The opinion of the Court was delivered by

This appeal concerns the use of expert opinion testimony to aid jurors in the criminal trial of a child-sexual-abuse case. The specific issue concerns expert-opinion evidence premised on the Child Sexual Abuse Accommodation Syndrome (CSAAS), and whether there is a reliable scientific explanation for certain exhibited characteristics of an abused child, such as acceptance of the abuse or delayed reporting, that would help jurors understand why a child victim would not complain to a parent or other authority figure about the abuse. We hold that CSAAS has a sufficiently reliable scientific basis to allow an expert witness to describe traits found in victims of such abuse to aid jurors in evaluating specific defenses. In this case, the expert’s opinion went beyond that limited scope and included opinions on commonplace issues, such as credibility assessments derived from conflicting versions of an event and not-yet scientifically established opinions on the ultimate issues that are for jury resolution. Although the evidentiary questions arise as a matter of plain error, we agree with the Appellate Division that the introduction of such evidence was clearly capable of producing an unjust result and we thus affirm the Appellate Division’s judgment ordering a new trial.

I

Facts and Procedural History

The background to the case is regrettably familiar, a story of childhoods unhinged by events so traumatic that even the participants cannot contemplate them. When first confronted with the possibility that defendant might have sexually abused his daughters, their mother was incredulous. After the girls *557first told their mother of the alleged abuse, she cautioned them that it was important to tell the truth about their father. The girls said that they were telling the truth. Their mother testified that she had warned her daughters never to let a stranger touch them but had never told them about sexual abuse from a father, because she never thought it could happen to her children. Defendant was equally insistent that committing sexual acts with children and, in particular, his daughters is unimaginable. He testified that anybody who engaged in sexual relations with his own children would have to be “sick” and that he did not believe he was sick. Whether defendant is sick or not, the jury has found that abuse occurred.

Rather than use the initials mandated by law, we shall use fictitious names to describe the parents and children involved. We shall refer to the mother as “Karen,” the father as “John,” and the two children as “Connie” and “Norma.” The parents appear to be of different cultures — Karen is from the midwest and John is a recent arrival to the continental United States. They met in Indiana in 1973 or 1974 and started their life together there when Karen was thirteen and John was nineteen or twenty. John already had a child at that time. They soon moved to Brooklyn and later settled in Newark. Connie and Norma were bom in 1977 and 1979, respectively. John and Karen never married. Although John worked as an auto mechanic and a golf-greens attendant during their years in New Jersey, he did not provide sufficient financial support for the household; thus, Karen received public assistance for the two children. Theirs was a tempestuous union marked by recriminations that each had been unfaithful. Their relationship deteriorated in late 1984 when John separated from Karen. During the separation, which lasted approximately four months, Karen became pregnant by another man. The breakup became final in late 1985 or early 1986 when John went to Brooklyn to live in an apartment with his brother. At the time of the breakup, Connie and Norma were approximately eight and six years old, respectively.

*558After the breakup, John customarily picked up the children and took them to Brooklyn for weekend visits. John was then living in a one-room apartment with another woman, whom he married in 1987. About two years after the separation, Karen learned that Norma, during play, had attempted to pull down her younger sister’s underwear and touch her buttocks. Karen asked Norma where she had ever learned of such things and Norma reluctantly identified the person who had initiated her into such conduct by spelling out the word “D-A-D”.

Although at first disbelieving, Karen consulted a family counsellor and eventually reported the incident to the police. Both Connie and Norma reported that they had been the victims of repeated acts of sexual abuse by their father in the Newark apartment in 1984 as well as during their visits to Brooklyn. An Essex County grand jury returned an indictment charging John with acts of criminal sexual abuse in New Jersey on his children between January 1, 1984, and December 31, 1984.

Before trial, the prosecutor moved, pursuant to N.J.S.A. 2A:84A-32.4, to have Connie’s and Norma’s testimony taken on closed-circuit television. In ruling on the motion, the trial court heard testimony from Madeline Milchman, a Ph.D. in psychology with a concentration in developmental or child psychology. Dr. Milchman had interviewed the children and had reviewed other information on them as well. The court first qualified Dr. Milchman, without objection, as an expert witness in the areas of child psychology and child sexual abuse. Dr. Milch-man essentially testified that both infant witnesses would suffer severe emotional distress if forced to testify before spectators, jurors, and especially their father in the courtroom setting. The trial court granted the motion, ordering that Norma and Connie testify through a closed-circuit system outside the presence of jurors, spectators, and defendant.

At trial, both Connie and Norma described, in graphic detail, the abuses committed on them involving sexual penetration and *559oral sexual contact. (Presumably, the instances in Brooklyn were testified to under an Evidence Rule 55 analysis.) A pediatric resident who conducted a genital examination of Norma testified that she found that the child had a stretched hymenal opening, an abnormal condition for a seven-year-old girl. The medical evidence relating to Connie, entered by way of stipulation, also revealed a stretched hymen. Karen described discharges that she had observed on Norma’s underwear but said that she had attributed them to Norma’s not changing her underwear.

Dr. Milchman was called to the stand again at trial and qualified, without objection, as an expert witness on child sexual abuse. She identified the child sexual abuse accommodation syndrome “as a pattern of behavior that is found to occur again and again in children who are victims of incest.” She described the various aspects of CSAAS and related them to behavior she had observed in Connie and Norma. Dr. Milchman also testified about how she assesses the veracity of an alleged victim of child sexual abuse. At the conclusion of her direct testimony, Dr. Milchman stated that in her expert opinion, Connie and Norma had been sexually abused.

The theory of the defense was that Karen had put the children up to this story to avenge her loss of John. Defendant offered evidence that the Newark apartment housed eight people and that there was no isolated occasion during which such abuse could have occurred.

The jury convicted defendant of multiple counts of first-degree aggravated sexual assault on Connie and Norma for various acts of penetration and oral sex, and of two counts of endangering the welfare of a child. The court sentenced defendant to thirty years’ imprisonment, with ten years of parole ineligibility. The Appellate Division reversed the convictions, finding that the trial court had committed plain error in permitting the use of the CSAAS testimony to establish the credibility of the witnesses rather than for other limited purposes for *560which it is generally reliable, i.e., to explain secrecy, belated disclosure, and recantation by a child-sexual-abuse victim. State v. J.Q., 252 N.J.Super. 11, 599 A.2d 172 (1991). The Appellate Division specifically held that

syndrome evidence, including CSAAS, is not reliable to prove the occurrence of sexual abuse, and that absent a question of capacity, a social science expert lacks the qualifications to render an opinion as to the truthfulness of a statement by another witness. Because the expert in this case testified before the jury as to syndrome evidence to prove that sex abuse occurred; opined as to the truthfulness of the children (and their mother), and rendered the opinion that the children were abused based in great measure upon these two interdicted classes of evidence, we are satisfied that the admission of her testimony was error clearly capable of producing an unjust result. [Id. at 15, 599 A.2d 172.]

One member of the panel dissented, believing that a case could be made in an Evidence Rule 8 [hereinafter Rule 8] hearing for the scientific reliability of CSAAS evidence as substantive evidence of abuse, thus obviating the need for a reversal of the judgment. The dissent impliedly requires consideration of categories of expert opinion evidence other than CSAAS testimony in child-sexual-abuse cases.

We granted the State’s petition and defendant’s cross-petition for certification, 127 N.J. 562, 606 A.2d 372 (1992). The State’s petition exceeded the scope of an appeal as of right under Rule 2:2 — 1(a)(2) on those issues as to which there is a dissent in the Appellate Division. Rather, the State requested us to review “all issues relating to psychological and social science expert testimony regarding familial child sexual abuse complainants.” Defendant’s cross-petition argues against admission of expert testimony based on behavioral science theories.

II

Types of Expert Testimony in Child-Sexual-Abuse Prosecutions

In recent terms of Court we have had to consider various evidentiary problems in criminal prosecutions of child-sexual-abuse cases. See State v. Crandall, 120 N.J. 649, 577 A.2d 483 *561(1990) (finding constitutional N.J.S.A. 2A:84A-32.4, which permits closed-circuit television testimony outside physical presence of alleged abusers); State v. D.R., 109 N.J. 348, 537 A.2d 667 (1988) (proposing new hearsay exception for child’s out-of-court statement concerning acts of sexual abuse); State v. R.W., 104 N.J. 14, 514 A.2d 1287 (1986) (setting forth standards for use of expert psychological evaluation of child victim witnesses). Whether society has changed or whether society is finally confronting its deepest flaws is something that we cannot answer. Freud mistakenly believed that his female patients’ experiences of sexual abuse as children were merely fantasies and attributed their complaints to hysteria. Chandra Lorraine Holmes, Child Sexual Abuse Accommodation Syndrome: Curing the Effects of a Misdiagnosis in the Law of Evidence, 25 Tulsa L.J. 143, 144 (1989) [hereinafter Holmes]. We know now, from the many accounts, that such abuse in the home is regrettably all too real. “Since the 1970s * * *, there has been a growing awareness of child sexual abuse and reports of it have skyrocketed.” Diana Younts, Evaluating and Admitting Expert Opinion Testimony in Child Sexual Abuse Prosecutions, 41 Duke L.J. 691, 693 (1991) [hereinafter Younts] (citing L.B. Suski, Child Sexual Abuse: An Increasingly Important Part of Child Protective Service Practice, 3 Protecting Children 3 (1986), which shows an increase to 200,000 reports nationwide in 1986, from 7,559 reports in 1976). At the same time, we know that moving from the child’s world into the courtroom presents legal issues of varying dimension. For example, in State v. D.R., supra, 109 N.J. 348, 537 A.2d 667, we recommended that the Legislature adopt an exception to the hearsay rule for the admission of out-of-court statements by a child witness provided that the child, if available, testify at trial (either through closed-circuit television or in the courtroom) or that there is other corroboration if the child is unavail*562able.1 We believed that the availability of the child for questioning would afford the jury the opportunity to assess the child’s credibility while safeguarding the defendant’s right of confrontation and cross-examination. Id. at 370-71, 537 A.2d 667. Society must tread a measured path that avoids ignoring the reality of child sexual abuse and avoids as well the possibility of unjust conviction of this most shameful of crimes. In Maryland v. Craig, 497 U.S. 836, 110 S.Ct. 3157, 111 L.Ed.2d 666 (1990), four members of the Supreme Court cautioned that courts should be particularly insistent in protecting innocent defendants in child-sexual-abuse cases because of the reliability problems created by children’s suggestibility in child-sexual-abuse prosecutions. Id., 497 U.S. at 868, 110 S.Ct. at 3175-76, 111 L.Ed.2d at 693-94 (Scalia, Brennan, Marshall & Stevens, JJ., dissenting).

A.

At the outset, we must carefully distinguish the issues presented in this case from those that are not presented. There are various categories of expert testimony on child sexual abuse. For purposes of this analysis, we draw on the survey of issues by John E. B. Myers et al., Expert Testimony in Child Sexual Abuse Litigation, 68 Neb.L.Rev. 1 (1989) [hereinafter Myers].2

*563The authors point out that some expert testimony is routinely accepted and presents no genuine evidentiary problem. For example, expert medical testimony plays an important role in child-sexual-abuse litigation. Such testimony is based on a physician’s clinical diagnostic examination and the child’s medical history. Courts have permitted expert medical witnesses to describe the results of the examination and to offer opinions as to the cause of any injuries, to establish penetration, and to answer questions whether injuries could have been inflicted in a particular way or whether a caretaker’s explanation for an injury is reasonable. Myers, supra, 68 Neb.L.Rev. at 48-49.

In the behavioral-science field, the authors identify one type of expert testimony that describes behaviors commonly observed in sexually-abused children. Id. at 51-69. By way of background, the authors describe a wealth of literature that documents general patterns of behavioral and emotional reactions found in clinical samples of sexually-abused children, typically victims of incest. Id. at 52-62. The authors’ review of the literature, however, leads them to conclude that the effects of sexual abuse vary among children. Id. at 61. Furthermore, children who exhibit symptoms of fear, anxiety, or avoidance are probably suffering the effects of some traumatic experience, of which sexual abuse is only one of many possible causes. Nevertheless, some behaviors, such as age-inappropriate sexualized responses, are more associated with sexual abuse than others. Ibid. Thus, the authors assert that situations in which sexual abuse is likely can be identified.

Given the prevalence of sexual abuse, and its documented association with anxiety symptoms, however, abuse should be considered and evaluated through direct inquiry. When symptoms of fear, anxiety, or avoidance accompany a credible report of sexual abuse, sexual abuse must be seriously considered.

While no symptom or set of symptoms is conclusive proof of sexual abuse, when symptoms evidencing abuse are present in conjunction with a report that bears indicia of reliability, the clinician is justified in forming a clinical opinion that a child has been sexually abused. [Myers, supra, 68 Neb.L.Rev. at 61-62.]

Despite the considerable basis for this behavioral-science evidence, “most courts do not approve such testimony as substan*564tive evidence of abuse.” Myers, supra, 68 Neb.L.Rev. at 65, 68 (citing People v. Bowker, 203 Cal.App.3d 385, 249 Cal.Rptr. 886, 890-92 (1988), Lantrip v. Commonwealth, 713 S.W.2d 816, 817 (Ky. 1986), State v. Black, 537 A.2d 1154, 1156-57 (Me. 1988)). However, this type of testimony has an important nonsubstantive purpose of which the majority of courts approve. It can be used on rebuttal “to rehabilitate” the victim’s testimony when the defense asserts that the child’s delay in reporting the abuse and recanting of the story indicate that the child is unworthy of belief. Myers, supra, 68 Neb.L.Rev. at 51, 86-92.

Another area of behavioral-science testimony seeks to address the ultimate question of whether a child was in fact sexually abused. Myers, supra, 68 Neb.L.Rev. at 69-86. Such testimony would be based on the clinical observations of a professional “trained in the patterns, effects, and dynamics of child sexual abuse.” Id. at 73. The authors say that many experts now believe that the study of child sexual abuse has advanced to such a point as to enable “qualified professionals” to “determine whether a child’s symptoms and behavior are consistent with sexual abuse.” Id. at 73-75. For example, in 1985 the Council on Scientific Affairs of the American Medical Association published a set of guidelines that listed, in particular, physical and behavioral signs identified with child sexual abuse. Id. at 74 (citing AMA Diagnostic and Treatment Guidelines Concerning Child Abuse and Neglect, 254 J.A.M.A. 796 (1985) [hereinafter Guidelines]). The behavioral signs listed include:

—Overt or subtle and indirect disclosures to a relative, friend or teacher

—Highly sexualized play

—Withdrawal and excessive daydreaming

—Low self esteem, feelings of shame or guilt

—Palling grades

—Pseudomature personality development

—Sexual promiscuity

—Poor peer relationships

—Suicide attempt

*565—Positive relationship exhibited toward the offender

—Frightened or phobic, especially towards adults [Guidelines, supra, 254 J.A.M.A. at 798.]

Although the Myers article points to only two instances in which courts have approved the actual use of such testimony at a criminal trial for the purpose of establishing that sexual abuse had occurred, Myers, supra, 68 Neb.L.Rev. at 80-85 (citing Glendening v. State, 536 So.2d 212 (Fla. 1988), cert. denied, 492 U.S. 907, 109 S.Ct. 3219, 106 L.Ed.2d 569 (1989), and Townsend v. State, 103 Nev. 113, 734 P.2d 705 (1987)), we do not rule out the possibility that a qualified behavioral-science expert could demonstrate a sufficiently reliable scientific opinion to aid a jury in determining the ultimate issue that the abuse had occurred. This record does not support the conclusion.

The scientific community does not yet exhibit a consensus that the requisite degree of scientific reliability has been shown. Although some argue that “ ‘under no circumstances should a court admit the opinion of an expert about whether a particular child has been abused * * * [,]’ [t]he majority of professionals believe qualified mental health professionals can determine whether abuse occurred; not in all cases, but in some.” 1 John E. B. Myers, Evidence in Child Abuse and Neglect Cases § 4.31, at 283-84 (2d ed. 1992) [hereinafter Myers, Evidence in Child Abuse and Neglect Cases ] (footnote omitted) (quoting Melton & Limber, Psychologists’ Involvement in Cases of Child Maltreatment, 44 Am. Psychol. 1225, 1230 (1989)). In evaluating the qualifications of a witness who seeks to offer substantive evidence of sexual abuse, the trial court may wish to consider the criteria suggested in Myers, supra, Evidence in Child Abuse and Neglect Cases § 4.31, at 284-85.

Assessing children for possible sexual abuse is a complex task requiring skill and experience. The expert must possess specialized knowledge of child development, individual and family dynamics, patterns of child sexual abuse, the disclosure process, signs and symptoms of abuse, and the use and limits of psychological tests. The expert is familiar with the literature on child abuse, *566and understands the significance of developmentally inappropriate sexual knowledge. The expert is able to interpret medical reports and laboratory tests. The expert also is trained in the art of interviewing children, and is aware of the literature on coached and fabricated allegations of abuse. Of tremendous importance is the expert’s clinical experience with sexually abused children. [Ibid. (footnotes omitted).]

Obviously, the more limited the purpose for which the evidence is to be used, e.g., were it not for a substantive but fór a rehabilitative purpose, the less demanding need be the qualifications of the witness.

Still another category of expert testimony, not presented in. this case, concerns inferences to be drawn from a child’s interaction with anatomically-detailed dolls during the investigative interview. In many child-sexual-abuse trials, the expert’s testimony is wholly or partially based on this methodology. Younts, supra, 41 Duke L.J. at 706. However, the social-science literature questions whether proponents of doll use have yet demonstrated that the dolls are a valid assessment tool for diagnosing abuse. Id. at 708-20. Younts suggests that before the reliability of this methodology can be established, researchers must show at least that children can be accurately categorized as abused or nonabused based on their interaction with the dolls. Id. at 719. Obviously, the use of the anatomical dolls may help a child in reciting the history of the event and that demonstration may be of aid to jurors.

B.

We must examine the scientific premises supporting the expert’s testimony and the purpose for which the testimony was used. We note first that testimony on the child sexual abuse accommodation syndrome has been placed within the category of behavioral-science testimony that describes behaviors commonly observed in sexually-abused children. Courts rarely permit the testimony for the purpose of establishing substantive evidence of abuse, but allow it to rehabilitate the victim’s testimony. See Myers, supra, 68 Neb.L.Rev. at 51, 66-69, 86-92. Roland C. Summit, M.D., has authored the most *567concise and seemingly most authoritative statement of CSAAS. Roland C. Summit, The Child Sexual Abuse Accommodation Syndrome, 7 Child Abuse & Neglect 177 (1983) [hereinafter Summit].3 Dr. Summit explained in 1983 that although “[c]hild sexual abuse has exploded into public awareness during a span of less than five years,” the awakening of interest creates new hazards for the child victim because it increases the likelihood of discovery “but fails to protect the victim against the secondary assaults of an inconsistent intervention system.” Id. at 178 (emphasis omitted). Dr. Summit believed that most adults who hear a distraught child accuse a “respectable” adult of sexual abuse will fault the child. Ibid. The “[disbelief and rejection by potential adult caretakers,” which in Dr. Summit’s view were the too-frequent responses to reports of child sexual abuse, “increase the helplessness, hopelessness, isolation and self-blame that make up the most damaging aspects of child sexual victimization.”4 Ibid.

To remedy the systemic injury to the child that results from disbelief, Dr. Summit undertook a scientific study of child-sexual-abuse victims.5 In publishing his results, Dr. Summit *568hoped “to provide a vehicle for a more sensitive, more therapeutic response to legitimate victims of child sexual abuse and to invite more active, more effective clinical advocacy for the child within the family and within the systems of child protection and criminal justice.” Summit, supra, 7 Child Abuse & Neglect at 179-80. In other words, the purpose of his study was to improve the health of the child, ensure that children receive adequate treatment for what they had suffered, and guarantee that society’s response be not flawed by misperceptions.

Dr. Summit’s study drew upon correlations and observations that “emerged as self-evident within an extended network of child abuse treatment programs and self-help organizations” and he tested the validity of his theory over a four-year period of his own practice. Summit, supra, 7 Child Abuse & Neglect at 180. The child sexual abuse accommodation syndrome, or CSAAS, “represents a common denominator of the most frequently observed victim behaviors.” Ibid. CSAAS includes five categories of behavior, each of which contradicts “the most common assumptions of adults.” Summit, supra, 7 Child Abuse & Neglect at 181. Of the five categories, he described two as “preconditions” to the occurrence of sexual abuse and the remaining three as “sequential contingencies” to the abuse “which take on increasing variability and complexity.” Ibid. Obviously, the “preconditions” continue into and characterize the period of abuse.

The first of the preconditions is secrecy: child abuse happens only when the child is alone with the offending adult, and the experience must never be disclosed. That secrecy is frequently accompanied by threats: “ ‘This is our secret; nobody else will understand.’ ” “ ‘Don’t tell anybody.’ ” “ ‘Nobody will believe you.’ ” “ ‘Don’t tell your mother; (a) she will hate you, * * * (c) she will kill you,' ” and the like. Summit, supra, 7 Child Abuse & Neglect at 181. From the secrecy, the child gets the impression of danger and fearful outcome. Ibid. In this case, Norma and Connie testified that they had not reported the alleged abuse because defendant had told them that if they did, *569he would hit them and they would get into more trouble than he.

The second precondition is helplessness. Dr. Summit explains that the abused child’s sense of helplessness is an outgrowth of the child’s subordinate role in an authoritarian relationship in which the adult is entrusted with the child’s care, such as the parent-child relationship. Summit, supra, 7 Child Abuse & Neglect at 182. The prevailing reality for the most frequent victim of child sexual abuse is a sense of total dependence on this powerful adult in the face of which the child’s normal reaction is to “play possum.” Id. at 182-83.

The third aspect of the syndrome, also the first of what Dr. Summit identifies as a sequential contingency, is a combination: the child feels trapped by the situation (entrapment), and that perception results in the behavior of accommodating the abuse (accommodation). Because of the child’s helplessness, the only healthy option left is to survive by accepting the situation. “There is no way out, no place to run.” Summit, supra, 7 Child Abuse & Neglect at 184. Adults find that hard to believe because they lack the child’s perspective, but “[t]he child cannot safely conceptualize that a parent might be ruthless and self-serving; such a conclusion is tantamount to abandonment and annihilation.” Ibid. The roles of parent and child become reversed: it is the child who must protect the family. The abuser warns, “ ‘If you ever tell, they could send me to jail and put all you kids in an orphanage.’ ” Summit, supra, 7 Child Abuse & Neglect at 185.

The fourth aspect, then, is delayed, conflicted and unconvincing disclosure. Id. at 186. Most victims never disclose the sexual abuse — at least not outside the immediate family. Dr. Summit found that family conflict triggers disclosure, if ever, “only after some years of continuing sexual abuse and an eventual breakdown of accommodation mechanisms.” Ibid.

Allegations of sexual abuse seem so unbelievable to most that the natural reaction is to assume the claim is false, *570especially because the victim did not complain years ago when the alleged abuse was ongoing. Ibid. Dr. Summit surmises that

[ujnless specifically trained and sensitized, average adults, including mothers, relatives, teachers, counselors, doctors, psychotherapists, investigators, prosecutors, defense attorneys, judges and jurors, cannot believe that a normal, truthful child would tolerate incest without immediately reporting or that- an apparently normal father could be capable of repeated, unchallenged sexual molestation of his own daughter. [Ibid.']

There are very few clues to such abuse. Most women (indeed, even this mother) do not believe it possible that a man whom she loved would ever- be capable of molesting his or her own children. Summit, supra, 7 Child Abuse & Neglect at 187.

The fifth and final aspect is retraction. Although this case does not involve retraction, that “[w]hatever a child says about sexual abuse, she is likely to reverse it” appears to be a fact. Summit, supra, 1 Child Abuse & Neglect at 188 (emphasis omitted). The post-disclosure family situation tends to confirm the victim’s worst fears, which encouraged her secrecy in the first place, i.e., her mother is disbelieving or hysterical, her father threatened with removal from the home, and the blame for this state of affairs placed squarely on the victim. Ibid. Once again, because of the reversed roles, the child feels obligated to preserve the family, even at the expense of his or her own well being. The only “good” choice, then, is to “capitulate” and restore a lie for the family’s sake. . Ibid.

Dr. Summit analogizes the gradual acceptance of the reality of child sexual abuse to society’s changing attitude towards adult rape with the emergence of the rape trauma syndrome theory (RTS). Summit, supra, 7 Child Abuse & Neglect at 189 (citing Ann W. Burgess & Linda L. Holmstrom, Rape Trauma Syndrome, 131 Am.J. of Psychiatry 981 (1974) [hereinafter Burgess & Holstrom]). Women were assumed to cause rape, in the absence of a consistent clinical understanding of the “psychological climate,” and reactions to such sexual attacks. Summit, supra, 7 Child Abuse & Neglect at 189. Thus, “Those who reported often regretted their decision as they found *571themselves subjected to repeated attacks on their character and credibility.” Ibid. The gradual departure from the mythology of women and rape so recently outlined by our Court in In re M.T.S., 129 N.J. 422, 432-33, 443-45, 609 A.2d 1266 (1992) (dispelling myth that victim’s silence in response to sexual assault equals consent), and State v. Hill, 121 N.J. 150, 157-66, 578 A.2d 370 (1990) (dispelling myth that victim’s failure immediately to report an alleged sexual assault tends to show that she was not assaulted at all), has slowly been extended to child-abuse victims.

Hence, the behavioral studies of CSAAS are designed not to provide certain evidence of guilt or innocence but rather to insure that all agencies, including the clinician, the offender, the family, and the criminal justice system, offer “the child a right to parity with adults in the struggle for credibility and advocacy.” Summit, supra, 7 Child Abuse & Neglect at 191. CSAAS achieves that by providing a “common language” for analysis and a more “recognizable map” to the understanding of child abuse. Ibid.

Ill

Standards Governing the Admission of Expert Opinion Evidence of CSAAS in Criminal Trials

We need not retrace the development of these standards in any extensive detail. In a long series of cases, we have outlined the general standards that govern the admissibility of such evidence. See State v. Spann, 130 N.J. 484, 617 A.2d 247 (1992); State v. Zola, 112 N.J. 384, 548 A.2d 1022 (1988), cert. denied, 489 U.S. 1022, 109 S.Ct. 1146, 103 L.Ed.2d 205 (1989); State v. R.W., supra, 104 N.J. 14, 514 A.2d 1287; State v. Kelly, 97 N.J. 178, 478 A.2d 364 (1984); State v. Cavallo, 88 N.J. 508, 443 A.2d 1020 (1982). A summary may be found in State v. R.W., supra, describing the accepted grounds for admitting expert testimony:

*572As provided in the Rules of Evidence, Evid.R.56, and reiterated by many cases, the testimony of an expert is allowed when it relates to a subject-matter beyond the understanding of persons of ordinary experience, intelligence, and knowledge. E.g., State v. Kelly, supra, 97 N.J. 178 [478 A.2d 364] Evers v. Dollinger, 95 N.J. 399 [471 A.2d 405] (1984). This applies as well to the field of child sex-abuse offenses. As we have seen, such testimony may be allowed to explain generally the behavior, feelings, and attitudes of such victims when it is shown that their condition is not readily understood by persons of average intelligence and ordinary experience; an expert or scientific explanation of their condition, one accepted as reliable by the scientific community that is involved in the diagnosis, treatment, and care of such individuals, can assist a jury in understanding the evidence. [104 N.J. at 30-31, 514 A.2d 1287.]

Acceptance within a scientific community can sometimes be an elusive premise to prove. There will always be some detractors to any theory or clinical approach. State v. Kelly, supra, 97 N.J. 178, 208-14, 478 A.2d 364, illustrates the kind of inquiry that must be made concerning the definition of the scientific community, the degree of acceptance within that community, and the method of proof of that acceptance.6 See also Rubanick v. Witco Chemical Corp., 125 N.J. 421, 593 A.2d 733 (1991) (discussing special difficulties encountered in certain forms of scientific analysis). Some courts have admitted expert testimony that a child has been sexually abused without any reference to scientific reliability. In re Cheryl H., 153 Cal.App.3d 1098, 200 Cal.Rptr. 789 (Ct.App. 1984). However, even a qualified expert’s opinion must have a reliable basis. Landrigan v. Celotex Corp., 127 N.J. 404, 417, 420, 605 A.2d 1079 (1992) (stating that whether an expert may testify to cancer causation depends on degree of acceptance within relevant scientific community of the methodology of analysis). Certainly, we *573should be even more hesitant to depart from the general-acceptance requirement in a criminal case. State v. Spann, supra, 130 N.J. at 510, 617 A.2d at 259-260.

Finally, we have consistently recognized that juries have exclusive responsibility in the determination of criminal guilt or innocence. State v. Simon, 79 N.J. 191, 398 A.2d 861 (1979). Hence, in State v. Odom, 116 N.J. 65, 560 A.2d 1198 (1989), we permitted a State’s expert witness to testify that, in his opinion, based on his experience and specialized knowledge, the quantity of drugs was there possessed with the intent to distribute them. We therefore concluded that “as long as the expert does not express his opinion of defendant’s guilt but simply characterizes defendant’s conduct based on the facts in evidence in light of his specialized knowledge, the opinion is not objectionable.” Id. at 79, 560 A.2d 1198. We realize that the line drawn in Odom may be thin at times but believe that properly charged juries can see the distinction.

There does not appear to be a dispute about acceptance within the scientific community of the clinical theory that CSAAS identifies or describes behavioral traits commonly found in child-abuse victims. See, e.g., Myers, supra, 68 Neb. L.Rev. at 66-69; Holmes, supra, 25 Tulsa L.J. at 158-59. The most pointed criticism of the theory is that the same traits may equally appear as the result of other disorders. Holmes, supra, 25 Tulsa L.J. at 158, 162-63. Even extreme poverty or psychological abuse can produce the sense of entrapment or accommodation. In other words, the existence of the symptoms does not invariably prove abuse.

That would be a valid criticism if the CSAAS evidence were offered for a purpose beyond the scope of the scientific theory. An analogy may be drawn from State v. Cavallo, supra, 88 N.J. 508, 443 A.2d 1020 (1982). In that case, defense counsel offered to prove that the defendant did not have the characteristics common to all or most rapists and thus to disprove the fact of rape. After reviewing the literature and cases in other *574jurisdictions, the Court was convinced that the medical or legal communities do not generally accept the view that psychiatrists possess the knowledge or capability to state the likelihood that an individual behaved in a particular manner on a specific occasion.

In contrast, State v. Kelly, supra, 97 N.J. 178, 478 A.2d 364, explained that expert scientific evidence concerning “battered-woman’s syndrome” does not aid a jury in determining whether a defendant had or had not behaved in a given manner on a particular occasion; rather, the evidence enables the jury to overcome common myths or misconceptions that a woman who had been the victim of battering would have surely left the batterer. Thus, the evidence helps the jury to understand the battered woman’s state of mind. Id. at 190-97, 204-05, 478 A.2d 364. Because the State in Kelly had reinforced those myths by repeatedly asking the victim why she had taken her husband back after the battering, the Court ruled such evidence admissible to counter the myths if reliability of the evidence of a counter-intuitive behavioral pattern were established within the scientific community. Id. at 205-06, 211-14, 478 A.2d 364.

IV

The Use of CSAAS Expert Testimony in This Case

Turning then to the application of the standards to the record of this trial, we find that the CSAAS evidence was not presented to the jury in accordance with its scientific theory, i.e., the evidence was not offered to explain the conflicting behavioral traits in this case either of accommodation or delayed disclosure. Rather, the evidence was presented to the jury as though it were to prove directly and substantially that sexual abuse had occurred. Dr. Milchman, the prosecution’s expert witness on child sexual abuse, described CSAAS as a pattern of behavior found to occur consistently in children who are victims of incest, and she outlined Dr. Summit’s five-part *575syndrome: secrecy, helplessness, entrapment and accommodation, delayed disclosure, and retraction.

The prosecutor asked whether she had examined Connie and Norma and whether, in the course of examination, Dr. Milchman had found the children to “suffer symptoms of the child sexual abuse accommodation syndrome.” (Recall that Dr. Summit describes two aspects of CSAAS, secrecy and helplessness, as “preconditions.”) Dr. Milchman said that Norma had exhibited four of the symptoms: secrecy, helplessness, accommodation, and delayed disclosure. She said that Norma’s crying, shaking, rubbing her hands in her eyes, and covering her face during the interview were manifestations of her feeling of helplessness towards the abusive situation and her fears, anxieties and anger which she had to suppress in order to accommodate the abuse. Technically, these interview observations do not seem to be the symptoms that Dr. Summit describes. Rather, they appear to us to be generic post-traumatic symptoms. Supra at 563, 617 A.2d at 1201. Dr. Milchman pointed out that Norma had revealed the abuse only after questioning initiated by her mother. She also explained that Norma had stuck to her story, instead of recanting it, because nobody was pressuring or threatening her. Connie, on the other hand, had presented herself differently as a “very brassy, very assertive, very outgoing child on the surface” but “[u]nderneath there was a lot of fear and anxiety.” Dr. Milchman described Connie as being entrapped and accommodating to the abuse — she complied and accepted the abuse for a long time. She also believed Connie had given a delayed disclosure. Dr. Milchman believed that Connie had kept the abuse a secret for a long time because John had threatened her and her mother with physical violence, a threat she believed because she had seen her father be violent with her mother.

The prosecution then made a transition into areas not covered by CSAAS. Dr. Milchman was presented with a series of questions based on a child’s testimony concerning (1) the experience of feeling someone putting a finger inside the child’s *576vagina, (2) the experience of feeling someone putting his tongue inside the child’s mouth, and (3) graphic details concerning oral sexual contact. She was asked whether that testimony would be consistent with a child who had been sexually abused or would reflect a child’s exposure to outside sources, such as watching others or watching an adult movie. Dr. Milchman said that such testimony was consistent with sexual abuse and added that the graphic details concerning oral sexual contact are not within a child’s normal imagination.

The three questions presented to Dr. Milchman obviously had no relationship to CSAAS. They pose evidentiary difficulties in the absence of a more detailed record.7 Yet counsel did not object to them. Perhaps counsel in such cases prefer to give a wide latitude to expert testimony rather than to appear to shield the jury from such opinion evidence. Had the expert’s opinion gone no further, we would not find clear error capable of bringing about an unjust result. The issues would perhaps be viewed as somewhat akin to the issue posed in State v. Zola, supra, 112 N.J. 384, 548 A.2d 1022, in which a State forensic expert was permitted to give an opinion on a probable source of saliva found within the victim’s body cavities. We ruled there that expert testimony may play a limited role in such issues as they are “common sense” deductions on subjects about which jurors may not have much familiarity, and such testimony does not deprive the jury of its qualitative function. Id. at 414-16, 548 A.2d 1022.

*577However, Dr. Milchman then proceeded to describe how one could “tell whether a child is lying.” For example, a child’s speaking in a mechanical way, “like it was by rote memory rather than by their own feelings,” could raise the suspicion that the child was trying to remember what someone else had told him or her, thereby undermining the child’s credibility. Or a child’s perfectly consistent narration of all details of the story would be inconsistent, she said, with a child’s natural tendency to forget minor detail. Certainly the prefatory basis of CSAAS has nothing to do with those areas of opinion.

This type of testimony equates with the kind of expertise said to relate to eyewitness testimony. Courts are frequently requested to permit expert opinion testimony on the reliability of eyewitness identification. Although People v. McDonald, 37 Cal.3d 351, 208 Cal.Rptr. 236, 690 P.2d 709 (1984), allowed expert testimony from a psychologist concerning the ability of eyewitnesses to perceive, remember, and relate accurately and the distorting effects of fear and excitement, courts have sometimes hesitated to allow such testimony because it may interfere in the truthfinding function or create an unwarranted aura of expertise. Thus, if the child-sexual-abuse expert testifies that the child is “believable” or “truthful,” the “courts will most likely exclude the evidence.” Linda E. Carter, Admissibility of Expert Testimony in Child Sexual Abuse Cases in California: Retire Kelly-Frye and Return to a Traditional Analysis, 22 Loyola L.Rev. 1103, 1118 (1989).

Yet, in this case Dr. Milchman was asked, “How can you tell when a child or a victim is telling the truth about the fact that they have been sexually abused?” Although the court cautioned the jurors that it was up to them ultimately to determine that, the expert proffered a theory — again unrelated to CSAAS:

Okay. I look for — I look for many different things. I look for whether the child appears to be sincere. I look for whether or not the feeling that they have at the time goes with what they are saying or whether it contradicts what they are saying. I go for whether there are a lot of different behaviors that all point to the same conclusion. For example, is what the child saying, does that match the demonstrations that they give when they try to explain it with their hands *578or with dolls, does it match the pictures that they draw for me? Does it match what they told the mother; does it match what they told the DYFS worker; does it match what they told the Prosecutor or investigator; or does it match what they told me? I look for consistency across a lot of different kinds of behaviors.

* * * I look for * * * realistic, concrete, specific, kinds of details that are not the kinds of things that you would tend to see on a television so that any kid could pick up or any cable T.V. or movie that any kid could just pick up * * *.

The-final question to the witness was: “Doctor, based on your examinations of the girls can you give this jury your expert opinion as to whether or not both [Connie] and [Norma] were sexually abused?” Answer: “I believe that they were sexually abused.”

At this point, whether Dr. Milchman had reached that opinion on the basis of her credibility assessments or on the basis of her understanding of CSAAS evidence is not clear to us and could not have been clear to the jury. If it were the former, it would be improper opinion evidence bécause it would introduce an unwarranted aura of scientific reliability to the analysis of credibility issues. If it were the latter, it would be improper opinion evidence because CSAAS is not relied on in the scientific community to detect abuse.

There has not been a showing in the record in this case, nor seemingly in other scientific literature or decisional law, of a general acceptance that would allow the use of CSAAS testimony to establish guilt or innocence. See David McCord, Expert Psychological Testimony About Child Complainants in Sexual Abuse Prosecutions: A Foray into the Admissibility of Novel Psychological Evidence, 77 J.Crim.L. & Criminology 1, 24, 38 (1986). Such use of CSAAS evidence would present the analog to State v. Cavallo, supra, 88 N.J. 508, 443 A.2d 1020, and would require a study of the reliability of psychiatric or psychological testimony on the likelihood that the traits found in the victim will establish that another had engaged in the conduct that had caused the symptoms. It strikes us that the premise would be strained, at least on the basis of the Summit studies. As Myers noted:

*579Summit did not intend the accommodation syndrome as a diagnostic device. The syndrome does not detect sexual abuse. Rather, it assumes the presence of abuse, and explains the child’s reactions to it. Thus, child sexual abuse accommodation syndrome is not the sexual abuse analogue of battered child syndrome, which is diagnostic of physical abuse. With battered child syndrome, one reasons from type of injury to cause of injury. Thus, battered child syndrome is probative of physical abuse. With child sexual abuse accommodation syndrome, by contrast, one reasons from presence of sexual abuse to reactions to sexual abuse. Thus, the accommodation syndrome is not probative of abuse.

Unfortunately, a number of mental health professionals, lawyers, and commentators drew unwarranted comparisons between battered child syndrome and child sexual abuse accommodation syndrome. This error led to considerable confusion. First, some professionals misinterpreted Summit’s article, believing Summit had discovered a “syndrome” that could diagnose sexual abuse. This mistake is understandable, if not forgivable. Mental health and legal professionals working in the child abuse area had long been accustomed to thinking in terms of syndrome evidence to prove physical abuse. Battered child syndrome was an accepted diagnosis by the time Summit’s accommodation syndrome came along in 1983. It was natural for professionals to transfer their understanding of battered child syndrome to this new syndrome, and to conclude that the accommodation syndrome, like battered child syndrome, could be used to detect abuse.

‡‡‡‡‡‡‡‡

* * * [T]he accommodation syndrome was being asked to perform a task it could not accomplish.

The accommodation syndrome has a place in the courtroom. The syndrome helps explain why many sexually abused children delay reporting their abuse, and why many children recant allegations of abuse and deny that anything occurred. If use of the syndrome is confined to these rehabilitative functions, the confusion clears, and the accommodation syndrome serves a useful forensic function. [Myers, supra, 68 Neb.L.Rev. at 67-68 (footnotes omitted).]

This we believe is the most concise summary of the proper use of CSAAS and will serve as a useful road map in the trial of such cases. Another commentator agrees.

Much of the legal controversy about CSAAS is a product of legal misuse and misunderstanding which is a direct result of the fact that CSAAS is a misnomer. The term “syndrome” may refer to two different things. In laymen’s terms it is defined as either “a group of signs and symptoms that occur together and characterize a particular abnormality” or “a set of concurrent things (as emotions or actions) that usu[ally] form an identifiable pattern.” Medically, however, the term refers to the aggregation of symptoms associated *580with a morbid process which forms a disease. CSAAS, as it is currently defined, is neither a disease nor a pattern of abnormality.

********

* * * [T]he syndrome seeks to define a coping process and not behavior that will identify the existence of sexual abuse.

* * * Since this “syndrome” is only a piece of the child sexual abuse machinery, testimony concerning CSAAS may only be offered for the purpose for which it was defined — to explain the child’s irrational behavior. [Holmes, supra, 25 Tulsa L.J. at 157-59 (footnotes omitted).]

The California Court of Appeals has thus refused to admit testimony about the general characteristics of molested children for the purpose of allowing the jury to conclude that a particular child is a victim of abuse. People v. Bowker, supra, 203 Cal.App.3d 385, 249 Cal.Rptr. 886. Instead, because CSAAS has a limited, therapeutic purpose and not a predictive one, “the evidence must be tailored to the purpose for which it is being received.” Id. at 891, 203 Cal.App.3d 385. “[A]t a minimum the evidence must be targeted to a specific ‘myth’ or ‘misconception’ suggested by the evidence” and limited to explaining why “the victim’s reactions as demonstrated by the evidence are not inconsistent with having been molested.” Id. at 891-92, 203 Cal.App.3d 385. The court must also explain to the jury that the expert’s testimony is not intended to address the ultimate question of whether the victim’s molestation claims are true and must admonish the jury not to use the testimony for that purpose. Id. at 892, 203 Cal.App.3d 385. That use of CSAAS testimony is consistent with the use to which we put the battered-woman-syndrome evidence in Kelly, supra, 97 N.J. 178, 478 A.2d 364.

Other jurisdictions follow that approach. See State v. Reser, 244 Kan. 306, 767 P.2d 1277, 1283 (1989) (qualifying clinical specialist with training in child sexual abuse to testify to “common patterns of behavior” resulting from abuse and that this victim had symptoms consistent with those patterns); People v. Beckley, 434 Mich. 691, 456 N.W.2d 391, 407 (1990) (finding appropriate expert testimony limited to providing jury with background information, relevant to specific aspect of *581child’s conduct at issue, which it could not otherwise bring to its evaluation of child’s credibility); State v. Middleton, 294 Or. 427, 657 P.2d 1215, 1220 (1983) (explaining “superficially bizarre behavior” of victims of abuse helps jury to assess credibility). But see State v. Bachman, 446 N.W.2& 271, 276 (S.D. 1989) (allowing opinion testimony that victim’s allegations were truthful).

Such use accords with the use now generally afforded to rape trauma syndrome (RTS) evidence most often in the context of adult rape. RTS describes symptoms frequently experienced by rape victims, e.g., phobic reactions and sexual fears. Because RTS was developed as a therapeutic tool, not as a test to determine the existence of a past event, the California Supreme Court in People v. Bledsoe, 36 Cal.3d 236, 203 Cal.Rptr. 450, 681 P.2d 291 (1984), questioned the reliability of RTS in determining whether a rape has occurred. Thus, the court held that, given the history, purpose and nature of RTS, testimony on the concept was inadmissible to prove that a rape occurred, but recognized that RTS testimony has been admitted in cases in which the alleged rapist suggests that the victim’s conduct after the incident was inconsistent with her claim of rape. Id. 203 Cal.Rptr. at 457-460, 681 P.2d at 298-301. In the latter context, expert testimony of RTS may play a particularly useful role by disabusing the jury of widely-held myths and misconceptions about rape and rape victims. Id. at 457, 681 P.2d at 298; see also People v. Taylor, 75 N.Y.2d 277, 552 N.Y.S.2d 883, 552 N.E.2d 131 (1990); State v. Saldana, 324 N.W.2d 227 (Minn. 1982) (both holding that RTS is not reliable as substantive proof of rape).

Y

Conclusion

Expert opinion testimony has a vital role to play in the trial of child-sexual-abuse cases. At one end of the spectrum is the clearly admissible evidence of the qualified expert with *582respect to the physical manifestations of sexual abuse or the child’s out-of-court statements relating to a sexual offense under Evidence Rule 63(33) (the tender-years exception). As courts and counsel proceed further from that clearly admissible end of the spectrum, they must focus on the research basis for the proposition that the expert witness seeks to establish.

The State has argued before us that it is appropriate to admit Dr. Milchman’s testimony describing CSAAS and concluding that Norma’s and Connie’s symptoms were consistent with sexual abuse and rendering an expert opinion that they had been sexually abused. Obviously, scientific evidence exists to aid a jury in determining whether sexual abuse has occurred. As we understand CSAAS, however, it does not purport to establish sexual abuse but helps to explain traits often found in children who have been abused. Hence we believe that in this case the “accommodation syndrome was being asked [by the State] to perform a task it could not accomplish.” Myers, supra, 68 Neb.L.Rev. at 68.

In this case the theory of the defense was that the children had been put up to their stories by a vengeful scorned lover. The CSAAS evidence would have served well to counter the mythology that if the abuse had occurred, the children surely would have complained sooner and would not have put up with repeated visits to the apartment in Brooklyn. However, when the expert, without a reliable foundation, went on to offer opinions with respect to the basic factual issues, including truth-telling, she transgressed the purpose for which CSAAS testimony is admissible.

We realize that these cases are extremely difficult to prosecute. They should not be made more difficult by evidentiary issues that surface in the midst of trial. Our rules permit a criminal defendant to have access to the contents of a State expert’s report in advance of trial. Pressler, Current N.J. Court Rules, R. 3:13-3(a)(4) (1992). We understand that the State routinely furnishes defendants with such discovery. *583Speedy-trial practice contemplates that the defendant will have the report by arraignment. If there is to be objection to the testimony, a defendant may request a Rule 8 hearing in advance of the trial at which the court will assess the qualifications of the witness as well as the research basis for the expert proposition to be stated. For example, in a case such as this the trial court should have been given the opportunity before trial to determine the qualifications of the witness and whether under any of the three methods of establishing reliability, i.e., expert testimony, authoritative literature, or decisional law, the research (here only CSAAS was mentioned) sustains an expert opinion that the children were telling the truth or were reliably diagnosed as victims of sexual abuse.

At such a Rule 8 hearing, the court will conduct an inquiry into whether the witness possesses the “education, knowledge, training, and experience” necessary to render an opinion on the proposition to be presented to the jury. Rubanick, supra, 125 N.J. at 449-53, 593 A.2d 733. If qualified, the witness should then establish at a minimum that the methodology used to form his or her opinion is generally accepted within the relevant scientific community to establish the proposition to be stated by the witness. In this case, the only scientific basis or methodology referred to by the witness was CSAAS and as we have seen, qualified mental health professionals do not regard CSAAS as a sufficiently reliable scientific indicator of the substantive fact of abuse. Obviously, a qualified witness may wish to incorporate the CSAAS research into the totality of his or her methodology for assessing the substantive fact of abuse.

In setting forth these requirements, we do not intend in any sense to mystify the trial of child-sexual-abuse cases. All that is required is close attention to existing precedent: (1) are the factual matters “beyond the ken” of the jurors, State v. Kelly, supra, 97 N.J. at 208, 478 A.2d 364, and will expert testimony aid the jury in resolution of the matter?; (2) is the proposed *584expert witness qualified by education, training, experience, and knowledge to express an opinion on the factual matter in dispute?; (3) is there general acceptance of the scientific theory or even the methodology used to establish the factual proposition?; and (4) will the jury be given proper instructions limiting the evidence to the purpose for which it is offered? The study of child sexual abuse is rapidly evolving, and we may expect that behavioral scientists will continue their efforts to develop reliable criteria to detect child sexual abuse.

Because this is the first time that our Court has addressed those issues, we affirm the judgment of the Appellate Division. In future cases, we shall assume that the failure of informed defense counsel to object to such expert testimony may reflect a tactical decision by the defense to let the jury hear all available information pertaining to the case.

In view of the disposition that we make, we do not address the remaining questions raised in defendant’s cross-petition. Obviously, on any remand the child-witnesses will be older and the court will have to reassess whether they should testify by closed-circuit television, pursuant to N.J.S.A. 2A:84A-32.4. See State v. Crandall, supra, 120 N.J. 649, 663-64, 577 A.2d 483. The admissibility of the hearsay statements of the victims will be governed by Evidence Rule 63(33).

The judgment of the Appellate Division is affirmed.

Opposed — None.