We granted the defendant’s application for *838direct appellate review to consider his request that we change this court’s position on the admissibility of expert testimony concerning the capacity of eyewitnesses to make reliable identifications in particular circumstances.
In Commonwealth v. Francis, 390 Mass. 89, 98-101 (1983), this court held that, although expert testimony on the capacity of eyewitnesses to make identifications is not admissible as of right, such testimony is admissible in the trial judge’s discretion. We recently reaffirmed that principle in Commonwealth v. Hyatt, 419 Mass. 815, 818 (1995). Although we see no reason to change our position, we shall take this opportunity to provide comments that may assist judges in the exercise of their discretion in this area. But first, we shall set forth evidence that warranted the jury verdict that the defendant was guilty of rape.
On the evening of April 21, 1992, while walking through Stoughton center on her way home from a friend’s house, the sixteen year old victim was raped. She testified that, as she walked past a pharmacy in a well-lit area, she glanced at a man as he passed by her. She first thought that the man was an acquaintance, but, after looking again, she realized he was not. She testified that, during the second look, she noticed thé man’s build, his face, and his height. The victim also noted that he was white, had brown hair, dark eyes, a big nose, and no facial hair. The man was wearing a leather jacket and a dark colored baseball cap, and was carrying a black case.
As she walked toward the train station, again in a well-lit area, the victim noticed that the man who had passed her earlier was walking toward her from the other side of the street. The man then grabbed the victim and steered her into a dark alley behind a nearby liquor store. When she screamed, and then began to cry, the man told her to shut up or he would kill her.
The man, who smelled of alcohol, ordered the victim to lie down next to a dumpster in the alley. He put the black case and a beer bottle from his jacket pocket on the ground. The assailant then ripped down her pants and raped her for ten to fifteen minutes. The man’s face was right next to hers. The man then picked up his black case and the beer bottle, told the victim to get out of there, and left. The victim ran home and reported the rape to her father, who promptly called the police.
*839Officer Tracey Sisco arrived at the home about five minutes later. The victim described the assailant to the officer, who broadcasted the description by radio at 10:44 p.m. As a result of that broadcast, two detectives stopped the defendant in the parking lot of a Shaw’s supermarket at 10:49 p.m. When they stopped him, the defendant was wearing a Bruins baseball cap with tags hanging from it, carrying a black pool cue case, and smelled of alcohol.
Officer Sisco, who was still at the victim’s house, informed her that the police had stopped a suspect matching the description of her assailant and that Sisco was going to bring the victim to look at the person. As they drove into the Shaw’s parking lot, the defendant was standing next to a police cruiser in a well-lit area. Officer Sisco testified that, as she slowly drove past the defendant, the victim, who had been given a cover of clothing to prevent her being identified, stated three times, “that’s him.” Officer Sisco asked her whether she was sure, and told her that it was important that she be certain. The victim again responded affirmatively.
It is significant, in deciding the principal issue on appeal, that the Commonwealth’s case was not built solely on the eyewitness testimony of the victim. Chemical tests conducted on items worn by the defendant and the victim at the time of the incident showed that seminal fluid residue and blood found on the defendant’s underwear and shirttail were consistent with a mixture of body fluids from the victim and the defendant. A similar mixture of fluids was found on the victim’s T-shirt, sanitary napkin, and vaginal swabbing. The defendant was a Type O secretor with a PGM sub-type of 1 +. Approximately thirty-six per cent of the population are Type O secretors. The victim, who was menstruating at the time of the incident, was found to be a Type A secretor, with a PGM sub-type of 1 +. Approximately thirty-two per cent of the population are Type A secretors like the victim. Moreover, while the defendant was in custody, the police observed and photographed dried blood on the defendant’s penis and scrapes on his knees.
The defendant’s defense was one of mistaken identity. He testified that, prior to his arrest, he had been at several bars in Stoughton with a friend. He watched a Bruins hockey game until it ended at about 10:30 p.m. After the game, he and his friend left one bar and headed to another. The defen*840dant then realized that he had left his pool cue case at an earlier bar. He then left his friend, returned alone to one of the bars they had been at earlier, retrieved his pool cue case, and then headed toward the Shaw’s supermarket where he planned to buy a sandwich. The police stopped him in the supermarket parking lot. At trial, the defendant testified that he did not commit the crime. He explained that the blood on his penis was the result of an accident while attempting to scale a fence a few days earlier and the body fluids on his clothing were the consequence of intercourse on the night before the crime with a woman who did not testify.
The defendant’s misidentification claim focused on asserted discrepancies between the defendant’s actual height and the estimated height the victim gave to Officer Sisco; the difference in length between the black case seized from the defendant and the black case described by the victim; the victim’s failure to notice the Bruins logo on, or the tags hanging from, the defendant’s cap; and the fact that the victim did not describe any facial features to Officer Sisco, but did so at trial.
To aid in his misidentification defense, the defendant sought to introduce the expert testimony of a psychologist, Alexander Daniel Yarmey. Dr. Yarmey was presented as an expert on the application of perceptions and memory to eyewitness identification and testimony. His testimony, it was said, was intended to disabuse jurors of common misperceptions they might have regarding eyewitness identifications. The judge held an extensive voir dire, during which Dr. Yarmey testified that when a person is under high stress, less attention is given to cognitive skills and to the examination of details, circumstances that he said are contrary to the beliefs commonly held by lay persons. Dr. Yarmey testified further, among other things, that the presence of a weapon tends to divert attention from the perpetrator and to the weapon; that postevent suggestions can alter a witness’s memory but lay people do not appreciate it; that people are subject to suggestions from an authority figure; and that a witness’s expressed confidence level in an identification and the accuracy of that identification are not significantly related.
Following the voir dire, the judge ruled that he would exclude the expert’s testimony pursuant to the discretion afforded him by Commonwealth v. Francis, 390 Mass. 89 (1983). He ruled, among other things, that “[i]t is my judg*841ment that the subject matter which [the expert] would testify to falls within the general realm of knowledge of jurors.” The judge stated that he was “specifically not ruling that evidence of this nature can never come into the court. What I am doing here is exercising my discretion that in this particular case there are no special circumstances that would justify an exception to the general proposition that this proposed testimony cannot be received in court.”
The defendant argues to us that the judge abused his discretion in excluding the expert testimony which, it is claimed, was carefully tailored to address identification issues in this case. He argues that, “[h]aving been terrorized and raped in a dark alley by an individual whom she believed had a weapon, the victim was told by the police that a suspect matching her description had been stopped nearby. Expecting to be shown her assailant, the victim was given a cover of clothing to prevent herself from being seen by the suspect, and then driven past the defendant, who was displayed alone next to a police officer and cruiser. Her expectations confirmed, the victim then expressed great confidence in positively identifying the defendant, a local resident whom she could easily have encountered previously under innocent circumstances.” In our recent opinion in Commonwealth v. Hyatt, 419 Mass. 815 (1995), we noted that the weight of authority was against the admission of expert testimony of the sort involved here, but recognized that some jurisdictions favor its admission in particular circumstances. Id. at 818. Federal Courts of Appeals treat the question as one of discretion in the trial judge. See United States v. Daniels, 64 F.3d 311, 315 (7th Cir. 1995), cert. denied, 516 U.S. 1063 (1996); United States v. Brien, 59 F.3d 274, 277 (1st Cir.), cert. denied, 516 U.S. 953 (1995); United States v. Rincon, 28 F.3d 921, 923 (9th Cir.), cert. denied, 513 U.S. 1029 (1994); United States v. Harris, 995 F.2d 532, 534 (4th Cir. 1993); United States v. Blade, 811 F.2d 461, 465 (8th Cir.), cert. denied, 484 U.S. 839 (1987); United States v. Serna, 799 F.2d 842, 850 (2d Cir. 1986), cert. denied sub nom. Cinnante v. United States, 481 U.S. 1013 (1987); United States v. Downing, 753 F.2d 1224, 1226 (3d Cir. 1985). Cf. United States v. Holloway, 971 F.2d 675 (11th Cir. 1992) (expert testimony on eyewitness identification per se inadmissible), cert. denied sub nom. Rudder v. United States, 507 U.S. 962 (1993). In State courts, judges have gen*842erally excluded expert testimony on eyewitness identification, and those decisions have been upheld on review. See, e.g., State v. Kemp, 199 Conn. 473, 477-479 (1986) (no abuse of discretion in excluding expert testimony because these matters “have generally been found to be within the realm of common experience and can be evaluated without expert assistance”); People v. Enis, 139 Ill. 2d 264, 288-290 (1990) (no abuse of discretion in excluding eyewitness expert testimony), cert. denied, 516 U.S. 827 (1995); State v. Rich, 549 A.2d 742, 743 (Me. 1988) (upholding trial judge’s exclusion of expert testimony and stating that effects of stress on eyewitness reliability are “not beyond the common knowledge of the ordinary juror”); Commonwealth v. Simmons, 541 Pa. 211, 230-231 (1995) (trial court properly excluded proposed expert testimony on reliability of eyewitness identification on ground that it would intrude upon jury’s basic function of deciding credibility. “Such testimony would have given an unwarranted appearance of authority as to the subject of credibility, a subject which an ordinary juror can assess”), cert. denied, 516 U.S. 1128 (1996).
In a relatively small number of cases, the exclusion of expert testimony on eyewitness identification has been characterized as an abuse of discretion. The opinions that rule that the exclusion of the expert testimony was or may have been error are typically those where there was little or no evidence to corroborate the eyewitness identification. See State v. Chapple, 135 Ariz. 281, 285, 293-297 (1983) (where there was “[n]o direct or circumstantial evidence of any kind” connecting the defendants to the crime, it was error to refuse expert testimony on eyewitness identifications. However, the court expressed “no quarrel with the result reached in the vast majority of cases” where such testimony was excluded); People v. McDonald, 37 Cal. 3d 351, 377 (1984) (“When an eyewitness identification of the defendant is a key element of the prosecution’s case but is not substantially corroborated by evidence giving it independent reliability, and the defendant offers qualified expert testimony on specific psychological factors shown by the record that could have affected the accuracy of the identification but are not likely to be fully known to or understood by the jury, it will ordinarily be error to exclude that testimony”); State v. Whaley, 305 S.C. 138, 141-143 (1991) (abuse of discretion to exclude expert *843testimony where only evidence against defendant was the testimony of two eyewitnesses). See also United States v. Downing, 753 F.2d 1224, 1226, 1230, 1238-1242 (3d Cir. 1985) (vacating conviction because defendant was convicted solely on basis of eyewitness testimony and suggesting that judge may have abused discretion in failing to properly evaluate helpfulness of testimony; providing extensive discussion of guidelines under which discretion pursuant to Fed. R. Evid. 702 should be exercised and remanding for hearing1).
These latter cases were decided on fact-specific grounds. For example, in State v. Chapple, supra at 285, 297, the court held that the expert should have been allowed to testify based on the “peculiar facts of this case,” which included a delay of more than one year between the commission of the crime and the identification of the defendant and the absence of any corroborating evidence. See United States v. Downing, supra at 1226 (no corroborating evidence); People v. McDonald, supra at 375-377 (failure or uncertainty of several witnesses in identifying defendant, cross-racial identification discrepancies, and lack of any corroborating evidence); State v. Whaley, supra at 143 (assailant’s features partially obscured during incident and cross-racial nature of identification). See also State v. McCutcheon, 162 Ariz. 54, 58 (1989) (distinguishing State v. Chapple, supra, because “there was prompt, positive, unambiguous eyewitness identification supported by considerable other physical evidence tying defendant to the crime”); State v. Poland, 144 Ariz. 388, 399 (1985) (no abuse of discretion in excluding expert testimony on eyewitness identification because “[t]he peculiar facts of [State v. Chapple, supra] were not present in the instant case. The question of guilt did not hinge solely on the testimony of eyewitnesses”). The facts of this case, particularly the physical evidence, provide significant corroboration of the victim’s identification. In such a case, we defer to the trial judge’s discretion. There was no error in excluding the evidence.
We grant that the rule to which we adhere may result in disparate treatment in similar cases. That is inherent in any *844grant of discretion to trial judges. Appellate courts have done little to guide the exercise of discretion in this area, except that a few courts have stated that some evidence corroborating the eyewitness identification is required if the expert testimony is to be excluded. The absence of appellate guidelines will likely persist unless and until learning about the reliability of eyewitness identifications becomes more developed.
Certainly, the admissibility of expert testimony about the reliability of eyewitness identification must be justified on general principles related to the admission of expert testimony. For example, the offered opinion must be relevant to the circumstances of the witness’s identification. As usual, the judge must conclude that the subject of the opinion is one on which jurors need assistance and can be helped, and will not be confused or misled, by the expert’s testimony. See United States v. Serna, 799 F.2d 842, 850 (2d Cir. 1986) (proffered expert’s testimony did not relate to conditions under which witness made identification and many of his conclusions coincided with common sense). Also, the opinion must be reliable in the sense that the tests and circumstances on which the opinion rests must provide a basis for concluding that the opinion is rehable. See Commonwealth v. Lanigan, 419 Mass. 15, 24-26 (1994); United States v. Brien, 59 F.3d 274, 277 (1st Cir. 1995) (“we sustain the district court’s ruling on the ground that the district judge made clear his need for some proffer of data or literature underlying the expert’s assumptions and conclusions” and “practically nothing” was offered). Another consideration of relevancy is whether the expert testimony offered is sufficiently tied to the facts of the case so that it will aid the jury in resolving the matter. United States v. Downing, supra at 1242 (detailed proffer required, “including an explanation of precisely how the expert’s testimony is relevant to the eyewitness identifications under consideration”). See also United States v. Rincon, 28 F.3d 921, 926 (9th Cir. 1994) (applying Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 [1993]).
Obviously, the circumstances under which a crime victim made an identification cannot be precisely, or perhaps even approximately, replicated. What tests were conducted and whether they lead to a reliable and helpful expert opinion will have to be considered. See Commonwealth v. Lanigan, supra *845at 25. The acceptability of the expert’s conclusions in the relevant community of experts remains a factor, but not necessarily a controlling one, bearing on the admissibility of an opinion. Id. at 26. Finally, we recognize that a principle concerning eyewitness identifications may become so generally accepted that, rather than have expert testimony on the point, a standard jury instruction stating that principle would be appropriate. See Commonwealth v. Hyatt, 419 Mass. 815, 818-819 (1995).
The defendant also challenges a portion of the judge’s charge concerning eyewitness identification. At the charge conference, the defendant objected to the use of that portion of the standard charge (see Commonwealth v. Rodriguez, 378 Mass. 296, 311 [1979], and Commonwealth v. Pressley, 390 Mass. 617 [1983]), that states: “You may take into account . . . the strength of the identification.” The judge adhered to the standard charge, and the defendant renewed his objection. In Commonwealth v. Jones, 423 Mass. 99 (1996), which was decided after the trial in this case, we stated:
“The instruction approved in Commonwealth v. Cuffie, [414 Mass. 632 (1993)], indicates that the jury should be told that they could ‘take into account . . . the strength of the identification.’ We would not object to the omission of that provision from a jury charge because it may suggest that the confidence with which a person makes an identification is a valid indicator of the accuracy of the recollection. There is doubt as to the soundness of that assumption. See State v. Long, 721 P.2d 483, 490 (Utah 1986). See also State v. Ramirez, 817 P.2d 774, 781 (Utah 1991).” Id. at 110 n.9.
We now conclude that, in cases tried hereafter, the challenged language should be omitted from the standard instruction concerning eyewitness testimony.
It is probably true that the challenged instruction has merit in so far as it deals with the testimony of a witness who expressed doubt about the accuracy of her identification, whether that identification was made during her testimony, or at a “showup” or lineup. Where, however, the witness has expressed great confidence in her identification of the defendant, the challenged instruction may pose a problem because, *846as we noted in Commonwealth v. Jones, supra, there is significant doubt about whether there is any correlation between a witness’s confidence in her identification and the accuracy of her recollection.
In this case, it was not error to give the challenged instruction. In the future, however, the significance, if any, of a witness’s confidence in an identification should be left to cross-examination and to argument of counsel and should not, in the normal course, be a subject of a jury instruction.
Judgment affirmed.
On remand, the District Court judge concluded that the expert testimony was not sufficiently reliable to warrant its admission (United States v. Downing, 609 F. Supp. 784, 791 [E.D. Pa. 1985]), and the Appeals Court affirmed without a published opinion (United States v. Downing, 780 F.2d 1017 [3d Cir. 1985]).