Opinion of the Court
SEITZ, Circuit Judge.The government appeals a pretrial order of the district court granting defendant Alan Grecco’s (“Grecco”) motion to suppress certain prospective eyewitness trial testimony of Frank Galimi (“Galimi”), that would identify Grecco as being present at the murder scene.1
On July 20, 1989, defendant Grecco and seven other defendants were indicted for alleged violations of the Racketeer Influenced and Corrupt Organizations Act (“RICO”). Count Two of the indictment enumerated the alleged racketeering acts, and among them was one charging Grecco and another defendant with conspiracy to murder Vincent Mistretta (“Mistretta”). On September 4, 1990, as part of its order disposing of numerous pretrial motions, the district court granted Grecco’s motion to suppress certain of Galimi’s prospective trial testimony 746 F.Supp. 432. The government appeals that order.
The district court had jurisdiction over this matter under 18 U.S.C. § 3231 (1988). This court has jurisdiction pursuant to 18 U.S.C. § 3731 (1988).
I. FACTUAL BACKGROUND
The facts surrounding the Mistretta murder, as they have been memorialized in various police reports, are as follows.
On the night of April 26, 1979, Mistretta was stabbed to death. At the time of the murder, Galimi, his roommate, was sitting in their second floor apartment watching television. Responding to a scream, Galimi *493went to the window of the apartment overlooking the street and saw “a few guys standing over somebody punching him.” Galimi yelled out the window, and the assailants backed away from the victim. Gal-imi, recognizing that it was Mistretta being beaten, went down to the street and began to chase the assailants. The assailants, who were at least a half block ahead of Galimi, jumped into a ear parked at the end of the block and drove away. Galimi then returned to Mistretta, and eventually realized that Mistretta had been stabbed.
An incident report, filed by Sergeant Puzio of the Garfield, New Jersey police on April 27, 1979, recounts the following additional details surrounding the evening of April 26, 1979. Puzio and two other police officers responded to the dispatch directing them to the scene of the crime. When he arrived he found Galimi who “appeared very upset and visibly shaken.” Puzio also reported as follows:
I approached Frank and asked him, “Frank did you see who did it? Who did it?” He yelled “Wallsnak” [sic] I asked did you see Wallsnak do it. He mumbled something. I then asked “What is Walls-nak first name?” He said Al. “Again I asked who did it.” Did you see Wallsnak do it. Putting his hands to his head, shouting Vinnie yelled out “Wallsnak, I heard him say Wallsnak.”
“Wolshonak” is the name by which Galimi and the deceased, Mistretta, knew the defendant Grecco. Puzio’s report states that Galimi was placed in an ambulance and taken to the hospital.
The district court record includes the report of Detective R. McGill who investigated the Mistretta murder. That report, dated April 27, 1979, gives the following version of Galimi’s description of the events on the evening of April 26, 1979:
He [Galimi] then heard the victim scream, “FRANK GALIMI help me!” GALIMI ran to the window, looked out, and could see two males dressed in brown leather jackets hovering over and punching someone on the stairs in front of his apartment. He yelled to the two men ... and then ran down the stairs. During this time he saw the two assailants running up the street get in a vehicle, and drive away, fleeing the scene. GALIMI states at this time that he did not get a good look at the assailants ... GALIMI asked the victim, “Who did it to you?” Mistretta stated, “Frank, Al Wol-shonak did it.”
McGill also reported that:
Galimi was interviewed at length at the Garfield Police Department and indicated that he feared for. his life. He stated that if he assisted us in this investigation by naming names, his life would be in jeopardy. He did, however, repeat the conversation between himself and MIS-TRETTA when MISTRETTA told him that AL WOLSHONAK has assaulted him.... He describes the assailants as being the same height as AL WOLSHO-NAK, however, he did not affirm or deny that AL WOLSHONAK was one of the assailants. FRANK GALIMI appeared in this office on the morning of April 27, 1979 and again appeared reluctant to discuss what he had witnessed. GALIMI was subsequently taken to DR. WAGLE ... to have GALIMI hypnotized to obtain additional information.
Further, the record contains the report of Investigator Alan Grieco filed April 30, 1979. Grieeo’s report states that on April 27, 1979, he accompanied Galimi and Gali-mi’s father to the offices of Dr. Wagle, a hypnotist, in Teaneck N.J. The purpose of this visit, the report states, was to place Galimi under hypnosis in order to gain further information about his observations on the night of the murder.
Grieco’s report of Galimi’s description of the incident, as recounted at the session with Dr. Wagle, was consistent with those in the two other police reports. It also included the following statement:
“He [Galimi] stated that he did not get a good look at their faces only the back of their heads. He was particularly drawn to the stature of their bodies and stated T have seen these guys before.’ He stated that he was almost certain that one of the assailants he observed was AL WOLSHONAK.”
*494On May 3, 1979, Galimi was interviewed by Investigator John Scioli and Detective McGill. Scioli’s report, filed on May 18, 1979, does not state that Galimi was able to positively identify either assailant.
The record contains Galimi’s grand jury testimony about the Mistretta murder. That testimony was given on March 31, 1988, nearly nine years after the incident, and it repeats the story Galimi told police in April 1979. The transcript reflects the following exchange:
Q. Now, you say from the look you had, you weren’t able to positively identify the people. Is that correct?
[Galimi]. If I said I was 100 percent sure, I’d be lying.
Q. Do you have a feeling as to who one of these people was based on your observation?
[Galimi]. I felt I knew both of them. Q. And who do you feel they were? [Galimi], One was Al Wolshonak. The other guy — I don’t know his name. I knew his face.
Grecco filed a motion to suppress Gali-mi’s prospective trial testimony about the murder. Only three witnesses testified at the suppression hearings, Dr. Orne, Dr. Conte and Galimi.
Dr. Orne testified as an expert on hypnosis. On the basis of an interview with Galimi in June 1990, a telephone conversation with Dr. Wagle, and a review of the police records, Dr. Orne opined that Galimi was not hypnotized by Dr. Wagle at his session on April 27, 1979. Dr. Conte, the physician who treated Galimi on the evening of the murder, testified that he was called to the hospital because the emergency room personnel could not bring Galimi under control. He stated that he could not recall the content of his discussion with Galimi, but that it took him approximately fifteen to twenty minutes to calm Galimi down.
At the suppression hearing, Galimi stated that he saw the two assailants’ faces and that he recognized one of them to be the defendant Grecco. Galimi further testified that while at the hospital on the evening of the murder, he told Dr. Conte that he saw Grecco attack Mistretta, and that Dr. Conte responded by telling him: “Keep your mouth shut. A1 will kill you.” Galimi stated that he took Dr. Conte’s advice, and that this explains his refusal to identify Grecco when the police interrogated him years earlier.
At the suppression hearing, Galimi described the police interrogation on April 27, 1979. Galimi stated that he believed that the interrogation lasted about 6.5 hours. He testified that during the interrogation, the police put Grecco’s picture on the wall about six inches away from his head and “kept pointing at it [and said:] He did it, didn’t he? Tell us he did it. Tell them.” Galimi said the police told him that they could not protect him, so he “clammed up.”
Galimi further testified that on April 27, 1979, the police had requested that he submit to a lie detector test, but that he refused because he could not beat it. Galimi stated that he accepted the police request to submit to hypnosis because he thought his prior experience with hypnosis would enable him to fake being hypnotized and convince the police that he was not an eyewitness.2
At the suppression hearing, Galimi explained his present willingness to testify as an eyewitness. He stated that after he appeared before the grand jury but before Grecco’s indictment, a news report surfaced stating that there was an eyewitness to the Mistretta murder. Galimi said that Grecco would know that he was the only possible eyewitness to the murder thus, believing that his life was in danger, he immediately went to the police to seek protection. He was then placed in the government’s witness protection program.
*495II. THE LEGAL PROCEEDINGS
At the suppression hearing, Grecco asserted that admitting Galimi’s prospective trial testimony would violate his rights under the confrontation clause and the due process clause. In support of his assertions, Grecco contended as he does on appeal that Galimi’s testimony was based on hypnotic suggestion, not personal knowledge. Thus, Grecco argued, admitting Gal-imi’s prospective trial testimony would deprive him of his right to confront adverse witnesses and his right to a fair trial.
The district court first concluded that the government had not carried its burden to show that Galimi was not hypnotized. It then stated that it was unable to anticipate the extent of cross-examination available to Grecco at trial, and thus turned to his due process challenge. The district court concluded that Galimi’s prospective trial testimony was so unreliable that its admission at trial would violate due process. It, therefore, barred Galimi’s prospective trial testimony. In view of its due process determination, the district court concluded that it need not decide the merits of Grec-co’s confrontation clause challenge.
The government moved for reconsideration or clarification of that portion of the district court’s order suppressing Galimi’s testimony. After denying the motion for reconsideration, the district court issued an order suppressing:
[T]he testimony of Frank Galimi regarding the events surrounding the murder of Vincent Mistretta to the extent that it is inconsistent with information memorialized in writing or other fixed form pri- or to Galimi’s hypnosis session with Dr. Wagle on or about April 27, 1979.
The order does not bar Galimi from testifying “to those things he told the police or others which were memorialized in writing or other fixed form prior to the hypnosis session.”
The government filed a timely notice of appeal.
III. DISCUSSION
The government urges that the district court erred when it determined that Gali-mi’s prospective trial testimony, particularly his identification of Grecco, would violate Galimi’s due process right to a fair trial. The government contends that Gali-mi’s prospective trial testimony was sufficiently reliable for a jury to weigh.
A. Due Process
In analyzing Grecco’s due process challenge, the district court first concluded that the government had not carried its burden to show that Galimi was never hypnotized, and then turned to whether Galimi’s “posthypnotic testimony” was barred because it was so unreliable that its admission would result in an unfair trial. Because Grecco contended that Galimi’s identification resulted in part from a coercive six and one half hour police interrogation, the district court considered the suggestiveness of both that interrogation and the hypnotic session that occurred the next day. After reviewing the totality of the circumstances, the court stated that it could not “conclude that the strength of the independent basis for identification overcomes the suggestiveness of the initial interrogation and the potential suggestiveness of the hypnosis session.” Consequently, believing that “the hypnotic session was impermissibly suggestive when coupled with the suggestiveness of the interrogation,” it determined that admitting Galimi’s prospective trial testimony, “would violate Grecco’s due process right to a fair trial.”
The government contends that two of the district court's factual findings were erroneous, and that in any event it erred by concluding that the independent basis for Galimi’s prospective trial testimony did not overcome the suggestiveness of the six and one half hour police interrogation and the hypnotic session with Dr. Wagle. We first address the government’s contentions of factual error as they may bear on our review of the district court’s ultimate conclusion that the police procedures were im-permissibly suggestive. Then we turn to the government’s contention that the independent basis for Galimi prospective trial *496testimony overcame the suggestiveness of the police procedures.
1. The Contested Factual Findings
The government asserts that the district court made two factual findings that were clearly erroneous: (1) that Galimi was hypnotized; and (2) that the hypnotism affected Galimi’s recollection of the events surrounding the Mistretta murder.
The district court did not make an explicit finding that Galimi was hypnotized. What it did was determine that the government had not carried its burden of showing that Galimi was not hypnotized. We believe the court treated the government’s failure as the equivalent of a finding that Galimi was in fact hypnotized. Although the government argues to the contrary, we shall assume for purposes of this appeal that the court’s finding is not clearly erroneous. We indulge in this assumption because we are satisfied that under the present circumstances it is not incompatible with our resolution of the government’s two other contentions of error.3
The government’s second contention of factual error is the district court’s finding that the hypnotism had an effect on Galimi’s recollection of the Mistretta murder. No such factual finding is explicit in the district court’s opinion. The district court did, however, conclude that Galimi’s prospective identification testimony was “the product of suggestive police procedures_” Although the phrase “suggestive police procedures” refers to both the initial interrogation and the hypnotic session, we believe that a fair reading of the district court’s opinion supports the government’s contention that the district court made a factual finding as to the possible effects of hypnosis. Thus, we shall review that finding under the clearly erroneous standard. Reversal is warranted only if our “review of the record leaves us with ‘the definite and firm conviction that a mistake has been committed.’ ” United States v. McMillen, 917 F.2d 773, 776 (3d Cir.1990) (quoting United States v. United States Gypsum, 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1948)).
The district court recounted Galimi’s identification statements as follows:
Before the session Galimi told the police that he recognized the assailants but he could not identify them. He told police that Mistretta’s dying words were “A1 Wolshonak did it.” ... During the session, Galimi stated that he could not be certain, but he thought the assailant was Wolshonak. Testifying before the court [on July 30, 1990] Galimi stated that he was absolutely certain that Wolshonak was the assailant.
These three statements, the district court believed, showed that “Galimi’s increasing certainty in the correctness of identification is consistent with hypnosis and suggestion.” Thus, it concluded that it could “not ignore that the exact phenomenon, that is memory hardening, that makes post-hypnotic testimony so suspicious is consistent with events here.”
Even if we were to agree with the district court that Galimi’s certainty increased, that would not render Galimi’s prospective trial testimony constitutionally infirm. Cf. Chaussard v. Fulcomer, 816 F.2d 925, 929 (3d Cir.) (“But to say that hypnosis may have increased the prosecu-trix’s certainty that her subsequent identification of Chaussard was accurate does not necessarily imply that the testimony violated the confrontation clause.”), cert. denied, 484 U.S. 845, 108 S.Ct. 139, 98 L.Ed.2d 96 (1987); Clay v. Vose, 771 F.2d 1, 4 (1st Cir.1985), cert denied, 475 U.S. 1022, 106 S.Ct. 1212, 89 L.Ed.2d 324 (1986); United States v. Harrelson, 754 F.2d 1153, 1180 (5th Cir.1985). More importantly, although the relevant post-hypnotic change in Galimi’s account of the Mistretta murder might be logically consistent with the effect of memory hardening, the change in Galimi’s account of the incident cannot be *497understood as simply a change in his degree of certainty.4
At the suppression hearing Galimi testified that at the time of the incident he recognized Grecco as one of the assailants. At the six and one half hour police interrogation, he would neither affirm nor deny recognizing either of the assailants. During the hypnotic session, he stated that he was “almost certain” that one of the assailants was Grecco. Thus, the relevant change is that at the suppression hearing Galimi testified that he was absolutely certain at the time of the incident that he recognized Grecco as one of the assailants. Consequently, under the circumstances of this case, the fundamental factual issue is whether hypnotic suggestion caused that substantive change.
Grecco did not introduce any evidence in support of the proposition that hypnotic suggestion caused the substantive change in Galimi’s post-hypnotic account of the events. See Beachum v. Tansy, 903 F.2d 1321, 1326 (10th Cir.) (defendant has burden to show that hypnosis caused a change in the witness’ post-hypnotic testimony), cert. denied, — U.S.-, 111 S.Ct. 269, 112 L.Ed.2d 225 (1990). Rather, he only adduced evidence supporting his claim that the hypnotism by Dr. Wagle was suggestive. But even if we were to assume that Grecco established that the hypnotism was suggestive, that does not prove that the hypnotism caused the relevant substantive change in Galimi’s account of the events.
The government concedes that certain suggestive elements surrounded the hypnotic session. It nonetheless contends that the evidence shows that the hypnosis did not affect Galimi’s recollection. Specifically, the government argues that the expert testimony and a full and accurate chronology of the statements made by Galimi show that hypnotic suggestion was not responsible for Galimi testifying at the suppression hearing that at the time of the incident he recognized Grecco as one of the assailants.
Dr. Orne, a preeminent expert on hypnosis, testified that hypnotic suggestion is most effective immediately after hypnosis, and its effectiveness decreases significantly over time. More importantly, Dr. Orne opined that it was “just almost impossible” for hypnosis to produce a suggestive effect ten years later. Therefore, if the hypnosis were suggestive, Galimi would normally have changed his account of the events and stated that he recognized Grecco as one of the assailants at the session with Dr. Wa-gle or shortly thereafter. He did not. As the record clearly shows, Galimi offered substantially the same account of the Mis-tretta murder to Investigator Scioli on May 3, 1979, and to the grand jury on March 31, 1988, as he did to the hypnotist on April 27, 1979. Thus, the government’s unrebutted testimony is inconsistent with a finding that the hypnotism had an effect on Gali-mi’s account of the Mistretta murder.
Furthermore, our independent review of the record reveals nothing to support the district court’s finding that the hypnotic session affected Galimi’s recollection. In addition, the brute fact is that between the time of the hypnotic session and his appearance before the grand jury, Galimi’s certainty did not increase. Therefore, the district court’s characterization of the evolution in Galimi’s statements is incompatible with the fact that for nearly nine years Galimi did not experience memory hardening.
We believe, therefore, that the government’s contention that hypnosis did not, in fact, affect Galimi’s post-hypnotic testimony is uncontradicted. Moreover, our review of the entire record has left us with “the definite and firm conviction that a mistake has been committed.” United States Gypsum, 333 U.S. at 395, 68 S.Ct. at 542. Consequently, we think that the district court committed clear error in finding that hypnosis affected Galimi’s recollection of the event’s surrounding the Mistretta murder. See McMillen, 917 F.2d at 775-76.
*4982. The Independent Basis
As we have already stated, the district court did not rest its due process conclusion solely upon the suggestiveness of the hypnosis. Rather, the district court concluded that the initial six and one half hour police interrogation coupled with the potential suggestiveness of the hypnotic session was impermissibly suggestive. Thus, contrary to the government’s assertion, our conclusion that the hypnotism did not affect Galimi’s recollection does not justify reversing the district court’s suppression order. We turn, therefore, to the government’s contention that the district court erred when it determined that, under the totality of the circumstances, the suggestiveness of the police procedures overcame the independent basis for Galimi’s prospective trial testimony.
In support of its position, the government asserts that the district court erred by (1) attaching significance to the absence of a recording of the hypnotic session with Dr. Wagle; and (2) overlooking the substantial evidence establishing an independent basis for Galimi’s prospective trial testimony. Consequently, the government argues, the district court erred by concluding that the suggestiveness of the police procedures overcame the independent basis for Galimi’s prospective trial testimony.
The government challenges the district court’s assessment of the constitutional significance to be attached to the evidence. Thus, our review of the district court’s legal determination is plenary. Neil v. Biggers, 409 U.S. 188, 193 n. 3, 93 S.Ct. 375, 379 n. 3, 34 L.Ed.2d 401 (1972); Landano v. Rafferty, 856 F.2d 569, 571 (3d Cir.1988); see also United States v. Frank, 864 F.2d 992, 1001 (3d Cir.1988) (plenary review of district court’s application of a legal standard in denying suppression request), cert. denied, 490 U.S. 1095, 109 S.Ct. 2442, 104 L.Ed.2d 998 (1989).
With respect to the government’s first contention, we note that recordings are procedural safeguards used to determine whether a hypnotic session was impermissi-bly suggestive. They are not constitutionally required, see Chaussard, 816 F.2d at 930-31, and, as the government points out, New Jersey state courts did not require tape recordings of hypnotic sessions until 1981, two years after Galimi met with Dr. Wagle. See State v. Hurd, 86 N.J. 525, 432 A.2d 86 (1981). Thus, the failure of the government to produce a recording might have an entirely innocent explanation, and in any event that failure shows nothing about the suggestiveness of the hypnotic session itself.
The government’s second contention is that the district court overlooked the substantial evidence demonstrating an independent basis for Galimi’s prospective trial testimony. In assessing that basis, the district court followed Biggers, where the Supreme Court stated:
[T]he factors to be considered in evaluating the likelihood of misidentification include the 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 at the confrontation, and the length of time between the crime and the confrontation.
Biggers, 409 U.S. at 189, 93 S.Ct. at 377.
The district court went on to determine that Galimi had an opportunity to view the assailants, possessed a high degree of attention when he responded to the screams, and offered a description of the assailants that could include Grecco. It also found that Galimi unequivocally identified Grecco at the suppression hearing in July 1990, and that approximately eleven years had passed between the time of the incident and the identification. We emphasize the fact that Galimi knew Grecco for fifteen years prior to the incident because that factor supports admitting Galimi’s prospective trial testimony and is not undermined by hypnosis. Cf. United States v. Wade, 388 U.S. 218, 241 n. 33, 87 S.Ct. 1926, 1940 n. 33, 18 L.Ed.2d 1149 (1967) (noting that “how well the witness knows the suspect” is one of the factors having “an important *499bearing upon the true basis of the witness’ in-court identification”).5
We believe that these findings constitute substantial evidence demonstrating that Galimi’s prospective trial testimony, including his identification of Grecco as one of the assailants, rested on an independent basis. We next consider whether the suggestiveness of the police procedures overcame that independent basis, thus making Galimi’s prospective trial testimony so unreliable that its admission would violate due process. See Manson v. Brathwaite, 432 U.S. 98, 114, 97 S.Ct. 2243, 2253, 53 L.Ed.2d 140 (1977) (“reliability is the linchpin in determining the admissibility of identification testimony”).
We have determined that the government’s failure to provide a recording of the hypnotic session shows nothing about the suggestiveness of the hypnotic session with Dr. Wagle. Thus, we believe that the record contains little, if any, evidence showing that Galimi was subjected to hypnotic suggestion. This fact, and our determination that the district court erred by finding that hypnosis had an effect on Galimi’s recollection, demonstrate that the suggestiveness of the hypnotism was negligible for purposes of evaluating Grecco’s due process challenge. Consequently, the issue upon which that challenge must rest is whether the initial six and one half hour police interrogation was impermissibly suggestive, thus overcoming the independent basis for Galimi’s prospective trial testimony.
If the initial police interrogation were impermissibly suggestive, then at that interrogation, the hypnotic session or shortly thereafter Galimi would have been expected to change his account of the Mistretta murder, presumably stating that he recognized Grecco and identifying him as one of the assailants. The three police reports filed immediately before and after the hypnotic session, however, show that Galimi did not change his account of the Mistretta murder.
Detective McGill’s report of April 27, 1989 states that at the police interrogation Galimi “described the assailants as being the same height as AL WOLSHONAK, however, he did not affirm or deny that AL WOLSHONAK was one of the assailants.” Investigator Grieco’s report of April 30, 1979 states that at the hypnotic session Galimi said “that he did not get a good look at their faces only the back of their heads ... [and] that he was almost certain that one of the assailants he observed was AL WOLSHONAK.” Finally, Investigator Sci-oli’s report of May 18, 1979 repeats Gali-mi’s statement that Mistretta told him that Grecco was one of the assailants, and repeats Galimi’s description of the assailants’ appearance. Scioli’s does not report that in his interview with Galimi, Galimi stated that he recognized Grecco as one of the assailants.
Thus, there is no evidence in the record showing that the initial six and one half hour interrogation caused Galimi to change his account of the events and state that he recognized Grecco as one of the assailants. As we have already pointed out, that change did not occur until nearly ten years later. Thus, even if the initial six and one half hour interrogation were unnecessarily suggestive, we do not think that it overcame the independent basis for Galimi’s prospective trial testimony. See Manson, 432 U.S. at 112-14, 97 S.Ct. at 2252-53 (under totality approach, testimony can be sufficiently reliable despite an unnecessarily suggestive police procedure).
We conclude that the police procedures were not impermissibly suggestive. Furthermore, we do not think that the facts of this case support the conclusion that the *500initial interrogation, the hypnotism, or the hypnotism “coupled with” the interrogation “created a very substantial likelihood of misidentification.” United States v. Dowling, 855 F.2d 114, 117 (3d Cir.1988) (quotations omitted) (emphasis added), aff'd on other grounds, — U.S. -, 110 S.Ct. 668, 107 L.Ed.2d 708 (1990). Instead, we believe that the government introduced clear and convincing evidence showing that Galimi’s prospective trial testimony had an independent basis, and thus it was sufficiently reliable to withstand Grecco’s due process challenge. Therefore, the district court erred by barring the government from introducing that testimony, and thereby preventing the jury from according it appropriate weight. See Manson, 432 U.S. at 116, 97 S.Ct. at 2254 (“Juries are not so susceptible that they cannot measure intelligently the weight of identification testimony that has some questionable feature.”).
B. Prejudice
Grecco contends that “without regard to the constitutional claims,” the district court suppressed Galimi’s prospective trial testimony under Fed.R.Evid. 403.6
The government argues that the district court did not suppress Galimi’s prospective trial testimony under Rule 403. Alternatively, the government contends that the potential prejudicial impact of Galimi’s prospective trial testimony did not substantially outweigh its probative value.
We believe that a fair reading of the district court’s opinion supports Grecco’s contention that, as an alternative holding, the district court suppressed Galimi’s prospective trial testimony because of its undue prejudicial impact. Therefore, we address the government’s alternative contention that the district court could not properly suppress that testimony under Rule 403. We review the district court’s prejudice ruling for abuse of discretion. United States v. Driggs, 823 F.2d 52, 54 (3d Cir.1987); see also McQueeny v. Wilmington Trust Co., 779 F.2d 916, 922 (3d Cir.1985) (“substantial deference” is given to district court’s balancing of prejudice and probative value).
The district court rested its prejudice analysis on United States v. Valdez, 722 F.2d 1196 (5th Cir.1984). In that ease, the court stated that an identification may be more potentially prejudicial than probative when it is the result of identification procedures that are “unduly suggestive.” Id. at 1203.
Even if testimony resulting from unduly suggestive police procedures may cause the type of “unfair prejudice” that is cognizable under Rule 403, we do not believe that it was proper for the district court to suppress Galimi’s prospective trial testimony under Rule 403.
The probative value of an eyewitness identification is always substantial. Moreover, even if Galimi were hypnotized, the probative value of his identification could not be significantly discounted in light of the district court’s findings that Galimi had an opportunity to view the assailants, possessed a high degree of attention when he responded to Mistretta’s screams, offered a description of the assailants that could include Grecco, and had known Grecco for approximately fifteen years at the time of murder. Furthermore, we have already stated that the district court erred by concluding that hypnosis caused any change in Galimi’s recollection of the events.
Under these circumstances, we do not believe there would be anything unfairly prejudicial about permitting Galimi to testify at trial. See McQueeny, 779 F.2d at 923 (unfair prejudice is not simply testimony adverse to opponent); see also United States v. Harrelson, 754 F.2d 1153, 1179-80 (5th Cir.) (where witness did not positively identify defendant before she was hypnotized, but also did not fail to identify defendant before she was hypnotized, the testimony was beyond the exclusionary *501rule formulated in Valdez), cert. denied, 474 U.S. 908, 106 S.Ct. 277, 88 L.Ed.2d 241 (1985). We conclude that the potential prejudice that Grecco might suffer did not substantially outweigh the probative value of Galimi’s prospective trial testimony. Thus, excluding Galimi’s prospective trial testimony on the basis of Rule 403 cannot be reconciled with a sound exercise of discretion. Driggs, 823 F.2d at 55; see also McQueeny, 779 F.2d at 923 (order excluding evidence reversed where district court “misconstrued both elements of the Rule 403 balance”).
C. Motion To Strike The
Government’s Brief
Grecco has moved this court to strike the government’s brief because it referred to matters not in the record before the district court. The materials in question are eight affidavits that the government included as a supplemental appendix on appeal. The government conceded that the affidavits were not properly presented to the district court under the local rules and offered to excise the few sentences in its brief that cite the affidavits.
We have not relied on any of the information in the government’s supplemental appendix. The same is true of the assertions in the government’s brief that rest solely upon those affidavits. Grecco’s motion to strike will therefore be denied as moot.
IV. CONCLUSION
We conclude that Frank Galimi’s prospective trial testimony concerning the events surrounding the murder of Vincent Mistretta does not violate defendant Alan Grecco’s (a/k/a/ Alan Wolshonak) due process right to a fair trial. We will, therefore, reverse the order of the district court suppressing the prospective trial testimony of Frank Galimi on that ground. We note that our order does not foreclose the district court from addressing defendant Grec-co’s confrontation clause challenge. Furthermore, we will deny defendant Grecco’s motion to strike the government’s brief as moot.
. We use the phrase “prospective trial testimony” to refer to the testimony that was suppressed by the district court’s order of September 20, 1990.
. Galimi testified that he was taught self-hypnosis in the military, and used the technique to help him fall asleep at night. He also told the court that while he was working as a bouncer in a bar, the entertainment included a hypnotist. Galimi testified that he believed that one evening he was hypnotized and that he was so disoriented as a result that he sought out the hypnotist and requested that he undo the effects.
. This assumption renders it unnecessary for us to evaluate Galimi's testimony at the suppression hearing that he feigned being hypnotized.
. Memory hardening results in a subject having "great confidence in both true and false memo-ries_” Rock v. Arkansas, 483 U.S. 44, 60, 107 S.Ct. 2704, 2713, 97 L.Ed.2d 37 (1987).
. Although the district court held a "small Wade hearing,” the district court's opinion relies on Biggers and does not address whether Galimi’s acquaintance with the defendant might have provided a sufficient "independent origin” for Galimi’s prospective testimony. Compare Wade, 388 U.S. at 242-43, 87 S.Ct. at 1940 (factors relevant to independent origin for in-court identification testimony) with Biggers, 409 U.S. at 199-200, 93 S.Ct. at 382 (factors for assessing reliability of testimony when confrontation procedure was suggestive). Since the government does not argue that such an independent origin exists under Wade, but only that the Biggers factors showed that the Galimi’s testimony was sufficiently reliable, we do not reach the Wade issue.
. Although relevant, evidence may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of evidence.
Fed.R.Evid. 403.