducted by Professors Lindsey and Wells in 1983, which revealed that there was no cross-racial effect. Id. at 144, 157-58.

The Chance and Goldstein Study involved two separate studies which displayed forty photographs to the subjects. 13 May Hearing Tr. at 235-36. Penrod stated viewing conditions were impaired to avoid the subjects from making one hundred percent accurate identifications. Id. at 235. He stated one impairment used in the Chance and Goldstein Study was to show the photographs for only twenty seconds. Id. at 236. After the subjects viewed the forty photographs for twenty seconds, twenty photographs were removed and replaced with twenty new photographs. The subjects were then asked to state how many and which photographs were not in the original photographic array. Id. at 236-37.

Penrod conceded that twenty seconds was a short period of time and if the rate of exposure were increased the rate of accuracy would increase as well. Id. at 236. In addition to rate of exposure, Penrod stated other factors which could be used to impair conditions are the amount of stress, viewer awareness of the need to make a subsequent identification, use of disguise and increase in retention interval. Id. at 242-43. Penrod stated the existence of these factors would promote a situation where a face would become over learned or identification reaches one hundred percent. 8 May Hearing Tr. at 166.

With respect to increasing the retention interval, the time between the initial viewing and subsequent viewings, Penrod stated an increase to twenty-four hours was too great an impairment to obtain useful data. 13 May Hearing Tr. at 243-44. He stated that when the retention interval was placed at around two hours, "the identification rates were about 50 percent overall." Id. at 244. He explained, the closer in time the identification is to the original viewing, the more accurate the identification will be. Id. at 246.

Penrod also stated that familiarity with a race reduced the cross-racial impairment. 8 May Hearing Tr. at 159-60. Penrod was previously of a different conviction as recently as September 1991, when he testified in Stevens. He stated then that contact or familiarity with race could not reduce the impairment. Id. at 160-61. He stated his position changed with respect to this issue after reviewing the Chance and Goldstein Study. Id. He stated he was comfortable with this new research and would not change his position on this issue in the future. Id. at 161.

With respect to the witness' degree of attention, Penrod stated he did not believe an increased level of attention would impact on the cross-racial impairment. Id. at 162. He stated, however, if the opportunity to view was increased there would be an increase in the performance in a cross-racial situation. Id. at 163. He explained, except in a situation where the subject is over learned, a difference will remain between accuracy of same race and cross-racial identifications given an increased opportunity to view. Id. at 163-64. He stated he was not aware at what point during a confrontation a subject became over learned, but he recognized a three to five minute opportunity to view would promote an increased ability to make cross-racial identification. Id. at 165-67.

Penrod testified he thought expert testimony regarding the cross-racial impairment, in part, would be helpful to jurors because of jury misperception. 13 May Hearing at 246. He stated there are some studies by Professors Army and Tersillian-Jones about people's beliefs concerning cross racial identifications. Id. at 247. He stated Professors Dethenbocker and Loftus have shown there is "some disagreement amongst members of the public about whether cross race identification has a deleterious impact on eye witness identification ...." Id. Penrod stated "there is no question in [his] mind that there is a perception among some portion of the public that it is more difficult to identify a member of a different race than of the same race." Id.

b. Admissibility — Rule 702

In this case, Peterkin, an African-American undercover agent, identified Nguyen, a Vietnamese man. As mentioned, Nguyen proposed to introduce expert testimony that African-Americans are not accurate at identifying Asians. 2 April 1992 Letter Brief at 4. The Government argued the proffered Penrod testimony does not fit with the facts of this case. 8 April 1992 Letter Opp. at 5. The Government argued, unlike the studies and the Stevens case, Peterkin was not faced with an isolated incident of having to identify Vietnamese individuals. Id. Indeed, the Government stated Peterkin had encountered five Vietnamese targets during the investigation. Id. The Government argued Peterkin had an exceptional opportunity to study a number of individuals and distinguish characteristics of individuals with the same national origin. Id.

The reliability of the proffered Penrod testimony is questionable. Although Penrod is well educated and has written a review of studies on the issue of cross-racial identification, he has not conducted the studies himself and was not able to articulate the underlying data and circumstances of the studies. Penrod agreed that in order to determine whether a particular study is reliable it is necessary to consider "all of the data, how it was performed [and] what kind of subjects [are used]." 8 May Hearing Tr. at 143. Nor did Penrod testify that the studies are the type reasonably relied upon by experts in the field of human perception. Indeed, different studies have concluded that there is no cross-racial impairment.

Penrod has, furthermore, changed his position (merely on the basis of reading a new study) regarding the impact of familiarity with a race on the cross-racial impairment. In Stevens, two caucasian victims of a holdup were required to identify an African-American assailant. 935 F.2d at 1384. The trial court admitted Penrod's testimony regarding cross-racial identification. Id. at 1397. At the time of the Downing hearing in Stevens, Penrod testified about the studies which had been conducted and the pattern of results from those studies. Penrod testified that familiarity with the race would not positively effect the cross-racial impairment.

Since Stevens, the Chance and Goldstein Study, involving caucasians, African-Americans and Asians, indicated that familiarity with the race reduces the cross-racial impairment. At the pretrial hearing, Penrod admitted to having changed his position regarding the impact of familiarity with a race on cross-racial identifications; he now believes it has a positive impact. 8 May Hearing Tr. at 160-161.

Furthermore, Penrod did not establish his expertise in the area of cross-racial impairment and what expertise he could provide to the jury that they could not otherwise ascertain by reading the same studies he has read. Penrod simply summarized studies he has read and reviewed; he did not specify what expert qualifications he has, that the jury lacks, in reading studies. Penrod testified there is a perception among some of the public that it is more difficult to identify someone from another race than it is to identify someone from the same race. By his own admission, therefore, Penrod would not be educating the jury with respect to something they do not know.

In addition, the scientific reliability of the studies showing a cross-racial impairment between African-Americans and Asians is uncertain. Only two studies have been conducted on a cross-racial impairment between African-Americans and Asians and it appears they have no non-judicial uses. Penrod did not testify that the studies are the type reasonably relied on by scientists in the field of human perception. Similarly, with respect to the reliability of the conclusions, Penrod testified that the number of studies conducted reaching the same result is an indicator of reliability. Because only the Chance and Goldstein Study has found a cross-racial impairment between African-Americans and Asians and other studies have concluded that there is no impairment in cross-racial identifications, the reliability of the studies on cross-racial impairment is questionable.

As a result of the fact that Penrod has no first hand knowledge of the studies about which he would have testified, it appears he would be no better able to explain the underlying data or effectively communicate the studies to the jury than the jury members themselves. Indeed, Penrod testified that of the eighteen studies he reviewed he can only discuss some of the data in about ten of the studies. Penrod has only read and has not conducted a thorough review of the Chance and Goldstein Study which studied African-Americans and Asians. Accordingly, the danger of confusing the jury as a result of Penrod's inability to articulate the data underlying the studies at issue would outweigh the reliability of the testimony which is itself questionable.

In addition, the facts of this case did not fit with the proffered Penrod testimony. Many of the variables which are attributable to the cross-racial impairment were positive forces in this case. A large number of the studies on cross-racial impairment were conducted on college campuses where the caucasian population had a reduced contact with African-Americans. Here, Peterkin was familiar with the Vietnamese subjects. He had been assigned to this undercover investigation since November 1991 and had contact with five other Vietnamese men prior to the 2 February 1991 meeting.

Additionally, Peterkin had a higher degree of attentiveness and a much longer time to observe than the average person in the studies. The studies involved brief viewings of twenty seconds to two minutes of photographic arrays of forty to one hundred photographs; Peterkin viewed the man in the Park face-to-face, at arm's length for three to five minutes. As previously stated, he was an undercover agent going to the Park with the intent of completing the purchase and observing the seller of the Weapon. Peterkin also knew in advance, unlike the students in the studies, that he would have to identify the seller of the Weapon and he made the identification within two hours from the meeting. As admitted by Penrod, these factors — familiarity with the race, length of observation time, proximity to subject, attentiveness to subject and awareness of need to identify — would promote a significant reduction in the cross-racial impairment.

In light of the existence of positive variables which did not exist in the cross-racial studies, Nguyen failed to show how the proffered Penrod testimony would have been relevant to Peterkin's identification.^[29] The expert testimony on cross-racial impairment would have been inadmissible under Rule 702.

3. Stress Factor

Nguyen proposed to introduce expert testimony that a person who is under stress at the time of observation is less accurate than a person who is not under stress. 2 April 1992 Letter Brief at 4. Nguyen argued Peterkin was under very stressful conditions during the 2 February 1991 meeting in the Park. Id. Nguyen argued the facts — that it was dark, Peterkin was alone and buying an illegal gun, from an alleged member of a violent street gang while four other individuals were sitting in the nearby parked Camaro — gave rise to a stressful situation for Peterkin. Id.

a. Penrod's Testimony

At the Downing hearing, Penrod testified about various studies he has reviewed relating to the affect of stress on the accuracy of identifications.^[30] Penrod has not conducted any of these studies. 8 May Hearing Tr. at 86. He stated the studies were designed to create an increased level of arousal to determine how the subjects performed under stress.^[31] 13 May Hearing Tr. at 248-49. Penrod explained the focus of the studies is not on whether stress exists, but rather on the level of stress and the impact that stress has on the accuracy of a person's identification. Id. at 249, 252. He stated he believed the conclusions of these studies were scientifically reliable. 8 May Hearing Tr. at 60-61. Penrod neither testified that the studies on stress were of a type reasonably relied upon by experts in the field of human perception, nor did he explain what expertise he has in reading and summarizing the studies that the jury would not have if given an opportunity to read the studies.

Penrod stated typical studies regarding stress involved the artificial affectation of stress by telling the subjects they had to make a public speech or receive an inoculation. Id. at 56. He stated the studies did not involve the kind of stress involved in real life crimes because of the limitations on researchers. Id. at 56-57. Penrod testified the studies concluded that the common perception that people under stress have heightened awareness and are better able to remember faces is incorrect. Id. at 57-58. Penrod stated, "in those kinds of studies, we do observe an impairment in performance under heightened levels of stress and arousal." Id. at 57 (emphasis added). He explained as stress increases it impairs one's ability to encode information. Id. at 58-59.

Penrod testified certain amounts of stress are good because it increases performance, but as stress levels further increase, it hits a point where stress decreases performance. Id. at 60; 13 May Hearing Tr. at 249. He described this phenomenon as a performance bell curve. Id. However, he could not say at what point in the so-called bell curve stress turned from assisting to impairing performance because it depends on the individual and the task involved. Id. at 250; 8 May Hearing Tr. at 60.

Penrod testified that a variety of factors could put one under stress such as fear or the pressure to perform well. Id. at 59. Similarly, Penrod testified how one performs under stress is dependent on the nature of the situation. 13 May Hearing Tr. at 250. He explained, however, that everyone has a different level of susceptibility to the effects of stress. 8 May Hearing Tr. at 57. Indeed, Penrod did not deny the common perception that some people handle stress well and perform better than other people under stress. 13 May Hearing Tr. at 253.

He further agreed that experience with a stressor will generally decrease the impact of the heightened levels of stress. 8 May Hearing Tr. at 57; 13 May Hearing Tr. at 253. Penrod agreed that undercover experience would be such an experience attributable to a lower level of stress in a particular situation. Id. at 254. Indeed, he agreed it was possible that a person with undercover experience may have heightened attention when put in an undercover situation and may have the amount of stress necessary to make his or her performance as good as it could be. Id.

Penrod stated there would be no way of knowing whether that person would be on the low or the high end of the performance bell curve. Id. at 255. Penrod stated he could only make such an evaluation after considering "the same information that the jury would in terms of acquiring knowledge about the experience of the person in that situation and their ... reports about what they were experiencing during that situation." Id.

With respect to the facts of this case, Penrod stated the stressor was the fact that the undercover transaction involved guns. Id. at 256. Penrod agreed experience in this situation was not limited to a situation of gun purchasing, but included situations "where an officer went into an undercover situation to purchase narcotics, and based upon his experience had to be concerned about the drug seller or buyer being armed with a weapon." Id. at 257.

Penrod also testified that even in a stressful situation, an increase in the duration of exposure time will increase the ability to make an accurate identification. Id. at 258. Specifically, he stated he "would expect performance to be better in [a] three to five [minute viewing] than in [a] twenty [second viewing]." Id. He stated many of the studies conducted on the affect of stress involve viewings in the range of twenty to thirty seconds to two minutes. Id. at 258-59.

Penrod further testified that "all things being equal," the awareness of the need to make a subsequent identification would reduce the effect of stress. Id. at 259. Similarly, he stated the time interval between the observation and the identification would have an impact upon the affect of stress. Id. Penrod testified that the "vast majority of studies ... show that the loss of information begins immediately, and an immediate identification or immediate statements are going to be more accurate than if you wait for a passage of hours or days." Id. at 262.

Penrod conceded, however, that a couple of studies have shown better performance a couple of hours after the original observation than an identification made immediately after the original observation. Id. at 261. Penrod stated he would rely on these studies to say the phenomenon exists, but he does not know why it appears. Id.

Penrod testified that proximity to the event would also have a positive impact on the ability to make accurate identifications, even in a situation where there is stress. Id. at 262. Although he was aware that studies have been done on this variable, he stated his conclusion was based "more on logical grounds than on the basis of empirical research." Id. at 262.

b. Admissibility — Rule 702

The Government attacked the proffered Penrod testimony on the ground that it is unreliable, it would confuse the jury and it does not fit with the facts of this case. Specifically, the Government argued Peterkin was able to handle any stress accompanying the purchase of the Weapon. The Government stated the meeting was not threatening or hostile but rather business-like. 8 April 1992 Letter Brief at 5. The Government argued Nguyen was, in fact, eager to please Peterkin. Id. The Government further argued the overwhelming amount of stress necessary for a decline in performance was not present in this case. 14 May Hearing Tr.

The reliability of the proffered Penrod testimony is questionable at best. Although Penrod is well educated and has testified he has reviewed studies of the issue of stress impairment, he has not conducted the studies himself and cannot articulate the underlying data and circumstances of the studies. Significantly, Penrod neither explained his expertise with respect to either the stress factor or statistics nor did Penrod testify that the studies conducted on this issue are the type reasonably relied on by scientists in the field of human perception. Moreover, Nguyen, as with other aspects of the proffered Penrod testimony, did not establish why the jury could not read the studies on which Penrod relies and draw the same conclusions.

Penrod would not have been able to explain the underlying data or effectively communicate the studies to the jury. Penrod would have merely read or related to the jury a summary of the studies he has read. Absent a thorough explanation, the jury would have been called upon to accept the accuracy of his conclusions without benefit of the underlying data. As indicated in Downing, if the jury does not have access to the underlying data to determine whether to give credence to the expert's testimony, there is a danger the testimony will mislead the jury. 753 F.2d at 1239. Weighing the strong potential of the proffered Penrod testimony to mislead the jury against the questionably reliable evidence, the balance would weigh in favor of excluding the proffered Penrod testimony on the stress factor.

In addition, Nguyen would have failed to satisfy the fit requirement. In Stevens, expert testimony relating to the effect of stress on the ability to make an accurate identification was admitted. 935 F.2d at 1397. The victims in Stevens were accosted by an African-American man during night at an isolated bus stop. The assailant held the male victim at gun point and directed the female victim to perform fellatio or he would shoot the male victim. Id. at 1384-85. The trial court found because the victims were under a great deal of stress, the Penrod testimony relating to the stress factor was sufficiently tied to the facts of the case. Id. at 1397-98.

In Sebetich, a police officer responded to a call reporting that a bank robber had fled in a pick-up truck. 776 F.2d at 412. During a twenty-two mile car chase at speeds of up to seventy-five miles an hour, the passenger in the truck extended his upper-body outside the window and fired a semi-automatic pistol at the police officer. Id. at 415. The police officer ducked as a bullet penetrated the windshield and lodged in the headrest of the driver's seat. Id. When the passenger continued to fire at the police officer, the officer allowed about forty yards as he followed the truck. Id. at 419. Eighteen months after the chase, the police officer identified the passenger from a photograph posted on a wall in the police station. Id. at 418. Two months later, the police officer identified the passenger from an array of five photographs. Id.

The trial court denied the admission of expert testimony regarding the effect of the stressful conditions under which the police officer observed the passenger. Id. at 418-19. The Third Circuit reversed the trial court's ruling and remanded the case for a hearing pursuant to Downing. Id. at 419. The Circuit stated the circumstances under which the police officer observed the passenger were stressful. The court indicated a Downing hearing was necessary because "[t]here is evidence that stress decreases the reliability of eyewitness identifications, contrary to common understanding." Id.

In this case, Penrod testified the stressor was being on an undercover assignment involving guns. The facts of the case, however, did not support a conclusion that Peterkin was under heightened levels of stress at the time he observed the seller of the Weapon. Peterkin had previously arranged the sale of the Weapon, had the necessary amount of money and the seller was accommodating. Unlike the situation in Stevens, Peterkin was not a victim of a crime or force, but rather was engaged in a business transaction with the seller. Peterkin stated he expected the man to pull out the Weapon and he was not surprised when he did. This case is also distinguished from the facts of Sebetich because Peterkin was not under gun fire and was not engaged in a high speed chase of a bank robber. Indeed, at the time the Weapon was handed to Peterkin a round was not put in the chamber, therefore, the Weapon could not fire. 8 May Hearing Tr. at 34.

Significantly, Peterkin was an experienced investigator who had been on many undercover assignments before. Id. at 16. The fact that this investigation was Peterkin's first federal investigation did not increase his stress; he stated this investigation did not differ from others because he always wanted to do a good job. Id. at 17. Penrod has stated that a person doing something for the first time may use that as a retrieval cue which enhances one's ability to correctly retrieve information. Id. at 94. Moreover, Penrod has conceded that a experience in a situation, such as undercover investigations involving guns, would reduce the effect of stress on the ability to make an accurate identification. 13 May Hearing Tr. at 254.

Nguyen's reliance on the fact that Peterkin was aware there were individuals in the nearby Camaro does not give rise to stressful conditions. Peterkin testified he could not see the other individuals in the Camaro during the transaction with the seller. 8 May Hearing Tr. at 23. As stated, the transaction was businesslike and the seller of the Weapon was accommodating and sought to arrange future sales. 8 April 1992 Letter Opp. at 5.

In light of the absence of stressful circumstances at the time of Peterkin's observation of the seller of the Weapon and the fact that Peterkin's experience in the situation would reduce the deleterious effect of stress, the proffered Penrod testimony on stress does not fit with the facts of this case. Accordingly, if Nguyen had sought to introduce the expert testimony relating to the effects of stress, it would have been inadmissible under Rule 702.

4. Effects of Weapon Focus

Nguyen proposed to introduce expert testimony on the issue that when there is a weapon present in a crime, the eyewitness focuses on the weapon and not on the face of the perpetrator, thus reducing the eyewitness's ability to make an accurate identification.

a. Penrod's Testimony

At the Downing hearing, Penrod testified about various studies he has reviewed on the weapon-focus phenomenon,^[32] which indicated that the presence of a weapon during an observation has a deleterious affect on the ability to make an accurate identification. 13 May Hearing Tr. at 263-69. He agreed the most accepted reason for this phenomenon is that during the incident in which a gun was visible, the person focuses on the weapon rather than the face of the perpetrator. Id. at 267.

Penrod explained that the studies involve either staged crimes or videotaped crimes and focus on whether there is impaired eyewitness performance when a weapon has been visible. Id. at 263. The studies compare the situation where no weapon is visible and where a weapon is visible for the same or similar periods of time as the person wielding the weapon. Id. at 264.

The studies have not manipulated the length of time the weapon is visible. Id. at 265. Therefore, Penrod stated he could not specify the distinction between a situation where a gun was visible for five seconds of a three to five minute event and one where a gun was visible for the full length of the event. Id. at 266. He concluded that where a weapon was visible only for a relatively small portion of the event, he would expect a "reduced weapon focus effect." Id. at 267. When confronted with the situation where a weapon was visible for only five to ten seconds of a three to five minute encounter, Penrod stated the weapon focus phenomenon would have result in only a minor impairment on the ability to make an accurate identification.^[33]Id. at 269.

b. Admissibility — Rule 702

The Government pointed out that the Weapon was only wielded for five to ten seconds during a three to five minute meeting and therefore, the weapon-focus phenomenon would not be a significant factor affecting the ability of Peterkin to make an accurate identification. 14 May Hearing Tr.

The reliability of the proffered Penrod testimony on the weapon-focus phenomenon is uncertain. Penrod merely stated he has reviewed studies on the weapon-focus phenomenon.^[34] He neither specified how many studies have been conducted, nor did he specify whether the studies are of a type reasonably relied on by experts in the field of human perception. In addition, it does not appear that the studies relating to weapon-focus have any non-judicial uses. Thus, reliability of the studies has not been established.

Penrod could not articulate with specificity the underlying data and circumstances of the studies. Significantly, Nguyen failed to establish why a jury could not reach the same conclusion by reading the studies. Absent a thorough explanation of the studies, the jury would have had to accept Penrod's conclusions as accurate. Thus there would be a strong possibility of misleading the jury. Moreover, the weapon-focus phenomenon is a rather logical conclusion and it was not explained why the same conclusion cannot be brought out on cross-examination of Peterkin.

The weapon-focus phenomenon also failed the fit requirement. In United States v. Dowling, 855 F.2d 114 (3d Cir. 1988), aff'd, 493 U.S. 342, 110 S. Ct. 668, 107 L. Ed. 2d 708 (1990), the defendant, a fleeing armed bank robber, was observed by one Messer who stated he was about fifteen feet away when he saw the defendant's face. Id. at 116. The defendant was also observed by one Pichardo when the bank robber drove up on her yard where her children were playing. Id. The defendant sought to introduce expert testimony on the weapon-focus phenomenon. Id. at 118-119. The Circuit affirmed the trial court's decision to exclude the expert testimony because neither Messer or Pichardo testified the defendant was wielding a weapon when they observed him. Id. at 119.

In this case, Peterkin only viewed the Weapon for approximately five to ten seconds of a three to five minute meeting. As stated by Penrod, the studies compare only the effect of a weapon being visible and do not compare the effect of the length of time during which the weapon is visible. In addition to the length of visibility not having been studied, Penrod conceded that in the circumstances of this case there would be only a minor impairment in accuracy as a result of the presence of the Weapon. Because the presence of the Weapon was for an insubstantial period of time of the meeting, the facts do not fit with the testimony. If Nguyen had sought to introduce the proffered Penrod testimony on the weapon-focus phenomenon, it would have been inadmissible under Rule 702.

5. Effects of Relation-Back

Nguyen proposed to introduce expert testimony on the effects of relation-back on eyewitness identifications. 8 May Hearing Tr. at 64-66. Specifically, Nguyen proposed to introduce expert testimony on the fact that a witness who is asked to make a subsequent identification will tend to relate back to the initial identification rather than to the crime itself. Id. at 65.

a. Penrod's Testimony

At the Downing hearing Penrod testified he was familiar with and had reviewed around eight to ten studies relating to the effects of relation-back; a Meta-analysis has not been conducted in this field. 13 May Hearing Tr. at 298-99. Penrod has not conducted any studies on this issue.^[35] 8 May Hearing Tr. at 64-65, 86. Penrod explained relation-back focuses on the issue of what information people draw upon in memory as a basis for making subsequent identifications. Id. at 64. He stated that these "studies indicate that the interposing of exposure to faces between the initial viewing of an event and a later effort to make an identification can impair the accuracy of the later identification." Id. at 65. He stated that, as a result of seeing a set of faces after an encounter, the eyewitness will have a heightened level of making a misidentification at a later point. Id. Penrod stated these results, in his opinion, were scientifically reliable. Id. at 66. Penrod did not testify, however, whether these studies are the type reasonably relied on by experts in the field of human perception.

Penrod explained information which is seen between an encounter and an identification can "either in some ways displace, supplant the information acquired at the original viewing, ... and substitute for that information or ... may be more readily retrieved at a later point in time...." Id. at 65-66; see also 13 May Hearing Tr. at 272.

Penrod stated: "There have been a variety of techniques, all of which, in one way or another, bear upon the question of what does this interpolated task or intervening information have on the quality of memory at a later point in time." Id. at 273-74. Penrod testified that the studies conducted involved situations where memory of the subjects is verbally modified by asking the subjects questions that imply the existence of objects they could not have seen or were not in the event observed. Id. at 272-73.

Penrod described other studies which involved a situation where subjects, who have viewed a crime,^[36] overhear conversations making suggestions about the characteristics of the perpetrator. Id. at 273. Another technique he described was used in studies in which subjects were asked to make an identification from an array of people or photographs which did not contain the perpetrator and then subsequently asked to make an identification from a second array which did contain the perpetrator. Id. at 274.

Penrod agreed that in all of the studies on the relation-back effect, the subjects were introduced to incorrect information after the original observation of the perpetrator and the studies focused on whether the subjects made a subsequent incorrect identification. Id. He agreed if the first identification was correct, the effect of relation-back on subsequent identifications would be moot. Id. at 275.

b. Admissibility — Rule 702

In Stevens, the victims of the crime initially identified Stevens from a wanted board and subsequently identified Stevens from a lineup. 935 F.2d at 1385. At trial, Stevens sought to introduce expert testimony that "once a witness makes an identification, he or she will tend to stick with that initial choice at subsequent photographic arrays or lineups, even if it was erroneous." Id. at 1399. Stevens argued the "testimony on the `relation-back' issue would have prompted the jury to discount the corroborative value of the victim's identifications of him from the photographic array and lineup." Id.

The Circuit affirmed the exclusion of the expert testimony on relation-back inadmissible because it is a "pedestrian" comment "susceptible of elucidation without specialized scientific knowledge." Id. It stated the thrust of the testimony "could have been fleshed out adequately by counsel through probing cross-examination and arguments pitched to the common sense of the jury." Id. at 1300-1400.

Penrod neither proffered testimony as to his expertise on the issue of relation-back, nor did he testify that the studies conducted were of a type reasonably relied upon by experts in the field of human perception. Significantly, Nguyen failed to establish that the jury would not be capable of understanding the concepts underlying the relation-back theory and could not reach the same conclusions through a "probing cross-examination." Nguyen failed to establish that the proffered Penrod testimony on the relation-back factor would have been helpful. If Nguyen had sought to introduce such expert testimony, the motion would have been denied.

6. Accuracy of Police Officer Eyewitnesses

Nguyen proposed to introduce Penrod's testimony that police officers are subject to the same limitations of a normal eyewitness. 2 April 1992 Letter Brief at 5. Nguyen argued several studies performed indicate that the ability of a police officer to make identifications is no greater than the ability of a civilian to make identifications. Id. (citing R. Bull, Psychology's Contribution to Policing, Psychology and Law, 419-21 (1984); B. Clifford, Police as Eyewitnesses, New Society, 176-77 (22 April 1978)).

a. Penrod's Testimony

At the Downing hearing Penrod testified he has reviewed various studies conducted by other researchers on police officers as eyewitnesses; he has not conducted studies on this issue.^[37] 8 May Hearing at 61, 86. He stated around five or six studies have focused on "whether police officers are better than civilians at making identifications or remembering other types of information" and "whether police officers with more experience are better at making identifications or recalling other types of information" and around twenty studies have focused on whether there are "ways to train people to be better, more effective eye witnesses." Id. at 61; 13 May Hearing Tr. at 299. He stated these studies were of the type routinely relied on by scientists analyzing these issues. 8 May Hearing Tr. at 64; 13 May Hearing Tr. at 299.

He testified police officers are no better at identifying persons than civilians because police officers have the same fundamental cognitive processes and are subject to the same limitations as civilians. 8 May Hearing Tr. at 64; 13 May Hearing Tr. 275-76. In addition, he stated a police officer's experience does not make him a better eyewitness. 8 May Hearing Tr. at 61-63. Penrod referred to a study conducted by Professor Malpas, which concluded there is no way to train people to be better eyewitnesses. Id. at 64. He testified that he believed these studies were scientifically reliable. Id.

Penrod agreed that the studies concluded that "a police officer is a human being just like you or I or anyone else and they don't have any innate ability to make identifications that are better than persons in the general public." 13 May Hearing Tr. at 276. He conceded, however, that in a situation where stress is present and a police officer has been trained or is experienced in dealing with the specific situation, he would expect a police officer to perform better than persons in the general public. Id. at 277.

b. Admissibility — Rule 702

The issue of the accuracy of an identification made by a police officer has not been addressed by Downing or its progeny. Pursuant to the teachings of Downing, crucial to the inquiry to the introduction of such expert testimony is whether it is helpful to the jury. As was the case on the relation-back theory, Penrod's comments on police officer eyewitnesses were pedestrian. Penrod offered nothing beyond the notion that police officers are human beings and their abilities to observe, retain and recollect information was no different than those of a civilian. Special scientific expertise is not necessary to convey this principle to the jury, but rather it could be thoroughly divulged through cross-examination.

As previously stated, expert testimony that cannot provide information the jury does not possess or cannot otherwise obtain is inadmissible under Rule 702. If Nguyen had sought to introduce the proffered Penrod testimony on the issue of identifications by police officer eyewitnesses, it would have been inadmissible under Rule 702.

Conclusion

For the reasons set forth above, the motion for a bifurcated trial for elements of Count Three of the Indictment was denied and the motion to exclude in-court and out-of-court identification testimony was denied. If Nguyen had sought to call Penrod at trial, the motion to introduce expert testimony on eyewitness identification would have been denied.

NOTES

[1] Defendants Steven Wu and Thoung Van Le, a/k/a "Alvin," entered pleas of guilty on 23 March 1992. Defendant Tuan Quoc Tran, a/k/a "Danny," is currently in Canada facing criminal charges. He will not be released until after the trial in Canada and after he serves any sentence imposed.

After the entry of the guilty pleas, the Indictment was redacted to include only those counts against Nguyen.

[2] Nguyen submitted the following in support of the pretrial motions: Brief in Support of Defendants' Pretrial Motions ("Moving Brief"); letter, dated 25 March 1992, regarding bifurcated trial and suppression of tape recording (the "25 March 1992 Letter Brief"); letter brief, dated 2 April 1992, in support of admission of expert testimony (the "2 April 1992 Letter Brief").

The Government submitted the following in opposition to the pretrial motions: Government's Response to the Defendants' Pretrial Motions (the "Opp. Brief"); letter brief, dated 8 April 1992, in opposition to 2 April 1992 Letter Brief (the "8 April 1992 Letter Opp.").

Nguyen did not file a reply brief to either the Opp. Brief or the 8 April 1992 Letter Opp.

A pretrial hearing was held on 8 May 1992 (the "8 May Hearing Tr."), 13 May 1992 (the "13 May Hearing Tr.") and 14 May 1992 (the "14 May Hearing").

Nguyen also filed motions for the suppression of references to the Asian street gang name Born to Kill, references to an arrest for robbery on 15 April 1991 and a tape recording of a conversation on 4 June 1991, which referred to the 15 April 1991 arrest, for a ruling to compel the Government to preserve all notes, drafts and final reports, for the production of all exculpatory evidence and impeachment evidence pursuant to Brady v. Maryland, 373 U.S. 83, 83 S. Ct. 1194, 10 L. Ed. 2d 215 (1963) and Giglio v. United States, 405 U.S. 150, 92 S. Ct. 763, 31 L. Ed. 2d 104 (1972), and a request by Nguyen to file additional motions if necessary. Because of pretrial stipulations and representations of the parties, these motions were withdrawn.

[3] Nguyen made this representation one day before the trial commenced.

[4] Nguyen was previously convicted on two separate robberies. 13 May Hearing Tr. at 197.

[5] The Hudson County Prosecutor's Office special investigation team set up surveillance of the Park. 13 May Hearing Tr. at 173. Lieutenant George Domanski ("Lt. Domanski") of the Hudson County Prosecutor's Office was one of the persons on the surveillance team. Id. He stated he observed the events from a distance; he was not close enough to identify the seller of the Weapon. Id. at 174-75.

[6] Identifiers such as name, social security number and aliases were recorded on the back of the photographs. 13 May Hearing Tr. at 179. Only the arrest number, if any, appeared on the front of the photograph. Id. at 180.

[7] The photographs were arranged in a book kept by the investigation unit investigating Asian street gangs. 8 May Hearing Tr. at 28. Peterkin stated he had not looked through the book prior to 2 February 1991. Id. at 29. He testified that on prior occasions, he had merely looked at individual pictures. Id. at 29-30. Lt. Domanski testified that he recalled seeing Peterkin look through the book prior to 2 February 1991. 13 May Hearing Tr. at 179.

[8] Officer Wisniewski did not recall if Peterkin was going for a lineup or a one-on-one identification. 13 May Hearing Tr. at 186.

[9] Officer Wisniewski did not recall seeing other Asian-Americans in the corridor. 13 May Hearing Tr. at 188.

[10] Nguyen's argument — that the introduction of the prior armed robbery convictions would prejudice the jury on Counts One and Two, conspiracy to sell weapons and the sale of the Weapon —might also have been made as a motion for severance pursuant to Fed.R.Crim.P. 14. For the reasons set forth below, however, the motion would have been unsuccessful.

Rule 14 of the Federal Rules of Criminal Procedure provides: "If it appears that a defendant is prejudiced by a joinder of offenses or of defendants in an indictment ..., the court may order ... separate trials of counts, grant a severance of defendants or provide whatever other relief justice requires." Fed.R.Crim.P. 14. When deciding a motion for severance, Rule 8 of the Federal Rules of Criminal Procedure merits consideration. Rule 8 permits joinder of offenses and defendants. Joinder of offenses and defendants promotes economy of judicial and prosecutorial resources, as well as the public interest in avoiding expensive and duplicative trials. United States v. Lane, 474 U.S. 438, 449, 106 S. Ct. 725, 732, 88 L. Ed. 2d 814, reh'g denied, 475 U.S. 1104, 106 S. Ct. 1507, 89 L. Ed. 2d 907 (1986); United States v. Gorecki, 813 F.2d 40, 42 (3d Cir.1987); United States v. Jackson, 649 F.2d 967, 973 (3d Cir.), cert. denied, 454 U.S. 1034, 102 S. Ct. 574, 70 L. Ed. 2d 479 (1981).

When distinct offenses have both a logical and temporal relationship, joinder permits the Government to present its evidence in an efficient manner. Such evidentiary overlap "strongly counsels in favor of joinder." United States v. McDonnell, 699 F. Supp. 1348, 1351 (N.D.Ill.1988) (citing United States v. Shue, 766 F.2d 1122, 1134 (7th Cir.1985), cert. denied, 484 U.S. 956, 108 S. Ct. 351, 98 L. Ed. 2d 376 (1987)). This rationale applies with equal force to the joinder of defendants. United States v. Dickens, 695 F.2d 765, 778-79 (3d Cir.1982), cert. denied sub nom, 460 U.S. 1092, 103 S. Ct. 1792, 76 L. Ed. 2d 359 (1983).

When, as here, the literal requirements of Rule 8 are met, a presumption arises in favor of joinder. See United States v. Swift, 809 F.2d 320, 322 (6th Cir.1987) (Rule 8(b) "can, and should, be `broadly construed in favor of initial joinder'....") (quoting United States v. Isaacs, 493 F.2d 1124, 1158 (7th Cir.), cert. denied, 417 U.S. 976, 94 S. Ct. 3184, 41 L. Ed. 2d 1146 (1974)). Indeed, there is a presumption against severance because it is "assum[ed] that closely related charges are being tried together...." United States v. Velasquez, 772 F.2d 1348, 1355-56 (7th Cir.1985), cert. denied, 475 U.S. 1021, 106 S. Ct. 1211, 89 L. Ed. 2d 323 (1986); see also United States v. Serubo, 604 F.2d 807, 819 (3d Cir.1979).

If, however, offenses or defendants have been improperly joined, severance is required as a matter of law under Rule 8. United States v. Andrews, 765 F.2d 1491, 1496 (11th Cir.1985) ("Misjoinder under Rule 8(b) is prejudicial per se...."), cert. denied sub nom., 474 U.S. 1064, 106 S. Ct. 815, 88 L. Ed. 2d 789 (1986); United States v. Bledsoe, 674 F.2d 647, 654 (8th Cir.) ("[M]isjoinder of defendants is inherently prejudicial."), cert. denied sub nom., 459 U.S. 1040, 103 S. Ct. 456, 74 L. Ed. 2d 608 (1982); United States v. Vastola, 670 F. Supp. 1244, 1261 (D.N.J. 1987) (citing United States v. Somers, 496 F.2d 723, 729 (3d Cir.), cert. denied, 419 U.S. 832, 95 S. Ct. 56, 42 L. Ed. 2d 58 (1974)), aff'd in part, rev'd in part, 899 F.2d 211 (3rd Cir.1990).

When a number of offenses are joined in one indictment or multiple defendants are jointly charged with a single offense, "[s]ome prejudice almost necessarily results." Vastola, 670 F.Supp. at 1261 (quoting Cupo v. United States, 359 F.2d 990, 993 (D.C.Cir.1966), cert. denied, 385 U.S. 1013, 87 S. Ct. 723, 17 L. Ed. 2d 549 (1967)). This level of prejudice, however, is permissible so long as the technical strictures of Rule 8 are met. Thus, Rule 8(b) is generally satisfied if the indictment charges a single conspiracy. United States v. Cole, 717 F. Supp. 309, 317 (E.D.Pa.1989), aff'd, 958 F.2d 365 (3d Cir. 1992); United States v. Di Pasquale, 561 F. Supp. 1338, 1347 (E.D.Pa.1983), aff'd, 740 F.2d 1282 (3d Cir.1984), cert. denied, 469 U.S. 1228, 105 S. Ct. 1226, 84 L. Ed. 2d 364 (1985).

Severance under Rule 14 is within the discretion of the trial court. United States v. Eufrasio, 935 F.2d 553, 568 (3d Cir.), cert. denied sub nom, ___ U.S. ___, 112 S. Ct. 340, 116 L. Ed. 2d 280 (1991); United States v. Boyd, 595 F.2d 120, 125 (3d Cir.1978). Rule 14 authorizes a trial court to sever counts or defendants where, despite an indictment's technical compliance with Rule 8, joinder would result in a "manifestly unfair trial." Vastola, 670 F.Supp. at 1261 (citing United States v. Reicherter, 647 F.2d 397, 400 (3d Cir.1981)). A defendant bears a heavy burden when he or she moves for severance under Rule 14. See Eufrasio, 935 F.2d at 568; United States v. Sandini, 888 F.2d 300, 305 (3d Cir. 1989), cert. denied, 494 U.S. 1089, 110 S. Ct. 1831, 108 L. Ed. 2d 959 (1990); United States v. De Peri, 778 F.2d 963, 983 (3d Cir.1985), cert. denied sub nom., 475 U.S. 1110, 106 S. Ct. 1518, 89 L. Ed. 2d 916 and 476 U.S. 1159, 106 S. Ct. 2277, 90 L. Ed. 2d 720 (1986); United States v. Di Pasquale, 740 F.2d 1282, 1293 (3d Cir.1984), cert. denied, 469 U.S. 1228, 105 S. Ct. 1226, 84 L. Ed. 2d 364 (1985).

Mere allegations of prejudice are insufficient to meet this burden. A defendant "must demonstrate `clear and substantial prejudice.'" Gorecki, 813 F.2d at 43 (quoting United States v. Sebetich, 776 F.2d 412, 427 (3d Cir.1985), cert. denied, 484 U.S. 1017, 108 S. Ct. 725, 98 L. Ed. 2d 673 (1988)); United States v. Wright-Barker, 784 F.2d 161, 175 (3d Cir.1986). "`[N]either a disparity in evidence nor the introduction of evidence more damaging to one defendant than another entitles the seemingly less culpable defendant to severance.'" Sebetich, 776 F.2d at 427 (quoting United States v. Simmons, 679 F.2d 1042 (3d Cir.1982)).

In determining whether to sever, the "court should balance the public interest in joint trials against the possibility of prejudice inherent in the joinder of defendants." Eufrasio, 935 F.2d at 568 (citing De Peri, 778 F.2d at 984). Joint trials "conserve [public] funds, diminish inconvenience to witnesses and public authorities, and avoid delays in bringing those accused of crime to trial." Bruton v. United States, 391 U.S. 123, 143, 88 S. Ct. 1620, 1631-32, 20 L. Ed. 2d 476 (1968). Moreover, "[j]oint trials generally serve the interest of justice by avoiding inconsistent verdicts and ... advantages which sometimes operate to the defendant's benefit." Richardson v. Marsh, 481 U.S. 200, 210, 107 S. Ct. 1702, 1708-09, 95 L. Ed. 2d 176 (1987).

A motion for severance in this case would have been unsuccessful for many of the same reasons the motion for bifurcation was denied. The Indictment charged Nguyen in three separate counts for conspiracy to sell firearms, the sale of the Weapon without the required license and possession of the Weapon while being a convicted felon. All three counts of the Indictment were centered around the same transaction. With the exception of the prior felony conviction, all of the evidence presented was relevant to each count. To sever the trial would have resulted in a duplicative trial and waste of judicial resources. Significantly, it appears Nguyen would not have been able to show clear and substantial prejudice would have resulted from trying Counts One and Two with Count Three of the Indictment. As discussed below, proper limiting instructions were given with respect to the prior felony conviction. Additionally, as a result of an agreement between the Government and defense, the substance of the prior felony convictions was not introduced into evidence; the jury was only aware that Nguyen was a felon at the time of the sale of the Weapon.

[11] Penrod has a BA. in political science from Yale College, a J.D. from Harvard Law School and a Ph.D. in psychology from Harvard University. 8 May Hearing Tr. at 41. Penrod is currently on the law school staff at University of Wisconsin. Id. at 82.

[12] As previously indicated, Nguyen did not call Penrod at trial and a ruling was not made on this motion. If Nguyen had called Penrod, the motion to introduce expert testimony on eyewitness identification would have been denied.

[13] Penrod acknowledged, however, that there is no agreement as to whether expert psychologist testimony is helpful to the jury. 8 May Hearing Tr. at 77-82.

[14] Rule 702 provides:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.

Fed.R.Evid. 702. In addition, Rule 703 provides:

The facts or data ... upon which an expert bases an opinion ... may be those perceived by or made known to the expert at or before the hearing. If of a type reasonably relied upon by experts in the particular filed in forming opinions ... upon the subject, the facts or data need not be admissible in evidence.

Fed.R.Evid. 703. "Rule 703 was designed to broaden and liberalize the permissible bases for expert testimony." DeLuca v. Merrell Dow Pharmaceuticals, Inc., 911 F.2d 941, 952-53 (3d Cir.1990). The party seeking to introduce expert testimony must demonstrate that the "expert's testimony is based on the type of data a reasonable expert in the field would use in rendering an opinion on the subject at issue." Id. at 953.

Despite the liberal breadth of Rule 703, the Third Circuit strictly imposes the foundational requirements of the Rules of Evidence. See e.g., United States v. Pelullo, No. 91-1792, Slip. op. at 8-29, 964 F.2d 193, 199-207 (3d Cir. 12 May 1992) (court reversed admission of business records because "some indicia of trustworthiness" insufficient to satisfy foundational requirement for business records and parties failed to satisfy notice requirement to invoke residual hearsay exception).

[15] He stated additional studies have been done in which confidence was a subsidiary issue to the study. 8 May Hearing Tr. at 44.

[16] Penrod was not able to testify from recollection about an example of a study he conducted until a recess was taken in order for him to refresh his memory. 8 May Hearing Tr. at 113-16.

[17] Penrod testified as follows:

Q In a real life situation, someone who takes the stand and is actually under the potential penalty of perjury, that's a real thing that you know that they have to consider that, correct?

A Yes.

Q But you can't tell us that the person that you're asking, you know, would you be confident to sign a sworn statement is really measuring it against that standard?

A I cannot tell you that. That's not a statement that we try to make or that I would pretend to make on the basis of defining this particular study.

8 May Hearing Tr. at 125.

[18] Field studies conducted by Penrod involved a situation where subjects went into a convenience store and purchased items with travelers' checks. The students showed a piece of identification and called attention to the photograph on the identification. 8 May Hearing Tr. at 46. A person posing as a law clerk went to the convenience store, either two or twenty-four hours later, to obtain an identification from the sales person. The law clerk would also ask the sales person how confident he or she was as to the accuracy of the identification. Id.

[19] The following is excerpted from the six page transcription of Penrod's explanation.

[T]his just illustrates what these numbers mean.

We had 10,000 people and knew in advance half were incorrect and we found — we measured their confidence, we divided them into two equal groups based on confidence, 5,000 high, 5,000 low.

The question is how much classification accuracy could we attain with that knowledge. If we knew everything, if confidence and accuracy were correlated 100 percent, we put 5,000 people up here and 5,000 people down here. Does that make sense?

. . . . .

I'm going to assign them on that basis and the question is how many times will I be right?

. . . . .

[T]he question ... could be anchored by very confident, not at all and nine point scale. What's the probability you have made a correct identification, clearly zero to 100 percent.

. . . . .

[T]he measurement scale depends on the question that you ask.

8 May Hearing Tr. at 118-122.

[20] Penrod described the cognitive process as one involving four stages: perception which involves a person's ability to focus on the subject, encoding which involves the ability to process the image, storage which is impacted by the time between the entry of the image and the retrieval of the image, and retrieval which considers whether there were cues to enhance the ability to retrieve information such as a birth-mark or the first time of seeing or doing something. Id. at 92-94.

[21] Penrod stated, at this point, that he could not recollect whether knowledge of the need to make a subsequent identification would have an impact on accuracy. 8 May Hearing Tr. at 100. However, later he conceded it would have an impact on accuracy. Id. at 128-29.

[22] Penrod could not recall whether this study used real life witnesses. 8 May Hearing Tr. at 139.

[23] Penrod explained a "Meta-analysis is a standard technique for combining the results of independent studies in order to arrive at a general conclusion, in order to figure out what the general implications of a body of research, what general are [sic] and also to determine the reliability of those findings." 13 May Hearing Tr. at 285. He stated Meta analysis is "an accepted method of analysis." Id. at 288. However, aside from his "assertion as to the accuracy of his conclusions," Downing 753 F.2d at 1239, Penrod did not offer any data upon which he relied for this, or indeed, other conclusions. Id.

[24] Out of the fifty studies conducted in this area only one review and one study has concluded there is no correlation between confidence and accuracy.

[25] The Government conceded under Stevens when an eyewitness asserts he is confident that his or her identification is correct, there is a fit with the confidence/accuracy studies. 935 F.2d at 1400.

[26] Penrod, with Professor Shapiro, conducted a Meta-analysis on eighteen studies where they looked at the "magnitude of the effect of the cross racial on identification accuracy." 8 May Hearing Tr. at 141.

Penrod did not explain what expertise he has in reading and summarizing studies that the jury would not have if given an opportunity to read the studies.

[27] Penrod stated the differential between generally being able to identify someone of the same race and someone of another race could be explained in one of two ways. 13 May Hearing Tr. at 233. He explained:

One would be to look at the relative magnitude of the effects, that is the differences for whatever group you're interested in, whites on blacks, blacks on whites, so on and there are several different measures that are used to gauge that correlation coefficient that I mentioned last week.

. . . . .

You can also ... look at the average correct identification rates and false alarm rates of cross studies that report them in percentages and compute mean percent for the difference between same versus cross race.

Id.

[28] Penrod agreed photographs of African-Americans may be more difficult to view than photographs of caucasians because of the reflection of light. 8 May Hearing Tr. at 146.

[29] The conclusion that the studies do not fit the facts of the case is not based on the difference of simulated studies and real life crimes, but rather is based on the fact that the circumstances in this case rise to the situation that is not studied because it is too easy a case. Because of the magnitude of positive factors in this case, the cross-racial impairment is sufficiently reduced so as to have little impact, if any, on accuracy.

[30] Penrod stated around twenty to twenty-five studies have been conducted in this area. 13 May Hearing Tr. at 248.

[31] None of these studies involved the level of stress an undercover agent is under when buying firearms. 13 May Hearing at 252.

[32] It is unclear how many studies have been conducted in this field.

[33] The relevant colloquy was as follows:

Q I'm asking you to quantify for us, if you can, whether or not you believe there is a difference between a situation where the event lasted between three and five minutes contrasting that situation to a situation where the event lasted three minutes and ten seconds to five minutes and ten seconds?

Do you see or would you expect to see a significant difference in those two situation in a person's ability to subsequently make an accurate identification?

A Ten seconds on top of three minutes is going to be fairly impact [sic] as compared to three seconds on top of one minute.

Q Wouldn't it also be fair to conclude then that in an event that lasted between three and five minutes, the exposure of a gun for a period of five to ten seconds is going to have a minimal, if any, effect on the person's subsequent ability to make an identification?

A That's what I said. Much less of an effect than if it was visible for the entire period of time.

Q You are still willing to conclude that given that kind of situation, you would expect to see a weapon focus phenomenon. Is that correct?

A I would expect that there would be some impairment, but it would be minor and it would probably require a large sample size in order to detect the impairment.

Q Which you haven't done.

A No.

Q And which you do not know exists?

A To the best of my knowledge, that type of study does not exist.

13 May Hearing Tr. at 269.

[34] Again, Penrod did not explain what expertise he has in reading or summarizing these studies that the jury would not have if given an opportunity to read the studies.

[35] Penrod did not explain what expertise he has in reading and summarizing these studies that the jury would not have if given an opportunity to read the studies.

[36] He did not state whether these were re-enacted, videotaped or real crimes.

[37] Penrod did not explain what expertise he has in reading and summarizing these studies that the jury would not have if given the opportunity to read the studies.