dissenting.
Today, in straining to affirm a deeply flawed conviction, the majority rationalizes a series of profound trial errors that undermine any confidence in the correctness of the jury’s verdict. First, the majority finds harmless the testimony of a State’s expert who gave his assessment of the collective credibility of child-abuse victims, calculating that ninety to ninety-five percent tell the truth. The expert had never met D.L., the putative victim, who recanted her statements implicating defendant. Yet, with his statistically based testimony—without the need to be burdened by the evidence in this case—the jury was left to ponder whether, if there was a ninety to ninety-five percent probability that the putative victim was telling the truth, then there must be a ninety to ninety-five percent probability of defendant’s guilt. The majority validates as harmless an expert’s opinion that denied defendant an individualized assessment of his guilt.
Second, the majority has grossly distorted the doctrine of fresh complaint to justify the admission of a hearsay statement incriminating defendant. In this new jurisprudential world, a victim’s statement made to her boyfriend more than fifteen months after alleged sexual encounters with defendant is deemed a “fresh” complaint.
Last, despite the violation of our court rules, the majority sees no problem in the court’s giving to the jury during its deliberations a videotaped statement of defendant that was not placed in evidence.
Because I believe that these errors individually and cumulatively denied defendant a fair trial, I respectfully dissent.
*625I.
Guilt by Statistics
The trial in this ease was about the individual credibility of both D.L., who on the stand recanted out-of-court statements implicating defendant, and defendant, who protested his innocence and claimed that the police coerced a confession from him through psychological pressures. The jury had to determine whether D.L. was telling the truth, under oath, when she testified or when she spoke to the police. Under our system of justice, we do not permit the jury to assess the credibility of a model victim, a statistical stereotype, and on that basis to infer guilt, no more than we would permit a jury to infer that the defendant on trial must be guilty because most defendants are guilty. Allowing group assessments of credibility to replace individualized assessments of credibility is antithetical to our system of justice. The deleterious impact of the type of testimony presented in this case, which was sanctioned by the trial court, cannot be ignored or wished away.
The State presented Dr. Coco as an expert in Child Sexual Abuse Accommodation Syndrome (CSAAS). The purpose of Dr. Coco’s testimony was to explain, generally, the reasons a child would delay reporting sexual abuse or, once reporting the abuse, recant the accusation. See State v. J.Q., 130 N.J. 554, 579, 617 A.2d 1196 (1993) (citation omitted). CSAAS testimony is not offered as proof of a defendant’s guilt. As Dr. Coco conceded: “[I]t cannot be offered as a tool to decide whether sexual abuse had happened.” See id. at 578, 617 A.2d 1196 (“CSAAS is not relied on in the scientific community to detect abuse.”). Yet, Dr. Coco’s testimony departed from the strict limitations our law places on CSAAS testimony.
During cross-examination, the defense attorney elicited from Dr. Coco that some children falsely report sexual abuse. Then, under the pretext of counsel’s having opened the door, the court permitted the prosecutor on redirect examination to introduce statistical evidence that approximately ninety to ninety-five per*626cent of child-sexual-abuse complainants are truthful. Even if the defense attorney cracked the door open slightly into this impermissible area, the trial judge, whose obligation is to ensure the fairness of the proceedings, should not have kicked the door wide open.
The jury was permitted to convict defendant based on a simple syllogism totally unrelated to the evidence: if ninety to ninety-five percent of sexual-abuse complainants tell the truth, then D.L. by the laws of statistical probability must have been telling the truth when she reported the sexual abuse to the police; and if D.L. was therefore truthful, then defendant must be lying and guilty of the crimes charged. The defense loudly objected; however, no correction was made by the trial court. The jury was never told that it could not draw the obvious damning inferences that flowed from Dr. Coco’s use of statistics to bolster D.L.’s initial complaint. The majority concedes that this testimony was improper and did not fall within the realm of CSAAS evidence, but claims that the error was harmless.
Police officers and social-science experts are not allowed to vouch for the credibility of witnesses. See State v. Frisby, 174 N.J. 583, 591-92, 595, 811 A.2d 414 (2002) (finding plain error in police officer’s testimony that potential suspect in child-abuse case was “more credible” than defendant).1 In J.Q., we specifically disapproved of a CSAAS expert touting the credibility of an alleged sex-victim whom the expert had interviewed. J.Q., supra, 130 N.J. at 556, 575, 617 A.2d 1196. Opining on the alleged victims’ credibility was an error that went to the heart of the very integrity of the proceedings, and therefore was not deemed harmless. Id. at 556, 617 A.2d 1196. How much worse in this ease where the expert—based on mere statistics—placed his authoritative imprimatur on the credibility of D.L.
*627The majority does not deny that the expert misused statistics to support the credibility of D.L. in this case. We disagree about the gravity of that error. For me, the presentation of a statistical, truth-telling, stereotypical victim—that presumably encompassed the alleged victim in this case—cannot be harmless.
Appellate courts are not required to put blinders on when assessing whether patently inadmissible evidence has the clear capacity to taint the jury. Here, the State’s case hinged on the credibility of witness testimony, not physical, objective evidence. The alleged victim testified under oath that her previous statements implicating defendant were false, and defendant on the stand testified that he was innocent and that the police elicited from him a false confession. Given this less than overwhelming evidence of guilt, it is impossible to say that Dr. Coco’s impermissible testimony did not tip the scales of justice unfairly against defendant. See Frisby, supra, 174 N.J. at 596, 811 A.2d 414 (holding that “[a]ny improper influence on the jury that could have tipped the credibility scale” could not be harmless); R. 2:10-2 (stating that new trial should not be granted unless error was “clearly capable of producing an unjust result”).
Because the error in this case could not be harmless, defendant should be entitled to a new trial.
II.
A Fifteerir-Month-Delayed Report Posing as Fresh Complaint
That the majority sanctions as a “fresh” complaint a report of sexual abuse that occurred fifteen months earlier indicates that the fresh-complaint doctrine has now become completely untethered from its historical underpinnings. The fresh-complaint doctrine is derived from the common-law requirement that a sexual-assault victim utter a “hue and cry” “to dispel any suspicion” of her consensual involvement in the act. See State v. Hill, 121 N.J. 150, 157, 578 A.2d 370 (1990). Under that requirement, it was assumed that “a ‘normal’ woman would ‘naturally’ complain after *628having been raped.” Id. at 162, 578 A.2d 370. Outmoded views about how “normal” women would react to a sexual assault reflected in the “hue and cry” requirement gave way to the more nuanced, modern version of the fresh-complaint rule.
Under the present fresh-complaint doctrine, a victim’s out-of-court report is admissible to support her in-court testimony for the purpose of negating the inference that her silence would be an indication she was not sexually assaulted. Id. at 163, 578 A.2d 370 (citation omitted). To qualify as a fresh complaint, the victim’s statement must have been made to someone she ordinarily would have called for support, “must have been made within a reasonable time after the alleged assault and must have been spontaneous and voluntary.” Ibid, (citations omitted) (emphasis added). The fresh-complaint rule serves the narrow purpose of rebutting the negative inference the jury would draw from silence. Ibid. (citation omitted). “Only the fact of the complaint” is admissible, not the particulars of the accusation. Ibid, (citation omitted).
Reporting a sexual offense within a “reasonable time” is a precondition to the admission of fresh-complaint evidence. The fifteen-month delay now permitted by the majority completely eviscerates the “reasonable time” requirement and will make every complaint—regardless of when it is reported—fresh. While “courts allow children additional time to make a fresh complaint” when they claim sexual abuse, State v. Bethune, 121 N.J. 137, 143, 578 A.2d 364 (1990), there must be some logical end to that period. Here, the “child” was a young adult, age sixteen, when she made her initial complaint. Today’s opinion is the first time that the Court has expanded the fresh-complaint doctrine to permit allegations made so long after an alleged assault. Cf. State v. Smith, 158 N.J. 376, 378-79, 730 A.2d 311 (1999) (complaint one day after assault admissible); State v. Mann, 132 N.J. 410, 415, 426, 625 A.2d 1102 (1993) (complaint within one day after assault admissible); Hill, supra, 121 N.J. at 152, 154, 169, 578 A.2d 370 (complaint to friend within six weeks after assault admissible); Bethune, supra, 121 N.J. at 140-41, 145-46, 578 A.2d 364 (permitting *629two-week-delayed complaint, but questioning whether statement was voluntary); State v. Tirone, 64 N.J. 222, 225-26, 314 A.2d 601 (1974) (complaint within one day after assault admissible); State v. Balles, 47 N.J. 331, 334-35, 339, 221 A.2d 1 (1966) (complaint within one day after assault admissible).
That some lower courts have allowed much longer periods to report a sexual assault, while still characterizing the complaint as “fresh,” is no reason for this Court to allow the doctrine to become so distorted and muddled that any result affirming a conviction can be rationalized. See ante at 617-19. Moreover, the “fresh” complaint in this case was not admitted to support the in-court testimony of the alleged victim but to impeach that testimony. We have never held that fresh-complaint testimony can be used for any purpose other than to show that the prior complaint is consistent with the witness’s testimony on the stand. Cf. Smith, supra, 158 N.J. at 378, 730 A.2d 311 (stating that fresh complaint was consistent with alleged victim’s trial testimony); Bailes, supra, 47 N.J. at 335, 221 A.2d 1 (same). An additional irony is that the State introduced the fifteen-month-late report as “fresh” at the very same time that it introduced CSAAS testimony through an expert to explain why the report was so delayed. One must wonder how a complaint can be admissible based on its “freshness” and yet require the extensive testimony of a CSAAS expert to explain why the complaint was not freshly made.
The importance of evidence to the State does not validate its admission. Process is an end in itself. It assures fairness in all proceedings and legitimizes a trial’s outcome. The majority’s new and improved fresh-complaint doctrine takes the “fresh” out of the doctrine.
III.
Playing Back Videotaped “Confession” Not Introduced in Evidence
Last, I cannot conclude, as the majority does, that the trial court properly exercised its discretion by playing back for the *630jury, during its deliberations, defendant’s videotaped “confession” that had never been moved into evidence. Indeed, in the circumstances here, the court was not invested with discretion. The plain language of our court rules permits a deliberating jury only to “take into the jury room the exhibits received in evidence.” R. l:8-8(a) (emphasis added). Here, the court allowed the jury to review an exhibit that had not been introduced into evidence. That the video itself had been played previously in open court is irrelevant. If a knife is marked only as an exhibit and observed by the jury during the trial, a request by the jury, during its deliberations, to view the knife would have to be denied because it was not moved into evidence. This is not a flexible rule. It is not applied pragmatically. Every trial attorney and trial judge knows that the jury can only receive items that have been moved into evidence.
State v. Burr, 195 N.J. 119, 948 A.2d 627 (2008), lends no support to the majority’s position. In Burr, the videotape played back to the jury “was admitted into evidence as an exhibit.” Id. at 133, 948 A.2d 627. In contrast, here the videotape was not admitted into evidence, although the jury did view the tape during trial. This was a videotaped trial. The jury was entitled to a replay of that portion of the trial during which defendant’s taped confession was played or to a reading of a transcript of that part of the trial—nothing more.
When a clear and sensible rule is breached with the blessing of this Court, it will be difficult to predict where and how the next incremental breach will occur.
IV.
Our ultimate charge is 'to ensure the integrity of the trial process. When that process has been compromised, as it was here, the public can have no confidence in a jury’s verdict. The errors in this trial were profound and undermined the fairness of the proceedings. They were not harmless. For these reasons, I respectfully dissent.
*631For affirmance—Chief Justice RABNER and Justices LaVECCHIA, RIVERA-SOTO and STERN (temporarily assigned)—4.
For reversal—Justices LONG, ALBIN and HOENS—3.
I do not suggest that character evidence that conforms to the Rules of Evidence would be impermissible. See N.J.R.E. 608.