Matter of Registrant GB

147 N.J. 62 (1996) 685 A.2d 1252

IN THE MATTER OF REGISTRANT G.B.: APPLICATION FOR JUDICIAL REVIEW OF NOTIFICATION AND TIER DESIGNATION.^[1]

The Supreme Court of New Jersey.

Argued September 10, 1996. Decided December 11, 1996.

The opinion of the Court was delivered by HANDLER, J.

This is another case concerning a set of bills collectively known as Megan's Law. The focus of the appeal is on the use of expert testimony in determining under that law the risk of reoffense posed by a paroled convicted sex offender.

We thus far have upheld the legislation requiring sex offenders to register with law enforcement, N.J.S.A. 2C:7-1 to -5, and the act requiring law enforcement, in certain circumstances, to notify community groups and individuals of the presence of sex offenders, N.J.S.A. 2C:7-6 to -11. Doe v. Poritz, 142 N.J. 1, 662 A.2d 367 (1995). More recently, we upheld the Attorney General's sex-offender classification guidelines, known as the Registrant Risk Assessment Scale ("RRAS" or "the Scale"). In re C.A., 146 N.J. 71, 679 A.2d 1153 (1996). Moreover, we have held that although the Scale is entitled to deference, it is not immune from challenge. Rather, it is merely a useful guide to determine the amount of notification that a community should receive. The responsibility for ultimately determining the proper scope of notification is left to the trial court after a hearing on the matter. Id. at 83, 679 A.2d 1153; Doe v. Poritz, supra, 142 N.J. at 12, 107, 662 A.2d 367.

The issue in the case presently before us is a relatively narrow one directed to the role of expert testimony in raising challenges to a registrant's Scale score and the scope of community notification. We hold that, in limited circumstances, expert testimony may be introduced at the judicial hearing in order to establish the existence of unique aspects of a registrant's offense or character that render the Scale score suspect. If believed, such evidence would lead to the conclusions that the Scale does not adequately represent the risk of recidivism for that particular registrant and that, therefore, in such circumstances the scope of notification should be more limited than that indicated by the registrant's Scale score and attendant tier classification.

I

Registrant G.B. was indicted for aggravated sexual assault, sexual assault, endangering the welfare of a child, and child abuse. The indictment stemmed from allegations that registrant had engaged in multiple sexual encounters with his minor female cousin over a seven-year period when she was between the ages of five and twelve; the encounters were alleged to have occurred in the family home. In a statement to the police in November 1991, registrant admitted to several of the alleged encounters, including instances of fellatio and one incident of intercourse. Subsequently, he pleaded guilty, pursuant to a plea agreement, to one count of second-degree sexual assault in violation of N.J.S.A. 2C:14-2b; the prosecutor agreed to dismiss the remaining charges. Registrant was sentenced to five years at the Adult Diagnostic and Treatment Center at Avenel ("Avenel").

In 1994, during registrant's sentence at Avenel, the Legislature enacted the Registrant Community Notification Laws ("RCNL" or "Megan's Law"). N.J.S.A. 2C:7-1 to -11. The overall purpose of the law was to protect the community from the dangers of recidivism by sex offenders. N.J.S.A. 2C:7-1a. The RCNL requires that convicted sex offenders found to be "repetitive and compulsive" register with local authorities. N.J.S.A. 2C:7-2. The law also specifies that the community be notified about certain sex offenders classified as moderate- or high-risk. N.J.S.A. 2C:7-8c. In 1995, this Court upheld the constitutionality of the RCNL. Doe v. Poritz, supra, 142 N.J. 1, 662 A.2d 367.

Registrant was released from Avenel in June 1995. Invoking the RCNL, the county prosecutor undertook to classify him according to his risk of reoffense. As a means of determining registrant's risk of reoffense, the prosecutor applied the Scale promulgated by the Attorney General pursuant to the RCNL. The Scale is composed of four categories: Seriousness of Offense, Offense History, Characteristics of Offender, and Community Support. Encompassed in these categories are thirteen factors deemed relevant by the Attorney General to the risk of reoffense. The Attorney General assigns a score to each registrant for each of the thirteen factors. A registrant is given a score of zero in each category if he is low risk, a one for moderate risk, and a three for high risk. The thirteen factors are then assigned varying weights, which are totalled to compute an overall Scale score. Those registrants receiving a score of at least seventy-four (out of a maximum of 111) are initially considered by the Attorney General to be "high-risk" or Tier Three. Registrants receiving scores between thirty-seven and seventy-three are considered "moderate-risk" offenders and fall within Tier Two, while those registrants scoring below thirty-seven are considered "low-risk" and are included within Tier One. See In re C.A., supra, 146 N.J. at 82-83, 679 A.2d 1153; see also id. at 111, 679 A.2d 1153 (reprinting Scale calculations sheet).

Under the Scale, the prosecutor ascribed to G.B. a numerical score of fifty-eight, resulting in a Tier Two or "moderate-risk" classification. Under the RCNL, moderate-risk offenders trigger targeted or selective community notification by the State. N.J.S.A. 2C:7-8c. Accordingly, the prosecutor determined that all schools within a two-mile radius of registrant's residence, as well as several other community organizations, would be notified of G.B.'s identity and presence.

The prosecutor based the classification on registrant's Avenel records, police reports, and witness statements. Several factors contributed to the Tier Two classification. First, the prosecutor determined that because of the nature of the contact between registrant and the victim and the victim's age, he posed a high risk to the community. Specifically, the prosecutor relied on the victim's allegation (and registrant's subsequent admission to the police) of penetration and the significant duration of the sexual relationship (approximately seven years). Second, the prosecutor cited registrant's prior convictions for burglary, theft, and illegal possession of a firearm as evidence of his "anti-social" behavior. Finally, the prosecutor pointed to registrant's poor treatment record at Avenel and his current lack of community and family support.

Registrant sought judicial review of the prosecutor's determination. At an in camera judicial hearing, he challenged the factual underpinnings of the State's calculation of his Scale score and the proposed scope of community notification.^[2] First, he argued that his degree of contact with the victim should not include penetration because he had pleaded guilty only to second-degree sexual assault and because the only evidence of penetration consisted of unconfronted allegations by the victim and registrant's uncounseled admission to the police. Second, he argued that because he had pleaded guilty only to one count of sexual assault, the duration of the sexual contact should be classified as less than one year (the lowest classification). Third, he asserted that his conviction for unlawful possession of a firearm was a regulatory offense and not indicative of anti-social behavior. Fourth, he contended that his community support was strong in that he had lived in the area for years. Finally, he asserted that the State's proposed community notification was overly broad and arbitrary in its blanket inclusion of all schools within a two-mile radius of his home.

The trial court rejected these arguments. It concluded that ample evidence existed of penetration and multiple years of sexual encounters and that actual convictions were unnecessary to justify reliance on these factors. Moreover, it accepted the description of unlawful possession of a firearm as "anti-social" and the State's contention that registrant had little community support. Lastly, the court found that notification of schools and select other organizations within two miles of registrant's home was not "arbitrary, capricious, unfair or unreasonable."

Registrant also sought to challenge the predictive value of the Scale for determining the risk of reoffense and the correctness of the Scale score as applied to the circumstances of his offense. To support these claims, he sought to introduce evidence from three experts: a psychiatric expert to evaluate his actual risk of reoffense; a statistical expert to analyze reoffense by sexual offenders, to evaluate whether registrant is the type of sexual offender likely to reoffend, and to determine the proper scope of community notification; and a human factors expert to study whether a nexus exists between community notification and the harm that the RCNL seeks to prevent. After considering registrant's request, the trial court ruled that expert testimony was unwarranted. Specifically, the court found that "if we are going to place great weight on the risk assessment scale, the testimony of these experts would be irrelevant because the factors in the Scale, at least the ones that are material here, do not call for expert testimony." Registrant then appealed.

The Appellate Division concluded that "the registrant should be permitted to retain an expert and to present testimony at the hearing to show that the variable factors [in the Scale calculations] as related to him, should result in a lesser Tier classification." 286 N.J. Super. 396, 407, 669 A.2d 303 (1996). Thus, the Appellate Division held that "registrant may attempt to prove that the variable factors applicable to him, demonstrate that he is so unlikely to reoffend, that he should be classified as a Tier One offender, notwithstanding his actual Scale score." Ibid. (emphasis added).

The Attorney General petitioned this Court for certification alleging that the Appellate Division's decision was too broad in permitting expert testimony and too lenient in allowing a registrant's Scale score to be overridden. We granted certification limited to those issues, 144 N.J. 174, 675 A.2d 1122 (1996), and now affirm.

II

We must consider the use of expert testimony and the weight to be accorded Scale score calculations under the RCNL within the framework of the role of the judiciary in overseeing this statutory scheme. As we have previously made clear in both Doe v. Poritz, supra, 142 N.J. 1, 662 A.2d 367, and In re C.A., supra, 146 N.J. 71, 679 A.2d 1153, because the RCNL implicates significant personal rights, the judiciary has an important nondelegable responsibility to ensure a proper balancing of the rights of registrants with the interests of the community.

A

In upholding the constitutionality of the RCNL, we concluded in Doe v. Poritz that (1) society may protect itself from convicted sex offenders; (2) it may do so as long as the protective means chosen are reasonably designed to safeguard the public and are not punitive; and (3) the laws in their application must in fact serve to protect society and not be punitive. 142 N.J. at 12, 662 A.2d 367.

In evaluating the statute, we determined that the liberty interests implicated by it were significant:

We believe a privacy interest is implicated when the government assembles those diverse pieces of information into a single package and disseminates that package to the public, thereby ensuring that a person cannot assume anonymity — in this case, preventing a person's criminal history from fading into obscurity and being wholly forgotten. Those convicted of crime may have no cognizable privacy interest in the fact of their conviction, but the Notification Law, given the compilation and dissemination of information, nonetheless implicates a privacy interest. The interests in privacy may fade when the information is a matter of public record, but they are not non-existent.

[Id. at 87, 662 A.2d 367.]

Due to the significant liberty and privacy interests at stake, the Court concluded that the judiciary had a special responsibility to ensure their adequate protection. We acknowledged the genuine threat to those interests posed by Tier Two and Tier Three classifications. Because "the statute sufficiently impinges on liberty interests to trigger both procedural due process and the fairness doctrine in our state," we held that "those [individuals] subject to the statute are entitled to the protection of procedures designed to assure that the risk of reoffense and the extent of notification are fairly evaluated before Tier Two or Tier Three notification is implemented." Id. at 30, 662 A.2d 367.

The judiciary's role was defined as overseeing the classification and notification of those registrants considered moderate- or high-risk who raised objections to their classification. To ensure due process, the Court required that sex offenders receive written notice of their tier classification and proposed scope of notification, and information concerning their right to retain counsel. Id. at 30-31, 662 A.2d 367. We also required prosecutors to make their files available to registrants and provided that offenders would have two weeks to object to the proposed classification and notification. Ibid.; see Registration and Community Notification Laws Bench Manual 46 (October 16, 1995) (hereinafter "Bench Manual").^[3]

For registrants who raised objections to the notification, the Court provided a judicial hearing at which a judge would be able to evaluate the merits of the parties' contentions. At the hearing, the State was given the burden of going forward with its prima facie case, consisting of that evidence justifying the proposed risk level and manner of notification. Doe v. Poritz, supra, 142 N.J. at 32, 662 A.2d 367. Once the prosecutor met the burden of going forward with the prima facie case, the offender bore the burden of persuading the court by a preponderance of the evidence that the proposed tier designation and notification did not conform with the laws and the Scale. Ibid.; Bench Manual, supra, at 60. As long as the prosecutor satisfied his or her burden, the trial court had to "affirm the prosecutor's determination unless it [was] persuaded by a preponderance of the evidence that it [did] not conform to the laws and Guidelines." Doe v. Poritz, supra, 142 N.J. at 32, 662 A.2d 367. If the court overruled the prosecutor's proposed tier designation, it had to state on the record the reason why the proposed designation did not conform to the law. Id. at 31-32, 662 A.2d 367; Bench Manual, supra, at 60.

The admission of evidence at the hearings was left to the sound discretion of the trial court, and the court was granted "broad powers in controlling the conduct of the summary proceedings." Bench Manual, supra, at 54.

The court shall control the manner of the summary proceeding, which shall be in camera, including determining whether and to what extent production of witnesses and cross examination shall be required or allowed, basing its determinations on the apparent complexity of the matter, the extent of doubt concerning the correctness of the level and manner of notification selected by the prosecutor, as well as the apparent need for prompt determination, presumptively present in all cases. The rules of evidence shall not apply and the court may rely on documentary presentations, including expert opinions, on all issues.

[Doe v. Poritz, supra, 142 N.J. at 31-32, 662 A.2d 367.]

In order to ensure that the laws were applied in a manner consistent with procedural due process and sufficiently protective of the liberty interests at stake, the Court imposed several important procedural requirements on the statutory scheme, including the following: (1) that registrants' behavior in the community following imprisonment be a factor in tier classifications; (2) that risk of reoffense, as demonstrated by psychological profiles, be used to decrease tier classifications as well as to increase such classifications; (3) that Tier Two classifications occur only after individualized determinations of which organizations will be notified; (4) that, for Tier Two classifications, only organizations that care for or supervise women or children be notified (as opposed to all who are likely to encounter the registrant); (5) that, for Tier Three classifications, only those likely to encounter the registrant be notified (as opposed to broad, public notification); and (6) that judicial review follow all Tier Two and Tier Three classifications.^[4]Id. at 29-30, 662 A.2d 367. The Court concluded that judicial review was necessary to protect the substantial interests at stake and was to be based on the trial court's independent review of the merits of the case.

B

In order for the State to meet more easily its prima facie case at the hearing and in order to comply both with the statute, N.J.S.A. 2C:7-8a, and with this Court's mandates in Doe v. Poritz, supra, 142 N.J. at 24 n. 2, 662 A.2d 367, the Attorney General promulgated the Registrant Risk Assessment Scale. We upheld that Scale in In re C.A., supra, 146 N.J. at 100-09, 679 A.2d 1153.

The Court in In re C.A. reiterated the point made in Doe v. Poritz that reviewing courts have an "obligation of providing procedural due process to ensure the appropriateness of a tier classification." Id. at 94, 679 A.2d 1153 (citing Doe v. Poritz, supra, 142 N.J. at 39, 662 A.2d 367). That obligation requires courts to hold an evidentiary and investigatory hearing, ibid., at which any trustworthy evidence is admissible. Id. at 95, 679 A.2d 1153.

The Court reaffirmed the structure of the hearing. The State has the initial burden of presenting a prima facie case, which may be built on reliable hearsay. Id. at 95-97, 679 A.2d 1153.

Once the State has proven its prima facie case, the burden of proof and persuasion shifts to the registrant, and the court "shall affirm the prosecutor's determination unless it is persuaded by a preponderance of the evidence that it does not conform to the laws and Guidelines." Doe v. Poritz, supra, 142 N.J. at 32, 662 A.2d 367. If the [registrant] produces proof, whether in the form of reliable hearsay, affidavit, or an offer of live testimony, that is sufficient to raise a "genuine issue of material fact," that the tier classification and the manner of notification are inappropriate, then the trial court should convene a fact-finding hearing and permit live testimony. Brill v. Guardian Life Ins. Co. of America, 142 N.J. 520, 534, 666 A.2d 146 (1995).

[In re C.A., supra, 146 N.J. at 96-97, 679 A.2d 1153.]

Once the registrant produces sufficient evidence to raise a question of material fact, the court must conduct a full evidentiary hearing at which time the State can supplement its proofs or rely solely on its hearsay allegations. Ibid. The court then must weigh the evidence and make a determination as to whether the registrant has met his burden of persuasion.

In In re C.A., we recognized that the Scale score was an important tool that the State could utilize both in meeting its initial burden of coming forward with a prima facie case and subsequently in rebutting evidence introduced by a registrant contrary to the State's position. Moreover, "[t]he Scale gives prosecutors a reasonably objective measure on which to assign registrants to the low, moderate, or high tier classification." Id. at 108, 679 A.2d 1153. The Scale, however, was found to be only one possible consideration. "The Scale ... is not a scientific device. It is merely a useful tool to help prosecutors and courts determine whether a registrant's risk of re-offense is low, high, or moderate." Ibid.

Because the Scale is merely a tool, we held in In re C.A. that "a court should not rely solely on a registrant's point total when it conducts a judicial review of a prosecutor's tier level classification or manner of notification decisions." Ibid. Therefore, "a court must review the correctness of the level and manner of notification selected by the prosecutor." Ibid. Even though "the Scale provides a useful guide for the prosecutors and court to evaluate risk of re-offense," the court must still make "a value judgment" in determining the proper tier classification and scope of community notification. Id. at 109, 679 A.2d 1153. Thus, courts are not "to blindly follow the numerical calculation provided by the Scale, but rather to enter the appropriate tier classification" based on all of the evidence available to them. Ibid. The determination of tier classification and scope of notification "are best made on a case-by-case basis within the discretion of the court." Ibid.

In sum, in In re C.A., we explained how the classification process would work. Sex offenders who the State initially determines, through the use of the Scale or other tools, should be classified as moderate- or high-risk are to receive notice informing them of the opportunity to object. At a hearing, the State must introduce evidence justifying its initial determination; in most instances that evidence will consist of the Scale score calculations with the factual bases underlying those calculations. A registrant then has the opportunity to present contrary evidence to persuade the court that the proposed tier classification is inappropriate. Based on all the evidence adduced and any further evidence secured by the court itself, see id. at 109, 679 A.2d 1153, the court shall make appropriate findings and conclusions. In re C.A., therefore, affirmed the principal first enunciated in Doe v. Poritz that the ultimate determination of a registrant's risk of reoffense and the scope of notification is reserved to the sound discretion of the trial court.

III

No one disputes that a registrant must be permitted to challenge the factual bases underlying the Scale score calculations. Thus, as occurred with G.B. below, a registrant can introduce evidence to show, for example, that the duration of the offense was not as long as the State maintains in its Scale calculations. Of course, as we noted in In re C.A., the State is free to rely on hearsay statements to support its assertions and does not need to base its calculations surrounding the underlying offense solely on the facts of conviction. 146 N.J. at 88-93, 679 A.2d 1153; see In re G.B., supra, 286 N.J. Super. at 403, 669 A.2d 303. However, the State's evidence is not immunized from challenge simply because the prosecutor can rely on hearsay and non-record facts. Rather, as the Attorney General recognizes, the ultimate determination of the facts underlying the Scale score calculations is left to the trial court after considering all relevant evidence.

Clearly, a registrant also can raise challenges to the "variable" factors that go into determining the Scale score. As noted earlier, the Scale consists of four categories that are accorded varying weights: seriousness of offense, offense history, characteristics of offender, and community support. Each of these categories employs different factors. The first two categories, seriousness of offense and offense history, have been termed "nonvariable factors"; the other two categories, characteristics of offender and community support, have been described as "variable factors." In re E.A., 285 N.J. Super. 554, 561, 667 A.2d 1077 (App.Div. 1995); see also In re C.A., supra, 146 N.J. at 104-05, 679 A.2d 1153 (terming the different factors "static" and "dynamic").^[5] Thus, a registrant, like G.B. here, can assert that his community support is greater than that recognized by the State in its Scale calculations. Again, the ultimate determination is left to the trial court after considering all relevant evidence.

More contentious is whether a registrant can raise challenges to his tier classification beyond those challenges, already described, which go solely to the propriety of the Scale calculations. As should be evident by the discussion of Doe v. Poritz and In re C.A., a registrant can make such challenges. The Scale is only a tool, albeit a useful one. It does not graduate to an irrebuttable presumption simply because it is properly and accurately computed. Rather, the Scale provides a guideline for the court to follow in conjunction with other relevant and reliable evidence in reaching an ultimate determination of the risk of reoffense posed by the registrant and the appropriate notification due the community.

The Scale, though, is more than just a piece of evidence. As previously recognized, it is sufficiently probative and reliable to fulfill the State's burden of presenting a prima facie case. In re C.A., supra, 146 N.J. at 107, 679 A.2d 1153. Thus, the Scale is presumptively accurate and is to be afforded substantial weight — indeed it will even have binding effect — unless and until a registrant "presents subjective criteria that would support a court not relying on the tier classification recommended by the Scale." Id. at 109, 679 A.2d 1153. Moreover, "[a]ny classification that is inconsistent with the classification based on the Scale is subject to judicial review by either side through appeal and any finding will have to be supported on the record." Ibid.

Registrant here also attacks the Scale itself. He challenges the predictive value of the Scale score calculations. In fact, registrant maintains that the Scale does not accurately measure the risk of reoffense. Registrant's general contention has already been rejected by this Court in In re C.A., and we continue to find the claim to be without merit. We exhaustively reviewed the factors that went into the Scale and the weight that those factors were afforded. Id. at 100-09, 679 A.2d 1153. With one notable exception,^[6] we held that "the Scale is an appropriate and reliable tool," which provides "a reasonably objective measure on which to assign registrants to the low, moderate, or high tier classifications." Id. at 107-08, 679 A.2d 1153. Thus, a registrant ought not to be allowed to continue to press the claim that the Scale is presumptively unreliable. Rather, we continue to maintain that the Scale is presumptively reliable.

Of course, in a particular case, a registrant may be able to establish that the Scale score for that registrant does not accurately reflect the risk of reoffense. In such a case, however, the registrant is limited to showing that the Scale did not accurately weigh certain factors as related to him or that the Scale did not take into account certain peculiar factors of the registrant's offense or history that might be relevant in determining his risk of reoffense. Such evidence would certainly be admissible.

We believe that few cases will involve facts that render the Scale score suspect. In fact, as previously stated, a Scale score calculation based on reliable evidence and properly performed is presumptively reliable. Only in the unusual case where relevant, material, and reliable facts exist for which the Scale does not account, or does not adequately account, should the Scale score be questioned. Those facts must be sufficiently unusual to establish that a particular registrant's case falls outside the "heartland" of cases.

We cannot define what those facts might be, but we can provide some examples. Here, G.B.'s offense occurred in the family home. The Scale calculations did not take this circumstance into account when computing his risk of reoffense. Registrant contends, with some support, that sexual offenders who commit their offenses within the family home pose less risk to the community than do other sexual offenders. See Robert J. McGrath, "Sex Offender Risk Assessment and Disposition Planning: A Review of Empirical and Clinical Findings," 35 J. Offender Therapy & Comp. Criminology 328, 334 (1991) (stating that the rate of recidivism among most incest offenders is approximately 10 percent or less, that the rate for rapists is between 7.7 percent and 35.6 percent, but that the rate for exhibitionists is 40.7 percent); Frank Tracey, et al., "Program Evaluation: Recidivism Research Involving Sex Offenders" in Joanne G. Greer & Irving R. Stuart, The Sexual Aggressor: Current Perspectives on Treatment 211-12 (1983). Of course, we express no opinion here about the validity of that proposition. Rather, we note that arguments based on such evidence, if found persuasive by a court, may support a claim that the Scale calculations, although accurately performed, do not accurately establish the risk of reoffense for a particular registrant. In such circumstances, a Scale score may be "overridden."

Another, more common, example of facts not currently taken into account by the Scale that may warrant a lowering of a particular registrant's tier category concerns a registrant's psychological state. In some instances, an expert evaluating a registrant may believe that the registrant's psychological profile makes him substantially less likely to reoffend than the general sex offender.^[7] The Scale takes into consideration psychiatric profiles and other variable factors in such a way that the categories may only add points to an offender's overall score (albeit fewer than the points to be added from the nonvariable categories). In other words, positive psychiatric or psychological profiles and positive post-sentence behavior can never lower an offender's overall score. Rather, positive numbers attributable to such variable factors serve only to add points, albeit fewer points, to a registrant's score. See Bench Manual, supra, at 23-25. Given the Scale's failure to consider positive psychiatric profiles and positive post-sentence behavior as true mitigating factors that can reduce the projected risk of reoffense, expert testimony may be essential for an accurate tier designation, even to the point of overriding the Scale score.

The benefit of allowing testimony to override a tier designation is that it can assist a court in arriving at a fairer and more accurate tier determination. Although experts opine that actuarial predictors are the best indicators of recidivism, it seems incongruous, given the statute's allowance for registrants to present evidence, see Doe v. Poritz, supra, 142 N.J. at 30, 662 A.2d 367, to afford a seemingly rehabilitated offender no opportunity to alter his tier designation and corresponding scope of notification.

A separate challenge that potentially could be raised to the Scale score concerns the scope of community notification. Thus, even a registrant whose Scale score was properly computed and whose case does not fall outside the "heartland" of cases in terms of his risk of reoffense may seek to narrow the scope of community notification. Although we have allowed this type of challenge, In re C.A., supra, 146 N.J. at 109, 679 A.2d 1153, we foresee few cases in which such a challenge will be successful. Indeed, the scope of notification for each tier categorization has been strictly defined by the Attorney General. See Bench Manual, supra, at 62-63. We uphold these notification provisions. However, in the unusual case, facts may exist that warrant a narrowing of the notification (or, perhaps, even the expansion of the notification).

We also conclude, in respect of the scope of notification, that the variable factors should contribute, perhaps through expert testimony, to narrow tailoring of community notification to each registrant's individualized situation. In cases of incest-type offenders, for example, registrants may be able to show that normal Tier Two or Tier Three notification is inappropriate, given the intra-familial nature of the offense.^[8]

In sum, we hold that a registrant is entitled to lodge three distinct challenges to his tier designation. First, a registrant may introduce evidence that the calculation that led to the Scale score was incorrectly performed either because of a factual error, because the registrant disputes a prior offense, because the variable factors were improperly determined, or for similar reasons. Second, a registrant may introduce evidence at the hearing that the Scale calculations do not properly encapsulate his specific case; or phrased differently, a registrant may maintain that his case falls outside the "heartland" of cases and, therefore, that he deserves to be placed in a tier other than that called for by the prosecutor's Scale score. Finally, a registrant may introduce evidence that the extent of notification called for by his tier categorization is excessive because of unique aspects of his case. Challenges to the Scale itself, or challenges to the weight afforded to any of the individual factors that comprise the Scale, are not permitted. Instead, all challenges must relate to the characteristics of the individual registrant and the shortcomings of the Scale in his particular case.

IV

We consider finally the admissibility, form, and weight of expert testimony in making any of the types of challenges that may be brought. For many of the challenges, expert testimony will be neither necessary nor helpful. For the remaining challenges, expert testimony in the form of reports or affidavits will suffice. The trial court has the ultimate authority to decide what weight to attach to the Scale and what weight to attach to expert testimony. It has the nondelegable responsibility to determine a registrant's tier classification and the scope of community notification.

This allocation of authority between courts and experts is firmly grounded. In In re D.C., 146 N.J. 31, 679 A.2d 634 (1996), the Court considered how trial courts should view their role in the presentation of expert testimony when a liberty interest is at stake. That case concerned the involuntary civil commitment of a paroled convicted sex offender, D.C. The Court, in concluding that D.C.'s commitment had comported with due process and fundamental fairness, was careful to warn trial courts not to allow their decisional processes to be overwhelmed by expert testimony:

The final determination of dangerousness lies with the courts, not the expertise of psychiatrists and psychologists. Courts must balance society's interest in protection from harmful conduct against the individual's interest in personal liberty and autonomy. The ultimate decision on dangerousness is, therefore, a legal one, not a medical one, even though it is guided by medical expert testimony.

[Id. at 59, 679 A.2d 634.]

We went on to state that "[courts must] reach a reasoned determination informed by and founded on [expert] evidence, but [are] not required to accept all or any part of the expert opinions." Id. at 61, 679 A.2d 634. Similarly, in State v. Krol, 68 N.J. 236, 344 A.2d 289 (1975), the Court defined the scope of civil involuntary commitment proceedings and periodic review hearings applicable to criminal defendants acquitted by reason of insanity. The Court made clear that the trial court, not the psychiatric expert, was the decisionmaker:

It should be emphasized that while courts in determining dangerousness should take full advantage of expert testimony presented by the State and by defendant, the decision is not one that can be left wholly to the technical expertise of the psychiatrists and psychologists. The determination of dangerousness involves a delicate balancing of society's interest in protection from harmful conduct against the individual's interest in personal liberty and autonomy. This decision, while requiring the court to make use of the assistance which medical testimony may provide, is ultimately a legal one, not a medical one.

[Id. at 261, 344 A.2d 289.]

See also State v. Michaels, 136 N.J. 299, 321-22, 642 A.2d 1372 (1994) ("[T]he relevance of expert opinion focusing essentially on the propriety of the interrogation should not extend to or encompass the ultimate issue of the credibility of the individual child as a witness."); State in the Interest of C.A.H. & B.A.R., 89 N.J. 326, 343-44 n. 5, 446 A.2d 93 (1982) ("[E]xpert opinion on [a juvenile's] rehabilitative prospects ... in a [juvenile] waiver hearing.... cannot be a substitute for the court's ultimate, highly discretionary decision, reached through an application of all of the statutory criteria to all of the relevant evidence, that the waiver of juvenile court jurisdiction is appropriate."); State v. Fields, 77 N.J. 282, 307-08, 390 A.2d 574 (1978) ("The final decision on the need for and appropriate extent of restrictions on the committee's liberty is for the court, not the psychiatrist.") (emphasis added).

In addition to taking care not to abdicate decision-making responsibility to experts, trial courts must ensure that tier designation hearings are not converted into prolonged battles of the experts. A significant number of registrants may attempt to introduce expert testimony tending to rebut their Scale scores. Routinely allowing expert testimony to dominate the proceedings and to override tier designations could compromise the goals of reliability and consistency in tier designations.

However, trial courts are quite capable of restricting such battles and keeping control of the proceedings. As the Court stressed in Doe v. Poritz:

[t]he procedures we have adopted are intended to assure fairness in implementing the law; they are not to thwart its implementation, and they should not be converted into long drawn-out contests between experts. Certainly we do not foreclose or discourage the production of expert testimony on both sides, but we grant to the court substantial power, beyond that permitted or used in ordinary litigation, to allow, reject, control, and limit expert testimony in order to render these proceedings administratively effective, practical, and timely.

[142 N.J. at 35, 662 A.2d 367 (emphasis added).]

Accordingly, we hold that a registrant shall be permitted to introduce expert evidence about his tier classification: (i) if such evidence tends to establish that the Scale score does not accurately or adequately take into account significant aspects of the registrant's character or prior offense; (ii) if such aspects would be relevant and material to the trial court's determination of tier classification; and (iii) if such evidence would, in the trial court's discretion, assist in the disposition of the case. Such evidence can be introduced, at the trial court's discretion, in the form of hearsay, written opinion, deposition-type testimony, or live testimony. If the trial court determines that the proffered evidence is not relevant to the issues raised, nor essential to their resolution, then it shall explain on the record the decision to exclude the evidence.

V

Here, registrant attempted to introduce the testimony of three experts. As explained by registrant's attorney:

[W]e seek three types of experts. I'd like to obtain a psychiatric expert to look at my client and evaluate the risk of recidivism, particular with respect to him, and also you'll see our objections later going to notification, who should be notified under tier two, if Your Honor finds him to be tier two classification. In that vein, I'd like to hire a statistical expert to review statistical analysis of recidivism and whether or not the studies and analyses indicate that my client is the type of person to recidivate, and more importantly, with respect to tier two notification, whether statistically, if he were to recidivate, or a person like him were to recidivate, whether it would be with persons who are members of the organizations that the Prosecutor's Office seeks to notify. And finally judge, I'd like to hire a human factors expert who will — who will be able to conduct a study and analysis of whether or not notification at all bears any nexus to the prevention of the type of harm that the law seeks to preclude.

Application of the standards we have set forth to the expert testimony at issue here indicates that the proffered testimony of the psychiatric expert could tend to establish that registrant's case includes important factors not addressed or not adequately addressed by the Scale score calculations. Thus, if the psychiatric testimony is believed, registrant's case may not be accurately or fairly characterized by the Scale score. The statistical expert's testimony potentially could strengthen the psychiatric expert's testimony by showing that the factors the psychiatrist relies on are statistically relevant and important factors that should be considered in weighing the registrant's risk of reoffense. That possibility, and its probative worth, however, would be subject to the trial court's assessment. The human factors expert, on the other hand, apparently would have no insights regarding any peculiar aspects of registrant's case. Rather, the human factors expert's proffered testimony merely would go to the accuracy of the Scale score in predicting the general risk of recidivism. We have upheld the Scale's predictive potential, In re C.A., supra, 146 N.J. at 107, 679 A.2d 1153, and the trial court would have no reason to revisit that decision.

This case, in particular, is one in which registrant attempted to overcome a principal deficiency in the Scale — the failure to give any meaningful role to the variable factors and the failure to take into account the risk of recidivism posed by a sexual offender whose offenses all occurred in the family home. Presumably, registrant's experts would have testified that registrant does not pose a substantial risk of reoffense because certain variable factors, as applied to him, make him far less likely to reoffend than other sex offenders. Also, registrant's experts could have testified that the scope of notification, as applied to him, was unreasonable given the fact that incest-type offenders are not as likely to reoffend in the community-at-large. That testimony, if believed, could warrant a lower tier classification or more limited community notification, and, therefore is admissible.

VI

For the foregoing reasons, we affirm the judgment of the Appellate Division and remand the case to the trial court for further proceedings not inconsistent with this opinion.

STEIN, J., concurs in result.

For affirmance and remandment — Justices HANDLER, POLLOCK, O'HERN, GARIBALDI, STEIN and COLEMAN — 6.

Opposed — None.

STEIN, J., concurring.

Although I have a separate view concerning the constitutionality of the retroactive application of the Community Notification Law, L. 1994, c. 128, see Doe v. Poritz, 142 N.J. 1, 111-47, 662 A.2d 367 (1995) (Stein, J., dissenting), I am in substantial accord with the Court's disposition of the issues presented by this appeal.

NOTES

[1] The initials used in the caption are fictitious. The Court has, for the purposes of confidentiality, refrained from identifying the names of those involved as well as the attorneys, municipality, and county in question.

[2] Registrant also challenged on constitutional grounds the validity of his classification and notification. He contended that the community notification portion of the RCNL was unconstitutional because, as applied to him, it violated due process and the ex post facto and double jeopardy bars. He also requested a stay of the classification hearing until a specialized Megan's Law unit of the Public Defender's Office was able to represent him and argued that the denial of a stay would violate his right to equal protection. Further, he argued that the Attorney General's guidelines were void because they had not been promulgated in accordance with the Administrative Procedure Act. The court summarily rejected these arguments, and the Appellate Division affirmed the trial court's rulings. 286 N.J. Super. 396, 669 A.2d 303 (1996). Our grant of certification did not encompass these issues, and, therefore, we do not address them.

[3] The notice would inform the offender of the date of a conference between 21 and 24 days from the date of anticipated service. Bench Manual, supra, at 46. The notice would also inform the registrant that at the conference, the court could decide the matter or require that there be further hearings to do so. Ibid.

The initial conference/hearing was intended to "narrow the issues," to address all discovery requests, and to decide on the use of experts. Id. at 56. If there were no material issues to decide, then the court could enter the appropriate order and summarily dispose of the case, placing its findings as to tier designation and manner of notification on the record. Id. at 56-57.

[4] The Court in Doe v. Poritz specifically instructed the Attorney General to revise the Guidelines in accordance with its opinion, particularly as to the manner in which the Scale considers psychological or psychiatric profiles. 142 N.J. at 24 n. 5, 662 A.2d 367; see In re C.A., supra, 146 N.J. at 106, 679 A.2d 1153. The Attorney General has not yet made those revisions.

[5] As the In re E.A. court explained:

Pertinent factors are both nonvariable and variable. Nonvariable factors are those that relate to all registrants. Variable factors are those that relate to the specific registrant and his community. More specifically, the nonvariable factors recognize the commonality of circumstances that relate to all registrants. They are factors based on common sense; factors that any reasonable person would recognize relate to the likely whereabouts of an adult in today's society. Variable factors recognize case-by-case circumstances that relate to the specific registrant before the court. They also relate to individualized circumstances of a community that may be encountered in the process of affording the public with proper notice.

[285 N.J. Super. at 561, 667 A.2d 1077.]

[6] As previously noted, supra at 77 n. 4., 685 A.2d at 1260 in In re C.A., we directed the Attorney General's Office to revise the guidelines or to establish that it had adequately responded to the Court's directive in Doe v. Poritz, supra, 142 N.J. at 24 n. 5, 662 A.2d 367, that psychological or psychiatric profiles be considered in support of low-risk assessments as well as high-risk assessments. In re C.A., supra, 146 N.J. at 106, 679 A.2d 1153.

[7] Even the drafters of the Scale realized that in certain circumstances the level of predictability of reoffense might be increased by using clinical interviews. However, the drafters determined that "the high cost associated with conducting such interviews for each registrant" could not justify including such interviews within the Scale calculations. In re C.A., supra, 146 N.J. at 106, 679 A.2d 1153. That decision was eminently reasonable. Ibid.

[8] Even the State has recognized the legitimacy of narrowly tailoring notification in such circumstances. During a hearing in In re E.W., the transcript of which was made part of the record in this appeal, the State's expert, Dr. Witt, commented that the issue of incest, and the high scoring of incest offenders in the Scale deeply troubled the committee drafting the Scale. Dr. Witt testified that "[t]he solution we would have liked is that we would still rate them wherever they fall, but the notification could be more individually tailored to the particular registrant." In re E.W., notification hearing (Law Division Feb. 26, Mar. 18-22, 1996) (sealed transcript). In such circumstances, notification may be limited to those parties most likely to be affected, namely, the family unit, family members' schools, and the like.