The opinion of the court was delivered by

COBURN, J.A.D.

This appeal from a pretrial ruling in favor of defendant concerns the admissibility before a jury in a murder case of opinion testimony from a social psychologist, offered by defendant as an expert in police interrogation and false confessions, on the subject of the credibility of defendant's confessions. We reverse.

I

The issue arose after a N.J.R.E. 104 hearing, resulting from defendant's motion to suppress his confessions as involuntary, Jackson v. Denno, 378 U.S. 368, 84 S.Ct. 1774, 12 L.Ed.2d 908 (1964), and as having been obtained in violation of Miranda v. Arizona, 384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694 (1966). The judge rejected the motion, ruling in favor of the State on both issues. Defendant then served the State with a report from the social psychologist, Dr. Saul M. Kassin, which contained his proposed trial testimony bearing on the circumstances of the interrogation and the credibility of defendant's statements.

The report read as follows:

February 27, 2001 REPORT State v. Patrick Free Analysis of the Defendant's Statement Saul M. Kassin, Ph.D. Professor of Psychology

I. Materials Reviewed Defendant's school records and reports Defendant's account of his interrogation and confession Report by psychologist Dr. Robert Gordan Catherine Suopys interview by Det. McGovern Agent Morgan Eric Blair interview by Dets. D'Ascentis Maahs Jason Abel interview by Dets. D'Ascentis Scalici Det. Leitenberger's polygraph exam questions and charts Tapes and/or transcripts of Defendant's three 1/9/98 statements Transcript of Jason Abel's 1/9/98 statement

Various police reports on the Suopys investigation Pretrial Miranda Hearing testimony, 7/10-7/13/00, 8/1/00. Interview training materials written and compiled by Det. Ryan

II. General Background

Confession evidence is generally assumed to be reliable. Over the years, however, numerous cases have been documented involving people who were erroneously convicted and imprisoned on the basis of persuasive confessions to crimes they did not commit. Recent DNA exoneration cases in the United States have confirmed that the elicitation of false confessions continues unabated. Among the first sixty-two prisoners exonerated by DNA evidence, fifteen had given full or partial false confessions.

It is not possible to discern the statistical prevalence of this problem. It is clear, however, that modern police interrogation techniques are psychologically powerful, that false confessions occur with some regularity, and that there are three distinct types of false confessions: voluntary, coerced-compliant, and coerced-internalized.

A voluntary false confession is a self-incriminating statement that is made without external pressure from police. Sometimes the goal is to protect a friend or relative. Other motives include pathological needs for attention, acceptance, recognition, or self — punishment. A coerced-compliant false confession occurs when a suspect confesses in order to escape an aversive situation, to avoid a threatened or implied harm, or to gain a promised or implied reward. This confession is thus an act of compliance by a person who has come to believe that the short-term benefits of confessing outweigh the long-term costs. A coerced-internalized false confession is one in which an innocent person who is anxious, tired, confused, and subject to certain suggestive methods of interrogation, actually comes to believe that he or she may have committed the crime. In this type of false confession, the suspect's memory of his own actions may be altered, a phenomenon closely related to studies involving the implantation of false "memories."

III. Patrick Free's Interrogations

In reviewing all the Defendant's statements, it is necessary to consider the conditions under which they were made — and how these conditions may have impacted upon his expectations, motives, and other behavior-relevant states of mind. Toward this end, the following aspects of this Defendant's interrogation raise serious cause for concern:

Interrogation time . Patrick Free was taken into custody 1/8/98 at 5:18 p.m. and concluded his last of three taped statements on 1/9/98 at 10:29 a.m., more than seventeen hours later. During that period of time, he was questioned, interrogated, and tested repeatedly, with only brief respites, by a team of investigators that included Captain King; Detectives Henry, Ryan, and Leitenberger; and perhaps Detective D'Ascentis-who, according to Det. Leitenberger, had brought Patrick in for the polygraph. For eleven of these hours, he consistently denied the charges in response to persistent accusations. He gave a statement pertaining

to the Suopys murder at 4:39-5:06 a.m., a revised statement at 6:02-7:09 a.m., and an unrelated third statement 10:02-10:29 a.m. At one point, according to Detective Henry, Patrick was confronted by four interrogators at the same time (7/13/00, p. 7).

By any normative or prescriptive measure, the length of this interrogation was excessive. Even Inbau, Reid, Buckley (1986), authors of Criminal Interrogations and Confessions (3e) — the manual on which the widely used Reid technique is based-advise investigators not to conduct "unreasonably long interrogations." They specifically advise against sessions that exceed four hours, and they describe as coercive the "participation of multiple interrogators over a lengthy time period."

Physical conditions. Patrick Free was held for 17 hours in a small, barely furnished room equipped with a two-way mirror, without contact with family or friends, and for a time frame that spanned an entire night. Regardless of how Patrick's interrogators may have perceived his level of alertness, he was, by the time he agreed to give his first statement, sleep-deprived — a state that causes physical and mental fatigue, heightens susceptibility to influence, and impairs decision-making. Other aspects of Patrick's physical state may also have increased his incentive to capitulate. Notably, he had not eaten since the morning of 1/8, was arrested before dinner, and was offered no meals, or even a snack, during the evening and early morning hours of 1/9.

Techniques used . Detectives involved in this investigation did not electronically record their interrogation sessions even though they had access to the tape recorder used to memorialize Patrick's three statements. When asked why they chose not to tape Patrick's consistent pre-statement story, which he told from 6-9:15 p.m. and from 10:20 p.m.-12:45 a.m., Detective Henry said that he and Detective Ryan "felt that Patrick was not being completely truthful with us" (7/13/00, p. 73).

As a result of this decision to selectively preserve what transpired in these sessions, there is no official record of Patrick's initial story, of the hours of prodding that elicited his first recorded confession, or of the prodding that elicited a revised second confession (e.g., whereas Patrick reports being threatened with Trenton State Prison, and being told of key details in Jason Abel's statement, Detective Henry could not recall these incidents). Despite this evidentiary limitation, however, it is possible to piece together from the various accounts some of what occurred during this lengthy interrogation.

Persistence. It is clear that Detectives Henry and Ryan presumed Patrick guilty at the outset despite his cooperative demeanor both before and after his arrest, despite his willingness to take a polygraph, and despite his repeated and consistent denials. Patrick describes in these sessions a relentlessness that is objectively confirmed by the mere passage of 17 hours. He claims he was interrupted whenever he tried to tell his story, a recollection that is highly consistent with common interrogation practices. Direct positive confrontation and the refusal to allow the suspect to mount a defense are designed to break the person down into a state of despair by communicating the futility of denial. Indeed, Detective Ryan specifically trains interrogators to "Use rhetorical questions that pre-suppose guilt or involvement" and "Use your hand to stop him from talking" (cf. Comprehensive Interview Program, Det. Edward B. Ryan, p. 56). Importantly, there is no evidence for the assumption that innocence protects people from the feelings of helplessness and despair that these tactics are designed to produce.

Presentation of false evidence . Detective Leitenberger administered a polygraph test that seemed designed more to extract a confession than to discern if Patrick was truthful or deceptive. There are three indications of this as the primary objective.

The first concerns the conditions under which the test was conducted. Detective Leitenberger commenced the pretest interview of Patrick at 1:30 a.m., started the polygraph at 2:20 a.m., and followed up with a posttest interview, which culminated in his 4:39-5:06 a.m. taped confession. The test was thus administered late at night, after eight hours of custody and interrogation, and with Patrick in a state of fatigue exacerbated by sleep deprivation. For the diagnostic purpose of obtaining valid results, examiners are required to ensure that examinees are psychologically fit. Toward that end, one wonders why Patrick was not permitted to sleep, eat, drink, and use the bathroom, before being tested — thus created an optimal diagnostic situation.

The second indication is that Detective Leitenberger concluded from the results-and reported to Patrick — that he had shown deception, an interpretation that is not supported by the data and has no basis in the logic or science of polygraphic lie detection. In her 8/1/00 testimony, Detective Leitenberger noted that Patrick exhibited arousal to both relevant questions and the control questions to which she presumed him to be non-truthful, a similarity that formed the basis of her assessment (8/1/00, pp. 125-131). (Question: "And are you saying that the physiological reactions would be the same if they were both lies?" Answer: "Correct"). Yet the Control Question Test is based on the specific and contradictory assumption that guilty examinees would exhibit a significantly larger response to crime-relevant questions, that innocents would exhibit the larger response to control questions, and that non-differences between relevant and control questions are inconclusive.

Third, assuming that Detective Leitenberger trusted her stated interpretation of the data, she expressed far more confidence than is warranted in her presentation of this assessment to Patrick. Polygraph experts who are familiar with the research literature know of the limitations (in particular, the unacceptably high rate of false positive errors), and so would be reluctant to imply that the results are conclusive (Iacono Lykken, 1997), as she did when querying Patrick after the test "To find out from him what he had not been truthful about" (p. 142). Whether by design or not, it appears that Patrick was confronted with false evidence in the form of a failed polygraph as a way to pressure a confession. This late disclosure, which was revealed to Patrick at about 3:30 a.m., triggered his eventual capitulation. Yet research shows that the presentation of false evidence can thrust a suspect — guilty or innocent — into a state of despair, lowering his expectation that it is possible to extricate himself from the situation through denial (as Patrick put it, "I figured I was already guilty because of the lie detector said I failed"). Not surprisingly, this form of trickery is implicated in numerous cases involving known coerced-compliant and — internalized false confessions.

IV. Patrick Free's Statements

To evaluate the validity of a disputed confession, it is necessary to closely examine the accompanying post-admission narrative, the conditions under which it was given, the techniques used to elicit it, and the extent to which the suspect recounted crime details that were accurate, knowable only to the culprit, and generative of new evidence. When a suspect-without any external sources of information-offers details that only the perpetrator could have known, or that lead to the discovery of new evidence, then his narrative can absolutely corroborate a guilty confession. However, when a suspect's story is incomplete or contains errors, when it is told in the language of uncertainty and inference, or when the facts recounted are knowable via second-hand sources, then it fails to corroborate the guilt implied by the admission.

Examination of the Defendant's post-admission narratives reveals many problems. The two murder statements are inconsistent with each other, they contain internal inconsistencies (e.g., in the 4:39 confession, Patrick described an incident that was not premeditated but then claimed to have armed himself for it), and they contain obvious external inconsistencies (e.g., with Jason Abel's 1/9 account). These confessions also contain odd gaps in knowledge (as when Patrick could not describe the knife he said he used and could not recall what happened to it afterward) and the kind of qualified language people use when describing a guess, theory, or inference, as opposed to an actual first-hand memory (e.g., "I may have left it in Jay's car," "Probably two or three times," "Wiped it off, probably, and left it in the car, "On the ground, maybe").

There is also no indication that Patrick's statements contained crime details that were novel or knowable only by first-hand experience. To be sure, his final narrative does describe an event that generally fits the facts of the crime. But it does not contain details not already known by the police and did not lead to the discovery of new evidence. To the contrary, despite the taped confession, Patrick was unable to direct the police to the weapons they sought that he said had been discarded.

To further contaminate matters, Patrick received details about the incident from second hand sources outside the interrogation room (e.g., the newspaper) and, even worse, from sources inside the interrogation room as well. Detective Leitenberger disclosed in her 8/1/00 testimony that before taking Patrick's first statement (in which he said for the first time that he took a knife from home), she and Patrick had an off-tape conversation about the knife and where it came from (pp. 156-7, 162). Then when Patrick did not respond to a more specific question about where in the house the knife was from, Detective Leitenberg asked, "Was it a regular kitchen knife?" Similarly, Detective Henry testified that some time after his 4:39 a.m. confession, Patrick was given a summary of Jason Abel's account, which was inconsistent, and the ensuing "comparison" of statements was used to shape a revision in the form of Patrick's 6:02 a.m. confession (7/13/98, pp. 111-115). These types of disclosures influenced and contaminated Patrick's narrative, making it difficult to discern what, if anything, he knew, and how he knew it.

V. Conclusion

The highly coercive circumstances under which Patrick Free agreed to confess and the production of a post-admission narrative that was redundant with what the police already knew and contaminated by his exposure to external sources of information, render his statements ambiguous and potentially unreliable. In many ways, this case resembles others involving documented coerced-compliant and -internalized false confessions. On the subject of the Adam Suopys murder, it is my opinion that both recorded statements should be treated with extreme caution.

The State moved to preclude Kassin from testifying, and the judge, after considering Kassin's report and the parties' legal arguments, ruled almost entirely for defendant, entering an order, which reads in pertinent part as follows:

Kassin may testify as an expert witness . . . on the issue of whether the defendant's statements were knowingly and voluntarily given, said testimony to be limited to the general confines of the report . . . and to the extent that opinions can be given with a reasonable degree of psychological certainty. The witness may offer opinions based on the length of the interrogation, the physical conditions under which the questioning took place, the availability of food and drink, and the techniques used by the interrogating detectives, including the administration of a polygraph examination, and including his opinion that the contents of the statements should be treated with extreme caution. . . . [He] may not testify as to his opinion with respect to accuracy of the results of the polygraph examination. . . .

[Emphasis added.]

The State moved for leave to appeal, arguing that the proposed testimony was inadmissible under N.J.R.E. 702 because the credibility of a defendant who confesses is not properly the subject of expert testimony and because defendant had "failed to establish that the claimed area of expertise of the witness is one which has been demonstrated to be reliable." We granted the State's motion, and, while retaining jurisdiction, remanded for a N.J.R.E. 104 testimonial hearing.

The hearing took eight days. Dr. Kassin testified for three days. The State responded with the testimony of Dr. Michael Welner, a psychiatrist, who testified for five days.

In Dr. Welner's opinion, the study of false confessions by social psychologists, such as Dr. Kassin, currently lacks a reliable scientific foundation. While he conceded that Dr. Kassin's "typology" of false confessions (voluntary false, coerced-compliant, and coerced-internalized, described above in Dr. Kassin's report) may serve as a useful description of the process which has led historically to a number of documented false confessions, he stated that it has never been accepted as a predictor of whether a particular confession is true, false, or questionable. We will not detail his testimony because the judge did not find that Dr. Kassin's evidence was based on science. Instead, without expressly rejecting Dr. Welner's testimony, he found that it was admissible as specialized knowledge.

Dr. Kassin's testimony, despite its length, added little to the contents of his report. He discussed in general some of the strategies used by police interrogators, such as: isolating the defendant; subjecting the defendant to prolonged questioning; confronting the defendant with evidence of guilt, whether real or fabricated; and minimizing the consequences of confession. According to him, the presence of those factors increases the likelihood to an unknown degree of procuring a false confession. He then discussed at length the circumstances surrounding Free's interrogation, concluding that despite the presence of the cited factors, he could not say that the confessions were false or unreliable. Instead, he phrased his conclusion, in weaker terms than those contained in his written report, as follows:

What I do think, that there are risk factors present which leads me to be — to recommend being cautious about the statement. I don't know if it's an accurate statement or not, but there are certainly risk factors present that would give me pause.

However, he conceded that in 1997, he had himself written, while addressing the psychological work in this area, that the "`current empirical foundation may be too meager to . . . qualify as a subject of `scientific knowledge'. . . .'"

At the conclusion of the hearing, the judge made the following determinations. He found that both experts were qualified in their respective areas. As to Dr. Kassin, he found that "he has devoted a fair portion of his academic and professional career to the study of forced and false confessions." He also found that Dr. Kassin had been qualified in other jurisdictions to offer opinions on police interrogation. He noted that Dr. Kassin had "dealt with the situation or circumstances of Patrick Free's statements but not with Patrick Free individually to any meaningful degree." He added that Dr. Kassin's opinions were "based on the setting in which the statements were given and not on Patrick Free as an individual."

After describing the testimony elicited from both experts at the hearing, he resolved the legal issue in the following manner. Without citation, he observed that "[s]ome courts have allowed expert testimony in the area of disputed confessions in appropriate cases." Then, based on his understanding of N.J.R.E. 702 and the principles set forth in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786,123 L.Ed.2d 469 (1993), the judge had this to say:

Not every case involves scientific principles or methodologies solely because of the area that's involved, for example, some business evaluations, some malpractice cases, evaluations of art. Those kinds of cases allow expert testimony and are not in and of themselves necessarily scientific. Clearly, expert opinions are not always hard science. There may be numerous non-scientific areas where experts are allowed to offer opinions, such as the examples I've given.

I find that these areas should also include the fields of psychology and psychiatry.

Turning to the subject of reliability, he then said:

I find that although the area of disputes confessions has been studied only a little and that there are not many studies yet published so that we cannot know the frequency of an error rate, but this is not a truly scientific area and the rules must be relaxed. This is an area involving people. The impact upon them of social stressors. Those stressors may be situational . . . This is, I find, not an area readily subject to investigatory techniques. It is, instead, an area which should be determined by the jury and not by a judge exercising a gate-keeping function.

Without explaining why, the judge concluded that Dr. Kassin's testimony was sufficiently reliable, and that he could discuss

the general circumstances surrounding interrogations, those specifically involving Patrick Free and opine that the circumstances of Patrick Free's interrogation should be treated with caution as they involve the voluntary aspects of his statements. He may not comment on [the] truth or falsity of those statements.

That determination was not embodied in an order. However, we surmise, and the parties agree, that by that statement the judge intended to reiterate the contents of the order previously filed.

II

We begin by noting that under the United States Constitution, a criminal defendant is certainly entitled to present to a jury "competent, reliable evidence bearing on the credibility of a confession. . . ." Crane v. Kentucky, 476 U.S. 683, 690, 106 S.Ct. 2142, 90 L.Ed.2d 636, 645 (1986) (emphasis added). Furthermore, we acknowledge that some forms of psychological testimony on the credibility of a defendant's confession have been found to be admissible because they were considered to be reliable. See, e.g., U.S. v. Shay, 57 F.3d 126 (1st Cir. 1995); Beagel v. State, 813 P.2d 699 (Alaska Ct.App. 1991); andCommonwealth v. Banuchi, 141 N.E.2d 835 (Ma. 1957). However, in each of those cases the psychological testimony concerned scientifically recognized mental disorders relevant to each defendant's confession, rather than, as here, testimony about the effects, in general, of police interrogation techniques. We also acknowledge that a few decisions from other jurisdictions have permitted psychological testimony on the effects of police interrogation. See infra. For now, we merely note that none of them went as far as the trial judge did here, and that, in any case, as shall appear, we do not find them persuasive.

In determining the admission of expert testimony, the first reference point is N.J.R.E. 702, which provides:

If scientific, technical or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education may testify thereto in the form of an opinion or otherwise.

The commentary following N.J.R.E. 702, after discussing the need for expert testimony to be reliable, notes that "[t]his requirement of reliability applies to all scientific fields, including the social and psychological sciences." Biunno, Current N.J. Rules of Evidence, N.J.R.E. 702 comment 3 (2002). The New Jersey cases have consistently considered psychological testimony as falling within the category of scientific evidence. See, e.g., State v. Papasavvas, 163 N.J. 565, 613 (2000); State v. Michaels, 136 N.J. 299, 307-23 (1994); State v. J.Q., 130 N.J. 554, 562-74 (1993); State v. Long, 119 N.J. 439, 494-96 (1990); State v. Kelly, 97 N.J. 178, 208-12 (1984);State v. Cavallo, 88 N.J. 508, 516-20 (1982) and State v. Hurd, 86 N.J. 525, 536 (1981). Those cases contradict the trial judge's conclusion that Dr. Kassin's testimony could be admitted even though it lacked a scientific basis.

As scientific evidence, to be admissible in New Jersey, Dr. Kassin's testimony was required to satisfy the Frye test. See Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923). In State v. Spann, 130 N.J. 484, 509 (1993), the Court observed that "the `general acceptance by the relevant scientific community' test, established inFrye, substantially is still the law in New Jersey." (citations omitted). More recently, the Court said that "[i]n criminal cases we continue to apply the general acceptance or Frye test for determining the scientific reliability of expert testimony." State v. Harvey, 151 N.J. 117, 169 (1997), cert. denied, 528 U.S. 1085, 120 S.Ct. 811, 145 L.Ed.2d 683 (2000). In a footnote in his dissenting opinion in Harvey, Justice Handler observed: "Although the `general acceptance' test still governs the admission of scientific evidence [in criminal cases], I am not convinced that such a high standard would justify the exclusion of similar evidence proffered by a defendant in a capital prosecution." 151 N.J. at 117, n. 9. He then added, in pertinent part, "The notion of different standards for the admission by the prosecution and the defense is common in criminal law in general and death-penalty jurisprudence in particular." Ibid. That notion, however, has not yet been endorsed by the Court with respect to the receipt of scientific evidence. See, e.g.,Michaels, supra, 136 N.J. at 312, and Cavallo, supra, 88 N.J. Super. at 523, both applying the "general acceptance" test to defense offers of psychological evidence.

In this case, the trial judge did not rely on the Frye test, instead citing the United State Supreme Court's liberalization of the Frye test in Daubert, supra. Since that was the wrong course to follow under Spann and Harvey, and since there was insufficient evidence of general acceptance of Dr. Kassin's offered principles and methodology in the relevant scientific community, the judge's legal conclusion cannot be accepted.

Since one of the methods of proving general acceptance of a scientific theory is by reference to "judicial opinions that indicate the expert's premises have gained general acceptance," Kelly, supra, 97 N.J. at 210, we turn to the applicable cases from other jurisdictions.

Some support for defendant's position may be found in U.S. v. Hall, 974 F. Supp. 1198 (C.D.Ill. 1997) aff'd on other grounds, 165 F.3d 1095 (7th Cir. 1999), cert. denied, 527 U.S. 1029, 119 S.Ct. 2381,144 L.Ed.2d 784 (1999); Callis v. State, 684 N.E.2d 233 (Ind.Ct.App. 1997); People v. Lopez, 946 P.2d 478 (Colo.Ct.App. 1997); and People v. Page, 2 Cal.Rptr.2d 898 (Cal.Ct.App. 1991).

In Hall, supra, the defendant made a proffer of psychological evidence along the same lines as that offered in this case. The federal trial court, after a pretrial evidentiary hearing, applied Daubert and determined that the evidence would be admissible, not as scientific evidence, but as "specialized knowledge" under the identical federal version of our N.J.R.E. 702. Id. at 1200, 1204-05. However, the court also severely limited the scope of the testimony, ruling that the expert, Dr. Richard Ofshe,

will only testify to the correlation between false confessions and the various factors espoused by him. Thus, he can testify that false confessions do exist, that they are associated with the use of certain police interrogation techniques, and that certain of those techniques were used in Hall's interrogation in this case. Dr. Ofshe cannot explicitly testify about matters of causation, specifically, whether the interrogation methods used in this case caused Hall to falsely confess. Without experimental verification, such testimony would be speculative and prejudicial. Dr.

Ofshe will simply provide the framework which the jury can use to arrive at its own conclusions.

[Id. at 1205.]

The court also addressed the question whether the proposed evidence satisfied that portion of the federal rule which, like our N.J.R.E. 702, requires that the testimony must be helpful to the trier of fact to be admissible. It stated that "Dr. Ofshe testified that a common misperception among the public is that once a person confesses to his guilt, he must be guilty." Id. at 1206. In support of that proposition, the court merely noted that Dr. Ofshe's challenge to this misperception is "based on systematic observation of data to which the jury is not privy." Ibid. The supposed data is not, however, described at all in the opinion. Nonetheless, based on the assumption that this misperception is prevalent, the court concluded the testimony would be useful and admissible since "social scientists in particular may be able to show that commonly accepted explanations for behavior are, when studied more closely, inaccurate." Id. at 1205-06 (citation omitted).

We find Hall unpersuasive. First, it is based on the Daubert standard rather than the Frye test, and is thereby inconsistent with our state jurisprudence. See People v. Green, 683 N.Y.S.2d 597, 600 (N.Y.App.Div. 1998), lv. denied 711 N.E.2d 649, 689 N.Y.S.2d 435 (N.Y. 1999), rejectingHall for this very reason in a case in which a defendant asked to present a psychologist who would testify that he was particularly susceptible to providing a false confession. Second, our cases, see supra, do not support categorizing this evidence as "specialized knowledge" rather than "science." And third, there is neither sufficient evidence in Hall, or, for that matter, in the record in this case, to support the proposition that the general public believes that a person who confesses must be guilty.

In Callis, supra, the defendant also offered the testimony of Dr. Ofshe. Before trial, the court ruled the evidence admissible but "`limited to testimony regarding the circumstances surrounding the statements [made by the defendant to the police] as it may affect the weight to be given such testimony.'" 684 N.E.2d at 239. On appeal, the court noted that at trial,

Dr. Ofshe testified generally about false confessions and the State offered no objection as to his expertise. When asked his opinion about the interrogation process in this case, the trial court sustained an objection to the testimony stating "my problem with that particular question is that it does invade the province of the jury in terms of the reliability of the statement. . . . what he cannot do is take his opinion as to the reliability of this particular confession."

[Ibid.]

The defendant appealed, and the court sustained the ruling; however, the opinion contains no indication of the testimony that might have been submitted by Dr. Ofshe on the question of the admissibility of his opinions nor does it indicate whether it is employing Frye or Daubert. Moreover, to the extent the court found the general testimony admissible, the opinion is dictum since the State had neither objected nor attempted to raise the issue on appeal. We note in passing that even under this opinion, virtually all of the analysis contained in Dr. Kassin's written report would be inadmissible.

In Lopez, supra, the defendant had offered a psychologist's "expert testimony [that] would have described the psychological environment in which he made the inculpatory statements." 946 P.2d at 484. Without any description of the basis for the testimony or reference to any legal standard for the admissibility of scientific evidence, the court simply cited Crane, supra, as "dispositive," and therefore concluded that the evidence was admissible. Id. at 482. We find that reasoning entirely unpersuasive. Crane requires that reliable evidence be submitted, while the Lopez opinion contains nothing to support the proposition that the evidence in question was reliable.

In Page, supra, the court was also concerned with the testimony of a social psychologist who was permitted to testify about "the factors which can lead a person to give an inaccurate statement in an interrogation setting." 2 Cal.Rptr. at 908. The trial judge had permitted the expert to testify about the "general principles of social psychology," but had refused to let him "specifically relate these principles to Page's statements, or to give his opinion concerning the reliability of the confession." Id. at 909. Page complained about the restrictions on appeal, and the court sustained the lower court's ruling.

Although California appears to be a Frye state, id. at 909, the opinion, though it describes the expert's proposed testimony in detail, does not expressly resolve the admissibility issue by reference to Frye. Instead, the opinion relies primarily on People v. McDonald, 208Cal.Rptr. 236, 690 P.2d 709 (Cal. 1984). In that case the Supreme Court of California, reaching a result contrary to Long, supra, 119 N.J. at 494-96, and State v. Cromedy, 158 N.J. 112, 133 (1999), concluded that expert testimony on the psychological factors affecting the accuracy of eyewitness identification was admissible. The court reasoned that "the body of information now available on these matters is `sufficiently beyond common experience' that in appropriate cases expert opinion thereon could at least `assist the trier of fact.'" 208 Cal Rptr. at 248; 690 P.2d at 721 (citation and footnote omitted).

Apparently, the court in Page reached the same conclusion with respect to the psychologist's testimony regarding the general psychological factors that might lead to an unreliable confession. However, that determination does not meet our Supreme Court's jurisprudence, which, as we have noted, requires adherence to Frye. In passing, we note that, as in Callis, supra, the limitations placed on the evidence in question would eliminate virtually all of the material contained in Dr. Kassin's report, which is almost entirely focused on analyzing the particular circumstances surrounding Free's confessions.

Support for the State's position may be found in State v. Tellier, 526 A.2d 941 (Me. 1987); Kolb v. State, 930 P.2d 1238 (Wyo. 1996); andState v. Ritt, 599 N.W.2d 802 (Minn. 1999), cert. denied, 528 U.S. 1165, 120 S.Ct. 1184, 145 L.Ed.2d 1090 (2000).

In Tellier, supra, the defendant was prevented from eliciting before the jury a psychologist's testimony designed "to provide background and to educate the jury on the phenomenon of false confessions. . . ." 526 A.2d at 942. At a hearing to determine admissibility, the psychologist

offered no opinion with respect to whether Tellier's admissions were in fact false, nor did he examine Tellier or inform himself of Tellier's psychological profile. He merely testified that based on his understanding of the Tellier murder case, he could not rule out the possibility that the admissions were false. [The psychologist] knew of no empirical studies with respect to the frequency of false confessions in cases of this type.

[Id. at 943 (footnote omitted).]

In affirming the rejection of the testimony, which defendant indicated would have given the jury "background information about false confessions in general," ibid., the Supreme Judicial Court of Maine observed

[The psychologist's] testimony, however, amounted to nothing more than an assertion that false confessions do occur. His testimony was so abstract, vague and speculative that its relevance and probative value was virtually nil.

[Id. at 944.]

In Kolb, supra, the Wyoming Supreme Court sustained the rejection of testimony which the psychologist described initially as involving "`False Confession Syndrome.'" 930 P.2d at 1241. However, in the evidentiary hearing before the trial court, he indicated it was "not a syndrome, but is better understood as a collection of reasons why people may give false confessions." Ibid. Based on the rather severe inadequacies in the presentation by the psychologist, the trial court found that the testimony was "scientifically unreliable and would not assist the jury with understanding the evidence nor in deciding a fact in issue." Id. at 1242.

In Ritt, supra, the Supreme Court of Minnesota was asked to review the exclusion of a psychologist's testimony about the reliability and effect of a particular, but undescribed, technique of interrogation. 599 N.W.2d at 810. More specifically, the defendant had asked to be allowed to have her expert review the videotaped confession, pointing out those aspects that were coercive. Ibid. Since there appears not to have been an evidentiary hearing on the issue, any scientific justification the expert might have supplied is not in the opinion. The court sustained the ruling under a rule virtually identical to N.J.R.E. 702, in essence, finding that the proposed testimony was "unlikely to add either precision or depth to [the jury's] evaluation." Id. at 812. However, the discussion is brief and not particularly enlightening.

After reviewing the opinions from other jurisdictions, we are not satisfied that Dr. Kassin's "premises have gained general acceptance,"Kelly, supra, 97 N.J. at 210, and, therefore, we are convinced that the opinions offered in his report are inadmissible as not scientifically reliable.

We are also convinced that the proposed evidence failed to satisfy the evidentiary requirement that the nature of expert testimony offered must be such as "will assist the trier of fact to understand the evidence or to determine a fact in issue." N.J.R.E. 702. This aspect of the direct predecessor to this Evidence Rule was described in Kelly, supra, as requiring proof that "the intended testimony . . . concern a subject matter that is beyond the ken of the average juror." 97 N.J. at 208.

In Long, supra, the Court found that psychological testimony on eyewitness identification was unnecessary and inappropriate since the problems implicated were well within the knowledge of ordinary people. 119 N.J. at 495-96. And in Cromedy, supra, the Court reached the same conclusion respecting cross-racial identification because of "`the widely held commonsense view that members of one race have greater difficulty in accurately identifying members of a different race. . . .'" 158 N.J. at 133 (citations omitted).

Although our rules of evidence recognize that people do not usually make statements against their penal interest unless they are true,N.J.R.E. 803(c)(25), it does not follow that ordinary jurors believe that all confessions made by defendants subjected to police interrogation are true. Moreover, the coercive factors mentioned by Dr. Kassin, such as isolation, persistent questioning, confrontation with real or fabricated evidence of guilt, and minimization of the consequences of confession, are all matters that a jury would recognize as having a potential for causing a false confession.

Furthermore, since Dr. Kassin cannot identify the degree to which the presence of one or more of these factors might cause a false confession, his opinions, as limited by what we might describe as the most liberal cases, Hall, Callis, Lopez, and Page, supra, would be of no assistance to the jury. What the jury would be left with under those cases was accurately categorized by the Supreme Judicial Court of Maine as "nothing more than an assertion that false confessions do occur." Tellier, supra, 526 A.2d at 944. Also, as that Court further observed, the testimony would be "so abstract, vague and speculative that its relevance and probative value [would be] virtually nil." Ibid.

The burden of proving the admissibility of the proposed scientific evidence was on defendant. Harvey, supra, 151 N.J. at 167. In reviewing a decision to admit expert testimony, we recognize that generally such matters rest in the sound discretion of the trial court. State v. Berry, 140 N.J. 289, 293 (1995) (citation omitted). However, we are required to reverse when there has been "a clear abuse of discretion." Little Egg Harbor v. Bonsanque, 316 N.J. Super. 271, 278 (App.Div. 1998) (citations omitted). For the reasons stated above, we are satisfied that such is the case here, and that the trial judge erred in ruling that Dr. Kassin could testify as an expert in this case on the subject of coerced confessions.

Reversed and remanded for further proceedings not inconsistent with this opinion.