The opinion of the court was delivered by
STEIN, J.In this appeal we address the contention of the New Jersey State Parole Board (Parole Board or Board) that the Appellate Division committed error in reversing its decision to deny parole to respondent Thomas Trantino. Trantino v. New Jersey State Parole Bd., 331 N.J.Super. 577, 624-25, 752 A.2d 761 (2000) (Trantino V).
This Court has considered appeals from this inmate on three prior occasions. In 1965 we upheld his conviction for murder and his death sentence, later commuted to life imprisonment. State v. Trantino, 44 N.J. 358, 209 A.2d 117 (1965), cert. denied, 382 U.S. 993, 86 S.Ct. 573, 15 L.Ed.2d 479 (1966), reh’g. denied, 383 U.S. 922, 86 S.Ct. 901, 15 L.Ed.2d 679 (1966). In 1982, we set aside the Parole Board’s decision to grant parole, with restitution as a condition, and remanded the case to the Parole Board to “reassess the punitive aspects of Trantino’s sentence,” resulting in the Board’s decision to deny parole and impose a ten-year future eligibility term. In re Trantino Parole Application, 89 N.J. 347, 446 A.2d 104 (1982) (Trantino II). In 1998, we reversed the Appellate Division’s judgment upholding the Parole Board’s denial of parole on the ground that the Board’s decision was based on an incorrect standard and was not supported by sufficient evidence, and remanded the matter to the Board to redetermine Trantino’s eligibility for parole. Trantino v. New Jersey State Parole Board, 154 N.J. 19, 22-23, 711 A.2d 260 (1998) (Trantino IV).
*121We now affirm in part and modify in part the judgment of the Appellate Division, and thereby uphold that court’s unanimous reversal of the Parole Board’s decision to deny parole. The Parole Board is ordered to grant Trantino parole subject to the pre-release condition of satisfactory completion of a twelve-month halfway house placement and such other pre- and post-release conditions that it may impose. The Department of Corrections is ordered to place Trantino within thirty days in a halfway house facility within a reasonable proximity to the Camden/Cherry Hill area.
The length of this opinion reflects Trantino’s thirty-seven years of incarceration and his numerous prior attempts to achieve parole. Our opinion explains that if Trantino’s heinous crimes had been committed after 1997 he would never be eligible for parole, but that under the prevailing law he has been eligible for parole since 1979. We also explain in detail our conclusion that the Parole Board acted arbitrarily and capriciously, and not on the basis of a preponderance of the evidence in the whole record, in deciding to deny parole. We conclude that the Parole Board’s extensive reliance on evidence relating to a 1956 robbery, assaultive conduct in 1963 with his first wife, violations from 1961 to 1963 of conditions of his parole from a New York prison, efforts by attorneys in 1974 to remove a New York detainer, and other evidence relating to events occurring prior to the 1963 murders was arbitrary and capricious. That evidence provided no substantial support for the Board’s conclusion that Trantino was substantially likely to commit another crime if released on parole now. The Board’s reliance on evidence of such remote events was a makeweight to compensate for the lack of substantial evidence to support the Board’s conclusions. Moreover, the Board’s insistence on the relevance of such evidence substantially undermined the deference that courts ordinarily confer on agency decisions.
We also conclude that the Board’s virtually exclusive focus on the 1999 testimony of Dr. Glenn Ferguson to find that Trantino’s psychological profile made him substantially likely to recidivate *122further demonstrated the arbitrariness of the Parole Board’s decision. As our opinion explains, the Board’s determination was not based, as the law requires, on a preponderance of the evidence, but rather on the Board’s selective reliance on only that limited testimony that possibly could support a denial of parole. The Board completely disregarded substantial evidence in the record, including three prior recommendations of parole by Dr. Ferguson, that was significantly supportive of parole. It also ignored the conclusions contained in numerous other psychological evaluations — including that of the psychological expert retained by the Attorney General — that Trantino presented a low risk of recidivism. The seleetiveness of the Parole Board’s review of this extensive record further undermines the deference to which its decision ordinarily would be entitled.
We also explain that agencies of government, like the Parole Board, cannot be allowed to apply the rule of law selectively, exempting from its coverage those least favored by society.
We previously have expressed our abhorrence of Trantino’s crimes:
The brutality of those crimes, whose victims were a Lodi police sergeant and a police trainee, unquestionably is seared not only in the memories of the victims’ families and friends but also in the consciousness of society. From the standpoint of retribution, perhaps no prison sentence, whatever its length, is sufficiently severe.
[Trantino IV, supra, 154 N.J. at 43-44, 711 A.2d 260],
Thus, our decision today is mandated not by the belief that Trantino has been punished enough, but rather by the rule of law and the lack of substantial evidence in the record to support the Parole Board’s decision.
I
Citizens of New Jersey old enough to recall the circumstances of respondent’s brutal murder of Lodi Police Sergeant Peter Voto and Lodi police trainee Gary Tedesco in August 1963 may be astonished to learn that respondent is eligible under the law for parole consideration. Under today’s laws, a defendant convicted *123of the murder of a police officer in the line of duty would face a minimum sentence of life imprisonment without any possibility of parole. N.J.S.A. 2C:11-3b (2). That law, enacted by our Legislature in 1997, L. 1997, c. 60, reflects our contemporary society’s belief that the murderer of a police officer should never be released from prison irrespective of the extent of his rehabilitation during his years of incarceration. If that statute had been in effect in 1963 when those homicides were committed, Trantino never would be eligible to be paroled from prison.
Significantly, we note that most of our sister states also have enacted laws, all within the past thirty years, increasing sentences for those who murder police officers.1 That such laws did not *124exist in New Jersey or in other states in 1963 is attributable to the criminal sentencing philosophy that prevailed throughout the nation during the nineteen-fifties and sixties, and reflected the view then expressed by the United States Supreme Court that “[r]etribution is no longer the dominant objective of the criminal law. Reformation and rehabilitation of offenders have become important goals of criminal jurisprudence.” Williams v. People of State of New York, 337 U.S. 241, 248, 69 S.Ct. 1079, 1084, 93 L.Ed. 1337 (1949).
The Parole Board, represented by the Attorney General, acknowledges that the Ex Post Facto Clause of the United States Constitution, U.S. Const. art. I, § 9, par. 3, prohibits the application to respondent of New Jersey’s current statute mandating life imprisonment without parole for those convicted of murdering” a police officer in the line of duty. Accordingly, the law in effect in 1963 when the murders were committed, supplemented by the Parole Act of 1979, L. 1979, c. 441, N.J.S.A. 30:4-123-45 to -69, is the source of law that governs respondent’s eligibility for parole today.
Pursuant to the law then in effect, N.J.S.A. 2A:113-4 (repealed), Trantino was sentenced to death based on his conviction for first-degree murder. That statute had provided that the penalty for first-degree murder was death unless the jury recommended life imprisonment. In 1965 this Court unanimously upheld his conviction and death sentence. Trantino I, supra, 44 N.J. at 371, 209 A.2d 117. In 1972, on the basis of the United States Supreme Court’s opinion in United States v. Jackson, 390 U.S. 570, 88 S.Ct. 1209, 20 L.Ed.2d 138 (1968), this Court concluded “that the United States Supreme Court has declared the death penalty to be unconstitutional under our statute.” State v. Funicello, 60 N.J. 60, 67, 286 A.2d 55 (1972), cert. denied sub. nom. New Jersey v. Presha, 408 U.S. 942, 92 S.Ct. 2849, 33 L.Ed.2d 766 (1972). *125Because of the United States Supreme Court’s decision that effectively invalidated New Jersey’s death penalty statute, this Court determined that the defendant in Funicello, and the defendants “in all the other [cases] in which the death sentence was imposed,” 60 N.J. at 68, 286 A.2d 55, would be “sentenced to life imprisonment -... as of the date the death sentence was initially imposed,” and would be “entitled to the same credits as if initially sentenced to life imprisonment.” Id. at 67-68, 286 A.2d 55.2 Pursuant to the law prevailing in New Jersey at that time, no more severe sentence could have been imposed after the death penalty was invalidated.
The 1979 Parole Act, N.J.S.A. 30:4 — 123.51(j), expressly provides that the primary parole eligibility date for inmates sentenced to life imprisonment under the law in effect when Trantino’s crimes were committed is to be determined in accordance with the provisions of the former parole act, L. 1948, c. 84. Pursuant to that statute, defendants sentenced to life imprisonment became eligible for parole after twenty-five years, less deductions for commutation time and work credits. See N.J.S.A. 30:4-123.11 (repealed) (“Any prisoner serving a sentence of life shall be eligible for consideration for release on parole after having served twenty-five years of his sentence, less commutation time for good behavior and time credits earned and allowed by reason of diligent application to work assignments”). Accordingly, when this Court first had occasion to address Trantino’s eligibility for parole, *126Trantino II, supra, 89 N.J. 347, 446 A.2d 104, we acknowledged that Trantino first became eligible for parole in 1979. Id. at 352, 446 A.2d 104. That fact was uncontested.
We also recognized in Trantino II, id. at 366, 446 A.2d 104, that the standard governing the grant of parole under the 1979 Act is that inmates eligible for parole “shall be released on parole at the time of parole eligibility, unless [it is shown] by a preponderance of the evidence that there is a substantial likelihood that the inmate will commit a crime ... if released on parole at such time.” N.J.S.A. 30:4-123-53. Accordingly, the critical and controlling question in this appeal is whether a preponderance of the credible evidence in the record supports the Parole Board’s determination that there is a substantial likelihood that Trantino will commit a crime if he is released on parole.
II
A summary of the Parole Board’s dispositions of Trantino’s prior efforts to achieve parole status will provide a context for our review of the evidence considered by the Parole Board.
Trantino first became eligible for parole consideration in 1979. Operating under the former parole statute, the Parole Board denied his first parole application in 1979, and denied a second application in April 1980 on the ground that “the punitive aspect of this sentence has not been satisfied.” Trantino II, supra, 89 N.J. at 353, 446 A.2d 104. In June 1980, in a parole proceeding conducted under the current parole law, L. 1979, c. 441, a single Board member sitting as a hearing officer recommended parole, finding “no substantial likelihood [of future criminal conduct].” Id. Pursuant to N.J.S.A. 30:4-123.55b, the Board’s Chairman affirmed the hearing officer’s recommendation and approved parole release effective August 12, 1980, subject to conditions, including intensive supervision and restitution to the victims’ families in an amount to be fixed by the sentencing court. Trantino II, supra, 89 N.J. at 353-54, 446 A.2d 104.
*127After the Law Division declined to fix the amount of restitution, the Parole Board vacated its earlier release order and then established a new release date of December 23, 1980, again subject to restitution as a special condition. Simultaneously, the Board appealed the Law Division’s order declining to fix restitution, and Trantino appealed from the Board’s December 1980 decision reimposing restitution as a condition of parole. The Appellate Division affirmed the Law Division’s order declining to fix an amount for restitution and reversed the Board’s December 1980 decision reimposing restitution as a condition of parole. In re Parole Application of Thomas Trantino, 177 N.J.Super. 499, 521-23, 427 A.2d 91 (App.Div.1981).
This Court granted both the Board’s and Trantino’s petitions for certification. In re Parole Application of Thomas Trantino, 87 N.J. 385, 434 A.2d 1068 (1981). We held that although restitution lawfully may be imposed as a condition of parole under the Parole Act of 1979, the Board’s perception of the scope of restitution was “much too imprecise and broad” and that the Board was obligated to establish specific criteria to guide and limit the sentencing court’s determination of the amount of restitution to be paid. Trantino II, supra, 89 N.J. at 361-63, 446 A.2d 104.
We also noted the substantial difference in sentencing philosophies between the Code of Criminal Justice, N.J.S.A. 2C:1-1 to 104-8, and the provisions of Title 2A under which Trantino was sentenced, observing that although inmates sentenced under the Code presumptively will have satisfied the punitive aspects of their sentences when first becoming eligible for parole, that may not be the case for inmates sentenced under Title 2A. Id. at 370, 446 A.2d 104. Accordingly, in remanding the matter to the Parole Board we required the Board to “reassess the punitive aspects of Trantino’s sentence in considering the extent of his rehabilitation and his fitness for parole.” Id. at 373, 446 A.2d 104. We also noted that “public outrage over an imminent parole determination, such as that which has occurred in this case, has no place in a parole proceeding and is to be given no weight in a parole decision.” Id. *128at 376, 446 A.2d 104. Finally, we held that “the Parole Board is authorized to reconsider and redetermine Trantino’s fitness for parole---- If the Board determines that Trantino has not been punished sufficiently and, for that reason, as well as any others, it appears by a preponderance of the evidence that there is a substantial likelihood of future criminal activity if he is released, the Parole Board must deny Trantino parole.” Id. at 377, 446 A.2d 104.
On remand, the Parole Board held a new hearing resulting in the denial of Trantino’s parole application and the imposition of a ten-year future eligibility term (FET). In announcing its determination in October 1982, the Parole Board Chairman explained to Trantino that upon completion of the ten-year future eligibility term, less commutation and work credits, the punitive aspect of Trantino’s sentence would be considered to be fulfilled. According to the Board Chairman, the Board concluded that the punitive aspect of Trantino’s life sentence as a Title 2A prisoner should be approximately equal to the twenty-five year parole ineligibility period imposed on a Title 2C life-sentenced prisoner. See N.J.S.A. 2C:43-7(c). Accordingly, the Board Chairman encouraged Trantino to assume that after serving twenty-five years in prison (1963-1988) he would be presumptively eligible for parole unless “through [aberrant] behavior you were to do something that gave rise to its reconsideration.”3
*129In 1988, when the ten-year FET, less credits, had been served, a two-member panel of the Board recommended parole release. However, a majority of the Board voted to deny parole and imposed a six-year FET, apparently on the basis of Trantino’s unwillingness to participate in drug counseling or long-term psychotherapy. In an unpublished opinion, the Appellate Division upheld that determination by the Parole Board.
Trantino next became eligible for parole in November 1990, and the Parole Board denied parole and imposed a four-year FET. In May 1991, however, the Board vacated that decision, resulting in a new hearing in September 1991 after which the Board again denied parole and imposed a three-year FET, “with recommendation that DOC (Department of Corrections) place Mr. Trantino in a halfway house.” In its formal decision denying parole issued October 9, 1991 the Board acknowledged “the positive efforts Mr. Trantino has made during his incarceration, specifically his participation in psychological counseling and substance abuse programs since his last hearing.” However, the Board determined that *130Trantino “can still further benefit from individual and group counseling thus giving him a greater opportunity to achieve his rehabilitative potential,” concluding that “rehabilitation has not yet been sufficiently achieved and therefore the punitive aspect of his sentence has not been satisfied.” Board members Andrew Consovoy and Luis Garcia dissented from the denial of parole. In its October 1991 decision the Parole Board made the following recommendations:
It is suggested that Mr. Trantino continue his participation in supportive programs available within the institution to better prepare himself for his eventual return to the community and thereby reduce the likelihood of his return to criminal activity upon his release. The Full Board urges Mr. Trantino to ... make every effort to achieve halfway house status in order for the Board to evaluate his behavior in a less structured environment.
[Trantino III, supra, 296 N.J.Super. at 477, 687 A.2d 274 (emphasis added).]
Trantino promptly requested that the Superintendent of Riverfront State Prison authorize his transfer to a halfway house. The Superintendent orally disapproved that request in December 1991, and confirmed his denial in writing in April 1992. Id. at 449, 687 A.2d 274.
Trantino next became eligible for parole in November 1992. In its opinion addressing the Parole Board’s 1996 ruling denying Trantino parole and imposing a ten-year FET, Trantino III, supra, 296 N.J.Super. 437, 687 A.2d 274, the Appellate Division majority described the result of Trantino’s 1992 efforts to achieve parole. The court stated:
Trantino became eligible for parole again in June, 1992. The panel deferred decision until an in-depth psychological evaluation could be held at Avenel, and thereafter pending an in-person interview with the psychologist who performed the evaluation. On April 2, 1993, a hearing was conducted after which one member of the Panel, Andrew Consovoy, voted to release Trantino on parole and the other, Arthur Jones, voted to deny, believing that halfway house placement was essential to parole for someone incarcerated as long as Trantino. Consovoy was quite critical of the handling of Trantino’s case, asserting that “this case has never been treated the same way as any other case” and that, despite the fact there is no right to parole, the case “should be treated on the merits.” He insisted that Trantino’s progress as a prisoner and his program participation warranted parole because, as he explained to Trantino, “[y]ou’ve done what you’ve needed to do, and you’ve done all you can do,” and urged him to take legal action because the halfway house application “cannot be rejected by the Department of Corrections.”
*131The panel also agreed that they saw no written reasons for the denial of halfway house status and that Trantino’s halfway house application was not denied in a proper manner. Gonsovoy then announced the panel’s decision to deny parole and impose a thirty-six month FET because the DOC would not grant Trantino halfway house status. They also recommended that Trantino file suit against the DOC.4
[Id. at 449-51, 687 A.2d 274.]
In its December 17, 1993 decision imposing a three-year FET, the Parole Board panel stated:
Mr. Trantino has in his current restricted environment, in our opinion, done his best to address these and other issues identified by the Parole Board, and the most recent professional reports reflect this progress. This panel, therefore, acknowledges that Mr. Trantino has reached his rehabilitative potential within the confines *132of his current state prison setting. However, given the specific facts of this particular case; the absolute inability to function in society prior to this crime, even as a supervised parolee; his long and difficult path towards real and not superficial rehabilitation; with his only recently addressing some major issues; and the length of his incarceration.... [W]e believe that he cannot be judged to have reached his true and full rehabilitative potential until and unless he has achieved an intensive, therapeutic and rigorously supervised, gradual reintegration into society. In New Jersey, the only present means to achieve this crucial goal is through the placement by the Department of Corrections of Mr. Trantino in a halfway house while still an inmate.
The Parole Board firmly believes that this last and vital step must be attempted before Mr. Trantino could even be considered to be fully rehabilitated and granted parole. Although we believe that it is not unreasonable to conclude that Mi-. Trantino has made impressive strides in resolving his problems and internal conflicts that led to these homicides we will only have full knowledge of this man’s rehabilitation through the reintegration process of a community based halfway house setting. In that context we can evaluate Trantino’s readjustment to societal and not institutional stresses, to societal and not institutional failures, and to societal and not institutional temptations. Only through this process can the Parole Board judge if this man has been truly rehabilitated. The Adult Panel is of the opinion that the placement of Mr. Trantino to a halfway house should be done while he is an inmate to insure the legitimate interests of all parties.
The Adult Panel is of the opinion that if Mr. Trantino can successfully enter and complete a correctional halfway house program as an inmate he can achieve his full rehabilitative potential and. therefore will satisfy the punitive aspect of his sentence and meet the substantial likelihood test.
[Id. at 453-54, 687 A.2d 274 (footnote omitted)(emphasis added).]
Consistent with the Parole Board’s express findings, in November 1993, and again in January 1994, Trantino applied to the Department of Corrections for transfer to a halfway house. Those requests were denied in February 1994. A June ,1995 letter from the Administrator of Riverfront Prison explained 'that the denial was based on the receipt of threats to kill Trantino if he were placed in a halfway house, the circumstances of the offense, the risk of possible adverse community reaction, and the objection of a member of the Legislature.5
*133Trantino again applied for parole in September 1994, but the hearing panel denied parole and imposed a three-year FET. In its April 17, 1995 decision explaining the reasons for the September 1994 denial of parole, the panel acknowledged that Trantino “has made enormous progress toward reaching his full rehabilitative potential in the 30 years of his incarceration,” but concluded that Trantino “has not reached his full rehabilitative potential, therefore the punitive aspect of his sentence has not been satisfied, and that a substantial likelihood that he will commit a crime if released on parole at this time continues to exist, and parole is denied.” (We note that the Board’s finding concerning the punitive aspect *134of Trantino’s sentence directly contradicted Chairman Dietz’s statement in 1982 that after imposition of a ten year FET “the punitive aspect of the sentence ... is resolved.”) On April 26, 1995, the full Board denied Trantino’s appeals from the Board Panel determinations of November 12, 1993 and September 1, 1994. However, on August 30, 1995 the Parole Board vacated its April 26, 1995 decision and scheduled a new parole hearing during the following month.
After a hearing on September 14, 1995, a two-member Board panel denied parole and informed Trantino that it intended to “exceed the guidelines” and establish a future eligibility term greater than thirty-six months. Although the Panel acknowledged “the great strides you have made towards achieving your rehabilitative potential over the course of the last thirty-two years,” the Panel primarily based its denial of parole on Trantino’s failure to remember certain aspects of the crimes. The decision stated:
It is the Panel’s belief that your failure to remember certain details regarding the murder is inhibiting you from reaching your rehabilitative potential. In sum, the Adult Panel is of the position that until you can remember specific events regarding the murder, including firing the gun that killed Sgt. Voto, you will not be able to fully accept your role in the crime and will not achieve your rehabilitative potential. Therefore, the Panel believes there is a substantial likelihood you will commit a crime if released on parole. The Adult Panel is of the belief that based on the severity of the crime for which you were sentenced, your prior criminal record, and your need for long term counseling, a future eligibility date established pursuant to N.J.A.C. 10A:71-3.21(a) is clearly inappropriate. Therefore, the Adult Panel is referring your case to a three member panel for establishment of a future eligibility date beyond code guidelines, pursuant to N.J.A.C. 10A:71-3.21(d). It is the Adult Panel’s belief that a future eligibility term pursuant to N.J.A.C. 10A:71-3.21(d) will allow you the opportunity to undergo long-term psychological counseling. Hopefully, this counseling will aid you in your attempt to remember specific details regarding the shooting of Sgt. Voto and other events that took place immediately before and after the murders.
The Adult Panel is aware that a different Board Panel determined at your parole hearing on November 12, 1993 that you had reached your rehabilitative potential within the confines of prison and that your progress toward real and not superficial rehabilitation could only be maintained by placement in a halfway house as an inmate. You have attempted on numerous occasions to be placed in a halfway house as an inmate. The Department of Corrections has continually denied you placement into a halfway house. While this Adult Panel believes placement of *135you into a halfway house would be beneficial to you in your goal to reach your rehabilitative potential, it is this Panel’s determination that certainly this is not the only means by which you can achieve this goal. It is this Panel’s position that you can eventually reach this goal through long term psychological counseling in an institutional setting.
[Trantino III, supra, 296 N.J.Super. at 458-59, 687 A.2d 274 (emphasis added).]
In April 1996 the full Parole Board voted to impose a ten-year future eligibility term. In the interim, the Appellate Division reinstated Trantino’s appeal from the Board’s 1993 and 1994 dispositions, and permitted Trantino also to challenge the Board’s most recent denial of parole in 1995.
A divided Appellate Division panel upheld the Parole Board’s denial of parole and imposition of a ten-year FET primarily on the basis of the Board’s conclusion that Trantino’s inability to recall details of the crimes evidenced a lack of candor or credibility that suggested a likelihood of future criminal activity if Trantino were released on parole. Id. at 466-71, 687 A.2d 274. However, the Appellate Division majority emphasized the “consistent position of the Board that it cannot prudently grant parole to a long-term prisoner, convicted of a crime such as murder, before performance in a halfway house or residential facility can be thoroughly evaluated.” Id. at 465, 687 A.2d 274. Accordingly, after concluding that the Department of Corrections’ denial of Trantino’s request for halfway house placement was not supported by a final determination with an adequate statement of reasons, the Appellate Division remanded the matter to the DOC for “consideration of an updated application for halfway house (or out-of-State transfer) and for findings and a statement of reasons with regard to such application.” Id. at 464, 687 A.2d 274.
Judge Pressler, concurring in part and dissenting in part, agreed with the remand to the DOC but would have directed the DOC “forthwith to place defendant Thomas Trantino in a preparóle halfway house or residential facility or to effect his out-of-State placement.” Id. at 471, 687 A.2d 274 (Pressler, P.J.A.D., concurring in part and dissenting in part). She added:
*136In the absence of such prompt action by the Department, I would remand to the Parole Board for its prompt consideration of parole-release conditions including, if a pre-release halfway house placement is not possible, then a closely monitored post-release placement as well as such other parole conditions as the Board concludes are necessary and appropriate. I would so do because I am persuaded that the record overwhelmingly demonstrates the arbitrariness and unreasonableness of both the Department’s denial of halfway house or out-of-state placement and the Parole Board’s decision, responsive to that denial, to deny parole and impose a ten-year future eligibility term.
[Ibid.]
Trantino appealed as of right to this Court, R. 2:2 — 1(a)(2), based on the dissent below. In a unanimous opinion, we affirmed that portion of the Appellate Division judgment holding that the Department of Corrections’ refusal to transfer Trantino to a halfway house was not supported by an adequate statement of reasons. Trantino IV, supra, 154 N.J. at 22, 711 A.2d 260 (1998) (Trantino IV). We reversed the Appellate Division judgment upholding the Parole Board’s denial of parole and fixing a ten-year FET based on our conclusion that the Board’s decision was not “based on a proper standard and supported by sufficient evidence and adequate findings of fact.” Id. at 23, 711 A.2d 260 (emphasis added). Accordingly, this Court clarified the standard governing Trantino’s parole eligibility and remanded the matter to the Parole Board to reconsider the evidence and redetermine Trantino’s eligibility for parole. Ibid.
We specifically noted that although the Parole Board did not ignore recidivism as a criterion of parole fitness, its most recent decisions appeared to focus on whether “Trantino had made sufficient progress toward ‘reintegration into society,’ was ‘fully rehabilitated,’ had realized ‘his real rehabilitative potential,’ had reached his ‘true and full rehabilitative potential,’ and had achieved ‘real and not superficial rehabilitation’ and ‘complete[] and total[ ] rehabilitation].’ ” Id. at 30, 711 A.2d 260. Accordingly, we determined that the record was unclear on whether
the Parole Board invoked a test that focuses primarily and essentially on the likelihood of criminal recidivism or whether it followed the more exacting and difficult test of full or complete rehabilitation that assures not only that an inmate *137will continue to lead a law-abiding life, but also that he or she will assume a responsible role in society consistent with the public welfare.
[Id. at 32, 711 A.2d 260.]
In our opinion we commented extensively on the evidence in the record suggesting that" Trantino’s rehabilitation in prison was sufficient to indicate a low likelihood of recidivism. We stated:
Trantino’s prison record plainly is material in determining whether he has achieved a level of rehabilitation such that he has been sufficiently deterred and there is no likelihood of recidivism. That record discloses substantial evidence of significant rehabilitation tending to demonstrate that Trantino has overcome any likelihood of recidivism. He has had no substance abuse violations in prison and no rules infractions since 1970. Trantino has already participated in sixty-nine work and recreation details that involved excursions into the community and has gone on overnight furloughs without incident. Further, he has completed seventeen programs designed to enable him to help fellow inmates and has been in individual and group psychotherapy for the past twenty-five years. Finally, Trantino’s supervisors in prison have consistently rendered favorable reports.
Psychological evidence also supports a determination of substantial rehabilitation militating against the likelihood of recidivism. Dr. James Bell, one of two evaluating psychologists, said in his July 1995 report: “[W]ithin the prison system, [Trantino] has been an exemplary inmate and not only has invested in several programs to improve himself, but also has started several programs to assist youthful delinquent offenders and substance abusers.” Dr. Bell further stated in his report:
Clinically, he impresses as a man who has reached a point of change come about through sincere self-inventory and aspirations to atone for the great wrong he has done in his life. He has adequate mental and emotional resources to live a socially responsible, self-reliant life. His PASS score of 66% suggests an above average possibility of post-release success.
An August 1995 evaluation by Glenn Ferguson, the other evaluating psychologist, states:
Presently, Thomas [Trantino] functions more as a staff member in the prison than as an inmate. He has developed a well-rounded and reputable support network which should enable Thomas to manage the increased stress level of independent living optimally. He has a high level of motivation for continued education, therapy and employment. His employment plans appeal realistic and attainable, including technical work with his wife’s copyrighting company and creative work in art and therapy____The combination of well-established social supports, motivation for ongoing treatment, and age will most likely lead to a successful parole outcome.
[Id. at 32-33, 711 A.2d 260 (citations omitted) (footnote omitted).]
Concerning the Parole Board’s determination that Trantino’s inability to remember details of the crime was preventing him *138from reaching his full rehabilitative potential, this Court observed that “there is evidence in the record that Trantino’s memory loss is consistent, long-standing and genuine, and, beyond the issue of recollection, his acknowledgment of responsibility is sincere and legitimate. That Trantino acknowledges and accepts responsibility for his crimes is disclosed in the record of his numerous parole proceedings.” Id. at 35, 711 A.2d 260. We noted that Trantino’s acknowledgment of full responsibility for the crimes was corroborated by two Board psychologists, one of whom, Dr. Glenn Ferguson, in his August 1995 evaluation of Trantino, stated: “ ‘[Trantino] has consistently claimed no recollection of the actual offense, although he accepts full responsibility for both murders, in light of eye witness testimony, and his recollection of holding a gun both prior to and after the offense.’” Id. at 35-36, 711 A.2d 260.
Accordingly, we held that the record did not provide an adequate evidentiary basis for the Parole Board’s rejection of the conclusions of those two psychologists. Id. at 36, 711 A.2d 260. Further, we held that the record “does not clearly sustain the conclusion that long-term psychotherapy will eventuate in a breakthrough recollection [of the details of the crimes] or that such a recollection is necessary, given Trantino’s repeated acceptance of responsibility, in order to ensure that he has achieved a level of rehabilitation that eliminates the likelihood that he will, if released, commit crimes.” Id. at 38, 711 A.2d 260.
Finally, in remanding the matter to the Parole Board to reconsider whether there is a likelihood that Trantino will again engage in criminal activity, this Court provided specific instructions concerning the remand proceedings before the Parole Board:
In evaluating Trantino’s fitness for parole under this standard, the Parole Board should give weight to the facts that Trantino is now sixty years old and, as already noted, has previously been released on sixty-nine work details and two furloughs without incident; has not violated a correctional rule in twenty-seven years; has successfully completed substance abuse counseling; has educated himself while in prison; has pursued the available vocations for prisoners; has a stable support network; was housed without incident at the Wharton Tract, a facility without perimeter guards; and has been deemed fit for transfer by the last two psychologists to evaluate him. Moreover, in light of that evidence, the length of time *139Trantino has served under his sentence, and the successive occasions on which he has been deemed eligible for parole, punishment is no longer a material consideration in the parole determination.
In its determination, the Board can devise pre-release conditions that will allow it to assess how Trantino handles the stresses of society that may induce or impel him to commit crimes. Those conditions could include community work details, furloughs, minimum security status, and other measures that would serve to reintroduce Trantino gradually into society and lead to his ultimate release. Alternatively, the Board may decide to impose post-release conditions, such as intensive supervision and drug testing, that would ensure that Trantino was not behaving in a way that indicates he will engage in criminal activity.
In its redetermination of parole fitness, the Parole Board may also consider halfway house treatment as a condition of parole.
[Id. at 39-40, 711 A.2d 260.]
Ill
Proceedings on Remand to the Parole Board
A. Introduction
Immediately following this Court’s May 1998 decision reversing the Parole Board’s denial of parole and imposition of a ten-year FET, and remanding for a new parole determination consistent with this Court’s opinion, Trantino renewed his request for transfer to a halfway house in anticipation of his new parole hearing. Department of Correction regulations require that, prior to transfer to a halfway house, inmates must undergo a “risk/needs” assessment at one of two residential substance-abuse treatment centers under contract with DOC, and be approved for transfer by the Institutional Classification Committee (I.C.C.). See N.J.A.C. 10A:20-4.4, 4.5 and 10A:9-31. After an evaluation by a DOC psychologist in July 1998 found Trantino suitable for halfway-house placement, the Riverfront State Prison I.C.C. recommended that Trantino be transferred to Talbot Hall in Kearny, one of the residential treatment centers that screened inmates prior to halfway-house assignment. The Assistant Commissioner of DOC approved the recommendation and Trantino was transferred to *140Talbot Hall6 on August 18, 1998.
The record reveals that immediately after Trantino’s assignment to Talbot Hall numerous public officials protested the transfer, and a statement issued by the Office of the Governor expressed her “surprise” over the transfer and her request that the transfer be reviewed by the Attorney General. The Attorney General’s office requested that a new psychological evaluation be conducted by Dr. Michael Weiner to determine whether Trantino’s assignment to Talbot Hall was appropriate. Dr. Weiner issued his report on November 9, 1998, finding among other things that Trantino’s risk of recidivism was low but that his prior history of substance abuse required that he again participate in an Alcoholics-Anonymous-type twelve-step program prior to community release. Three days later, on November 12, 1998, the DOC’s Community Classification Committee at Talbot Hall determined, on the basis of Dr. Weiner’s report and various psychological tests administered at Talbot Hall, that Trantino was unsuitable for halfway house placement in part because he presented an “escape risk” and had an “unresolved substance abuse problem.” Trantino was transferred that same day to South Woods, a medium security prison, with a recommendation that he be assigned to a substance abuse program. At South Woods, however, Trantino’s application for enrollment in its substance abuse program was rejected because his score was substantially below the minimum requirement on an Addiction Severity Test for admission to the program, and he asserts that that rejection significantly undermined the reliability of the Talbot Hall Committee’s determination. Trantino’s cross-petition for certification challenges the legality of his transfer from Talbot Hall to South Woods. In an unreported opinion *141the Appellate Division previously has dismissed as moot the appeal asserting that that transfer was illegal.
B. Evidence Adduced at Parole Board Hearing
In view of the magnitude of the record before us, which includes not only evidence and testimony presented at the Parole Board’s latest remand hearing but also the entire appellate record underlying this Court’s 1998 disposition, we will focus our attention on the evidence that relates specifically to the grounds relied on by the Parole Board in denying parole. The Board based its June 9, 1999 decision to deny parole primarily on four grounds:
1. “The Parole Board is primarily concerned with your psychological profile, as reflected primarily by the testimony of Dr. Ferguson. Dr. Ferguson’s testimony leads the Board Members to conclude that it is likely you would commit a crime if released on parole.”
2. “The Parole Board is also concerned* regarding your lack of a suitable parole plan, especially in light of Dr. Ferguson’s comments.”
3. “The Parole Boat’d is also concerned with your failure to address in psychological counseling the issues that led you to engage in domestic violence with your first wife.”
4. “[T]he Parole Board Members are very concerned about your long standing history of being less than candid with Board Members and psychologists about issues that are at the core of your case.”
Although not specifically characterized as a separate ground for denial of parole, Trantino’s professed intention to write a second book was sharply criticized by the Board in its decision denying parole. “Perhaps no other conversation at your hearing on June 9, 1999 exemplifies areas in which you have not made great changes, than your conversation with Mr. Gomez regarding the possibility of you writing a book if you were paroled.”
We summarize the evidence in the record that is relevant to each of the grounds for denial of parole advanced by the Parole Board. We do so in reverse order because the evidence adduced in connection with the first ground, Trantino’s psychological profile, is more complex than the evidence relating to the other grounds.
*1421. Trantino’s Plan to Write Another Book
At the June 9, 1999 hearing, Board member Gomez criticized Trantino’s intention to write a second book.
MR. GOMEZ: The problem is basically the victims, once again, are going to suffer. You know what it is, you are going to suffer. You say you suffer about Charlee [Trantino’s second wife], you don’t want to hear people talk about the [m], but you are willing to write a book—
MR. TRANTINO: No.
MR. GOMEZ: — And have people suffer again.
MR. TRANTINO: No, I am not willing to do that, and that is not my intention. If I try to write a book again, and it certainly wouldn’t be about their families, their family members.
MR. GOMEZ: You are missing the point.
MR. TRANTINO: I am getting the point, but it’s not—
MR. GOMEZ: You are profiting, you are profiting.
MR. TRANTINO: Yes, profiting from a crime.
MR. GOMEZ: You are talking about putting salt in people’s wounds. You don’t think that is putting salt in people’s wounds?
MR. TRANTINO: Yes.
MR. GOMEZ: It’s as simple as that.
MR. TRANTINO: I was not thinking about that.
MR. GOMEZ: The last question.
MR. TRANTINO: I was thinking—
MR. GOMEZ: Continue if you want, I only have one more question.
MR. TRANTINO: Go ahead.
Later during the June 9, 1999 hearing Trantino responded to Board Member Gomez’s criticism:
MR. TRANTINO: What I was thinking of was Charlee and Charlee’s debt that she has and trying to do something about that.
MR. CONSOVOY: Now—
MR. TRANTINO: And a book, now the book I’m taking very seriously what Mr. Gomez said and what you are saying here.
MR. CONSOVOY: Please do.
MR. TRANTINO: I never, it does — I have expressed this to you over many times about the families here, it really does hurt me, I had a mother, that could have been my mother, it could have just been the opposite thing, I know the sorrow that the kind of feelings they have, I wouldn’t want to hurt them. What I would be doing would not be to hurt them.
MR. CONSOVOY: But that’s one point, and as I say that’s why you have to in or out of the prison, where ever you may go when we are done with today, you must *143deal with this narcissistic diagnosis and the features thereof, because that’s what that is, Tom. I am not saying you meant to hurt them, you don’t mean to hurt them, but the way you look at things that’s how you looked at it. You were doing a good thing for Charlee. On the other hand, you were ruining—
MR. TRANTINO: I don’t know.
MR. CONSOVOY: That is my point.
MR. TRANTINO: Yes.
In addition, during Trantino’s final remarks at the June 9 hearing, he stated:
What you just said bothered me a lot, about the families. It just brought up stuff, you know, I work on that, and it hurts. I don’t want to hurt them. And now I have to — I am going to rethink about the book. After what you and Mr. Gomez pointed out, that bothered me a lot. I, more than anything, I’m sorry for what I did, I shot and killed two people. And it wasn’t only their lives I took, their families, their families as well were affected right to this day, as you are pointing out here. And that is something that I try to live with, I try to deal with, I try to heal, I am ashamed of that.
2. Lack of Candor
The Board listed seven “examples of when Trantino has been less than candid with authorities or psychologists or [has mitigated] central events [in his] life or aspects of [his] personality.”
(a) The Talbot Hall Incident
In February 1999 the Parole Board conducted an interview by telephone with Joseph Trabucco, the Director of psychological testing at Talbot Hall, whose qualifications include a master’s degree in psychology. In September 1998, after numerous public officials had protested Trantino’s assignment to Talbot Hall, Trabucco administered to Trantino a 175 question true-false test known as the MCMI-II, the result of which suggested that Trantino possessed a mild to moderate psychological dysfunction and stated that “[although violence is not a major characteristic of this inmate’s behavioral repertoire, there may be occasions when violence can be provoked, although not readily.” Trabucco described Trantino as “a very controlled individual, thinks about what he’s saying, thinks about who he’s speaking to, calculating if you will.” In response to Board Chairman Consovoy’s question concerning “any specific types of instances that might provoke him *144to violence,” Trabueco related an incident that the Board deemed significant in assessing Trantino’s likelihood of recidivism. Trabucco stated that in September 1998 he tested Trantino about substance abuse only on the basis of events during the preceding year, and the test result revealed no substance abuse problem. Trantino was then asked to retake the same test, this time responding on the basis of his entire lifetime. According to Trabueco, Trantino became “sort of agitated” and “definitely perturbed.” He said “Why are you making me do this?” Trabucco stated that Trantino’s “veneer ... started to — to crack a little bit,” and described the incident in the following manner:
His face was kind of red, he was slightly more animated and agitated. And he said, You know, I know it’s not you, Mr. Trabueco, but, you know, those guys at corrections, they’re out to get me. And he went into this kind of spiel about how it wasn’t me that was making him take this test, but other people, and they were going to take the results and do what they wanted to do anyway. And we got him to sit down again, and then he got up the third time and was then agitated enough, now again, I didn’t feel threatened, I didn’t feel, you know, that he was going to hit me or anything, but he — it was the only time in this whole experience here when I — when I sensed that that veneer had cracked to some extent. He was highly animated and agitated, took his glasses off, was sort of pacing, was a very different Mr. Trantino then [sic] I think most people get to see. And — and well, he didn’t take — the outcome was we didn’t — we couldn’t get him to sit down and take the other part of the test, and I’m not sure it mattered much in the long run anyway.
But as I look back and look at the results of the testing and his experience here, I mean I think what — what it indicated to me was, that when — when you put him in a situation, when you press him, or put him in a situation he doesn’t feel in control or totally comfortable with, you begin — you begin to see this other side of — of this person, which, you know, the cool calculated responses tend to vanish, and the agitation sets in.
[Trantino V, supra, 331 N.J.Super. at 589, 752 A.2d 761.]
Cynthia Smarook, an assessment counselor at Talbot Hall, who was interviewed by the Parole Board’s executive director, confirmed Trabucco’s characterization of the incident. Smarook also reported that during Trantino’s stay at Talbot Hall, he was “controlled, he was polite, he was cooperative, he was actually rather chatty,” and had not exhibited any prior agitation, other than the incident. When told that Trantino later denied the incident, Smarook responded that:
*145[H]e’s either lying or it — it happened fairly quickly and he just completely put it out of his mind that — that it occurred. I don’t — I can’t tell whether or not he is deliberately lying about it. It — it was fairly quick before he was told he didn’t have to take it based on the — the same stipulations as the rest of the population who does.
At his June 4, 1999 hearing before the Parole Board, Chairman Consovoy, without mentioning Trabucco, referred to an incident at Talbot Hall during which Trantino became “agitated” and “quite perturbed.” Trantino denied that any such incident had occurred. Between the June 4 and June 9, 1999 Parole Board hearings Trantino apparently was shown a copy of Dr. Rosenfeld’s report that referred to the Trabucco incident. At the June 9 hearing, when Trantino again was asked about the incident by Consovoy, with specific reference by name to Trabucco, Trantino explained his earlier denial and recounted his version of what had occurred: [On June 4] [y]ou didn’t mention Mr. Or [sic] Dr., I think he is a psychologist, Tribucco [sic].
I know the incident now, there was a woman psychologist, she went to give me a test, and I am saying this is not appropriate, and it was in this computer room. Other people were there, including Dr. Trabucco was in the back, and I just turned to him, I said, Doctor, could you help out over here. And he said what’s the problem? I stated the problem and he told her that this man is in remission, he does not have to take that test, that’s what he said, not whatever is said there actually, that’s not what happened.
Now I might have been perturbed, certainly I was perturbed, but not angry, I was able to talk to him and he straightened it out. He said I didn’t have to take it, that I am right. The man is in remission, he took the other test, all these other tests, he is okay, throw that away. That is what he said to that woman.
[Id. at 597, 752 A.2d 761.]
Nevertheless, in its decision denying parole the Parole Board relied on Trantino’s inconsistent responses on June 4 and June 9 concerning the Trabucco incident as evidencing a lack of candor that was germane to Trantino’s likelihood of recidivism:
The record indicates that you became upset when asked to participate in the SASSI Test at Talbot Hall in September, 1998. At your two member Panel hearing on June 4, 1999, you denied that this event ever occurred. At your full Parole [B]oard hearing less than a week later, you admit something did happen, but denied it happened as was reflected in statements Chairman Consovoy read to you (a statement provided by Mr. Trabucco who conducted the testing). This exchange is reflected in Pages 234 through Page 236 of the June 9, 1999 hearing *146transcript. In the opinion of the Parole Board you have been less than candid about this incident and the Board finds the statements of Mr. Trabucco and Dr. Smarouk [sic], the two individuals who conducted the test, to be more credible than your statement regarding this incident. '
[Id. at 612, 752 A.2d 761.]
(b) Trantino’s Lack of Candor Concerning the Murders of Voto and Tedesco
At the June 9,1999 Parole Board Hearing, Chairman Consovoy questioned Trantino about his responsibility for the murders.
MR. CONSOVOY: [W]hen [Rosenfeld] says regardless of whether or not you actually fired the gun, did you fire the gun?
MR. TRANTINO: Yes, I shot and killed both men.
MR. CONSOVOY: You’re sure?
MR. TRANTINO: I am sure.
MR. CONSOVOY: Okay. And whether or not there was ever any evidence that you shot Gary Tedesco it is your position that you did?
MR. TRANTINO: Yes.
In its decision denying parole, the Parole Board referred to prior occasions in 1964, 1967, 1980, 1982, 1995, and 1998 in which Trantino claimed that he did not recall pulling the trigger and committing the murders. The Board concluded that Trantino’s inconsistent testimony about whether he recalled committing the murders constituted an additional basis for denying parole:
This issue appears now to be a moot issue in light of your statement at your June 9, 1999 hearing that you áre now sure that you shot and killed both individuals. The Supreme Court also stated that there was evidence that your memory loss is consistent, long standing, and genuine. With all due respect to the State Supreme Court, the Parole Board, based on a review of your entire record, does not find your claims of past memory loss as consistent or genuine. The issue at this time for the Parole Board is not whether you have made a sincere avowal of responsibility for commission of these offenses. [It is] the [sic] opinion of the Parole Board that you have sincerely claimed responsibility. The issue is rather, in the opinion of the Parole Board, that you have not been candid in the past with psychologists or Board Members regarding your memory of that night and this lack of candidness on a central issue to you is part of an ongoing pattern of deception you have exhibited with authorities.
*147(c) Trantino’s Actions on Parole in New York
In its June 1999 decision denying parole, the Parole Board quoted from portions of Trantino’s May 25, 1961 hearing before the New York State Parole Board that resulted in his parole from prison in New York in June 1961. The Board observed that Trantino “would reward the faith the New York paroling authority placed in you by returning to committing burglaries, using and abusing alcohol and narcotics, associating with criminals wanted for murder (including harboring those individuals), and ultimately slaughtering two police officers in the Angel Lounge in Lodi, New Jersey on August 25, 1963.”
In his testimony before the Board in June 1999, Trantino acknowledged that while on parole in New York he failed to report as required, had quit his job, and was living at an unauthorized address. The Board also noted that during this period Trantino was consorting with known criminals and had on one occasion directed his wife to lie to his parole officer concerning his whereabouts. Based on that record testimony revealing Trantino’s breach of his obligations to New York parole authorities from 1961 to 1963, the Parole Board concluded that Trantino’s conduct during that period also reflected a lack of candor justifying denial of parole:
It is clear based on a review of your New York parole records, you did not follow conditions of parole placed on you by New York paroling authorities. In addition to resorting to criminal behavior with known criminal associates while on parole, records indicate you lied to parole officials in New York when you were hiding Falco and Cassarino, both wanted for murder, prior to the murder in Lodi. The Parole Board also notes your statements to New York paroling authority prior to parole release in 1961 and your subsequent failure to follow through on your commitment to. the New York paroling authority, as reflected earlier in this notice of decision.
(d) Trantino’s Account of the Payroll Robbery on February 16, 1956
The Parole Board also concluded that Trantino’s attempt to understate the violent nature of a robbery he committed in New York in 1956 reflected a lack of candor that supported denial of *148parole. The Board’s June 1999 decision denying parole described the circumstances of that robbery:
Records indicate on February 16, 1956, you and a co-defendant, George Fernandez, accosted a twenty-one year old dental assistant, Carmella DeVito, and stole $500.00 of payroll money from the dental office Ms. DeVito worked at. Records indicate that when Ms. DeVito approached you and your codefendant, your co-defendant placed his hands over Ms. DeVito’s eyes and mouth. You then pulled Ms. DeVito down and located a payroll envelope in her pocket. Ms. DeVito struggled and was pushed up against a wall. Ms. DeVito and your co-defendant then fell on the tile floor. Ms. DeVito screamed and then you and your co-defendant released her. You and your co-defendant would subsequently be arrested for this offense and your co-defendant would admit to committing twenty-five burglaries prior to this incident and several of the burglaries in your company. As a result of the assault, Ms. DeVito received a black eye, bruises and later suffered from somatic pains, lack of sleep and nightmares as a result of this incident. You would later be indicted for first degree robbery, first degree grand larceny and second degree assault. On June 18, 1956, as the result of a plea bargain, you were given a five to ten year sentence for second degree robbery.
[Trantino V, supra, 331 N.J.Super. at 616, 752 A.2d 761.]
At its June 9, 1999 hearing Chairman Consovoy questioned Trantino about whether the 1956 robbery was a violent crime.
MR. CONSOVOY: Now I am going to paraphrase, tell me if I am not right, accurate or fair, that this was a snatching of a payroll?
MR. TRANTINO: Yes.
MR. CONSOVOY: From a female?
MR. TRANTINO: Yes.
MR. CONSOVOY: Who was going to or from the bank and this was an envelope of cash?
MR. TRANTINO: Yes.
MR. CONSOVOY: Your partner was George Fernandez in this situation?
MR. TRANTINO: Yes.
MR. CONSOVOY: And you said repeatedly that this was not a violent crime.
MR. TRANTINO: That is [sic] a violent crime, but my role in it is [sic] as an accomplice, and I was guilty as the person who did the snatching, but there was [sic] no weapons.
MR. CONSOVOY: Right.
MR. TRANTINO: Or no one got hurt.
MR. CONSOVOY: You are [sic] sure?
MR. TRANTINO: Yes, absolutely, no one got hurt.
MR. CONSOVOY: No one was injured?
*149MR. TRANTINO: No, it was just a snatching. The woman — to the best of my recollection, it was dental offices, adjoining dental offices or doctor offices, and this woman picked up the payrolls for all of the offices. And we waited, George — when she went into the hallway George snatched it and we ran.
MR. CONSOVOY: And you’re very clear that’s how that happened?
MR. TRANTINO: To the best, I mean—
MR. CONSOVOY: And you have been very consistent about that.
MR. TRANTINO: Yeah.
MR. CONSOVOY: And you have always maintained that was not a crime of violence or the alternative you weren’t violent.
MR. TRANTINO: I ... in that context.
During that same hearing Consovoy read to Trantino excerpts from the pre-sentence report relating to the 1956 robbery. In that report the complainant acknowledged that Trantino’s accomplice, George Fernandez, restrained her while Trantino removed an envelope containing a payroll from her pocket. The complainant acknowledged struggling with Fernandez and then being pushed against a wall and falling to the floor with Fernandez on top of her. The pre-sentence report included statements by the parole probation officer that “the complainant advises she suffered a black eye and received black and blue bruises being caused by being thrown against the wall to the floor.” Trantino stated that this was the first time he had been informed the victim had been injured. Consovoy replied: “Well it’s in the pre-sentence report.” Trantino responded: “Until this day I hadn’t heard it.”
The Parole Board apparently determined that Trantino’s prior failure to disclose that the victim of the 1956 robbery sustained injuries reflected a lack of candor that supported denial of parole. The Board stated:
Your previous accounts to Board Members in which you mitigated the violence involved in this offense is detailed on Page 222 through Page 224 of the June 9, 1999 hearing transcript. As reflected in the official accounts of this offense, the offense was violent in nature and in the opinion of the Parole Board you have mitigated the extent of violence in this act. As discussed on Page 10 and Page 11 of notice of decision.
*150(e) Domestic Abuse of Trantino’s First Wife in 1963
In addition to concluding that Trantino’s “failure to address in psychological counseling the issues that led you to engage in domestic violence with your first wife” was a separate ground for denial of parole, the Parole Board also found that Trantino’s minimization of the extent of his assaults on Helene constituted another example of Trantino’s lack of candor with the Board.
Trantino denied that he had ever “minimized” the 1963 abuse of his first wife, Helene, as Trantino’s expert, Dr. Rosenfeld had reported. He admitted that he had abused her, for which he was now remorseful, and which he had addressed during many years of counseling. The following colloquy occurred at the Board’s June 4, 1999 hearing:
THE CHAIRMAN: In some of your early interviews, you talk about the fact that you had hit her.
MR. TRANTINO: Yeah.
THE CHAIRMAN: It disappeared off the charts for years. Hearing, after hearing, after hearing it’s gone. When you’re talking to Rachel — [a former Board Member]
MR. TRANTINO: Yes.
THE CHAIRMAN: —you brought it up.
MR. GOMEZ: (Inaudible).
THE CHAIRMAN: Yeah. I just (inaudible). She didn’t bring it up because she had no context in which to bring it. You had to bring it up. And maybe I’m wrong, maybe she brought it up. I think you did. Either way, it got brought up, and now it’s appearing in a couple of reports, because psychologists read transcripts, and they ask questions, and how often — how—the domestic violence issue, to my reading — to me, it’s a new issue. I never thought about it, hadn’t heard about it, hadn’t read the stuff. Maybe I should have, so let’s talk about that, because to me it’s been — it’s obviously not news, 38 years old or something, but how often did you actually hit your wife?
MR. TRANTINO: I — I didn’t hit her frequently, but I did raise my hands to her on a number of occasions, that I wouldn’t say it was often, but one time is too many. Maybe three or four times.
THE CHAIRMAN: Now in looking back, tell me — you’ve been involved with counseling and A.A., why? Why did you hit her?
MR. TRANTINO: At that time in my life, I was confused, as I said before, and there’s no way I can answer for my behaviors (inaudible), but I was confused and (inaudible). I’m able to control — felt out of control, and impulsive, and I reacted *151to her. She was just a sweet young girl. I look back at that now, I feel like I— I — I hit a child.
THE CHAIRMAN: I was going — you know, I was almost going to say it’s like you hit a teenage girl.
MR. TRANTINO: It was a teenage girl.
THE CHAIRMAN: Yeah.
MR. TRANTINO: And this something — I mean, I still deal with. I was able to tell her I was sorry, but that, as you know, at least for me, is not adequate. I mean, I — I just — you can’t change the way you were, and what you did, and what I did and the way I was, and I just feel so sorry for what I did to her. She was such an innocent girl, a very, very sweet, intelligent nice person. She really liked me, you know, and I look back at that and it’s just like — it’s just like part of all of my behaviors at that time. They just — there’s no excuse. I can’t come up with an excuse for that.
Trantino related that he never saw violence in his family; he never saw his father hit his mother. His explanation of why he hit Helene was that he took out his feelings of powerlessness on her. Trantino stated: “I just was not adequate, unable to control myself, not capable of coping with the stresses in my life, totally unable ... and I took it out on her, on a weaker person.” Trantino did not agree with Rosenfeld that he “minimized” the abuse, but maintained that Helene would not have called him abusive. Consovoy stated:
THE CHAIRMAN: You don’t think then she thought that it was particularly bad.
MR. TRANTINO: Yeah. I don’t think she would have thought that then. She was just a child. We never heard those words. And just like you said, domestic violence, you never heard of that. And now (inaudible) older, she (inaudible) abusive because it’s not — as you know, it’s not just the hands on somebody, it’s the (inaudible), the cheating, the yelling, disrespect, and it’s all of that. I believe those are violence — acts of violence as well, and are abusive.
THE CHAIRMAN: When you said I don’t think she would call me abusive, you meant back then.
MR. TRANTINO: Back then, yeah.
When the issue was raised again by Consovoy on June 9,1999, Trantino again disputed minimizing the domestic abuse that occurred:
MR. CONSOVOY: [Y]ou said the other day that the word minimized is not something that you feel is appropriate to your statements.
*152ME. TRANTINO: I don’t think I — I don’t know what he meant by minimize, so that’s hard, but I don’t think I do. I mean, I think back to what I did to her, I hit her, I beat her. And I mean, even a slap is a beating.
MR. CONSOVOY: [Y]ou don’t feel that [minimization is] appropriate to what you were telling [Rosenfeld]?
MR. TRANTINO: At least not what my intentions or what my feeling is about that____
MR. CONSOVOY: And I asked you did you punch her or slap her? You said open hand.
MR. TRANTINO: I — to the best of my knowledge I don’t remember ever punching her.
MR. CONSOVOY: Okay.
MR. TRANTINO: A smack could be just as hard as a punch.
MR. CONSOVOY: Oh, I understand that.
MR. TRANTINO: That’s why when I think about it I would say if I smacked a woman that’s — that would be a beating. So, um, I mean, there is no way to get around that.
MR. CONSOVOY: And it’s violent.
MR. TRANTINO: It’s violent, yeah.
Trantino’s New York parole file contained a parole officer’s report, prepared soon after the Lodi homicides, that referred to an interview with Trantino’s first wife:
The parolee’s wife reported to the writer on [sic] 8-28-63 that throughout most of their marriage, they got along well together. However, she reported that approximately 2¡é months ago, the parolee stuck her several times per week. She reported at that time that the parolee and she would quarrel and then make up. The parolee’s wife felt that the reason why the subject would hit her was that he no longer loved her and was interested in other women.
At the June 4, 1999 Parole Board hearing Chairman Consovoy characterized the information about Trantino’s physical abuse of his first wife as a “new” issue. In response, respondent’s counsel notes that the Parole Board had possessed in its files “for twenty years” a memorandum summarizing the New York parole file sent to then Bergen County Prosecutor Guy Calais. On the question of spousal abuse, that memorandum stated:
August 26, 1963 The wife was interviewed and she stated that her husband had left her two weeks before and she had removed her clothing and her son’s clothing from the apartment.
In an interview on August 28, 1963 with the wife it was revealed that for several months prior, the subject had physically abused her several times a week.
*153Trantino’s Appellate Division brief also quotes at length from his testimony at a 1995 Adult Panel parole hearing in the course of which Trantino acknowledged that he had on occasion physically abused his first wife during the period prior to the Lodi murders.
Nevertheless, the Parole Board concluded that Trantino’s lack of candor concerning the extent of his physical abuse of his first wife was an additional factor supporting denial of parole:
Records from New York indicate the extent of abuse your first wife suffered in the months prior to the murders in Lodi. When questioned about your relationship with your first wife by Dr. Rosenfeld, you claimed you assaulted your wife on two occasions but you stated it wasn’t a beating and you didn’t think she would call you abusive. Dr. Rosenfeld noted in his report that you appeared to be minimizing the extent of violence towards your wife. It is the Parole Board’s opinion that you have not been fully candid with authorities about your domestic violence to your first wife and have mitigated the extent of violence you showed towards her.
(f) Trantino’s Lack of Candor Concerning Efforts on His Behalf in 1974 to Remove a New York Parole Detainer Issued in Respect of New York Parole Violations Dating Back to 1963
At the June 9, 1999 hearing Chairman Consovoy asked Trantino if he was aware that in 1974 a New York attorney named Sanford Katz had attempted to remove the New York detainer so that Trantino could attain full minimum status in New Jersey. Trantino denied any knowledge concerning Katz’s efforts on his behalf. He acknowledged that an attorney named Jeff Fogel, who then was representing Trantino, had mentioned that “he was going to do something [about the detainer],” but stated that “I never saw the letter.”
In its June 9, 1999 decision denying parole, the Parole Board concluded that Trantino had not testified candidly before the Board on that issue:
You explained at your June 9, 1999 hearing that you were not aware your attorneys were attempting to have New York’s parole warrant removed, despite the fact that letters indicate that you did retain a Mr. Katz to represent you in this matter. In sum, the Parole Board believes you were being less than candid regarding this issue. Perhaps this was a miscommunication between you and your attorney so therefore if you can provide the Board an affidavit from Mr. Katz stating that his *154letter to New York parole authorities dated August 30, 1974, in which he claims he was retained by you, was false the Board will reconsider our position on this issue.
Subsequent to the Board’s June 9, 1999 decision attorneys Jeff Fogel and Sanford Katz submitted certifications to the Parole Board. Fogel’s certification stated that he requested Katz’s assistance because of his own lack of familiarity with New York parole law, and that he did not recall informing Trantino that Katz was attempting to remove the New York detainer. Katz’s certification stated that he assisted Fogel without fee, and never sent Trantino copies of his correspondence with New York parole authorities. Katz added that his efforts were-unsuccessful, but that the New York parole officials voluntarily removed the detainer two years later.
(g) The Parole Board Also Inferred a Lack of Candor from Trantino’s Assertion During His Interview with His Retained Psychologist, Dr. Barry Rosenfeld, That “He Took Pride in the Many Part-time Jobs He Held as a Youth” and That He Had ‘Worked Steadily from Age Seven until Approximately Fourteen Years Old. ”
In its decision denying parole, the Board observed “Your employment during your childhood —You claim on Page 10 of Dr. Rosenfeld’s report that you maintained steady employment from the time you were age seven to age fourteen. Kings County Probation Reports indicate otherwise.” However, Trantino’s comment to Dr. Rosenfeld was readily reconcilable with the New York probation report that merely noted that Trantino’s employment record was “spotty” after the age of sixteen.
3. Trantino’s Failure to Address in Psychological Counseling the Issues That Led Him to Engage in Domestic Violence with His First Wife.
The Court previously has summarized the evidence in the record relating to domestic abuse by Trantino of his first wife, Helene, in connection with the Parole Board’s finding that Trantino’s testimony about that issue was not candid. Supra at 150-53, *155764 A.2d at 962-65. In its decision denying parole, the Parole Board also observed that Trantino’s “lack of counseling in the area of domestic violence is disturbing,” relying primarily on the following colloquy that occurred between Trantino and Board member Washington at the June 9, 1999 hearing:
MS. WASHINGTON: Mr. Trantino, do you understand the cycle of violence in regards to domestic violence, have you studied that at all?
MR. TRANTINO: No.
MS. WASHINGTON: Or had any counseling on that at all?
MR. TRANTINO: I understand a little bit about it.
MS. WASHINGTON: Could you explain that?
MR. TRANTINO: Many of the counselors talk about it in particular to us, because most of the men who were incarcerated have been violent with women.
MS. WASHINGTON: Can you explain-can you tell us what the cycle of violence is, can you explain it?
MR. TRANTINO: No, I don’t know what that particular thing is.
MS. WASHINGTON: No counseling on the cycle of violence?
MR. TRANTINO: I don’t know specifically what that is. If I — my understanding, if I understand this correctly, I don’t know if it’s a correct understanding of this, is the way I have heard a cycle of violence explained in our groups by women counselors who want us to understand it is that our fathers did this, we did this, our children would do this. 7
MS. WASHINGTON: No, that’s not the cycle of violence.
MR. TRANTINO: That’s not it?
*156MS. WASHINGTON: No.
MR. TRANTINO: Okay.
MS. WASHINGTON: I just wanted to find out if you had any counseling at all on the cycle of violence.
Based on its finding that Trantino had not received psychological counseling while in prison about the causes of his abuse of his first wife, the Board determined that it was substantially likely that he would commit a crime if released on parole:
Records indicate in the months prior to your present criminal offense you engaged in domestic abuse, both physically and in the manner you described at your hearing. The abuse was so severe that your first wife moved out of your apartment in the weeks preceding the murder. Yet, in all your years of incarceration and counseling, you have never seriously explored this issue. In fact, as Dr. Rosenfeld notes, you have mitigated your abuse towards your first wife. Without exploring this issue in psychological counseling and the causes of the aforementioned abuse towards your first wife, the Parole Board is of the opinion it is substantially likely you would commit a crime if released on parole.
4. Lack of a Suitable Parole Plan
At the June 4 and June 9, 1999 hearings members of the Parole Board questioned Trantino about his current parole plan. Trantino acknowledged that his current plan was more uncertain than prior parole plans because of his divorce in 1997 from Charlee, his second wife. Trantino testified that Charlee sought the divorce because of her inability to cope with his continued incarceration and her fear and discomfort about the media attention focused on Trantino. He testified that he and Charlee “still love each other even though we’re divorced,” and that she remained willing to let him work for her in Pennsylvania if he was released. He stated that, other than the divorce, his parole plan remained the same, “to live in Camden, get an apartment, say in Cherry Hill. I would be working for my wife, my wife had a business, I would be going to Rutgers, I would be getting counseling, be going to AA meetings.” Acknowledging the possible difficulty in securing permission to work in Pennsylvania, Trantino stated that he would “work at McDonalds,” “scrub the floors,” “clean toilets,” or work in a library or bookstore. He expressed confidence in his ability to get work, and anticipated receiving assistance from his AA *157sponsor who lived in Camden, Charlee, his attorney, his sister, and his brother-in-law, a retired policeman. For recreation he would read, write and paint.
Chairman Consovoy disagreed that Trantino had a suitable parole plan, describing it instead as a “notion.” In its decision, the Parole Board expressed concern over Trantino’s parole plan, noting that the Board “would generally be concerned with the parole plan of any inmate, such as you, who has been incarcerated most of his adult life, has little regards of a support system, and has not had steady employment in his entire life.”
In concluding that Trantino’s inadequate parole plan constituted an additional ground for denying parole, the Board observed:
Your lack of a sufficient parole plan and the impact of likelihood of future recidivism is also a concern to the Parole Board in light of your clear lack of full understanding of the intensity of criticism you may face if paroled and how you may respond to that. It is the position of the Parole Board that you would be intensely scrutinized and criticized if you were released on parole. This would be heightened by the fact that if you were paroled to a placement plan, pursuant to the Department of Corrections policy you possibly would be paroled to Bergen County, the county of your underlying criminal commitment.
The Parole Board also would like to make two items clear for the record regarding your parole plan suitability. First, it is not the responsibility of the Parole Board to ensure that an inmate has available to him or her, a parole plan that will reduce the likelihood of future criminal behavior. Therefore, it is not the responsibility of the Parole Board to ensure that your parole plan is suitable to reduce the likelihood of future criminal behavior. The Parole Board can only evaluate the parole plan submitted by the inmate in the context of the Parole Board’s legal standards in assessing parole suitability. In your case your parole plan raises serious concerns in terms of likelihood of future recidivism if you are released on parole.
Second, in contrast to your past several parole hearings, you are now divorced from your wife Charlee, who in the past was a central part of your support system if paroled. You attempted to explain your current relationship with Charlee, and in all candor your explanation of your current relationship with Charlee did nothing to assuage the Board’s concerns regarding your parole plan.
5. Trantino’s Psychological Profile
The record before this Court includes numerous psychological evaluations of Trantino since 1979 when he first became eligible *158for parole and the vast majority of those evaluations reflect a positive assessment of Trantino’s suitability for parole. Among those positive psychological evaluations are 1995, 1996, and 1997 evaluations of Trantino by Dr. Glenn Ferguson, the Parole Board staff psychologist whose June 2 and June 9, 1999 testimony before the Board constituted the primary evidence on which the Board relied in denying parole. In its decision, the Parole Board gave “no weight” to the November 1998 psychological evaluation of Trantino by Dr. Michael Weiner who was retained by the Attorney General and on the basis of whose report Trantino was removed from Talbot Hall and transferred to South Woods Prison. Although Dr. Weiner expressly found that Trantino’s likelihood of criminal recidivism was “low,” the Board determined that his report was not useful because it did not focus on the central issue of recidivism and because Weiner was not provided with reports from New York parole authorities concerning Trantino’s background prior to the Lodi murders.
In addition, the Board also accorded “no weight” to the January 1999 psychological evaluation of Trantino by his own expert, Dr. Barry Rosenfeld, who from 1994 to 1999 served as the senior forensic psychologist of the “New York Criminal and Supreme Courts” in affiliation with Bellevue Hospital. Dr. Rosenfeld concluded that “the majority of predictive factors suggest that Mr. Trantino will be capable of successfully adjusting to parole and does not present a high risk of re-offending.” Dr. Rosenfeld’s report listed eighty-seven separate sources of information on which he relied. The Board determined that Dr. Rosenfeld’s report and evaluation should be given “no weight in this parole determination” because (1) Dr. Rosenfeld was given “a very limited amount of material to review,” (2) Dr. Rosenfeld “did little or nothing to challenge the veracity of statements made by the inmate,” and (3) Trantino “himself claimed at his hearings ... that a number of statements Dr. Rosenfeld attributed to him, he did not make.”
*159In addition to its primary reliance on the June 2 and June 9, 1999 testimony of Dr. Ferguson, the Board also relied to a limited extent on the testimony of Mark O’Sullivan, a “psychology consultant” employed by the Board. After issuance of its June 9, 1999 decision denying parole, the Board conducted additional interviews on November 8 and 9,1999 with O’Sullivan, Dr. Naftali Berrill, Dr. Timothy Brennan and Dr. Mario Papparozzi for the purpose of establishing Trantino’s future eligibility term. In its supplemental brief to this Court the Parole Board relies to a limited extent on the post-parole decision testimony of Drs. O’Sullivan and Berrill, and refers briefly to the November 1999 testimony of Drs. Brennan and Papparozzi. In the interest of completeness, we first summarize the testimony of all of the witnesses on whom the Board relied in denying parole based on Trantino’s psychological profile.
A. Evidence Unfavorable to Parole
(i) Dr. Glenn Ferguson — Testimony June 2 and June 9, 1999
Dr. Ferguson, the Parole Board’s chief psychologist and a Board employee since 1981, received his Ph.D. in psychology in 1997. Dr. Ferguson had served as a parole hearing officer from 1988 to 1995 and had performed inmate psychological evaluations for the Board beginning in 1995.
He began his June 2, 1999 interview by reviewing the history of Trantino’s psychological evaluations as an inmate. He stated that of Trantino’s fifty-six evaluations, he considered fourteen to be “in depth evaluations” and forty-two to be relatively brief evaluations. Of the fourteen “in-depth” evaluations, he determined that four were clearly supportive of parole and the balance were indecisive or not supportive of parole. He acknowledged that thirty-five out of the remaining forty-two shorter evaluations were supportive of parole for Trantino. Ferguson attributed the difference in those results to Trantino’s ability to establish “a very good rapport very quickly.”
*160Dr. Ferguson observed that Trantino’s psychological testing at Talbot Hall in 1998 was “one of the more thorough in-depth evaluations ... done in the recent past on Mr. Trantino.” Dr. Ferguson was questioned by Board Chairman Consovoy about the results of a Millon Clinical Multiaxial Inventory-III (MCM-III) test of Trantino performed at Talbot Hall. Dr. Ferguson acknowledged that although the test results revealed symptoms of three distinct personality disorders (narcissistic, antisocial and borderline), none of the scores could be considered high enough to be “clinically significant.” Ferguson stated that Trantino frequently had been diagnosed with narcissistic and antisocial personality disorders, but noted that none of the prior psychological evaluations had considered Trantino to possess a borderline personality disorder. Ferguson observed that in his opinion Trantino “meets the criteria for all three [disorders].”
Ferguson observed that certain aspects of Trantino’s personality tend to make him “potentially violent,” noting that it was unpredictable how Trantino would act in certain situations. Ferguson stated: “Because he doesn’t have a very clear understanding of who he is, and he hasn’t really done a very thorough self-examination, he’s not even real clear on what type of situations might present problems for him in the future. And that might set him off.”
Dr. Ferguson explained to the Board that, in isolation, Trantino’s narcissistic personality disorder was not a significant cause for concern. He explained that “narcissism is definitely a problem. But again, there’s plenty of people out there with narcissistic personality disorders that aren’t a threat to anybody else. As long as they get their egos fed and they get everything the way they want, you know, life goes on.” However, in Ferguson’s view the combination of Trantino’s narcissism and a borderline personality disorder could lead to unpredictable and potentially violent situations:
Mr. Trantino [is] a particularly dangerous individual. Because of his unpredictability and his unwillingness to clearly examine potential problems in the future. That, you know, makes it almost impossible for him to avoid situations where he is going *161to be faced with blows to his ego, blows to his narcissism that are going to trigger feelings of, you know, emptiness, and loneliness, and unworthiness, and, you know, wanting to fill those.
If he’s in a community setting where he doesn’t have a lot [of] external controls, you know, strict supervision, you know, serious consequences for behaviors, and there’s a liquor store or bar on the corner, you know, it’s going to take an awful lot of effort on his part to not to walk into that bar and sit down and start drinking.
And once that hurdle’s crossed, you know, the next guy that walks in and recognizes him and says, “Oh, that’s that cop killer.” And, you know, “What the hell are you doing here?” And, you know, gets in his face and starts pointing his finger and pushing his buttons, I don’t think it takes that big a leap to say that there is a potential for violence.
According to Ferguson, Trantino could not successfully overcome the risks inherent in his personality disorders until he acknowledged their existence and made a commitment to seek the treatment necessary “to change those parts of his personality.”
Dr. Ferguson also disputed the findings of Trantino’s expert, Dr. Rosenfeld. Ferguson accused Rosenfeld of bias and inexperience, stating that Rosenfeld had not previously evaluated an inmate for parole purposes. Ferguson and the Board asserted that Rosenfeld’s report was based on unsubstantiated and incorrect information provided solely by Trantino and his lawyer, and that Rosenfeld admitted he had not sought out objective data on his own.
Ferguson also accused Rosenfeld of scoring the Psychopathy Checklist-Revised test (PCL-R) incorrectly, which Ferguson described as his “biggest problem” with Dr. Rosenfeld’s report. Ferguson testified that the PCL-R test is “one of the leading tests being used ... by forensic psychologists for making predictions for future violence.” He explained that the test contains twenty questions about an inmate’s personality and behavior, with each question being scored as either “0” (behavioral trait does not apply), “1” (behavioral trait applies to a certain extent), or “2” (behavioral trait applies). According to Ferguson, a score of thirty or above is indicative of psychopathy, which also correlates with a high risk of recidivism.
*162Ferguson testified that when he administered the PCL-R to Trantino in August 1998, Trantino received a nineteen, but that after receiving the New York probation documents Ferguson re-scored Trantino as a 24, which is not in the psychopathic range (30-40), and reflects the average score of most inmates. Ferguson reported that Talbot Hall initially scored Trantino a 22, but after receiving updated information revised that to a 26. Ferguson admitted there are “always some judgement calls” in scoring psychological tests.
Ferguson noted that Dr. Rosenfeld scored Trantino at 14.7 on the PCL-R, a score he characterized as “indicative of a very low level of antisocial- functioning, let alone psychopathy.” Ferguson stated that Rosenfeld erroneously scored Trantino on the basis of his current ability to function rather than on the basis of Trantino’s “lifetime” experience as contemplated by the PCL-R scoring manual.
Ferguson also explained to the Board his change in opinion concerning Trantino since his 1995 evaluation, attributing his current views to his enhanced experience as a psychologist, newly-acquired information from the New York parole authorities, and the changes in Trantino’s support system. Ferguson explained that in 1995, Trantino had “an extensive support system” in Camden, “that he developed through his contacts at the Riverfront State Prison,” including his wife, Charlee, whom Trantino met in prison, and married while on furlough. Ferguson noted that in 1997, after almost twenty years of marriage, Trantino and Charlee divorced. Ferguson concluded that Trantino’s support network and parole plan had-gone “by the wayside.”
Concerning Trantino’s substance abuse problems, Ferguson noted Trantino’s extensive and on-going involvement in various programs including Alcoholics Anonymous, but observed that statements by Trantino, such as confidence that he would never relapse, revealed a failure to understand “the [Alcoholics Anonymous] model of substance abuse and alcohol treatment.”
*163Dr. Ferguson recommended that Trantino receive “more intensive treatment to deal with his personality issues,” which would require a therapist to see him “for a long period of time [and] then to see him in a less restrictive environment.” In response to a Board member’s question about how long it would take — “three years, five years, to really find out if he can function,” Dr. Ferguson replied that it would depend on the “level of freedom and ... supervision in a halfway house” noting that “within two years you can come across just about all the conceivable options and outcomes that he might see.”
In concluding the interview with Dr. Ferguson, Chairman Consovoy noted that “it would be fair to say that we would never put any pressure on anyone to give us an evaluation we want.”
After Trantino’s June 9, 1999 interview, at which Dr. Ferguson was present, the Board re-interviewed Dr. Ferguson. Dr. Ferguson stated that Trantino’s testimony had confirmed his assessment of Trantino’s unfitness for parole. He agreed with a Board member that Trantino’s desire to write another book about the crime evidenced his narcissism and lack of empathy for the families of the victims. He reemphasized Trantino’s lack of self-awareness and the deterioration of his support system. Finally, he agreed with Consovoy’s characterization that “in judging risk [Trantino] hasn’t made much progress” in the thirty-five years since the crime.
At the conclusion of the June 9, 1999 interview, Chairman Consovoy asked Dr. Ferguson whether “it would be fair to say that under the legal standard of substantial likelihood [of recidivism] that this Board has some legitimate concerns.” Dr. Ferguson’s reply was that “[i]t sounds like a legal question more than a psychological question.”
(ii) Mark O’Sullivan — November 9,1999
Mr. O’Sullivan is a psychology consultant employed by the Parole Board. His background and qualifications are not revealed by the record except for the statement in the confidential adden*164dum to the Board’s decision that “Mr. O’Sullivan has been a member of the Parole Board’s Psychology Unit for a number of years and has conducted hundreds of pre-parole psychological evaluations in his own right.”
On November 9, 1999, five months after the Parole Board’s decision denying Trantino parole, Chairman Consovoy interviewed O’Sullivan concerning a PCL-R test that he administered to Trantino prior to the Board’s parole decision. O’Sullivan testified that prior to administering the test he reviewed “some 40-odd years’ worth of psychological [reports],” court decisions, transcripts of parole hearings, and “media presentations.” O’Sullivan scored Trantino at 31 on the test, a score substantially higher than that received by Trantino when the same test was administered at Talbot Hall, twice by Dr. Ferguson, and once by Dr. Rosenfeld. Based on his scoring of the test, O’Sullivan expressed the view that Trantino was a “prototypical psychopath,” and that he considered him a poor risk for parole.
(in) Dr. Naftali Berrill — November 9, 1999
Several months after the Parole Board issued its decision denying Trantino parole, Chairman Consovoy and Mark O’Sullivan interviewed Dr. Berrill on November 9, 1999. Dr. Berrill, a Board-certified forensic psychologist, had never met nor interviewed Trantino but had reviewed materials from his parole file including psychological tests and reports, Parole Board hearings and evaluations. He noted that the prior psychological evaluations tended to characterize Trantino as possessing narcissistic and antisocial personality disorders, but noted that “I don’t think anyone has gone as far as to say, for example, that there was a form of borderline personality disorder....” When asked about the relevance of Trantino’s twenty-five years of infraction-free incarceration, Berrill remarked, “it’s impressive.... It’s interesting. It is noteworthy. It demonstrates this man’s capacity to perform this behavior on some level____ He’s not a total loose cannon in what he’s really doing.” Berrill added: “I just think it’s *165evident that ... when drugs and alcohol are not in the picture, he’s capable of formidable behavior, which is good, I mean that’s fine.” However, Berrill continued, it “absolutely” does not mean that there has been a “shift in the way he looks at the world____ It speaks in no way, shape or form to the core personality structure.” Berrill observed that a realistic assessment of Trantino is only possible in an unsupervised setting.
Berrill summarized his findings: “There is no reason for you to think that there’s any substantive change in the way that that man looks at himself or ... if left to his own devices, it’s entirely possible, that he may well drink and use drugs again.” Moreover, he opined, “there’s been no genuine treatment, the real treatment, substantive treatment.”
(iv) Timothy Brennan
On November 8, 1999, subsequent to the Board’s parole decision, Chairman Consovoy conducted a telephone interview with Dr. Brennan. Brennan observed that research in the area of risk assessment demonstrates that prior criminal history is probably the most important factor in predicting future risk. Other major factors are the length of criminal history, seriousness of current offense, and substance abuse, “[institutional behavior itself seldom is seen in our literature or even in our clinical work as the main factor.” He stated there are about ten major factors that most institutions use, and institutional behavior is not one. Brennan explained that institutional behavior is “in a constrained and supervised environment,” and thus unreflective of behavior in the community.
(v) Dr. Mario Papparozzi — -November 8,1999
Subsequent to the Board’s decision denying parole, Chairman Consovoy on November 8, 1999 conducted a telephone interview with Dr. Mario Papparozzi, who late last year was confirmed by the State Senate as the new Chairman of the Parole Board. Dr. Papparozzi, who served for twenty-six years in the Department of *166Corrections working in parole and community programs, also served as an assistant professor at the College of New Jersey and as the associate director of the College’s Criminal Justice Policy Center. Dr. Papparozzi had not interviewed Trantino nor did he testify that he had reviewed records of Trantino’s psychological evaluations. The focus of his interview concerned “the relation between long-term infraction-free or trouble-free institutional adjustment and behavior on the street — once out.”
Papparozzi stated that from personal experience, “just because someone is infraction-free in an institution for a period of time, one should not conclude that that is reason to believe that once released to the streets that they also remain infraction-free or remain in a pro-social mode.” He noted that inmates are constantly being watched, thus are less likely to act out, and prison rewards good behavior.
Papparozzi then asserted that his personal observations are supported by the academic literature on risk prediction. He stated that, according to the literature, the “most important point” to understand is that “no one factor can be held responsible ... for predicting the likelihood of an outcome once released to the street.” In fact, “institutional adjustment” is not even in the top three items “that are accounting for outcomes,” although it does account “for a portion of the outcome in terms of an actuarial risk assessment.”
The witness admitted that it is “virtually impossible ... to predict ... which particular person is going to do a bad thing. ” The risk models, which are “nowhere [sic] near 80 or 90 percent” accurate, only predict the percentage of inmates that will recidivate, not which inmates will do so.
B. Evidence Favorable to Parole
(i) Prior Psychological Evaluations by Dr. Glenn Ferguson
Dr. Ferguson, on whose June 1999 testimony the Board primarily relied in denying parole based on Trantino’s psychological *167profile, also had evaluated Trantino in August 1995 and August 1998, and had been interviewed by the Board in February 1996. Dr. Ferguson’s August 1995 evaluation was highly supportive of parole for Trantino. In that report Dr. Ferguson reviewed Trantino’s background, his “pattern of behavioral change” while in prison, and noted that he had consistently been diagnosed with an antisocial personality disorder. Ferguson stated that Trantino, although unable to recall the actual shootings, “accepts full responsibility for both murders, in light of eyewitness testimony and his recollection of holding a gun both prior to and after the offense.” He observed that Trantino “received a PASS score of 62% which places him in the high end of the mid-range of PASS recipients who are viewed as equally likely to succeed and fail on parole. PASS scores of 65% or higher are viewed as clear cut indicators of successful parole outcomes, while scores of 35% or lower are viewed as indicators of parole failures.” Ferguson’s 1995 report concluded with a strong recommendation for parole, preferably but not necessarily preceded by transfer to a halfway house:
Presently, Thomas functions more as a staff member in the prison than as an inmate. He has developed a well rounded and reputable support network which should enable Thomas to manage the increased stress level of independent living optimally. He has a high level of motivation for continued education, therapy and' employment. Elis employment plans appear realistic and attainable, including technical work with his wife’s copyrighting company and creative work in art and therapy.
A transitional setting prior to the final release would most likely aid Thomas in making a problem free adjustment on parole after thirty-two years of incarceration, but is not viewed as crucial. The combination of well established social supports, motivation for ongoing treatment, and age will most likely lead to a successful parole outcome. Psychotherapy at this point should focus on issues of conflict resolution, anxiety, and possible emerging depression. Any weakening in the support network or evidence of relapse into substance abuse could lead to disastrous results.
In a February 1996 interview with the Parole Board, Ferguson was asked whether further institutional programs could benefit Trantino. Ferguson replied, “[a]t this point it’s almost overkill as far as institutional programs” and “he’s gotten just about every*168thing he can get as far as institutional programs.” Ferguson also stated:
I think the substance abuse has been addressed. You know, there was really no psychopathology to begin with.
As far as the personality, I don’t want to say personality disorder, because yes, the antisocial personality disorder definitely holds true with Mr. Trantino. But in terms of a thought disorder like, you know, a major personality problem schizophrenia or (inaudible) disorders, you know, (inaudible) types of syndromes, those— those never existed as far as anybody can see.
In Ferguson’s opinion, although he believed that Trantino could “succeed on the street,” the more appropriate disposition would be to transfer Trantino to a halfway house to allow him to encounter day-to-day pressures, while providing him with continued outpatient treatment.
In Dr. Ferguson’s August 1998 report, he continued to recommend parole for Trantino preceded by a transitional placement in a halfway house. He observed that the result of a PCL-R test revealed antisocial personality traits, “although they are not of a magnitude to suggest that [Trantino] is an imminent risk for parole violation, or violent recidivism if released,” noting that “[h]is overall score falls at the 28th percentile, suggesting Thomas is less antisocially oriented than 72% of the 1192 prison inmates in the standardization sample.” Ferguson concluded:
If Thomas can successfully negotiate the difficulties inherent in community halfway house placement for one year, it would seem reasonable to believe that community living on parole would meet with similar success. Special care should be given to ensuring his sobriety through frequent unannounced urine monitoring and continued AA participation. Mental health counseling should also be a condition of parole with an emphasis on developing and maintaining positive coping strategies.
(ii) Dr. Barry Rosenfeld
Dr. Rosenfeld, a forensic psychologist retained by Trantino, submitted a psychological evaluation of Trantino on January 22, 1999 and an addendum on March 12, 1999. As noted, supra, at 157-58, 764 A.2d at 968 the Parole Board gave “no weight” to Dr. Rosenfeld’s conclusion that Trantino “will be capable of successfully adjusting to parole and does not present a high risk of re-offending.” Rosenfeld reviewed Trantino’s present psychological *169condition and concluded that he did not currently “suffer from any major mental disorder or personality disorder.” Rosenfeld conceded Trantino’s “grandiosity, a need for attention, and limited empathy,” but he deemed those traits to be “consistent with a narcissistic personality disorder [rather] than an antisocial personality disorder.”
According to Rosenfeld, “perhaps the most important predictor” of the low likelihood of Trantino’s recidivism was that he scored relatively low on the PCL-R. The PCL-R score was undoubtedly insightful, Rosenfeld noted, because the test is not “overly susceptible to deliberate attempts to mask psychopathy.” He stated that Trantino’s low score, a 15, was “a strong positive indicator of successful adjustment to parole.”
Rosenfeld recognized that Trantino possesses several of the traditional historical factors that correspond to a likelihood of recidivism, such as childhood delinquency (drug use), the young age at which he began his criminal behavior, the fact that he reoffended while on parole in the past, and a substance abuse history. However, Rosenfeld noted that such factors can never be altered or minimized by rehabilitation, and that regardless, research supporting those criteria do not consider inmates who have spent “thirty plus years in prison.” Rosenfeld stated that there is very little research on recidivism against which Trantino can be compared. As a result, Rosenfeld asserted that all traditional factors must be considered in context, including clinical data, current psychological adjustment, and behavioral controls.
Finally, Rosenfeld acknowledged the high rate of recidivism among paroled violent criminals, noting that, “the possibility of recidivism will never be eliminated.” He concluded that the risk of recidivism would be lessened by parole conditions designed to ease Trantino’s transition into society, including placement in a halfway house, life-skills training, outpatient psychotherapy, and outpatient substance abuse treatment.
*170(iii) Dr. Michael Weiner
As noted, supra at 157, 764 A.2d at 968, the Board gave “no weight” to the fifty-three page report of Dr. Weiner, the expert retained by the Attorney General in response to Trantino’s placement in Talbot Hall, who determined that Trantino’s likelihood of criminal recidivism was “low.” Applying a more stringent standard than the “substantial likelihood” of recidivism standard set forth in the 1979 Parole Act, Dr. Weiner’s report concluded with the observation that Trantino “has more to accomplish before I can say, with any reasonable degree of medical certainty, that he will not break the law again.”
C. Other Psychological Evaluations
Pursuant to the order of the Appellate Division entered February 4, 1999, the Parole Board was required to make available to Trantino’s counsel copies of various documents that were reviewed by the Parole Board, parole authorities, and state experts. Those documents included approximately fifty psychological evaluations of Trantino, in addition to those evaluations previously identified in this opinion. Although only a relatively small number of those evaluations were reproduced in the printed record submitted to this Court, all of the remaining psychological evaluations were provided to the Court by the Attorney General’s office and we treat those evaluations as part of the record in this appeal.
Because Trantino did not become eligible for parole consideration until 1979, we have examined closely only the thirty-five psychological evaluations from 1979 to date. We note that the Parole Board, in its decision denying parole, does not appear to take into account the conclusions reached in any of those thirty-five psychological evaluations, although their decision makes reference to findings concerning Trantino’s personality disorders in two of those evaluations.
Our review of those thirty-five psychological evaluations, spanning the period from 1979 to 1997, reveals that eight of the evaluations focused not on parole eligibility but rather on Tranti*171no’s suitability for full-minimum status, para-professional placement within the correctional system, or halfway house placement, and all such evaluations were favorable and supportive.
Of the twenty-seven psychological evaluations that focused on suitability for parole, twenty-three were favorable and supportive, one was not supportive, and three could fairly be described as guarded. Among the more comprehensive evaluations in the record are two by Dr. Kenneth L. McNiel, principal clinical psychologist at the Avenel Diagnostic and Treatment Center. In June 1991 Dr. McNiel observed:
Regarding personal changes during this period of incarceration, several factors suggest that Mr. Trantino had indeed made significant progress in managing emotional conflicts without manifesting impulsive, acting out behaviors. First, he has maintained sobriety for many years, and it is undeniable that his extensive history of drag and alcohol abuse contributed to his overall reckless and dangerous lifestyle as a young adult. Second, his institutional record clearly indicates steady improvement in institutional adjustment with decreased conflicts with authority figures over time. Third, a review of previous psychological evaluations also suggests steady improvement in emotional functioning over time. Fourth, Mr. Trantino has clearly achieved a new identity for himself, no longer the wild and careless outlaw, but now the respected inmate and published author. Finally, Mr. Trantino has demonstrated some improvement in specific aspects of his antisocial personality. For example, an inability to empathize or appreciate the viewpoints of others is one common feature of antisocial and narcissistic personalities. Consistent with this, when asked about his first wife during his first psychological evaluation by Joseph Fahey (4/15/64), Mr. Trantino expressed bitterness towards her for not visiting him, despite his acknowledgment that he had abused her and cheated on her. When asked a similar question during the present evaluation, Mr. Trantino reported, “She was a kid, only 19 years old. She was scared to death. I only treated her terrible, so it makes sense to me that she wouldn’t visit. At that time, I wasn’t really thinking about her, I was just self-absorbed and hurt.” Furthermore, Mr. Trantino is better able to recognize his inadequacies and ongoing problems than previously, and appears more capable of experiencing emotional satisfaction without the use of artificial stimulants or extreme stimulation.
In October 1992, after a second evaluation, Dr. McNiel observed:
As noted in the previous evaluation, Mr. Trantino’s fundamental personality is of a narcissistic and antisocial type, and this will likely remain unchanged throughout his life. However, as is usually the case with similar personality structures, there comes a time in the middle to late adulthood when the antisocial personality seems to “burn out”, leaving a less energetic, less impulsive individual. It would appear that this process has begun to happen for Mr. Trantino.'
*172As to what this means for community release, it would appear that critical factors remain similar to the previous evaluation, including an initial period of close supervision to include monitoring for substance abuse, mental health treatment, and vocational and social support as needed. How well Mr. Trantino will ultimately fare in response to specific life stresses in the future is not easily predicted, thus it would appear that the availability of appropriate support systems and monitoring would make the most sense. As to whether r-elease should be though a halfway house placement or direct community release, this frankly seems less important than whether appropriate treatment and parole supports are in place if and when full release is granted.
IV
Standard of Review
The New Jersey Constitution, Art. VI, sec. 5, para. 4, specifically authorizes judicial review of administrative agency determinations. See In re Senior Appeals Examiners, 60 N.J. 356, 363, 290 A.2d 129 (1972); Fisher v. Bedminster Twp., 5 N.J. 534, 538-10, 76 A.2d 673 (1950). In Trantino TV, supra, we set forth the standard that courts should follow in determining the validity of the Parole Board’s denial of parole. A reviewing court must examine:
(1) whether the agency’s action violates express or implied legislative policy, i.e., did the agency follow the law; (2) whether the record contains substantial evidence to support the findings on which the agency based its action; and (3) whether in applying the legislative policies to the facts, the agency clearly erred in reaching a conclusion that could not reasonably have been made on a showing of the relevant factors.
[Trantino IV, supra, 154 N.J. at 24, 711 A.2d 260.]
In this appeal we focus on the second prong of that three-part standard. We noted in Trantino TV, supra, the essentially factual nature of a Parole Board’s determination that ‘“there is a substantial likelihood that an inmate will commit another crime if released,’ ” and that accordingly a reviewing court is obligated to “ ‘determine whether [that] factual finding could reasonably have been reached on sufficient credible evidence in the whole record.’ ” Id. at 24, 711 A.2d 260 (quoting State Parole Bd. v. Cestari, 224 N.J.Super. 534, 547, 540 A.2d 1334 (App.Div.) (citation omitted), certif. denied, 111 N.J. 649, 546 A.2d 558 (1988)).
*173We previously have recognized that Parole Board decisions are highly “individualized discretionary appraisals.” Beckworth v. N.J. State Parole Bd., 62 N.J. 348, 359, 301 A.2d 727 (1973). Accordingly, the Board “has broad but not unlimited discretionary powers,” and its determinations “are always judicially reviewable for arbitrariness.” Monks v. N.J. State Parole Bd., 58 N.J. 238, 242, 277 A.2d 193 (1971); accord In re State in Interest of Steenback, 34 N.J. 89, 101, 167 A.2d 397 (1961); In re Smigelski, 30 N.J. 513, 527-28, 154 A.2d 1 (1959).
Although in Trantino IV, supra, we acknowledged “the inherent difficulty in gauging whether a parole determination constitutes an abuse of discretion,” 154 N.J. at 25, 711 A.2d 260, we emphasized that the judicial review of Parole Board determinations “does not engender a more exacting standard of judicial review than that applicable to other administrative agency decisions.” Ibid. As this Court observed in In re Parole Application of Hawley, 98 N.J. 108, 484 A.2d 684 (1984):
The Board is the administrative agency charged with the responsibility of deciding whether an inmate satisfies the criteria for parole release under the Parole Act of 1979. We find no reason to exempt the Parole Board from the well-established principle that a court may review the actions of an administrative agency to determine if its power is being exercised arbitrarily or capriciously.
[Id. at 112, 484 A.2d 684 (citations omitted).]
See also Cestari, supra, 224 N.J.Super. at 548 n. 6, 540 A.2d 1334 (rejecting the contention that a more restrictive standard of judicial review should apply to parole release than to other administrative decisions).
We also take note of the Parole Board’s statement in its June 9, 1999 decision denying parole that “the actual granting or withholding of parole is a function reposing exclusively in the Parole Board, and there is no such thing as judicial parole.” Although the instances are few in which courts have found Parole Board decisions denying parole to be so arbitrary that affirmative judicial intervention to grant parole was necessary, that relief clearly may be encompassed within the province of judicial review. See Williams v. State Parole Board, 336 N.J.Super. 1, 9-10, 763 *174A.2d 747 (App.Div.2000); Cestari, supra, 224 N.J.Super. at 551, 540 A.2d 1334; Mallamaci v. Dietz, 146 N.J.Super. 15, 22-23, 368 A.2d 947 (App.Div.1976).
In performing our judicial review function in the context of this record, which reflects extraordinary reliance by the Parole Board on the expert testimony and opinion of Dr. Ferguson, we must reiterate concerns that we previously have expressed about the perils of undue reliance by courts and agencies on expert testimony and the impropriety of abdicating decisional responsibility to experts. In State v. Krol, 68 N.J. 236, 344 A.2d 289 (1975), we addressed the nature of civil involuntary commitment and periodic review hearings for criminal defendants acquitted by reason of insanity, and made clear that the court, not the psychiatric expert, was required to decide the issues:
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.
[68 N.J. at 261, 344 A.2d 289.]
Similarly, in In re D.C., 146 N.J. 31, 679 A.2d 634 (1996), which involved the involuntary civil commitment of a paroled convicted sex offender, we cautioned that “[t]he final determination of dangerousness lies with the courts, not the expertise of psychiatrists and psychologists.” Id. at 59, 679 A.2d 634. See also State in the Interest of C.A.H. & B.A.R., 89 N.J. 326, 343, 344 n. 5, 446 A.2d 93 (1982) (“Obviously, the court was not required to give controlling effect to the testimony of the experts---- [W]hile expert opinion on rehabilitative prospects is helpful and important in a waiver hearing ... such opinion evidence cannot be a substitute for the court’s ultimate, highly discretionary decision, reached through an application of all the statutory criteria to all of the relevant evidence, that the waiver of juvenile court jurisdiction is appropriate.”); Williams, supra, 336 N.J.Super. at 9, 763 A.2d 747 (“Dr. Gibbons’ opinion in his report ... is entirely without founda*175tion and is contradicted by the empirical evidence. It does not provide sufficient credible evidence to deny parole.”); Cestari, supra, 224 N.J.Super. at 550, 540 A.2d 1334 (“[Dr.] Rotgers opinion that Cestari has ‘the potential to become involved in a violent incident’ does not mean that there is a substantial likelihood that he will commit another crime____Thus, the Adult Panel read more into Dr. Rotgers’ solitary negative evaluation of Cestari than actually is stated in his report.”
For purposes of this appeal, the principle that we extract from those decisions is that the Parole Board was obligated, in considering Trantino’s application for parole, to render its decision not on the basis of the testimony of a single expert, or selected experts, but rather by application of “the statutory criteria to all of the relevant evidence.” C.A.H. & B.A.R., supra, 89 N.J. at 344 n. 5, 446 A.2d 93 (emphasis added).
V
Sufficiency of the Evidence8
We now address the sufficiency of the evidence supporting the various grounds for denial of parole relied on by the Parole Board. Because of the complexity and controversy concerning the first ground, Trantino’s psychological profile, we shall proceed again by addressing the grounds for denial of parole in reverse order.
1. Trantino’s Plan to Write Another Book
Although the Parole Board did not identify this issue as a discrete ground for denial of parole, in its decision the Board *176treated Trantino’s statement that “I feel like I am going to have to write something to earn some money” as evidence that he had not been rehabilitated during his years in prison. The Board stated:
Perhaps no other conversation at your hearing on June 9,1999 exemplifies areas in which you have not made great changes, than your conversation with Mr. Gomez regarding the possibility of you writing a book if you were paroled.
The Board’s observations, however, did not appear to take into account Trantino's subsequent testimony on the issue, in which he expressed his intention to reconsider the appropriateness of writing another book: “What you just said bothered me a lot, about the families. It just brought up stuff, you know, I work on that, and it hurts. I don’t want to hurt them. And now I have to — I am going to rethink about the book.”
We agree with the Appellate Division’s conclusion that
[viewing Trantino’s comments in their entirety, the Board’s finding that Trantino’s plans for a second book evince a lack of empathy for the victims’ families could not be reasonably reached on this record.
[Trantino V, supra, 331 N.J.Super. at 616, 752 A.2d 761.]
Moreover, none of the evidence concerning that issue is at all relevant to the question whether Trantino would be “substantially likely” to commit another crime if he were released on parole.
2. Lack of Candor
(a) The Talbot Hall Incident
As noted, supra at 143-46, 764 A.2d at 958-60, the Parole Board concluded that Trantino had not been candid in his June 4, 1999 testimony to the Board in which he denied that any incident occurred at Talbot Hall during which he became “agitated” and “perturbed,” noting that Trantino later had recalled the incident during his June 9, 1999 testimony. However, the record reveals that between the two hearings Trantino was shown a copy of Dr. Rosenfeld’s report that referred to the incident, and he fully explained his initial failure to recall it by noting that at the June 4 hearing Consovoy did not identify Trabueco by name. We are fully in accord with the Appellate Division’s observation that “as soon as Consovoy mentioned Trabucco’s name at the June 9, 1999 *177hearing, Trantino remembered the incident and provided a reasonable explanation.” Id. at 613, 752 A.2d 761. Neither the incident itself, nor Trantino’s explanation of his failure to recall the Talbot Hall incident at the June 4 hearing, indicate that Trantino is substantially likely to recidivate nor do they provide any evidentiary support for the Board’s decision to deny parole.
(b) Trantino’s Lack of Candor Concerning the Murder of Voto and Tedesco
As noted, supra at 146, 764 A.2d at 968, the Parole Board determined that Trantino’s inconsistent testimony at trial, at a post-conviction relief hearing, at Board hearings and during psychological evaluations about the extent of his recollection of his role in the murders of Voto and Tedesco reflected a “pattern of deception” that justified denial of parole. According to the Board, Trantino’s “claims of past memory loss” were neither “consistent” nor “genuine,” leading the Board to conclude that Trantino had not been candid regarding his memory of the events on the night of the murders. That finding by the Board is contrary to this Court’s unanimous determination in Trantino TV that “there is evidence in the record that Trantino’s memory loss is consistent, long-standing and genuine, and, beyond the issue of recollection, his acknowledgment of responsibility is sincere and legitimate.” 154 N.J. at 35, 711 A.2d 260.
The Board’s heavy reliance on Trantino’s inconsistent testimony about the extent of his recollection of the homicides, and its finding that his claimed memory loss is not genuine, reflects the Board’s conclusion that Trantino can never be paroled until he sufficiently recalls the details of the Lodi murders. In Trantino IV, we expressly held that the Board could not deny parole until psychological treatment resulted in a restoration of Trantino’s recollection:
While it may be appropriate in other cases involving pre-Code inmates to insist upon evidence of subjective awareness of guilt before concluding that an inmate’s avowal of responsibility is sincere and that he is rehabilitated to the point that he *178will not commit crimes if released, such an insistence may be unwarranted in this case where, we repeat, the record supports the finding that Trantino cannot and will not ever be able to remember actually pulling the trigger. The record thus does not clearly sustain the conclusion that long-term psychotherapy will eventuate in a breakthrough recollection or that such a recollection is necessary, given Trantino’s repeated acceptance of responsibility, in order to ensure that he has achieved a level of rehabilitation that eliminates the likelihood that he will, if released, commit crimes.
[Trantino IV, supra, 154 N.J. at 38, 711 A.2d 260.]
Accordingly, the Board’s reliance on Trantino’s inadequate recollection of the details of his crimes to support its denial of parole constituted a clear abuse of discretion in view of this Court’s specific holding to the contrary in Trantino IV.
(c) Trantino’s Actions on Parole in New York
Our prior description of this ground for denial of parole, supra at 147, 764 A.2d at 960-61, demonstrates that the Board determined that because Trantino, from June 1961 to August 1963, violated the conditions pursuant to which he had been released on parole from a New York prison, those violations demonstrated a lack of candor in his dealings with New York parole authorities. Although the facts relied on by the Board are undisputed, we agree entirely with the Appellate Division’s observation that Trantino’s failure to comply with New York parole conditions from 1961 to 1963 was irrelevant to his fitness for parole in 1999:
The Board failed to reasonably explain how Trantino’s conduct in the early sixties could reveal his suitability for parole in 1999, when the only relevant inquiry before the Board was whether Trantino was likely to commit another crime----In the light of these findings and the record before us, the conclusion by the Board that Trantino’s conduct while on parole in New York some thirty-six to thirty-nine years earlier supports his denial of parole could not reasonably have been reached, and fails to apply the appropriate standard for parole.
[Trantino V, supra, 331 N.J.Super. at 619-20, 752 A.2d 761.]
(d) Trantino’s Account of the Payroll Robbery on February 16, 1956
As noted, supra at 147-49, 764 A.2d at 961-62, the Parole Board determined that Trantino’s failure to disclose, during his testimony before the Board, that the victim of a robbery that *179he committed with an accomplice in 1956 sustained personal injuries revealed a lack of candor that supported denial of parole. In our view the Board’s reliance on Trantino’s testimony concerning that 1956 robbery to support its denial of parole was both arbitrary and a clear abuse of the Board’s discretion. The record clearly indicates that Trantino had no direct knowledge of the victim’s injuries until Chairman Consovoy informed him of those injuries at the June 9, 1999 hearing. More importantly, his testimony concerned a crime he committed forty-three years earlier, and neither inaccuracies, if any, in his testimony nor the crime itself can constitute support for the Board’s decision to deny parole. We fully agree with the Appellate Division’s comments concerning that issue:
Additionally, Trantino’s lack of knowledge, before June 9, 1999, that the victim had been injured, renders his denial of that fact during prior proceedings irrelevant to his recidivism prospects. Moreover, even if he did know, his mitigation of that portion of an offense committed at age eighteen, some forty-three years earlier, fails to demonstrate he would probably commit another crime now," especially when viewed in the light of his consistent admission to the offense, and his concession that it was “violent.” Furthermore, the excerpt from the New York probation report relied upon by the Board suggests it was Trantino’s accomplice, Fernandez, who fell to the floor with the victim, and that Trantino’s own role was limited to patting her down and locating the payroll envelope. There is nothing in the report to support the Board’s conclusion that it was Trantino who “pulled” the victim “down.”
[Trantino V, supra, 331 N.J.Super. at 617, 752 A.2d 761.]
(e) Domestic Abuse of Trantino’s First Wife
We summarized in detail (supra at 150-53, 764 A.2d at 962-65) the evidence in the record concerning this ground for denial of parole. As we understand it, the Parole Board concluded that Trantino’s testimony that he struck his first wife, Helene, three or four times in 1963, and his similar statements when interviewed by Dr. Rosenfeld, minimized the extent of his actual physical abuse of Helene. That misrepresentation, the Board determined, reflected a lack of candor supporting denial of parole. In reaching that conclusion, the Board relied on a New York parole officer’s interview of Helene in August 1963 during which she stated that “approximately two and one-half months ago, the *180parolee struck her several times per week.” We note that that information has been included for many years in the Parole Board’s file on Trantino.
We are unpersuaded that the difference between Trantino’s 1999 testimony and the 1963 report of the New York parole officer clearly demonstrated a lack of candor. The parole officer’s 1963 report did not necessarily establish that Trantino struck Helene more often than Trantino had acknowledged. Even if it did, the Parole Board has failed to establish any material connection between Trantino’s alleged minimization of his 1963 assaults of his first wife and the “substantial likelihood” that if released on parole presently he would commit another crime. The Parole Board’s reliance on that testimony to support its denial of parole was arbitrary and improper.
(f) Trantino’s Lack of Candor Concerning Efforts on His Behalf in 197Uto Remove a New York Parole Detainer Issued in Respect of New York Parole Violations Dating Back to 196S
We previously summarized the evidence in the record relating to this issue. Supra at 153-54, 764 A.2d at 965. As we understand the record, the Parole Board has abandoned this ground for denial of parole based on certifications filed with the Parole Board demonstrating that, contrary to the Board’s belief, Trantino had no knowledge about efforts made in 1974 by lawyers on his behalf to remove a New York parole detainer that had prevented Trantino from attaining full minimum status in New Jersey.
(g) The Parole Board Also Inferred a Lack of Candor from Trantino’s Assertion During His Interview with His Retained Psychologist, Dr. Barry Rosenfeld, That “He Took Pride in the Many Part-time Jobs He Held as a Youth” and That He Had “Worked Steadily from Age ¡S'euere”
The Parole Board inferred a lack of candor from Trantino’s representation to Dr. Rosenfeld that he had maintained *181steady employment from ages seven to fourteen, noting that a probation report from Kings County contained information inconsistent with that representation. However, the Kings County probation report merely noted that Trantino’s employment record after age sixteen was “spotty,” a characterization that is totally consistent with Trantino’s representation that focused on ages seven to fourteen. Moreover, that evidence bears no relation to whether there exists a “substantial likelihood” that Trantino will commit another crime if released on parole, and the Parole Board’s inclusion of that ground to support denial of parole consisted a clear abuse of discretion.
3. Trantino’s Failure to Address in Psychological Counseling the Issues That Led Him to Engage in Domestic Violence with His First Wife.
We previously summarized the evidence relating to this ground for denial of parole, supra at 150-53, 764 A.2d at 962-65, noting that the Parole Board concluded that, because of Trantino’s lack of psychological counseling while in prison about the causes of his physical abuse of his first wife in 1963, it was substantially likely that Trantino would commit another crime if released on parole. We note that the Board did not acknowledge Trantino’s consistent admissions that he had abused his first wife in the months preceding the Lodi murders and that he considered his infidelity during that period to constitute a form of abuse. No evidence was produced to establish that domestic violence counseling had been offered to Trantino while in prison, nor that he had rejected such counseling. No psychological, behavioral, or domestic abuse experts were interviewed by the Board to establish a connection between domestic abuse and recidivism. The Board did not rely on evidence of recidivism rates in relation to prior untreated domestic violence, nor did it establish that the Board’s routine practice was to deny parole to inmates that had not received counseling for a prior history of domestic violence although incarcerated on an unrelated charge. In short, no evidence in this record provides any support for the Parole Board’s *182conclusion that Trantino!s lack of counseling for domestic violence that he engaged in more than thirty-seven years ago demonstrates a substantial likelihood that he would commit another crime if released on parole.
4. Lack of a Suitable Parole Plan
We previously summarized the evidence concerning Trantino’s current parole plan and its differences from prior plans. Supra at 156-57, 764 A.2d at 967-68. The primary change in his current parole plan, attributable to his divorce from his second wife Charlee in 1997, is that the extent to which Charlee would serve as a resource for Trantino in the event of his parole appears to be uncertain. Nevertheless, the record includes evidence indicating some likelihood that Charlee would continue to be an available resource. In a letter to the Parole Board Charlee stated that their divorce was “prompted by the Board’s capricious behavior in four times granting parole to him, then after pressure was applied, taking away that parole - which caused emotional devastation and tremendous financial hardship to me----With my own health at stake, removing myself from the situation seemed the only choice, and it was a heartbreaking one.” Trantino’s testimony was consistent with Charlee’s explanation, and he asserted that Charlee would be willing to offer him a job if he were released. Moreover, Dr. Weiner, the psychological expert retained by the Attorney General, noted in his 1998 report that Charlee remains part of his support system and stated: “Were Tom Trantino to walk out of prison tomorrow, despite their divorce, it is still my professional opinion that Charlee would quickly become part of his everyday life.”
In other respects Trantino’s parole plan was unchanged. He contemplated living in the Camden/Cherry Hill area, finding work, attending classes at Rutgers, going to AA meetings, and seeking counseling. To the extent, however, that the Board determined that this parole plan was not “sufficient,” we note the impracticability of the Board’s assumption that an inmate imprisoned by the *183State for thirty-seven years could be expected to develop on his own a detailed and precise plan for his daily activities in the event of parole. The Board’s own regulations appear to recognize that a parole plan should not be a decisive factor in the Board’s parole decision, including an inmate’s “parole plans” as only one of twenty-three factors to be considered by the Board in deciding whether to grant parole. See N.J.A.C. 10A:71-3.11(b)14.
In its decision the Parole Board disclaimed all responsibility for assisting inmates in developing a parole plan, ignoring this Court’s specific recommendations to the Board in Trantino IV, supra:
In its determination, the Board can devise pre-release conditions that will allow it to assess how. Trantino handles the stresses of society that may induce or impel him to commit crimes. Those conditions could include community work details, furloughs, minimum security status, and other measures that would serve to reintroduce Trantino gradually into society and lead to his ultimate release. Alternatively, the Board may decide to impose post-release conditions, such as intensive supervision and drug testing, that would ensure that Trantino was not behaving in a way that indicates he will engage in criminal activity.
In its redetermination of parole fitness, the Parole Board may also consider halfway house treatment as a condition of parole.
[Trantino IV, supra, 154 N.J. at 39-40, 711 A.2d 260.]
We conclude that the Parole Board acted arbitrarily in relying on the insufficiency of Trantino’s parole plan as a ground for denial of parole. The Board’s implication that Trantino could be paroled to Bergen County, supra at 157, 764 A.2d at 967-68, and its unwillingness to recognize its own authority to mandate material supportive requirements as conditions of parole, cast doubt on the soundness of the Board’s reasoning. Finally, in view of our disposition of this matter, which will require Trantino to reside for one year under the supervision of a halfway house under contract with the Department of Corrections, we are satisfied that any imprecision in Trantino’s current parole plan readily can be remedied in the ensuing twelve months.
5. Trantino’s Psychological Profile
Because we heretofore have rejected all of the other grounds relied on by the Parole Board in denying parole, the sufficiency of *184the evidence relating to Trantino’s psychological profile, set forth in detail earlier in this opinion, supra at 157-72, 764 A.2d at 968-76, is of critical significance to our disposition. Before reviewing that evidence, we restate the controlling statutory standard that governs the Board’s disposition:
An adult inmate shall be released on parole at the time of parole eligibility, unless information supplied in the report filed pursuant to ... this act or developed or produced at a hearing ... indicates by a preponderance of the evidence that there is a substantial likelihood that the inmate will commit a crime under the laws of this State if released on parole at such time.
[N.J.S.A 30:4-123.53 (emphasis added).]
We noted earlier the Board’s heavy reliance on the June 1999 testimony of Dr. Glenn Ferguson, the Board’s chief psychologist. Supra at 159, 764 A.2d at 968. Other than Dr. Ferguson’s testimony, the Board relied only marginally on the testimony of several other witnesses who testified in November 1999 before the Board concerning Trantino’s future eligibility term, five months after the Board issued its decision denying parole. We are convinced that the Board should have accorded only minimal weight to the testimony of those witnesses because of both the substantive content of their testimony and the fact that the Board had decided to deny parole to Trantino several months before any of those witnesses testified.
Mark O’Sullivan, a member of the Board’s psychology unit for many years, testified in November 1999 that he scored Trantino at 31 on a PCL-R test administered earlier in 1999. Because that score was so much higher than Trantino’s scores on the same test administered twice by Dr. Ferguson, once at Talbot Hall, and once by Dr. Rosenfeld, the Board could not reasonably accord significant weight to his testimony in the context of the entire record. Dr. Naftali Berrill, who had never met Trantino but had reviewed some of his records, noted that he frequently had been diagnosed with narcissistic and antisocial personality disorders, but never with a borderline personality disorder. Dr. Berrill stated that Trantino’s many years of infraction-free incarceration do not necessarily reflect any change in his “core personality structure,” *185noting that a realistic assessment of Trantino’s prospects for success on parole would “only [be] possible in an unsupervised setting.” Dr. Berrill expressed no opinion on the likelihood that Trantino would commit another crime.
Dr. Timothy Brennan also testified in November 1999, and that testimony essentially was limited to Dr. Brennan’s opinion that institutional behavior “in a constrained and supervised environment” was not reflective of behavior in the community. Similarly, Dr. Mario Papparozzi, who also had not interviewed Trantino, testified that infraction-free institutional behavior is not “reason to believe that once released to the streets ... [an inmate will] also remain infraction free____” He acknowledged that it is “virtually impossible to predict ... which particular person is going to do a bad thing.” Neither Dr. Brennan nor Dr. Papparozzi expressed opinions on the likelihood that Trantino would commit another crime.
Accordingly, only Dr. Ferguson’s testimony provides any direct evidentiary basis for the Board’s findings concerning Trantino’s likelihood of recidivism. On several counts we find the Board’s reliance on Dr. Ferguson’s testimony unpersuasive.
We note preliminarily that as of the June 2, 1999 hearing, Dr. Ferguson had “just recently passed the licensing exam, written exam. So I’m on my way to being a full-fledged licensed-practicing psychologist hopefully in the next couple of months.” Chairman Consovoy characterized Ferguson as an expert because he had performed almost 400 in-depth evaluations.
Nevertheless, we are highly skeptical of the Parole Board’s complete acceptance of Dr. Ferguson’s testimony that Trantino possesses a borderline personality disorder that “makes Mr. Trantino a particularly dangerous individual.” In its decision the Board referred to the borderline personality diagnosis as Dr. Ferguson’s “most serious concern.”
We note that of the fifty-plus psychological evaluations of Trantino that are included in this record, Dr. Ferguson is the only *186psychologist to diagnose him with a borderline personality disorder. Dr. Berrill’s testimony confirmed that no one “has gone as far as to say, for example, that there was a form of borderline personality disorder.” The Diagnostic and Statistical Manual of Mental Disorders, fourth edition, (DSM-IV) explains that the symptoms of borderline personality begin in early adulthood, so it is unlikely that the disorder could have remained undetected by the other psychologists over the past forty years. “The essential feature of Borderline Personality Disorder is a pervasive pattern of instability of interpersonal relationships, self-image, and affects, and marked impulsivity that begins by early adulthood and is present in a variety of contexts.” DSM-IV at 650. The DSM-IV also cautions that borderline disorder can be confused with narcissistic disorder:
The most useful feature in discriminating Narcissistic Personality Disorder from ... Borderline ... whose interactive styles are respectively coquettish, callous, and needy, is the grandiosity characteristic of Narcissistic Personality Disorder. The relative stability of self-image as well as the relative lack of self-destructiveness, impulsivity, and abandonment concerns also help distinguish Narcissistic Personality Disorder from Borderline Personality Disorder.
[Id at 660.]
Moreover, “Borderline Personality Disorder is diagnosed predominately (about 75%) in females.” Id. at 652.
We also note that Dr. Ferguson’s own 1998 report makes no mention of this new borderline diagnosis. The 1998 report observes that “PCL-R results reveal antisocial personality traits, although they are not of a magnitude to suggest that Thomas is an imminent risk for parole violation, or violent recidivism if released.” According to Ferguson’s 1998 report, Trantino “is less antisocially oriented than 72% of the 1992 prison inmates in the ... sample.” Moreover, “there is no evidence of serious mental illness in either Thomas’s history or present adjustment.” The report reveals the results of a Rorschach (inkblot) test that “make the Board’s previous insistence on halfway house placement prior to parole release all the more appropriate.” The report does not mention that Trantino is a “particularly dangerous individual,” nor *187does it state that Trantino presents a substantial risk of recidivism.
The reliability of Dr. Ferguson’s 1999 testimony before the Board also is diminished significantly when it is compared in detail with Dr. Ferguson’s prior evaluations of Trantino in 1995 and 1998, and his February 1996 interview by the Board. In his 1995 evaluation Dr. Ferguson noted only Trantino’s consistent diagnosis with an antisocial personality disorder, referred to Trantino’s relatively favorable score on the Parole Assessment and Selection Scale test, and concluded with a favorable recommendation for parole, observing that “[t]he combination of well-established social supports, motivation for ongoing treatment, and age will most likely lead to a successful parole outcome.”
In his 1996 interview, Ferguson acknowledged only Trantino’s antisocial personality disorder, observed that “there was really no psychopathology to begin with,” and concluded that although he believed Trantino could “succeed on the street,” a better disposition would be a transfer to a halfway house prior to parole. Moreover, Ferguson’s June 1999 conclusion that Trantino should receive “more intensive treatment to deal with his personality issues” stands in sharp contrast to his 1996 testimony that “he’s gotten just about everything he can get as far as institutional programs.”
In his 1998 evaluation, Dr. Ferguson scored Trantino at 19 on the PCL-R test, a highly favorable score, but in his 1999 testimony before the Board he stated that after reviewing the New York parole records he re-scored Trantino at 24 on the same test, which is the average score for State prison inmates. We find it improbable, in view of Dr. Ferguson’s familiarity with Trantino’s entire background, that information in New York parole records from the early sixties could so substantially have affected Ferguson’s 1999 scoring of Trantino’s PCL-R test. Nevertheless, in his August 1998 evaluation, less than a year before his testimony before the Board, Dr. Ferguson reiterated that Trantino’s antisocial personality traits “are not of a magnitude to suggest that *188[Trantino] is an imminent risk for parole violation, or violent recidivism if released.”
We also note that although in Dr. Ferguson’s 1999 testimony before the Parole Board he described Trantino as “potentially violent,” Dr. Ferguson expressly declined to respond to Chairman Consovoy’s question whether Trantino was substantially likely to commit another crime, observing that that was “a legal question more than a psychological question.”
In Trantino TV, although acknowledging that courts and agencies should not abdicate decision-making responsibilities to experts, this Court nevertheless criticized the Parole Board for its failure to explain adequately its reasons for rejecting the opinions of psychologists who recently had recommended that Trantino be released on parole. Trantino IV, supra, 154 N.J. at 36, 711 A.2d 260. That criticism applies with far greater force to this record. In relying almost exclusively on Dr. Ferguson’s 1999 testimony about Trantino’s psychological profile as the primary basis for denying parole, the Board’s decision effectively disregarded all of the other psychological evaluations in the record that supported parole. The Board’s decision to accord “no weight” to the evaluations of Dr. Weiner and Dr. Rosenfeld — the former retained by the Attorney General and the latter by Trantino — was totally unjustified, and strongly suggests that it was their conclusions about Trantino’s low risk of recidivism, not their qualifications, that motivated the Board. Inexplicably, the Board completely disregarded Dr. Ferguson’s three earlier positive evaluations, and relied only on his 1999 testimony. Similarly, the Board ignored almost completely the thirty-five psychological evaluations of Trantino from 1979 to 1997 contained in the record, the vast majority of which were supportive of parole. The Board’s highly selective focus only on the psychological evidence supportive of its denial of parole, and its total disregard of evidence favorable to parole, undermines the deference that a court ordinarily would confer on an agency determination.
*189Under the Parole Act, the Board was not permitted to deny parole unless it found “by a preponderance of the evidence” that Trantino was substantially likely to commit another crime if released. N.J.S.A. 30:4-123.53. On this record there exists no doubt that the Parole Board’s finding that Trantino was substantially likely to recidivate was based not on a preponderance of the evidence in the record, but rather on the Board’s selective and arbitrary reliance on only those portions of the record that could possibly support the Board’s conclusion. The Board failed to address, and in fact utterly disregarded, substantial evidence in the record, spanning many years of infraction-free incarceration and favorable psychological evaluations, that demonstrated Trantino’s likelihood of success on parole. See C.A.H. & B.A.R., supra, 89 N.J. at 344 n. 5, 446 A.2d 93 (“[W]hile expert opinion on rehabilitative prospects is helpful and important in a waiver hearing ... [s]uch opinion evidence cannot be a substitute for the [tribunal’s] ultimate, highly discretionary decision, reached through an application of all the statutory criteria to all of the relevant evidence, that the waiver of juvenile court jurisdiction is appropriate.”) (emphasis added); Williams, supra, 336 N.J.Super. at 9, 763 A.2d 747; (“Dr. Gibbons’ opinion in his report ... is entirely without foundation and is contradicted by the empirical evidence. It does not provide sufficient credible evidence to deny parole.”).
The Board also ignored other evidence in this record to which we specifically directed its attention in Trantino TV:
The Parole Board’s ultimate determination of parole fitness must be based on whether there is a likelihood that Trantino will again engage in criminal activity. In evaluating Trantino’s fitness for parole under this standard, the Parole Board should give weight to the facts that Trantino is now sixty years old and, as already noted, has previously been released on sixty-nine work details and two furloughs without incident; has not violated a correctional rule in twenty-seven year’s; has successfully completed substance abuse counseling; has educated himself while in prison; has pursued the available vocations for prisoners; has a stable support network; was housed without incident at the Wharton Tract, a facility without perimeter guards; and has been deemed fit for transfer by the last two psychologists to evaluate him. Moreover, in light of that evidence, the length of time Trantino has served under his sentence, and the successive occasions on which he has been deemed eligible for parole, punishment is no longer a material consideration in the parole determination.
[Trantino IV, supra, 154 N.J. at 39, 711 A.2d 260.]
*190Based on our detailed and comprehensive review of the evidence relied on by the Parole Board in support of each of the several grounds on which it based its denial of parole, we conclude that the Parole Board’s denial of parole is not supported, as the Parole Act requires, by a preponderance of the evidence in the record and cannot be sustained. As noted, we found to be arbitrary and an abuse of discretion the Parole Board’s purported reliance on events that occurred in the years prior to the 1963 Lodi murders to support its decision. That reliance by the Board on evidence of such distant events can be understood only as a makeweight to overcome the lack of substantial evidence to support the Board’s conclusions. In our view, the Board’s emphasis on that evidence diminishes significantly the deference courts ordinarily accord to agency decisions. Concerning Trantino’s psychological profile, we conclude that the Board viewed the evidence selectively rather than comprehensively, relying almost exclusively on Dr. Ferguson’s testimony and excluding from its consideration substantial, credible, and persuasive evidence that was highly supportive of parole. Accordingly, we affirm the judgment of the Appellate Division to the extent that it reverses the Parole Board’s June 9, 1999 decision denying Trantino parole.
Our dissenting colleague relies almost exclusively on Dr. Ferguson’s 1999 testimony to support his conclusion that the Parole Board’s decision can be sustained based on evidence in the record. He also relies in part on the testimony of Drs. Berrill, Brennan, and Papparozzi, and Parole Board psychologist O’Sullivan, four witnesses who testified before the Board five months after the Board decided in June 1999 to deny parole. The dissent does not acknowledge, however, that those witnesses were not called by the Board to testify on whether parole should be denied, but rather were asked to testify on the length of Trantino’s future eligibility term. Nor does the dissent disclose that neither Dr. Berrill, Dr. Brennan nor Dr. Papparozzi ever examined or interviewed Trantino and that none of those witnesses ever expressed an opinion about the substantial likelihood of Trantino committing another *191crime. In fact, Dr. Papparozzi, the newly-designated chairman of the Parole Board, admitted to the Board that it is “virtually impossible to predict” which inmates are likely to recidivate. Supra at 166, 764 A.2d at 972. Nor does the dissent acknowledge that the PCL-R score obtained by O’Sullivan, a Board employee, exceeded by more than 30% the highest scores on the same test when administered at Talbot Hall and by Dr. Ferguson.
Thus, the essence of the dissent’s reliance on the record to support the Board’s decision focuses on the 1999 testimony of Dr. Ferguson. That reliance is plainly flawed — not because Dr. Ferguson failed to testify to the conclusion the Board anticipated would be reached — but because the Board, contrary to law, utterly and completely disregarded all of the contradictory evidence on the record, including Dr. Ferguson’s own prior testimony and evaluations, that supported parole.
We previously have summarized at length Dr. Ferguson’s 1995 and 1998 evaluations of Trantino, and his 1996 testimony, all highly supportive of parole following halfway house placement. The Board fails to address or acknowledge that testimony, and the dissent does not explain or justify the Board’s failure to consider such highly relevant evidence. The dissent acknowledges and agrees with the Board’s failure to accord any weight whatsoever to the supportive opinion of Dr. Rosenfeld. Post at 209-10, 764 A.2d at 997. But the dissent offers no explanation or justification for the Board’s decision to accord “no weight” to the opinion of Dr. Michael Weiner — the psychologist hired by the Attorney General — who concluded that Trantino had a “low” likelihood of recidivism. Nor does the dissent attempt to explain or justify the Board’s failure to consider and weigh the thirty-five psychological evaluations in the record, spanning the period 1979 to 1997, the vast majority of which were favorable and supportive of parole.
We already have acknowledged that reviewing courts ordinarily do not disturb agency decisions unless they are arbi*192trary, capricious or unreasonable, or are not “supported by substantial credible evidence in the record as a whole.” Dennery v. Board of Educ., 131 N.J. 626, 641, 622 A.2d 858 (1993) (emphasis added). However, as Justice Pollock noted in P.F. v. New Jersey Div. of Developmental Disabilities, 139 N.J. 522, 530, 656 A.2d 1 (1995): “When an agency’s decision is manifestly mistaken ... the interests of justice authorize a reviewing court to shed its traditional deference to agency decisions.” This Court, in Clowes v. Terminix International, Inc., 109 N.J. 575, 589, 538 A.2d 794 (1988), explained that a “sense of [agency] ‘wrongness’ can arise in numerous ways — from manifest lack of inherently credible evidence to support the finding, obvious overlooking or undervaluation of crucial evidence, a clearly unjust result, and many others.” (quoting State v. Johnson, 42 N.J. 146, 162, 199 A.2d 809 (1964)) (emphasis added).
Of the eight judges who have meticulously reviewed this record, only our dissenting colleague is satisfied that the Parole Board did not act arbitrarily, and that there is sufficient evidentiary support for its decision. This Court has concluded, however, that the Board’s decision cannot be sustained because it failed to comply with the Parole Act’s mandate that parole be denied only if a “preponderance of the evidence” demonstrated that Trantino was substantially likely to commit another crime if released. N.J.S.A. 30:4-123.53. Because of the Parole Board’s unjustifiable and “obvious overlooking or undervaluation of crucial evidence,” Johnson, supra, 42 N.J. at 162, 199 A.2d 809, we have no doubt that its determination, based not on a preponderance of all evidence but on evidence arbitrarily selected to support a desired result, is manifestly mistaken and must be set aside.
Finally, we mention only briefly the other portions of the record noted by the dissent. We previously have addressed the Talbot Hall incident and Trantino’s vaguely expressed intent to write another book, supra at 175-77, 764 A.2d at 978-79, both relied on by the dissent, post at 217-18, 764 A.2d at 1001-02, and add only that the evidence relevant to those matters sheds little, if any, *193light on the issue of Trantino’s likely recidivism, and provides no support for the Board’s determination. Similarly, the dissent’s reliance on Trantino’s alleged lack of candor concerning the injuries to the victim of the 1956 robbery and regarding his assaultive conduct toward his first wife, post at 215-17, 764 A.2d at 1001, involve incidents that occurred, respectively, forty-five and thirty-seven years ago. Moreover, the record reveals that Trantino apparently had no prior knowledge of the extent of the 1956 robbery victim’s injuries, and that the Parole Board had in its possession for over twenty years the New York probation report dealing with his assaultive conduct toward his first wife. This inmate’s alleged lack of candor about events that occurred three or four decades ago simply has no bearing on his present likelihood of recidivism. We flatly reject the dissent’s assertion that that evidence is material to or supportive of the Board’s denial of parole.
Despite the dissent’s reliance on evidence of Trantino’s alleged lack of recollection of his specific role in the 1963 Lodi murders, post at 215, 764 A.2d at 1000, we reiterate that pursuant to this Court’s unanimous decision in Trantino IV, supra, 154 N.J. at 34-38, 711 A.2d 260, that evidence cannot constitute support for the Board’s determination of Trantino’s likelihood of recidivism. We clearly stated in that opinion that “the record supports the finding that Trantino cannot and will not ever be able to remember actually pulling the trigger.” Accordingly, we determined that
[t]he record thus does not clearly sustain the conclusion that long-term psychotherapy will eventuate in a breakthrough recollection or that such a recollection is necessary, given Trantino’s repeated acceptance of responsibility, in order to ensure that he has achieved a level of rehabilitation that eliminates the likelihood that he will, if released, commit crimes.
[Id. at 38, 711 A.2d 260.]
Our decision in Trantino TV should have been understood by the Board to preclude it from denying parole based on evidence of Trantino’s lack of adequate recollection of the details of the Lodi homicides. Both the Board’s and the dissent’s reliance on that *194evidence are inconsistent with this Court’s decision in Trantino IV.
VI
In several important respects we modify the judgment of the Appellate Division. That judgment reversed the Parole Board’s denial of parole and remanded the matter to the Parole Board with the direction to grant Trantino parole in thirty days. In addition, although dismissing as moot Trantino’s appeal from the decision of the Department of Corrections transferring him from Talbot Hall to South Woods State Prison, the Appellate Division remanded that matter to the Department of Corrections and directed that Trantino “be immediately placed in a pre-parole halfway house or residential facility” in preparation for his release on parole in thirty days.
We are fully persuaded that Trantino’s release on parole should not occur in thirty days as contemplated by the Appellate Division’s judgment, but should be subject to a pre-release condition of satisfactory completion of twelve months placement in a halfway house facility with parole release to occur on completion of such twelve-month period. Accordingly, our disposition of Trantino’s separate appeal from the Department of Corrections’ decision to transfer him to South Woods is to remand that matter to the Department of Corrections and to order the Department of Corrections to transfer Trantino within thirty days of the filing of this opinion to a halfway house facility selected by the Department of Corrections that is located within a reasonable proximity to the Camden/Cherry Hill area.
Our conclusion to impose satisfactory halfway house placement for twelve months as a pre-release condition of parole is supported by the Parole Board’s numerous attempts during the past decade to achieve that very same result. In its 1991 decision denying parole, the Parole Board specifically recommended that “Trantino make every effort to achieve halfway house status in order for the Board to evaluate his behavior in a less structured, *195transitional environment.” Trantino then promptly sought halfway house placement, which was denied by the Department of Corrections.
In 1993, a two-member panel of the Board denied parole “because the DOC would not grant Trantino halfway house status.” The panel members recommended that Trantino file suit against the Department of Corrections. Trantino III, supra, 296 N.J.Super, at 451, 687 A.2d 274. In its December 1993 notice of decision, the Panel again strongly recommended halfway house placement for Trantino prior to parole release:
The Adult Panel is of the opinion that if Mr. Trantino can successfully enter and complete a correctional halfway house program as an inmate he can achieve his full rehabilitative potential and therefore will satisfy the punitive aspect of his sentence and meet the substantial likelihood test.
[Id. at 453-54, 687 A.2d 274.]
As directed, Trantino again applied for halfway house placement, which again was denied by the Department of Corrections in 1994. A letter from the Administration of Riverfront Prison attributed the denial to threats of harm to Trantino, his offense, adverse community reaction, and objections by a member of the Legislature.
After an April 1995 decision denying parole was vacated, a two-member Board panel denied parole again in September 1995, noting that “[t]he Department of Corrections has continually denied your placement in a halfway house” and expressing its belief that placement “into a halfway house would be beneficial to you in your goal to reach your rehabilitative potential.... ”
Because the record contains only an abbreviated explanation of the Department of Correction’s refusal to transfer Trantino to a halfway house, we express no view on whether the Department’s refusal, despite strong and consistent statements by the Parole Board favoring the transfer, was based primarily on meritorious considerations. In Trantino IV, supra, 154 N.J. at 24, 711 A.2d 260 we expressed our emphatic agreement with the Appellate Division’s determination in Trantino III, 296 N.J.Super, at 465, 687 A.2d 274, that the Department of Corrections’ 1994 decision *196denying Trantino halfway house placement was deficient in that it lacked an adequate statement of reasons demonstrating that the denial was based on sufficient credible evidence. In any event, the recommendation of halfway house placement for Trantino prior to parole release is supported not only by prior Parole Board decisions, by numerous psychological evaluation reports in the record concerning Trantino, and specifically by Dr. Ferguson’s 1998 report. In addition, we note that Department of Corrections regulations subject inmates placed in halfway houses to urine monitoring, breathalyzer testing, and disciplinary rules that regulate, among other subjects, unauthorized absences, out-of-state travel and possession of alcohol or controlled dangerous substances. See N.J.A.C. 10A:20-4.19 and 4.20. We are convinced that a twelve-month halfway house placement prior to Trantino’s parole is an appropriate and necessary safeguard to minimize the likelihood that his release from prison will lead to the commission of a criminal offense. In addition, the Parole Board is authorized to impose reasonable and appropriate pre- and posL-release conditions of parole consistent with this opinion.
VII
A
The responsibility for resolving this appeal in accordance with law has weighed heavily on the members of this Court, as well as on other judges and Parole Board members who have ruled on prior but related appeals. We do not underestimate the pain and anguish that our disposition is likely to cause to the families and friends of the victims of the Lodi murders. Nor are we unaware that our disposition will not be readily understood by members of the public who will find it incomprehensible that the law requires párole release of an inmate who was responsible for the murder of two police officers. Our opinion explains that parole release would be impossible if the Lodi murders occurred today, because of significant changes in the severity of the sentencing provisions of our Criminal Code. Current law would mandate no less than life *197imprisonment without parole, if not the death sentence. The law in effect when these crimes were committed was different. Under that law Trantino became eligible for parole release in 1979.
The Parole Act of 1979 eliminated the conventional parole discretion to release a prisoner once the punitive aspect of a sentence has been served. That statute provided only that a prisoner “shall be released on parole at the time of parole eligibility, unless ... there is a substantial likelihood that the inmate will commit a crime under the laws of this State if released on parole at such time.” N.J.S.A. 30:4-123.53(a). “The legislation shifts the burden to the State to prove that the prisoner is a recidivist and should not be released.” New Jersey Parole Bd. v. Byrne, 93 N.J. 192, 205, 460 A.2d 103 (1983). We stated that “the Legislature recognized that under the Parole Act of 1979 a parole eligibility date creates a legitimate expectation of release ... absent findings that justification for deferral exists. Based upon this interpretation of our statute, we find that a federally-protected liberty interest exists.” Id. at 207, 460 A.2d 103. In short, because the punitive aspect of his sentence already has been served, Trantino IV, supra, 154 N.J. at 39, 711 A.2d 260, Trantino had a constitutionally protected right to parole unless the State could prove that there was a “substantial likelihood” that he would commit another crime. It is the absence of that proof that entitles Trantino to parole, not sympathy or compassion for him. No matter how much we may abhor the admitted killing of those two officers, the law must apply.
Portions of this record can be read to suggest that under the law Trantino was eligible for transfer to a halfway house and subsequent release several years ago, but that public pressure prevented that from occurring. No matter how great the pressure, agencies of government cannot ignore the law in special eases. At its core, this case is more about the rule of law than it is about Thomas Trantino. “The great strength of the rule of law in a democratic society is that it applies equally to all persons, the *198bad as well as the good.” Catena v. Seidl, 68 N.J. 224, 228, 343 A.2d 744 (1975).
Carved into the stone of the facade of the United States Supreme Court building are these words: “Equal Justice Under Law.” If ever courts permit agencies of government to create exceptions to the rule of law, applying it for the many but exempting the disfavored, we will have irreparably damaged the foundation of our democracy.
B
We affirm in part and modify in part the judgment of the Appellate Division. The June 1999 decision of the Parole Board denying parole is reversed, and the Board is ordered to grant Trantino parole subject to the pre-release condition of satisfactory completion of a twelve-month halfway house placement, and such other appropriate pre- and post-release conditions that it may impose. Concerning Trantino’s appeal from the Department of Corrections’ order transferring him to South Woods prison, we remand that matter to the Department and order that Trantino be placed within thirty days of this opinion in a halfway house facility within a reasonable proximity to the Camden/Cherry Hill area.
So ordered.
See Ala.Code § 13A-5-40 (enacted 1975); Alaska Stat. § 12.55.125(a)(1), amended by 1992 Alaska Sess. Laws ch. 79 § 23; Ariz.Rev.Stat. § 13-1105(A)(3), amended by 1996 Ariz. Sess. Laws Ch. 343, § 2; Ark.Code § 5-10-101(a)(3) (originally enacted as 1975 Ark. Acts No. 280, § 1501); Cal.Rptr. § 190(c), amended by 1987 Cal. Stat C. 1006, § 1 (Prop. 67, approved June 7, 1988); Colo.Rev.Stat. § 18-3-107(3) (originally enacted as 1988 Colo. Sess. Laws, H.B. 1165, § 5); Del.Code Ann. tit. 11, § 636(4), amended by 59 Del. Laws, c. 284, § 1 (1973); D.C.Code Ann. § 22-2406 (originally enacted as D.C. Law 70-256, § 2(d) (1995)); Haw.Rev.Stat. § 701-701(1)(b), amended by 1986 Haw. Sess. Laws c. 314, § 49; Idaho Code § 18-4003(b) (originally enacted as 1972 Idaho Sess. Laws ch. 336, § 1); Ind.Code § 35-50-2-9(b)(6) (originally enacted as 1977 Ind. Acts P.L. 340, § 122); Iowa Code § 707.2(4) (originally enacted as 1976 Iowa Acts ch. 1245, § 702); Kan. Stat. Ann. § 21-3439(a)(5) (originally enacted as 1994 Kan. Sess. Laws ch. 252, § 1); La.Rev.Stat. Ann. § 30(A)(2), amended by 1973 La. Acts No. 109, § 1; Mich. Stat. Ann. § 750.316(1)(c), amended by 1994 Mich. Pub. Acts No. 267, § 1; Minn.Stat. § 609.185(4), amended by 1981 Minn. Laws c. 227, § 9; Miss.Code Ann. § 97-3-19 (originally enacted as 1974 Miss. Laws Ch. 576, § 6(1),(2)); Nev.Rev.Stat. § 200.033(7) (originally enacted as 1977 Nev. Stat. § 1542); N.H.Rev.Stat. Ann. § 630:1(I)(a); N.Y. Law § 125.27(a)(i),(ii) (originally enacted as 1974 N.Y. Laws c. 367, § 5); Ohio Rev.Code. Ann. § 2903.01(E), amended by 1998 Ohio Laws S. 193, H. 5; Okla. Stat. tit. 21, § 701.12(8), amended by 1981 Okla. Sess. Laws c. 147, § 1; Or.Rev.Stat. § 163.095(2)(a) (originally enacted as 1977 Or. Laws c. 370, § 1); R.I. Gen. Laws § 11-23-2(5), amended by 1984 R.I. Pub. Laws ch. 221, § 1; Tenn.Code. Ann. § 39-13-204 (originally enacted as 1989 Tenn. Pub. Acts ch. 591, § 1); Tex. Penal Code Ann. § 19.03(a)(1) (originally enacted as 1973 Tex. Gen. Laws ch. 426, art. 2, § 1); Utah Code Ann. § 76-5-202(1)(e) (originally enacted as 1973 Utah Laws ch. 196, § 76-5-202); Vt. Stat. Ann. tit. 13, § 2311(a)(7) (originally *124enacted as 1987 Vt. Acts & Resolves No. 60, § 1); Va.Code Ann. § 18.2-31(7), amended by 1977 Va. Acts ch. 478.
In his Appellate Division brief Trantino notes that all other defendants whose death sentences were commuted to life imprisonment in Funicello, supra, have long since been paroled. He observes that defendants Russo and Bisignario, who killed a policeman in Newark during a robbery, were sentenced to death in 1961 and paroled in 1973; James Belton, who killed a Fort Lee policeman, was sentenced to death in 1967 and served fifteen years in prison; and Clarence Billingsley, who stabbed two young sisters to death in 1964, served sixteen years in prison. He also notes the prison terms served by the remaining death row defendants: Reynolds brothers (15 years); Edgar Smith (15 years); Ralph Davis (15 years); Martin Mathis (15 years); Jesse Wilson (15 years); Wilbert Sinclair (15 years); Charles Holland (15 years); Daniel Artis (20 years); Thomas Farrow, Jr. (16 years).
The Chairman stated:
Okay. The State Prison cases under the Parole Act, when there is a denial of parole the Board has an obligation to set a future eligibility term. What the Board did was set a future eligibility term that would carry with it a presumptive — in other words we've considered the punitive aspect, we’ve considered the rehabilitative integration into that, and we’ve come up with the future eligibility term. When that term has been served and your credit pattern earning that you’ve established and the fine efforts that you've made, you can then say — it's something for you to hope to. In other words we are saying that in approximately five to five and a half years when perhaps the 25 year mandatory minimum might have been served had that been imposed, you have under the current law the right to presume that there will be. Where that presumption was not available to you because this was a *129precode case. We’ve tried to move it to that. So that there is some degree of certainty for you. We felt that you shouldn’t leave this hearing today without understanding the hope — light at the tunnel, and understand that it will end, and society has that obligation also to understand. But the Board is not just saying we are denying now and maybe next time we are going to deny and maybe the next time we are going to deny. What we are saying is we dealt with the issue once and for all, of the punitive aspect of the sentence. That issue for us is resolved, unless of course through aborant [sic] behavior you were to do something that gave rise to its reconsideration. But if it weren't, and all those aspects have been done, we are now looking at a day when you will be released.
I think the Board felt very strong in trying to arrive at a certainty. Because we feel strongly that there should be certainty and hope for any individual who is incarcerated, so he can look to the day that he will be released at— provided his behavior is good and all the things are good and he has earned that privilege. And that's what the Board has tried to do. It was a very, very difficult decision to do, but it — we have resolved that for ourselves at this time.
[Trantino v. New Jersey State Parole Board, 296 NJ.Super. 437, 475, 687 A.2d 274 (1997) {Pressler, J., dissenting in part) (Trantino /!/).]
The panel’s determination included the following comments by panel member Consovoy:
If the Court tells us that in your case or any other case we cannot— especially in a 2A case, we cannot demand a halfway house placement as a pre-condition to parole approval, then frankly the board has to deal with it as it is.
And in my opinion, in the setting you’re in and it’s also the opinion of the professionals that have evaluated you the last two times, that you have, in fact, reached your rehabilitative potential. Any inmate who's reached his rehabilitative potential on a gratuity sentence must be paroled whether we want to or not.
But in this situation — you see, Tom, there is no legal reason for them to deny halfway house. We’ve been through that. You meet every criteria. I suggest you follow it. That you take this matter where it belongs, and I believe this matter belongs before the judicial body to referee the tug-of-war between the parole board and the DOC.
But it’s come — Tom, it’s come to the end. You’ve done your 30 years. You’ve done every program you can. You achieved full-minimum status, you’ve had full-time minimum status. If the Courts come back and tell the parole board that we can’t do what we’re trying to do, then you basically have my (indiscernible) I would to parole you. I can’t speak for anybody else. I voted to parole you last time, and you know, people (indiscernible) I think (indiscernible).
This is not what you wanted to hear, and you don’t understand this, but this is the best thing that can happen to you. It’s got to end. It’s got to end before a judge and the judge is going to tell the DOC you’re right or they’re going to tell the parole board you're right. And then you are on the way to get out.
[296 NJ.Super, at 451-52, 687 A.2d 274.]
That letter read in part:
On January 28, 1994, inmate Trantino made application for community release with Volunteers of America and Clinton House as his place of preferential assignments. His application was referred to the Institutional Classifica*133tion Committee, chaired by Donald E. Lewis on February 2, 1994. Mr. Trantino's request for halfway house assignment was discussed, and the committee rejected his request. The denial was is [sic] based upon two (2) factors:
1. Letters of threat, received by my office, warning that Mr. Trantino would be killed if paroled. The three (3) letters were unsigned and very crudely written. One letter was alphabetized, meaning constructed by letters cut out from magazines and newspapers to spell out the threat. The letters did not have a place of origin identifying the area where mailed. Letters were received on or about mid-January, and were shared with the Classification Committee in order to render an informed decision.
2. The committee also took into consideration the circumstances of the offense and the risk of possible adverse community reaction if inmate Trantino was permitted to participate in a residential community release program.
In considering all of the above factors, a unanimous decision was rendered by the Institutional Classification Committee, in keeping with the provisions of New Jersey Administrative Code 10A:20-4.12 which provides for Institutional Classification Committee review and disposition.
Further be advised, notwithstanding the letters of threat, inmate Trantino would have been denied based on other factors referenced under the above cited provision. As his case has high visibility and notoriety, through the news media and through Senator Kosco who vehemently objected to the parole and community release of inmate Trantino.
In concluding this report, the letters of threat have been misplaced, as it is my recollection that they were to be processed to Internal Affairs; however, this is a standard procedure, and it is possible that they were misdirected. For your further review and assistance, I have Mr. Trantino's application, classification blocks and reports to Senator Kosco all addressing the situation of inmate Trantino's parole.
Talbot Hall is operated by Education and Health Centers of America, Inc. pursuant to a 1995 agreement with the DOC for an initial term of ten years, subject to the DOC's right to terminate the agreement at any time without cause on thirty days’ notice. The basic compensation rate under the contract is $65 daily for each DOC inmate housed at Talbot Hall.
The definition provided by Trantino, while not the one used most often by clinical specialists in the area of abuse, is an acceptable definition that is used by individuals and organizations working with children in domestic violence homes. See Women's Shelter Program of San Luis Obispo County (visited Dec. 18, 2000) <http://www.callamer.com/~wsp/children.htm> ("Stopping the Cycle of Violence____Unfortunately, children learn what they experience at home____Boys and girls who grow up in violent homes are more likely to cany these behaviors into their own relationships.”). The definition Ms. Washington apparently was looking for is:
Battering does not occur constantly, but rather in a cycle. The cycle consists of three phases: the tension building phase, the acute battering incident, and the kindness, contrite, loving behavior (sometimes referred to as the "honeymoon” stage.)
Domestic-Violence.net (visited Dec. 18, 2000) <http://www.actabuse.com/ cycleviolence.html >.
Before the Appellate Division and before us Trantino contended that the Parole Board exceeded the scope of this Court's remand by adducing evidence not in the record at the time of our decision in Trantino IV, or alternatively that if the Board could supplement the record it was limited to "new information” only. The Appellate Division rejected that contention. Trantino V, supra, 331 N.J.Super, 605-11, 752 A.2d 761. In view of our disposition we need not address the issue.