OPINION OF THE COURT
Read, J.Defendants in both of these cases unsuccessfully sought to introduce expert testimony on the reliability of eyewitness identification. The question for us to decide is whether, in light of our decisions in People v Lee (96 NY2d 157 [2001]), People v Young (7 NY3d 40 [2006]) and, especially, People v LeGrand (8 NY3d 449 [2007]), the trial courts abused their discretion when they disallowed this testimony. We conclude that the trial judge in Abney abused his discretion, but that the trial judge in Allen did not.1
*257I.
Abney
A. Facts and Trial
On June 2, 2005 at about 3:20 p.m., 13-year-old Farhana U., on her way home from school, was descending well-lit stairs into the subway station near the corner of Essex and Delancey Streets in Manhattan when a man whom she did not know approached her and asked for “some change.” This man stood face-to-face with Farhana, about two feet away. She initially did not think he intended to harm her and was not afraid. Looking him squarely in the face, she said she had no change.
After Farhana “took a couple of steps forward,” the stranger wheeled in front of her, placed a knife with a six-inch blade and “a big curve on the end” near her throat, and asked her “a couple of times” to hand over her necklace, a gold chain with a locket. As this man stood close by her, Farhana was “looking at his face”; she was “really scared” and “didn’t know what to do.” When Farhana refused his demand, screaming “No,” he ripped the chain off her neck, and fled up the stairs. This entire encounter was fleeting.
Farhana continued down the stairs into the subway station and reported the robbery to a token clerk. At about 4:30 p.m., Detective Samuel DeJesus interviewed her at his desk in the transit station house at Columbus Circle. According to Detective DeJesus, Farhana seemed frightened; she told him that she had been robbed at knifepoint by a stranger, a black man in his thirties who was over six feet tall, had “pinkish” lips, and wore a short-sleeved blue shirt and a blue bandana.
Farhana’s physical description of the robber and how he carried out the crime prompted Detective DeJesus to suspect defendant Quentin Abney: he was familiar with defendant on account of his arrest for an earlier subway-related robbery. Telling Farhana that he would be “right back,” the detective left his desk to put together an array of six photographs, including defendant’s. When he returned, he told Farhana that he “was going to show her a group of photos and, if she recognized anyone, to let [him] know which one and what number.” Pointing out defendant’s photograph, she responded, “that’s him, number six.”
On June 22, 2005, Detective Ernest Dorvil telephoned Farhana at her home to ask her to view a lineup. Upon arrival at *258the station house, she waited in an office with the door shut while the lineup was being put together. From the office, Farhana could not see defendant, who was in a “cell area” on the other side of the building, or any of the “fillers” selected to participate in the lineup. Once the lineup was assembled, Detective Dorvil escorted Farhana to the viewing room, where they were joined by another police officer and defendant’s attorney. She identified defendant, in position number four (there were six men in the lineup); Farhana was “sure” that he was the man who had asked her for change and then robbed her at knife-point.
Defendant was arrested and charged with one count of robbery in the first degree (Penal Law § 160.15 [3]). At the subsequent jury trial, Farhana testified on direct examination that the man who robbed her had a dark brown complexion, “puppy dog eyes,” and “pinkish-purplish lips”; she did not remember whether he was wearing anything over his head, but thought his shirt was blue. Defense counsel cross-examined Detective DeJesus about the details of Farhana’s initial description of the robber; Detective Dorvil about the lineup, eliciting testimony that he let Farhana know ahead of time that a suspect was included (which she contradicted, saying that the detective told her only that he wanted her to view “a couple of people”); and Farhana about the lineup and the appearance of the knife.
Defendant presented an alibi defense, contending that, at or around the time of the robbery, he was picking up his girlfriend Mary Nimmons’s daughter from preschool in Brooklyn and that, before and after doing this, he was at Nimmons’s home at 64 MacDougal Street in Brooklyn. His alibi witnesses were Nimmons (the mother of two children by defendant) and her sister and two cousins, and Carolyn Murphy, an assistant teacher at the preschool.
Murphy testified that, on June 2, 2005, defendant picked up Nimmons’s daughter. She claimed to recall the date because the very next day—June 3, 2005—Nimmons came to the school to obtain a copy of the sign-in sheet reflecting that fact. While Nimmons also initially testified that she got the sign-in sheet on June 3, she later professed to have picked it up at defendant’s request some time after his arrest on June 22, 2005. Murphy further testified that she was one of the employees responsible for making sure the sign-in sheets were filled out properly, and that the sheet was checked for accuracy every time it was signed.
*259On rebuttal, the parties stipulated that “the logbook [of sign-in sheets] from March through June [would] be entered into evidence and made available to the jury and [could] be referenced in the closing arguments.” In general, the sign-in sheets were filled in irregularly—some days children were signed in, but not signed out; and in some places where a signature appeared, no time was filled in. The sheet dated June 2, 2005 from the March-June logbook exhibit differed from the sheet dated June 2, 2005 admitted on defendant’s behalf. The former stated that defendant picked up Nimmons’s daughter at 3:04 p.m.; the latter pegged the time at 3:00 p.m.
The jury convicted defendant of first-degree robbery, and Supreme Court sentenced him as a persistent violent felony offender to a prison term of 20 years to life. He appealed.
B, Proffer of Expert Testimony
Before jury selection, defendant made a motion in limine to present expert testimony concerning “psychological factors of memory and perception that may affect the accuracy of eyewitness identifications.” Specifically, he sought to call Dr. Solomon M. Fulero, who would “educat[e] the jurors on many counterintuitive findings that bear directly on the reliability of the identification evidence in [the] case,” which were beyond the average juror’s ken. Defendant identified 15 such factors: stress, exposure time, color perception under monochromatic light, event violence, cross-racial accuracy, similarity of lineup fillers, lineup instructions, rate of memory loss, postevent information, the wording of questions posed to an eyewitness, unconscious transference to the crime scene of someone seen in another situation or context, the witness’s preexisting attitudes and expectations, simultaneous and sequential lineups, the lack of correlation of confidence and accuracy, and confidence malleability. He specifically noted that the robbery was brief, the victim was under stress, and a weapon was used.
Supreme Court denied the motion as premature, with leave to renew at the close of the People’s direct case. The trial judge observed that while he was aware that the People’s case hinged on the eyewitness testimony of the 13-year-old victim, “[a]t this [juncture], . . . [he], in the exercise of discretion, [did] not consider [the] case an appropriate one for an expert identification witness” for several reasons.
The trial judge reasoned that “[a]s a threshold matter,” defendant’s papers did not “appropriately narrow the scope of *260the expert’s proposed testimony,” which therefore threatened to turn into “a full-fledged seminar . . . which could lead to hours of academic discussion and speculation.” Second, the proffered testimony about how police investigative techniques might influence a lineup identification was not relevant because the victim had previously picked out a photograph from an array, and so must have realized that a suspect would be included in the lineup. Third, two of the proposed subjects of testimony— postevent information and unconscious transference—“[had] not passed the Frye test” in other courts.2 Fourth, evidence about simultaneous versus sequential lineups was “unmanageable in a trial setting,” because juries were “not experts on constitutional law and procedure and [could not] be educated about those topics during a trial.”
Finally, the trial judge commented that “jurors know that, as a matter of common sense, a person’s memory does fade as time passes.” In his view, “[mjany” of defendant’s concerns relative to the accuracy of Farhana’s eyewitness identification could be adequately addressed by tailoring cross-examination and the jury charge. He indicated that defendant was “free to renew his motion at the close of the People’s case, at which time he [should] narrow his proffer to the specific topics that he believes are relevant to the facts of this case.”
After the People rested, defendant did, in fact, renew his motion, proposing a more limited range of subjects to be covered. Specifically, he asked to elicit expert testimony about the effect of event stress, exposure time, event violence and weapon focus, cross-racial identification, and lineup instructions, in addition to a new topic, double blind lineups. Defendant pointed out specific circumstances of his case to which the proposed testimony would be relevant. While acknowledging the trial judge’s earlier ruling that witness confidence could be addressed through jury instructions, defendant specifically took exception to it.
The trial judge denied defendant’s renewed motion on the ground that “having had the benefit of the witness’ testimony,” there was “nothing unique about [the] case . . . presenting] issues that are beyond the ken of the ordinary juror.” In his view, the relevant issues had been explored adequately during *261cross-examination, and could be argued in summation and covered in the jury charge.
C. Appellate Division
The Appellate Division, with two Justices dissenting, affirmed the judgment of conviction and sentence. The court observed that LeGrand—where the People’s case rested on identifications made nearly seven years after the crime—did not mandate admission of Dr. Fulero’s testimony. The court commented that “[t]he unusual fact pattern presented in LeGrand raise[d] a genuine question as to whether that case’s rule . . . applie[d] in cases, like this one, where the circumstances create much less doubt about the reliability of the identification testimony” (People v Abney, 57 AD3d 35, 43 [1st Dept 2008]).
The Appellate Division saw no need to answer this question, however, because of significant evidence corroborating defendant’s guilt, such that “by the terms of the LeGrand rule itself, the exclusion of the proffered expert testimony was within Supreme Court’s discretion” (id.). This corroborative evidence was the testimony of defendant’s own witnesses (Nimmons and Murphy), which suggested that he sought to document an alibi long before he was arrested for the robbery. The court recognized that, because this testimony was not elicited before the close of the People’s direct case, Supreme Court might have erred in refusing to permit Dr. Fulero to testify. In the Appellate Division’s view, however, defendant’s alibi evidence ultimately rendered any such error harmless.
The two dissenting Justices concluded that the trial judge abused his discretion when he denied defendant’s renewed motion at a time when there was “not a scintilla” of evidence corroborating the eyewitness’s identification (id. at 49). In their view, he should have considered the four factors enumerated in LeGrand, which they thought were generally met in this case. They acknowledged that certain of the subjects that Dr. Fulero proposed to cover were not recognized by the courts as generally accepted within the relevant scientific community, but faulted the trial judge for summarily rejecting this evidence without first conducting a Frye hearing. A dissenting Justice subsequently granted defendant permission to appeal to us.
*262 Allen
A. Facts and Trial
At around 3:30 p.m. on March 10, 2004, two masked men barged into a busy barbershop in Queens, located across from the Woodside Housing Project. The mask worn by one of the men, who wielded what appeared to be a knife, left exposed the top portion of his face from his top lip to above his eyebrows. His accomplice (who has never been apprehended) displayed a gun and stated, “This is a holdup.” The gunman grabbed Juan Almonte, one of the barbers, and pulled back the slide on his gun. He next struck the shop owner in the face with the gun; he chased down a barber who tried to escape to the basement, dragged him back, put a gun into his mouth, and rifled his pants pockets; he demanded money from another barber, who emptied his pockets of about $30 in tips.
The knife-wielding intruder asked Almonte for money, checked around his neck for a chain, tried to open an apparently locked drawer at his barber station, and then called out to the gunman, “It’s time to go.” The two robbers instructed everyone present to lie down on the floor. They fled the shop together, and someone called the police. This entire incident lasted only a few minutes.
Gabriel Bierd, a customer in the shop, quickly realized that the knife-wielding robber was defendant Gregory Allen, whom he encountered regularly in the neighborhood. Bierd had heard defendant’s speaking voice several times during the previous six months, and he recognized him from both his “[b]ody type” and his voice.
When Sergeant Ronald Buell arrived at the barbershop to investigate the robbery, Bierd provided him with defendant’s nicknames—“Junior” and “J.R.”—and described him as a black male, about five feet, eight inches or five feet, nine inches tall. Bierd accompanied Sergeant Buell to the squad house, and looked through a “photo mug book” with 180 photographs, one to a page. Pedigree information was covered up. He picked out defendant’s photograph (at page 64), identifying him as the knife-wielding robber. At Sergeant Buell’s instruction, though, he looked through the remainder of the book. Buell then prepared a photographic array with six photographs, and Bierd selected defendant’s. Sergeant Buell drove to the barbershop and showed this photographic array to Almonte, who likewise identified defendant as the knife-wielding robber.
*263Defendant was arrested at about 10:30 p.m. on the day of the robbery. He was charged with four counts of robbery in the first degree (Penal Law § 160.15 [3], [4]), two counts of attempted robbery in the first degree (Penal Law §§ 110.00, 160.15 [3], [4]), two counts of robbery in the second degree (Penal Law § 160.10 [1]), and three counts of attempted robbery in the second degree (Penal Law §§ 110.00, 160.10 [1], [2] [a]).
At defendant’s trial, Bierd and Almonte recounted the robbery, and explained how they immediately recognized defendant as the knife-wielding robber. They testified to having picked defendant out in a lineup on July 7, 2004; they identified him in court.
Detective David Beutel testified that he arrested defendant on March 10, 2004 and, when he asked him his name, defendant replied that his nicknames were “Junior” and “J.R.” Detective Beutel attempted to put together a lineup the next day, but defendant would not cooperate; he pulled his shirt over his head, got into the fetal position on a table, and refused to sit and hold up a number unless all the men in the lineup wore masks. Pursuant to court order, Detective Beutel finally organized a lineup on July 7, 2004, four months later. Bierd, Almonte and the barbershop’s owner each viewed the lineup separately. Bierd and Almonte identified defendant as the knife-wielding robber; the shop owner did not recognize anyone.
Defendant called five defense witnesses. An investigator with the Legal Aid Society recounted her interview with Almonte, stating that he told her that he could see the skin around the knife-wielding robber’s eyes and the bridge of his nose, and that his mask did not have a hole for the nose. Another investigator testified that he was present at the lineup on July 7, 2004, and that defendant’s attorney asked for all the lineup participants to wear ski masks; a police officer testified that he recovered six latent prints from the door of the barbershop; and a detective testified that none of those prints matched defendant’s. Finally, Danielle Allen, defendant’s sister, told the jury that her brother had a scar on his nose as a result of a dog bite suffered in childhood, and that he had long had a scar on his cheek. Photographs showing these scars were admitted into evidence. Allen further explained that, at the time of the robbery, defendant was dating someone who lived in the Woodside Housing Project.
The jury convicted defendant of all the charges submitted to it (the trial judge dismissed three counts). On February 28, *2642006, defendant was sentenced to concurrent, determinate prison terms of 15 years on each first-degree robbery conviction, 10 years on each attempted first-degree and second-degree robbery conviction, seven years on each attempted second-degree robbery conviction, and five years’ postrelease supervision. He appealed.
B. Proffer of Expert Testimony
On July 14, 2005, defendant moved in limine to admit the expert testimony of Professor Steven Penrod regarding 17 “psychological factors of memory and perception tbat may affect the accuracy of eyewitness identifications.” Attached to the motion was Dr. Penrod’s curriculum vitae, a list of his publications, and a document written by Dr. Penrod discussing many of the factors about which he proposed to testify, but not how these factors were relevant to the case. This document mentioned unconscious transference only once, describing this phenomenon as when “innocent persons seen in some other context can be mistakenly identified as having been seen at the crime or [where] a bystander at the scene is mistaken as the perpetrator.”
The People opposed the motion on the ground that defendant had not demonstrated that the ability to assess the accuracy of the eyewitnesses’ testimony was beyond the ken of the average juror, or that the psychological factors about which Dr. Penrod proposed to testify were relevant to this case, “particularly . . . where . . . defendant was known to one of the eyewitnesses [Bierd] after months of regular encounters.” Further, the People argued that this case was distinguishable from the one-eyewitness identification cases relied upon by defendant. Here, there were two eyewitnesses who corroborated each other, and there was additional corroborative evidence, including defendant’s refusal to participate in a lineup unless he and all the fillers wore masks. At the time, the People argued, defendant would have had no way to know that masks were involved in the robbery unless he was there when it took place.
Immediately before jury selection, defendant made additional arguments on the motion. According to defendant, the expert testimony would be “important” because the use of a gun made the robbery stressful, and Dr. Penrod would explain that high levels of stress could impair an identification. Additionally, Dr. Penrod would address “weapon focus,” and the concept that there is “not necessarily a correlation between certainty and *265accuracy” of an eyewitness identification, although the “general public . . . thinks there is.” Finally, defendant argued that Dr. Penrod’s testimony on the subject of “unconscious transference” was relevant because the eyewitnesses claimed that they had seen defendant in the neighborhood before, and thus it was possible that they identified defendant because “he’s the only one in the lineup that they’ve seen in the neighborhood.”
The trial judge denied the motion, finding that “this is not an area in which an expert is at all helpful” because the proposed testimony involved matters of
“common sense and life experience . . . [T]he jury knows the longer you look at something the easier it is to remember . . . , just like they know if a person is wearing a disguise, it makes it hard to recognize them. They don’t need an expert to testify to these things.”
He ruled that defendant should address the reliability of eyewitness identification during cross-examination and in summation.
After the People rested, defendant renewed his motion, reiterating his position that Dr. Penrod’s testimony regarding unconscious transference was relevant because “statistics show that people are much more likely to identify someone they’ve seen before.” The court denied the application, again deciding that the issue was a matter of common sense.
C. Appellate Division
The Appellate Division unanimously affirmed defendant’s judgment of conviction and sentence because “the facts . . . [were] sufficiently distinct from those of [LeGrand] that . . . Supreme Court did not err in denying the defendant’s request to adduce expert testimony regarding the reliability of eyewitness identification” (People v Allen, 53 AD3d 582, 584 [2d Dept 2008]). The court emphasized defendant’s confirmation that he used the nicknames given to the police by Bierd, and his knowledge that the robbers wore masks. Further, the court pointed out, two eyewitnesses identified defendant from a photographic array the day of the crime and then again, four months later, in a lineup; and one of the eyewitnesses had seen and heard defendant in the neighborhood regularly over a period of six months before the robbery. Defendant appealed, and a Judge of this Court granted his application for leave (12 NY3d 780 [2009]).
*266II.
In Lee, we held that expert testimony proffered on the issue of the reliability of eyewitness identification “is not admissible per se”; rather, “the decision whether to admit it rests in the sound discretion of the trial court” (96 NY2d at 160), which should be guided by “whether the proffered expert testimony would aid a lay jury in reaching a verdict” (id. at 162 [internal quotation marks omitted]). We also noted that although “jurors may be familiar from their own experience with factors relevant to the reliability of eyewitness observation and identification, it cannot be said that psychological studies regarding the accuracy of an identification are within the ken of the typical juror” (id.). And finally, we recognized that because “expert testimony of this nature may involve scientific theories and techniques, a trial court may need to determine whether [it] is generally accepted by the relevant scientific community” (id.).
Applying these principles to the facts in Lee, we “[could] not say the trial court’s denial of defendant’s motion [to introduce expert testimony regarding the reliability of eyewitness identification] constituted an abuse of discretion” (id. at 163). In reaching this conclusion, we observed that the trial judge entertained the defendant’s motion during the People’s case-in-chief, and so “was in a position to weigh the request against other relevant factors, such as the centrality of the identification issue and the existence of corroborating evidence” (id.). The crime in Lee was a carjacking, and the defendant was discovered driving the stolen vehicle, which corroborated the victim’s identification of him as the carjacker.
Next came our decision in Young. There we reiterated several of the key points we made in Lee: that the decision whether to admit or exclude expert evidence on the reliability of eyewitness identification lies within the bounds of the trial court’s discretion; in the exercise of this discretion, the trial court should consider whether “the expert [could] tell the jury something significant that jurors would not ordinarily be expected to know already” (Young, 7 NY3d at 45); and scientific studies of factors affecting the reliability of eyewitness identification are beyond the ken of the typical juror.
Calling the question a “close” one on the facts in Young (id. at 42), we nonetheless decided there was no abuse of discretion. Young was a home invasion case, and we commented that if the trial had “turned entirely on an uncorroborated eyewitness *267identification, it might well have been an abuse of discretion to deny the jury the benefit of [the expert’s] opinions” {id. at 45). As it was, though, two of the defendant’s female acquaintances were found to possess property stolen during the home invasion; neither of them could have been the robber (who was a male); and one of them said she got the property from the defendant. Under these circumstances, we considered it “reasonable” for the trial judge to decide that the eyewitness’s identification “was quite unlikely to be mistaken, and that [the expert’s] testimony would be an unnecessary distraction for the jury” {id. at 46).
We most recently considered the admissibility of expert testimony on the reliability of eyewitness identification in Le-Grand. There, we held that
“where [a] case turns on the accuracy of eyewitness identifications and there is little or no corroborating evidence connecting the defendant to the crime, it is an abuse of discretion for a trial court to exclude expert testimony on the reliability of eyewitness identifications if that testimony is (1) relevant to the witness’s identification of defendant, (2) based on principles that .are generally accepted within the relevant scientific community, (3) proffered by a qualified expert and (4) on a topic beyond the ken of the average juror” {LeGrand, 8 NY3d at 452).
In LeGrand—unlike Lee and Young—there was no evidence other than eyewitness identifications to tie the defendant into the crime, the stabbing death of a livery cab driver in 1991. Further, the identifications on which the People’s case depended were made nearly seven years after the crime, and the defendant was not convicted until 2002. In light of these facts, we decided that the trial judge abused his discretion when he disallowed expert testimony on those three factors where the defense expert’s testimony at the Frye hearing “confirm[ed] that the principles upon which [he] based his conclusions [were] generally accepted by social scientists and psychologists working in the field” {id. at 458): the correlation between confidence and accuracy, the effect of postevent information, and confidence malleability.
Finally, we noted in LeGrand that
“[although the trend has been of late to more liberally admit such [expert] testimony—as recognized *268in Lee and Young—the admissibility of such evidence would also depend upon the existence of sufficient corroborating evidence to link defendant to the crime. In the event that sufficient corroborating evidence is found to exist, an exercise of discretion excluding eyewitness expert testimony would not be fatal to a jury verdict convicting defendant” {id. at 459).
Applying our precedents in this area to the facts in Abney and Allen, we reach different answers to the question of whether the trial judge abused his discretion. While it was reasonable for the trial judge to deny defendant’s pretrial motion in Abney as premature and overly broad, another outcome was called for when defendant renewed the motion at the close of the People’s direct case. By that point, it was clear that there was no evidence other than Farhana’s identification to connect defendant to the crime, and she did not describe him as possessing any unusual or distinctive features or physical characteristics.
Defendant asked to elicit testimony from Dr. Fulero, who is a qualified expert on the subject of eyewitness identification research findings, about the following topics: the effect of event stress, exposure time, event violence and weapon focus, cross-racial identification, lineup instructions, double-blind lineups, and witness confidence. All but two of these subjects—lineup instructions and double-blind lineups—seem relevant to Farhana’s identification of defendant, given the particular circumstances of the case. And as we stated in LeGrand, the principles related to witness confidence upon which Dr. Fulero proposed to testify are generally accepted within the relevant scientific community. They are also counterintuitive, which places them beyond the ken of the average juror.
Accordingly, the trial judge in Abney abused his discretion when he did not allow Dr. Fulero to testify on the subject of witness confidence. As for the remaining relevant proposed areas of expert testimony—the effect of event stress, exposure time, event violence and weapon focus, and cross-racial identification—the trial judge should have conducted a Frye hearing before making a decision on admissibility.
Finally, we do not consider the trial judge’s error in Abney to have been harmless. While defendant’s muddled alibi evidence was no doubt unhelpful to his cause with the jury, it is not overwhelmingly inculpatory either. And, of course, it is possible that defendant would not have pursued an alibi defense in the first place if Dr. Fulero had testified.
*269Contrariwise, we consider Allen to be more akin to Lee and Young. We are unwilling to second-guess the trial judge’s exercise of discretion in Allen because the case did not depend exclusively on Bierd’s eyewitness testimony—i.e., Allen is not a “case [that] turns on the accuracy of eyewitness identifications [where] there is little or no corroborating evidence connecting the defendant to the crime” (LeGrand, 8 NY3d at 452). Critically, Almonte independently identified defendant as the knife-wielding robber who searched him and stood nearby throughout the course of the robbery. And defendant was not a stranger to either Bierd or Almonte.
Defendants’ remaining arguments, to the extent preserved, are without merit.
Accordingly, in People v Abney, the order of the Appellate Division should be reversed and a new trial ordered; in People v Allen, the order of the Appellate Division should be affirmed.
Judges Ciparick, Graffeo, Smith, Pigott and Jones concur; Chief Judge Lippman taking no part.
In People v Abney: Order reversed, etc.
In People v Allen: Order affirmed.
. The trials in Abney and Allen were both completed after our decision in Lee, but before our decisions in Young and LeGrand, which elaborated on the principles announced in Lee.
. We subsequently decided in LeGrand that studies about the potential effect of postevent information on the reliability of an eyewitness identification are, in fact, generally accepted as scientifically valid (8 NY3d at 458).