The defendant was convicted by a jury in the Superior Court on two indictments charging statutory rape, in violation of G. L. c. 265, § 23.1 The alleged victim was his then *24fifteen year old niece, Molly.2 The defendant appealed from his convictions, and we granted his application for direct appellate review. On appeal, the defendant argues that the judge erred by excluding expert psychological testimony regarding “dissociation” and the evidence from Molly’s past that suggested that Molly had that disorder. The defendant contends that the exclusion of that evidence was prejudicial because it denied him the opportunity adequately to explain to the jury the significant possibility that Molly’s allegations were either fabricated or arose from a distortion of memory. We conclude that the exclusion of this evidence was prejudicial error and therefore reverse the convictions and remand the case for a new trial.
Background. We summarize the evidence admitted at trial that is relevant to this appeal. Molly’s adoptive mother is the defendant’s sister. In August, 2007, Molly, her adoptive parents, and three of her four siblings traveled from their home in Minnesota and stayed with the defendant’s family at their home in Massachusetts for one week.3
Six months after the visit, on February 14, 2008, Molly attended a meeting at school led by her godmother that discussed the topic of sexual abuse. During the meeting, according to Molly, “all of a sudden, it just popped up in my head of what happened in Boston.” After the meeting, Molly told her godmother that “someone had touched her in ways that he shouldn’t have.” Her godmother asked her who had touched her, and she said it was her “Uncle Bill,” referring to the defendant, and then started to cry. When asked how he had touched her, Molly *25said that “he had sex with me.” After the revelation, the godmother took Molly home, where Molly told her parents, who took her to the Midwest Children’s Resource Center (center). At the center, she was interviewed by a nurse, and the interview was recorded on videotape. She was also examined by a physician. The results of this physical examination were normal, revealing no signs of trauma or sexually transmitted disease.
At trial, Molly testified that, on the next to last night at the defendant’s home (first night), she, the defendant, and Martha, the defendant’s then ten year old daughter, went upstairs to the first floor to watch a movie on the television, while Molly’s three siblings and two of the defendant’s five children stayed downstairs in the basement to watch another movie. The defendant and Molly sat on one couch; Martha sat alone on another. The defendant placed a blanket over both him and Molly, and then put his hand under her shirt and touched her breasts. Nothing else happened, and she later went to bed.4
The next night (second night), Molly and the defendant watched television alone on the first floor, while three of her siblings and two of the defendant’s children were in the basement television room. The defendant again touched Molly’s breasts and told her to go downstairs and take off her underwear, which she did.5 When she returned to the first-floor room, they kissed and the defendant, now on his knees on the floor, put his finger in, and later licked, her vagina. He then told Molly to sit on top of him, which she did. He took out his penis, which went into her vagina when she sat on him, but she said it hurt and they stopped. She then went to the downstairs bedroom, where her sister lay awake on the upper bunk. The defendant later entered the room to say good night, kissing Molly on her lips while she lay on the lower bunk.
On cross-examination, Molly was shown portions of her videotaped interview. With respect to the first night, Molly admitted that she had told the nurse at the center that Martha did not see the defendant touching her breasts because Martha from her vantage point on the other couch could not see Molly *26and the defendant. But Molly agreed at trial that she and the defendant, while on the couch, would have been within Martha’s field of vision. Molly also admitted that she had made no mention of a blanket during her videotaped interview at the center. When the assistant district attorney and a detective sergeant interviewed Molly in Minnesota in October, 2009, she told them, “I don’t remember the first night.”
At trial, Molly initially said that she did not remember whether the defendant touched her at any other time during the week. But when the detective sergeant had interviewed Molly in March, 2008, she told him that the defendant had swiped his arm against her breasts approximately ten times while sitting with her and Martha at the piano. When asked at trial whether she recalled that actually happening, she replied, “Yes, because of the video I watched,” but then admitted that she had not mentioned this touching during the videotaped interview. At trial, she said that the touching at the piano happened only when Martha had left the room, but she had earlier told the detective sergeant that it happened while Martha was in the room.
Molly had also told the detective sergeant in March, 2008, of another incident with the defendant during her week-long vacation at his home where, three or four times, the defendant bmshed her thigh with his foot while they leaned against a table watching television. At trial, she did not remember whether that touching happened, or whether she had spoken of it during her interview with the detective sergeant.
Defense counsel informed the judge that he wished to advance two separate theories as to why Molly’s allegations of sexual abuse were not credible. First, he contended that at the time she made her first complaint to her godmother, Molly was seeking “attention” and “sympathy” because her adoptive family was in turmoil as a result of the conduct of her adopted sister Jane.6 She sought attention during a group discussion of sexual abuse by accusing her uncle of sexual abuse, because, as a result of *27Molly’s prior exposure to abuse before her adoption at the age of nine, it had been “drilled into her in the most traumatic time in her life that this is what uncles do.” She did not expect her allegation to cause a “blow up” because sexual abuse by an uncle was a “pretty ordinary” event in her preadoption childhood and none of her uncles had ever been prosecuted for such abuse.
This theory required the admission of evidence as to what Molly had seen, said, heard, or experienced regarding the sexual abuse committed against her and others by Molly’s biological father’s brothers. Defense counsel made a proffer to the judge of the evidence of abuse in support of this theory. He proffered that Molly’s biological mother had testified in a deposition that Molly, when she was approximately five years old, said that an uncle had touched her “in between her legs” when she was taking a bath. In a voir dire, Molly did not recall being sexually molested in her bath, but she testified that she did remember that an uncle had anally penetrated her with his penis while she lay on a bed when she was younger than six years old.7 Defense counsel further proffered that the biological mother had told Molly of an incident of abuse of Molly’s cousin by another one of her biological father’s brothers, where Molly’s aunt “woke up to the bed shaking back and forth to find [the uncle] on top of his niece.” And defense counsel proffered that, after the biological mother recognized she could no longer care for her children, she told Molly that she wanted her placed outside the family “because all of your uncles are child molesters.”
While the first theory suggested that Molly had initially fabricated what she instinctively understood to be a familiar allegation of sexual abuse against an uncle without appreciating the consequences of such an allegation, the defendant’s second theory suggested that Molly’s memory might be impaired because of a “dissociative memory” disorder. Defense counsel intended to present this theory to the jury through the expert testimony of a licensed psychologist, Dr. Daniel Brown.
At voir dire and in his report, Dr. Brown declared that children *28with “disorganized attachment” at the age of twenty months have “significantly more frequent and a greater range of dissociative behaviors” during childhood and adolescence. Dr. Brown stated that persons with a dissociative memory can sometimes manifest episodes of restricted awareness, where they can have large gaps in memory, or “rigid compartmentalization,” where “they have significant memory gaps for important life events and memory sort of comes and goes.” He testified that many children grow out of these dissociative behaviors but children who were physically or sexually abused in later childhood do not, because they use dissociation to deal with their abuse, which changes the dissociation from restricted awareness to rigid compartmentalization.
Dr. Brown further opined that, while recovered memories of abuse by those with dissociated memory are not generally inaccurate, there is a subgroup of persons with major dissociative symptoms “whose memory is very uncertain.” Some in this subgroup start to speculate when they do not have the memory available to them, and then engage in “expressed uncertainty,” where they fill in the blanks of what they cannot remember, either by inferring the memory from other information or making it up, which is known as “confabulation.” Dr. Brown noted that “source misattribution error” is a common error of memory, where someone hears or reads a story, incorporates it into their memory, and then believes it is a memory about their own life. Dr. Brown reported that others in this subgroup have a factitious disorder, which is manifested by “the simulation of symptoms and/or Ufe stories, not necessarily conscious, for the purpose of adopting a sick role or gaining attention.”
According to Dr. Brown, a child is most likely to develop the disorganized attachment that can lead to dissociative memory when the child’s mother has had children at age fourteen or earlier, is frightening to her children, and is “sort of present but out of it.” Dr. Brown noted that Molly’s biological mother had all three risk factors.8 He opined that Molly “meets the criteria for dissociative amnesia” because “she has significant gaps in *29her memory specifically for alleged abuse, assuming that it happened,” based on her lack of memory of the sexual abuse in the bathtub and her recovered memory of the sexual abuse in the bed when she was a young child.9 Dr. Brown also opined that “there are some red flags” as to factitiousness, but his evidence of this focused primarily on social service records that reported lies by her biological mother regarding her life story. Dr. Brown made clear that it was not possible for him to interview Molly and have her perform standardized testing, and that he would not make a diagnosis that Molly has a dissociative disorder without having interviewed her.10 But he opined that the record was “quite consistent” with Molly’s having dissociative symptoms and that “there are dissociative behaviors with respect to memory, and she clearly has that.”
The judge excluded the testimony of Dr. Brown and the biological mother, and also excluded evidence regarding what Molly had seen, said, heard, or experienced as a young child regarding her uncles’ alleged sexual abuses, with one exception: the judge permitted Molly to testify on cross-examination about inconsistencies in her recollection of the rape committed in a bed by an uncle when she under six years old. As a result of the evidentiary ruling allowing this testimony, the jury learned from Molly that she had told the nurse during the videotaped interview that she had never been abused by anyone other than the defendant, but at trial she recalled having been raped by an uncle when she was six years old or younger. Molly explained at trial that “I didn’t remember at the time” of the videotaped interview but “start[ed] remembering” the incident “[wjhen I went and *30talked to [the detective sergeant and assistant] district attorney.” Molly told the jury, “There are just times where an image just popped into my head.”11
The judge ruled that this evidence was admissible solely as an inconsistent prior statement and instructed the jury, over the defendant’s objection, that it was admitted for the limited purpose of assisting the jury in their evaluation of Molly’s credibility as to what happened at the defendant’s home.12
The judge explained her reasons for denying the defendant’s motion to admit the expert testimony of Dr. Brown, the testimony of Molly’s biological mother, and evidence of prior instances of abuse. First, the judge found that there was no evidence that Molly was currently suffering from a psychiatric disorder and the records “indicate the opposite.”13 The judge declared that *31Dr. Brown’s opinion was “wholly speculative,” noting that he based his opinion on the premise that Molly suffers from disorganized attachment but “has no idea whether or not [Molly] has any current difficulty in forming attachments.”14 She also noted that he had never seen Molly and was basing his opinion on records from at least nine years earlier.15 Second, she declared that Dr. Brown’s opinion “falls clearly on that side of the line of forbidden testimony” by intruding on the jury’s function of determining the credibility of witnesses. Third, she declared that the defendant “has been given plenty of latitude and has fully exploited every opportunity to probe memory lapses as to this incident.” While she recognized that the defendant would not be able to “probe prior sexual abuse and memory lapses as to those events, . . . we’re not on trial here for those events.” For these reasons, she concluded that the risk of prejudice arising from this testimony outweighed its probative value.
Discussion. Where a party in a criminal trial seeks to offer an expert opinion, the judge, as gatekeeper, must first determine whether the proponent of the evidence has met the five foundational requirements for admissibility: (1) that the expert testimony will assist the trier of fact because the information is beyond the common knowledge of jurors; (2) that the witness is qualified as an expert in the relevant area of inquiry; (3) that the expert’s opinion is based on facts or data of a type reasonably relied on by experts to form opinions in the relevant field; (4) that the theory underlying the opinion is reliable; and (5) that the theory is applied to the particular facts of the case in a reliable manner. Commonwealth v. Barbosa, 457 Mass. 773, 783 (2010), cert. denied, 131 S. Ct. 2441 (2011). “Once admitted, the *32validity and credibility of the expert testimony is subject to challenge like any other testimony, including through the admission of opposing expert testimony, and it is for the jury to determine what aid it might provide to their deliberations.” Commonwealth v. Shanley, 455 Mass. 752, 762 (2010) (Shanley).
At trial, the Commonwealth conceded the reliability of the “general science” described by Dr. Brown.16 But it contended that Dr. Brown’s opinion regarding dissociative memory and confabulation was not relevant in this case, because there was not sufficient evidence that Molly had a dissociative disorder.
When a judge, in her role as gatekeeper under Commonwealth v. Lanigan, 419 Mass. 15, 25-26 (1994), determines the admissibility of an expert opinion regarding a psychological disorder, the judge must first decide whether the expert’s opinions regarding the disorder are reliable, i.e., whether the existence of the disorder and recognition of its symptoms are generally accepted in the relevant psychological community or are otherwise demonstrated to be reliably established. See Shanley, supra at 761-762. Where an opinion regarding a psychological disorder and its symptoms is reliable, the judge must then determine whether the opinion evidence is relevant to an issue in the case and, if relevant, weigh its relevance against the risk the evidence will be unfairly prejudicial; will confuse, divert, or mislead the jury; or will be unnecessarily time consuming. See generally Commonwealth v. Bonds, 445 Mass. 821, 831 (2006); Mass. G. Evid. § 403 (2012). In this case, the judge determined that Dr. Brown’s expert testimony regarding dissociative disorder was not relevant (or that any relevance was outweighed by the risk of juror confusion) because there was insufficient evidence to permit the jury to conclude that Molly had the disorder. We review her evidentiary decision under the abuse of discretion standard. See Shanley, supra at 762, citing Canavan’s Case, 432 Mass. 304, 312 (2000).
*33Where a defendant seeks to admit expert testimony regarding the credibility of an alleged victim’s testimony, especially where, as here, the Commonwealth’s case rests almost entirely on the credibility of the alleged victim, a judge’s evidentiary decision assumes a constitutional dimension because our “Constitution requires that a defendant be permitted to introduce evidence which may materially affect the credibility of the [alleged] victim’s testimony.” Commonwealth v. Ruffen, 399 Mass. 811, 816 (1987). See Commonwealth v. Joyce, 382 Mass. 222, 229 (1981), and cases cited. This right is grounded in both the Sixth and Fourteenth Amendments to the United States Constitution, as well as art. 12 of the Massachusetts Declaration of Rights, which provides that “every subject shall have a right to produce all proofs, that may be favorable to him.” See Commonwealth v. Louraine, 390 Mass. 28, 33 (1983), quoting Washington v. Texas, 388 U.S. 14, 19 (1967) (“few rights are more fundamental in our jurisprudence than that of an accused ‘to present . . . [his] version of the facts’ ”).
In the context of this case, the defendant was not required to offer in evidence an expert diagnosis that Molly had dissociative memory in order to establish the relevance of Dr. Brown’s expert opinion that those with dissociative memory sometimes have a distorted memory of past events. We found no abuse of discretion in the admission of comparable testimony by the Commonwealth regarding a dissociative disorder in Shanley without a diagnosis that the victim suffered from dissociative amnesia. See Shanley, supra at 757, 766. And we have rejected a defendant’s argument that expert testimony regarding the general behavioral characteristics of sexually abused children should only have been admitted by the Commonwealth in response to a hypothetical question related to the facts of the case, whether assumed or in evidence. Commonwealth v. Dockham, 405 Mass. 618, 627-628 (1989). See Commonwealth v. Goetzendanner, 42 Mass. App. Ct. 637, 641-646 (1997) (expert testimony regarding behavioral and emotional characteristics common to victims of battering without diagnosis that victim suffered from battered women syndrome properly admitted). If such a diagnosis were required, expert evidence regarding dissociative disorder would rarely be available to either a defendant *34or the Commonwealth, because a judge has no authority to order a psychological examination of a witness to assess the witness’s credibility. Commonwealth v. Widrick, 392 Mass. 884, 884-885 (1984). Without such an examination, a psychologist, as in this case, may not be able to provide a diagnosis, and may not even consider it ethical to attempt one. Nor will there commonly be evidence of such a diagnosis. A witness with a dissociative memory disorder may not have received psychological counselling and, where she did, those confidential records may not be discoverable. Commonwealth v. Bishop, 416 Mass. 169, 179-183 (1993). Even where, as here, the defendant has access to the witness’s psychological records, the absence of a diagnosis of dissociative memory in the records is of little consequence where, as Dr. Brown testified, the counselling was provided by social workers and therapists who did not appear to consider such a diagnosis.
In determining whether Dr. Brown’s expert testimony was relevant in this case in the absence of such a diagnosis, it must be remembered that the defendant need not prove that Molly had a dissociative disorder to prevail at trial; the defendant need only raise a reasonable doubt in a rational jury’s mind as to the reliability of her testimony, which he may accomplish with evidence permitting a reasonable inference that her memory may be distorted by a dissociative disorder. Therefore, a defendant may establish the relevance of reliable expert testimony regarding a dissociative disorder by presenting evidence regarding the alleged victim and her personal history that is consistent with the diagnosis, and that is sufficient to permit a reasonable inference that the alleged victim may have the disorder. See Commonwealth v. Sheehan, 435 Mass. 183, 189 (2001) (mental health records admissible where they “provide a basis for cross-examination about the possibility that these earlier events in [the alleged victim’s] life caused him to fantasize and have problems separating reality from fiction”). Cf. Commonwealth v. Wilson, 441 Mass. 390, 401 (2004) (expert witness may testify that certain quantity of drugs is consistent with possession with intent to distribute).
In Shanley, supra at 756-759, the Commonwealth sought to admit expert testimony to explain how the victim of sexual abuse in that case could have recovered memories of the sexual *35abuse he suffered between the ages of six and thirteen that had been lost for twenty years. The expert was allowed to testify that, among adults who had been seriously traumatized, approximately twenty per cent suffer from dissociative amnesia. Id. at 758-759. We concluded that the judge did not abuse his discretion in admitting this expert testimony to assist the jury in determining the credibility of the victim’s testimony of recovered memory, and the reliability of those memories. Id. at 757, 766.
Just as the Commonwealth in Shanley was permitted to offer expert testimony about dissociative memory disorder to educate the jury about this psychological condition in seeking to demonstrate the reliability of the victim’s testimony, id., so, too, here the defendant should have been able to offer comparable expert evidence to demonstrate the unreliability of Molly’s testimony. In Shanley, the victim’s testimony of recovered memory of childhood sexual abuse by a priest was consistent with such a diagnosis. Id. Here, Molly’s testimony about the defendant’s alleged rapes suddenly “popping into” her head, her loss of memory as to what happened on the first day of the defendant’s touching, her recovered memory of being sexually abused in a bed when she was a young child, and her lack of memory of having told her biological mother that she was sexually assaulted as a young child in a bathtub17 were consistent with a diagnosis of dissociative memory disorder.18 Because the evidence in the record was consistent with the diagnosis of a dissociative disorder, and was sufficient to permit a reasonable inference that Molly may suffer from the disorder, we conclude the judge abused her discretion in barring Dr. Brown from testifying about the risk of confabulation arising from dissociative memory.19 See State v. Lujan, 192 Ariz. 448, 452 (1998) *36(“Just as the prosecution . . . could use expert testimony about the behavioral characteristics of sexually abused children to explain the inconsistencies in a child’s statements, when appropriate under the facts of a particular case, the defense may use such testimony to show a child’s possible misperceptions”).
Properly limited, such expert testimony does not intrude on the jury’s function of determining the credibility of witnesses. No witness, expert or not, may offer an opinion as to the credibility of another witness. Commonwealth v. Montanino, 409 Mass. 500, 504 (1991), and cases cited. While the line “between permissible and impermissible opinion testimony in child sexual abuse cases is not easily drawn,” Commonwealth v. Richardson, 423 Mass. 180, 186 (1996), an expert witness does not cross that line where he educates the jury about the causes of dissociative disorder, the prevalence of the disorder, the symptoms of the disorder, and the risk of confabulation arising from “expressed uncertainty” and “source misattribution error.” See Shanley, supra at 758-759, 766 (expert testimony admissible regarding dissociative amnesia, how it works, its causes, and its prevalence among those with serious trauma); Commonwealth v. Frangipane, 433 Mass. 527, 535 n.12 (2001) (“no error in permitting the witness to testify to the symptoms, including dissociative memory loss and recovered memory, of sexually abused children”).
The line would be crossed if Dr. Brown, in testifying before the jury, were explicitly to opine that Molly’s memory of events at the defendant’s home was unreliable as a result of her dissociative disorder. See Commonwealth v. Richardson, supra (“impermissible vouching” where “witness explicitly links the opinion to the experience of the witness”); Commonwealth v. Ianello, 401 Mass. 197, 202 (1987) (“expert may not render an opinion on the credibility of a witness”). Because of this limitation, an expert’s opinion that touches on the credibility of a fact witness must be accompanied by other evidence that permits the *37jury to apply the information learned from the expert in deciding for themselves whether they find the fact witness credible. Here, the jury needed to learn what Molly said, saw, heard, and experienced regarding her prior sexual abuse to evaluate the risk that Molly had dissociative disorder and that her memory of what happened at the defendant’s home was unreliable because of confabulation. The judge erred in excluding this evidence.20 For the same reason, the judge also erred in limiting Molly’s testimony about her inconsistent memory regarding her sexual abuse by an uncle when she was a young child. This evidence was relevant to the defendant’s theory of confabulation arising from dissociative memory, so the jury should not have been instructed, in essence, that her lapse of memory is inconsequential unless the jury found it to be intentional.
The Commonwealth contends that evidence of the prior sexual assaults against Molly is inadmissible under the rape shield statute, G. L. c. 233, § 21B. Under that statute, “[ejvidence of specific instances of a victim’s sexual conduct. . . shall not be admissible except evidence of the victim’s sexual conduct with the defendant or evidence of recent conduct of the victim alleged to be the cause of any physical feature, characteristic, or condition of the victim . . . .”21
We have recognized, however, that where the rape shield *38statute is in conflict with a defendant’s constitutional right to present evidence that might lead the jury to find that a Commonwealth witness is lying or otherwise unreliable, the statutory prohibition must give way to the constitutional right. See Commonwealth v. Harris, 443 Mass. 714, 721 (2005) (defendant may introduce evidence of complainant’s past sexual conduct notwithstanding rape shield statute “where that conduct is relevant to the complainant’s bias or motive to fabricate”); Commonwealth v. Ruffen, 399 Mass. 811, 816 (1987) (“despite the general statutory policy prohibiting inquiry into a victim’s prior sexual experiences, the Constitution requires that a defendant be permitted to introduce evidence which may materially affect the credibility of the victim’s testimony”); Commonwealth v. Joyce, 382 Mass. 222, 229 (1981), quoting Commonwealth v. Haywood, 377 Mass. 755, 760 (1979) (“The right to cross-examine a complainant in a rape case to show a false accusation may be the last refuge of an innocent defendant. ‘A defendant has the right to bring to the jury’s attention any “circumstance which may materially affect” the testimony of an adverse witness which might lead the jury to find that the witness is under an “influence to prevaricate” ’ ” [emphasis in original]).
Here, the defendant sought to admit evidence of Molly’s sexual abuse by her uncles to demonstrate the significant possibility that Molly suffered from dissociative memory and confabulated her memory of the defendant’s alleged sexual assaults by recovering a dim memory of sexual abuse by her biological father’s brother, confusing the source of the abuse with her adoptive mother’s brother (the defendant), and inferring facts to fill in the blanks of her memory. Because such evidence, if credited, would materially affect the jury’s evaluation *39of Molly’s credibility and reliability, and because it was not cumulative of other admitted evidence, cf. Commonwealth v. Frey, 390 Mass. 245, 251-252 (1983) (no constitutional violation in excluding evidence under rape shield statute where other “ample evidence of the complainant’s bias and motive to he was presented to the jury”), we conclude that the defendant was constitutionally entitled to present the evidence regardless of the prohibition in the rape shield statute. See Commonwealth v. Baxter, 36 Mass. App. Ct. 45, 51-52 (1994) (evidence of prior rape improperly excluded under rape shield statute where defendant “was seeking to show that the complainant had been victimized, that she was suffering from psychiatric problems as a result of that assault, and that because of those problems and the many remarkable similarities of that trauma to the present incident, she was unable to distinguish between the two situations”).
The error was prejudicial and requires a new trial. The Commonwealth’s case rested wholly on the credibility of Molly, and the exclusion of evidence that may have significantly affected the jury’s evaluation of her credibility regarding the defendant’s alleged rapes was not harmless. See Commonwealth v. Sheehan, 435 Mass. 183, 190 (2001) (exclusion from evidence of complainant’s psychiatric records required new trial where Commonwealth’s case rested “almost entirely” on his testimony and evidence, if believed, might have had significant effect on outcome of trial).
Because the issue may arise at a new trial, we briefly address the defendant’s claim that the prosecutor’s closing argument was improper in asking the jury, “What motive does [Molly] have to lie?” The defendant relies on our statement in Commonwealth v. Beaudry, 445 Mass. 577, 587 (2005), quoting Commonwealth v. Riberio, 49 Mass. App. Ct. 7, 10 (2000), that “[t]elling the jury that the victims have no reason to lie is over the line of permissible advocacy . . . .” But in the Beaudry case we declared that a prosecutor may not argue that a victim is credible simply because she appeared to testify in court. Id. See Shanley, supra at 777. We recognized that a prosecutor may marshal the evidence in closing argument to “urge the jury to believe the government witnesses and disbelieve those testifying for the defendant.” Commonwealth v. Beaudry, supra. Where, *40as here, defense counsel in his closing argument challenged the credibility of the alleged victim, a prosecutor acts properly in inviting the jury to consider whether the victim has a motive to lie, and identifying evidence that demonstrates that the victim’s testimony is accurate and reliable. See Shanley, supra.
Conclusion. Because the judge erred in excluding the expert testimony of Dr. Brown and the evidence of childhood sexual abuse necessary to apply the expert opinion to the facts of this case, and because the error was prejudicial, we reverse the defendant’s convictions and remand the case to the Superior Court for a new trial.
So ordered.
The defendant was sentenced to from four to five years in State prison on the first indictment, and to five years’ probation, to commence on the conclusion of his incarceration, on the second indictment. In addition to the statutory conditions of probation, G. L. c. 265, § 47, the judge ordered the defendant to comply with five special conditions of probation: that he enter and complete a *24sex offender treatment program, submit to an evaluation for alcohol abuse and receive any treatment deemed necessary, have no unsupervised contact with children under sixteen years of age (except his own children), engage in no employment or volunteer work with children under sixteen years of age, and stay away from the alleged victim and her family. A single justice of this court allowed the defendant’s motion for a stay of execution, and we affirmed his decision. Polk v. Commonwealth, 461 Mass. 251 (2012).
We use pseudonyms for the children.
Molly and her brother were adopted by her adoptive parents in Minnesota when she was nine years old. Her adoptive parents also had separately adopted another girl, and had given birth to two children. Molly’s adopted sister, Jane, had serious behavioral issues and had not lived in Molly’s home since May, 2006. Jane did not accompany the family when they visited the defendant in August, 2007.
The defendant was not charged with any crime related to this alleged incident.
While at the defendant’s home, Molly shared a bedroom in the basement with one of her sisters.
Jane is two months younger than Molly and once had shared a bedroom with Molly. Jane had been running away from home, was jailed for stealing from the family, and later was placed with another family. In January, 2008, Molly’s adoptive father wanted to see Jane, but Molly’s adoptive mother did not want him to, and Molly’s performance in school had slipped because of the trouble at home.
Molly told the assistant district attorney and the detective sergeant that the sexual assault may have been committed by a cousin rather than an uncle but testified at trial that the abuse was committed by an uncle.
Molly’s biological mother testified at a deposition that she gave birth to Molly when she was twelve years old. The social service records show that the biological mother was prone to violent outbursts. A social worker’s observation of the biological mother with Molly’s younger sister showed that the biological *29mother was “sort of there but out of it” even when she knew she was being observed.
Molly testified at voir dire that she had forgotten about the bed rape after it occurred but then remembered the bed rape in “seventh or eighth grade,” but then had forgotten the memory again and did not mention it in her videotaped interview at the center, where she had instead stated that she had never been sexually abused by any other person in her life. Molly recalled the incident again three months before trial in October, 2009, when she was interviewed by the assistant district attorney and the detective sergeant. In that interview, Molly described her recollection of the bed rape as “something happens, and you don’t think about it, and then one day, you just remember it.”
Dr. Brown asserted that it would not be ethical to make such a diagnosis without having interviewed her.
The judge did not allow the defendant to elicit from Molly that she had remembered the incident in “seventh or eighth grade” and then forgot it before she remembered it again. When asked at trial, “[H]ow long have you remembered [the bed rape]?” Molly answered, “I don’t know.”
The jury appeared to struggle with the meaning of this limiting instruction during their deliberations. After the jury had informed the judge that they were at an impasse and the judge had instructed the jury in accordance with Commonwealth v. Rodriquez, 364 Mass. 87, 101 (1973) (Appendix), the jury informed the judge: “There is disagreement among the jury as to the extent we can use the testimony regarding [Molly’s] prior sexual abuse by an uncle. Are we limited to determining whether the interview with [the nurse] alone is credible or can we use it to question [Molly’s] overall state of mind?”
The judge instructed the jury, again over the defendant’s objection, that the significance of the inconsistency “depends largely on why you think [Molly] made that prior statement.” The judge explained:
“Was her statement to the interviewer that she had not been abused an intentional misstatement? In that case, you may ask yourselves whether, with regards to the events of August, 2007, she is not telling the truth there as well. Alternatively, was it simply that [Molly] did not remember the prior abuse at the moment she was asked that question in the interview? In that case, her testimony at trial may be seen simply as a correction of a past misstatement. . . that was due to an unintentional lapse of memory.”
The judge then added: “You are not to use this evidence of prior sexual abuse for any other purpose. More specifically, there is no evidence in this case that ties the events of ten or twelve years ago to the events of August, 2007. Moreover, [Molly’s] state of mind, apart from her credibility, is not relevant in this case.”
The records the judge appeared to be referring to were the records prepared *31by a social worker who was providing psychological counselling to Molly in 2008 after her disclosure of the defendant’s rapes.
The previous day, the judge articulated her concern with Dr. Brown’s opinion:
“You say, ‘Here are the characteristics of the disorder. I can’t make the diagnosis, but you the jury, you be the expert. You make the diagnosis and put it together.’ That is essentially . . . what you are asking the jury to do.”
Molly was eighteen years old at the time of trial, and nine years of age at the time of her adoption, so the judge appears to be referring to Dr. Brown’s reliance on records that preceded her adoption.
Dr. Brown had earlier testified on behalf of the Commonwealth at a hearing pursuant to Commonwealth v. Lanigan, 419 Mass. 15, 25-26 (1994), in the case of Commonwealth v. Shanley, 455 Mass. 752, 763 (2010), regarding the theory, conditions, and symptoms of dissociative amnesia and the reliability of recovered memory. The Commonwealth defended the admissibility of expert testimony on these subjects in the appeal in the Shanley case at the time of the defendant’s trial. We concluded that the trial judge in Shanley did not abuse his discretion in admitting the expert testimony. Id. at 766.
WhiIe no evidence was presented at trial that this abuse actually occurred, there is evidence in the record that Molly told her biological mother about the abuse in the bath, because at her deposition, the biological mother testified that she reported the incident to police immediately after the incident. Other records show she later reported the incident again to the State child services protective agency.
In addition, to the extent that disorganized attachment leads to dissociative memory, there was evidentiary support in the record that Molly had suffered disorganized attachment by the age of twenty months based on evidence from the biological mother and about the biological mother.
Because the evidence in the record was not consistent with a “factitious *36disorder” and was not sufficient to permit a reasonable inference that Molly may suffer from that disorder, we conclude that the judge did not abuse her discretion in barring testimony regarding factitious disorder. Our conclusion that the judge did not abuse her discretion as to this testimony does not bind the judge on retrial again to bar such testimony, especially if the defendant makes a stronger preliminary showing that Molly may suffer from this disorder.
The Commonwealth contends that the biological mother’s testimony as to what Molly said to her about an uncle sexually abusing her in a bathtub is inadmissible hearsay, but the evidence is admissible regardless of its truth, because it is relevant to Molly’s apparent belief at the time that her uncle had abused her and her consequent fear of her uncle. See Commonwealth v. Montanez, 439 Mass. 441, 447-448 (2003) (evidence of victim’s statement to friend admissible to establish her fear of defendant). See generally Mass. G. Evid. § 801(c) (2012). The Commonwealth also contends that Molly’s biological mother is incompetent to testify, but no competency hearing was conducted and no finding made as to competency. If there is evidence at retrial that the biological mother is incompetent to testify, the judge may conduct a hearing and make appropriate findings.
General Laws c. 233, § 21B, provides:
“Evidence of the reputation of a victim’s sexual conduct shall not be admissible in any investigation or proceeding before a grand jury or any court of the commonwealth for a violation of [G. L. c. 265, §§ 13B, 13B½, 13B¾, 13F, 13H, 22, 22A, 22B, 22C, 23, 23A, 23B, 24, and 24B,] or [G. L. c. 272, § 5], Evidence of specific instances of a victim’s sexual conduct in such an investigation or proceeding shall not be *38admissible except evidence of the victim’s sexual conduct with the defendant or evidence of recent conduct of the victim alleged to be the cause of any physical feature, characteristic, or condition of the victim; provided, however, that such evidence shall be admissible only after an in camera hearing on a written motion for admission of same and an offer of proof. If, after said hearing, the court finds that the weight and relevancy of said evidence is sufficient to outweigh its prejudicial effect to the victim, the evidence shall be admitted; otherwise not. If the proceeding is a trial with jury, said hearing shall be held in the absence of the jury. The finding of the court shall be in writing and filed but shall not be made available to the jury.”