The opinion of the court was delivered by
Following a jury trial, defendant was convicted of two counts of murder, N.J.S.A. 2C:11-3(a)(1) and (2); burglary, N.J.SA 2C:18-2; unlawful possession of an assault firearm, N.J.S.A 2C:39-5f; unlawful possession of a handgun, N.J.S.A 2C:39-5b; and two counts of possession of a weapon for unlawful purposes, N.J.S.A 2C:39-4a. These convictions arose from an incident that occurred on November 28, 1992, during which two people were murdered. Two other individuals, Charles Ankrah and Terrence Doe, were also indicted for the same offenses; Ankrah and defendant were tried together. At the time of that trial, Doe had not yet been tried. The same jury that convicted defendant acquitted Ankrah of all charges. At a subsequent trial, Doe was also acquitted of all charges.
Following his convictions, defendant was sentenced on the murder convictions to two consecutive life terms, each with a thirty-year parole disqualifier. Two concurrent four-year terms were imposed on the unlawful weapons possession convictions and a concurrent seven-year term for the burglary conviction was also imposed. The two remaining weapons convictions were merged into the murder convictions.
*516In the same indictment charging him with the November 28, 1992 murders and other related offenses, defendant was also charged, along with codefendant Doe, with attempted murder and related offenses arising from a November 21, 1992 incident. Following his convictions for the November 28, 1992 murders, defendant entered into a plea agreement regarding the November 21, 1992 charges to two counts of fourth degree aggravated assault and to one count of unlawful possession of a handgun. He received two concurrent 18-month terms without parole eligibility on the aggravated assault convictions and a concurrent five-year term for the weapons possession conviction. These sentences were to run concurrent with the life sentences imposed on the November 28,1992 murders.
Subsequent to his convictions and sentences, and while his appeal therefrom was pending, defendant became aware that the main identification witness against him, John Smith,1 who was eleven years old at the time of the murders, had a history of psychiatric problems. At that point, Doe was awaiting trial for his role in the November 28 events and a hearing was held to determine when John, who was then hospitalized, might be available to testify. Based on his treating doctor’s testimony, the judge postponed the Doe trial for another sixty days. Subsequent to this hearing, the voluminous medical records detailing John’s condition were provided by the prosecutor to defendant’s appellate counsel.
On defendant’s subsequent motions, we permitted the supplementation of the record with a July 5, 1995 transcript of the proceeding in the Doe trial regarding the then ability of John to testify, the discovered medical reports, and further ordered a limited remand for a hearing to determine whether the new evidence constituted grounds for a new trial. Thereafter, we entered an order expanding the scope of the remand to include a *517Wade hearing in light of the new evidence and to include in the record a December 8, 1995 transcript from the Doe trial containing evidence regarding the identification. Additional portions of the Doe trial were also considered by the trial judge on remand, along with the testimony of medical experts, John, and an Essex County Prosecutor’s Investigator who had participated in the murder investigation. Ultimately the trial judge rejected defendant’s Wade2 contentions, and denied his motion for a new trial based upon the “newly discovered” evidence relating to John.
I
In his original brief, defendant contends:
POINT I. DEFENDANT’S CONVICTIONS MUST BE REVERSED AND THE CASE REMANDED FOR A NEW TRIAL, BEFORE WHICH AN EVIDEN-TIARY HEARING IS HELD TO DETERMINE WHETHER JOHN SMITH, THE ONLY EYEWITNESS AGAINST DEFENDANT, IS COMPETENT TO TESTIFY, AND IF SO, TO WHAT EXTENT THE DEFENSE WILL BE PERMITTED TO USE HIS PAST PSYCHIATRIC HISTORY TO IMPEACH HIS CREDIBILITY.
POINT II. THE COURTS FAILURE TO DECLARE A MISTRIAL WHEN THE JURY COULD NOT REACH A VERDICT WAS REVERSIBLE ERROR IN VIOLATION OF DEFENDANT’S FEDERAL AND STATE CONSTITUTIONAL RIGHTS TO DUE PROCESS AND A FAIR TRIAL. U.S. CONST., AMENDS. VI AND XIV; N.J. CONST., ART. 1, PAR. 10.
POINT III. THE COURT’S FAILURE TO DISMISS COUNT FOURTEEN UNFAIRLY PREJUDICED DEFENDANT SINCE, CONTRARY TO THE EVIDENCE ADDUCED AT TRIAL THAT ANKRAII ALONE FIRED THE WEAPON, THE CHARGE AGAINST DEFENDANT AND NOT ANKRAH INDICATED TO THE JURY THAT THERE WAS A REASON UNKNOWN TO IT BUT KNOWN TO THE PROSECUTION WHY DEFENDANT AND NOT THE CO-DEFENDANT WAS CHARGED WITH UNLICENSED POSSESSION OF THE 9 MILLIMETER WEAPON.
POINT IV. THE TRIAL JUDGE’S REFUSAL TO INSTRUCT THE JURY ON THE FULL PANOPLY OF LESSER-INCLUDED OFFENSES SUPPORTED BY THE EVIDENCE, AND TO CHARGE THE ACCOMPLICE INSTRUCTION TO WHICH THE LESSER CRIMES GAVE RISE UNDER STATE V. MELKIEWICZ, DEPRIVED THE DEFENDANT OF DUE PROCESS AND A FAIR TRIAL UNDER THE STATE AND FEDERAL CON*518STTTUTTONS. (U.S. CONST., AMENDS. VI AND XTV; N.J. CONST. (1947), ART. 1, PAR. 10).
POINT V. THE SENTENCE OF THE COURT IS MANIFESTLY EXCESSIVE.
In his supplemental brief filed following the trial judge’s decision on remand, defendant contends:
POINT I. THE NEWLY DISCOVERED EVIDENCE OF JOHN SMITH’S MULTIPLE AND SEVERE PSYCHIATRIC DISORDERS WARRANTS A NEW TRIAL.
POINT II. THE REMAND COURT ERRED IN PROHIBITING DEFENDANT FROM ASKING JOHN SMITH, AT THE WADE HEARING, ABOUT THE RELIABILITY OF HIS PERCEPTIONS AND MEMORY AND ERRED IN FAILING TO EXCLUDE THE CHILD’S PRETRIAL AND IN-COURT IDENTIFICATIONS OF DEFENDANT AS UNRELIABLE.
Finally, in a second supplemental brief filed pursuant to our order granting defendant leave to appeal, nunc pro tunc, his guilty plea to the November 21,1992 charges, defendant contends:
POINT I. DEFENDANT’S CONVICTIONS MUST BE VACATED BECAUSE HE DID NOT PROVIDE AN ADEQUATE FACTUAL BASIS FOR CONVICTION OF ANY OF THE COUNTS TO WHICH HE PLED GUILTY.
POINT II. DEFENDANT’S CONVICTIONS MUST BE VACATED BECAUSE THE COURT DID NOT ADEQUATELY ADVISE HIM OF THE CONSEQUENCES OF ENTERING THE PLEA.
We have carefully considered all of these contentions in light of the entire record and applicable law. We are convinced point I of defendant’s first supplemental brief requires a reversal of the trial judge’s denial of defendant’s new trial motion premised upon the newly discovered evidence and point I of defendant’s second supplemental brief requires a vacation of his guilty plea to the November 21, 1992 charges and a reinstatement of the original charges. That determination renders moot points I, II, IV and V of defendant’s original brief and point II of his second supplemental brief. Point III of the original brief and point II of the first supplemental brief are without merit and require no further opinion. R. 2:11-3(e)(2). We add only as to point TV of the original brief that while we would not lightly dispose of defendant’s Bielkieuncz3 accomplice liability contention, we decline to *519consider the issue as it arose in the first trial particularly since a Bielkiewicz charge was not requested. Upon new trial, should defendant request such a charge, we assume the trial judge will fairly consider the request in light of the current decisional law. See, e.g., State v. Jackmon, 305 N.J.Super. 274, 702 A.2d 489 (App.Div.1997).
II
During the trial, the jury heard the following evidence. On November 28, 1992, at about 6:30 p.m., two men jumped out of a red car and, with one of these men shooting a 9-millimeter gun, chased two other men, who were returning to their townhouse located on Bergen Street in Newark. Clarence Black, the only adult eyewitness, was at the gate to his apartment on Bergen Street at the time. He heard the car make a loud screeching noise as it made an abrupt U-turn at a high rate of speed, and saw the two men carrying guns jump out of the ear, shooting at the victims. Black recognized one of the assailants, Ankrah, as someone he had previously seen in the neighborhood. Black ran inside his own home where he continued to observe the scene from his bedroom window on the second floor, hearing more shots coming from the house.
Among the people in the townhouse were John Smith, then twelve years old, and Quasim Clark, also then twelve years old. John testified that he lived with his grandmother at the house along with the two victims, who were his uncles, and other family members. Just before the shootings, he and his cousin Quasim were in the living room at the entrance of the house when their two uncles pushed the front door open and ran towards the back door while being chased by two men.
During his trial testimony, John said at first he hid behind a big-screen television, then he started looking at the shooters. He had grabbed Quasim to hide with him. John admitted that it was difficult to see what happened in the house because it happened so quickly and he was crouched behind the television. One of the *520shooters stayed at the door while the other ran after his uncles. At trial he said he never saw the man at the door fire any shots; according to his police statement, however, the man at the door was shooting his gun also. At trial, John identified defendant as the man at the door. During a police photo array following the murders, John had identified the photographs of defendant and codefendant Doe. At that time he identified defendant as the man who was chasing the victims, shooting at them. Yet at trial he insisted that it was Doe who was the one who chased the uncles into the kitchen. At Doe’s trial, on the other hand, he returned to the version in his police statement that had defendant as the chaser. After the men left, John ran upstairs and watched them get into the red car in which one other man was waiting. On cross-examination, John admitted that he had told the police he saw two other men in the car.
Gregory Clark, then thirteen years old and also in the house at the time, claimed that, after having hid under a bed on the second floor when he heard the gunfire, he ran to the hallway window and saw two men leave the house. In his police statement the night of the murders he told the police he was able to see one of the men clearly as he turned around and might be able to identify him. A week after the murders and after seeing a newspaper article containing the photos of Ankrah and Doe, Clark identified Ankrah as the man who turned around. He also selected the photograph of Doe but did so based upon the newspaper article. Although defendant’s photo was also in the photo array, Clark did not identify it. Yet at trial, he initially identified defendant as the man he saw, but then said it was Ankrah. His final testimony was that he had seen Ankrah leave the house, had never seen defendant and did not know why he had initially said he had seen him.
Black, who had seen the assailants exit their car and chase the victims into the house, also saw the assailants leave the house. When shown the photo array shortly after the murders, Black identified the photograph of Ankrah but not that of defendant. Black said that Ankrah was the taller of the two men who ran into *521the Clark house firing his gun. He described the taller man as 6’1”, 180 pounds with a light moustache and a beard, wearing red sweatpants and a red hooded sweatshirt under a black leather coat; the other man he described as 5’3”, 160 pounds and wearing a wool hat which partly obscured his face.'4 At trial, Black admitted that although he claimed to have recognized Ankrah from the neighborhood, in fact, he was not “absolutely sure” of his identification. He was never able to identify the second man.
On December 2, John and Quasim were shown a photo array that contained nine photographs. As we have said, John picked out photographs of defendant and Doe. Although included in the array, the photograph of Ankrah was not identified by John. The same procedure was conducted with Quasim resulting in the same two identifications. However, at trial, Quasim recanted his identification of both men. He said that he was not sure of the photographs which he selected, either at the time of trial or at the time he made the selections. When asked to survey the courtroom to pick out the shooters, he was unable to choose anybody. When asked why he had selected the photos of defendant and Doe, he said “I don’t know. I ain’t sure.”
Defendant was arrested shortly after the photo identification. At that time, he was twenty years old, 5’7” and weighed 125 pounds. Neither his height nor weight matched the descriptions of the two assailants given by Black and John.
It was only John who identified defendant during the trial as one of the gunmen. But the voluminous medical evidence adduced during the remand hearing would seem to almost overwhelmingly depict his cognitive and recollective abilities as impaired and unreliable, both at the time of the murders and during defendant’s trial. Indeed, that is apparently what the Doe jury, which heard the evidence, must have concluded in acquitting Doe.
*522That evidence includes the following. At the Doe July 5, 1995 pretrial hearing on John’s then competency to testify, Dr. Kaune, his treating psychiatrist, opined that John was not ready to testify. He explained that John “has poor reality testing when stressed” and that being compelled to testify “would cause this fragile child to further regress.” Dr. Kaune diagnosed John with impulse control disorder; disruptive behavioral disorder; attention deficit disorders with hyperactivity (ADHD), combined type; post traumatic stress disorder (PTSD), chronic type; communication disorder; mixed personality traits; and schizoid, paranoid, borderline and antisocial behavior. He was then being medicated with haldol, imipramine, buspar, thorazine, lithium citrate and cogentin.5 Dr. Kaune was hopeful that once the child was stabilized on medication and ready to leave the hospital for a residential facility, he would be able to appear in court. Five months later John did testify with the outcome an acquittal for Doe.
Defense called as an expert Dr. Frank J. Dyer, a psychologist and member of the Forensic Psychology Committee of the New Jersey Psychological Association with impressive credentials. In order to form an opinion as to the effect of John’s psychiatric and psychological conditions on his perception, recall and veracity, Dr. Dyer reviewed the numerous hospital records, psychiatric reports and psychological reports, as well as John’s testimony at defendant’s trial and at the subsequent Doe trial. The doctor also read the 1995 and 1996 reports prepared by the State’s expert, Dr. Daniel Greenfield. Dr. Dyer characterized the records as compre*523hensive and said they “contained such an extreme picture of psychopathology----” Dr. Dyer observed that John “has diagnoses from virtually every severe category of psychological condition in the DSM.” The doctor concluded that John is in “the most severe one percent” of the hundreds of children he has tested since he began as a consultant for the Department of Youth and Family Services in 1980.
John’s documented psychiatric problems had manifested themselves years before the November 28, 1992 murders and before the January 1994 trial. Dr. Dyer dated John’s first traumatic experience with psychiatric consequences from 1987, when, at the age of six, the child was deliberately burned by his mother, following which he was placed in a series of foster homes. In 1988, the Newark Child Study Team had classified him as emotionally disturbed. In April of 1991, at age 10, he began threatening to kill himself. Between May and June of 1991, he was admitted to St. Mary Hospital, Hoboken, where he was diagnosed as suffering from post-traumatic stress disorder (due to “a significant history of repeated physical abuse by [his] mother” which included the severe burning at age six), attention deficit disorder with hyperactivity (ADHD), and specific developmental disorder, and was treated for aggressive conduct disorder. While at St. Mary’s, John was medicated with mellaril, an anti-psychotic drug, and nortriptyline, a depressant. These medications were continued upon discharge. The St. Mary’s 1991 discharge summary related “cognitive distortions and poor reality testing____”
When the suicidal ideation which sent John to the hospital in April 1991 continued, John was admitted to St. Clares-Riverside Medical Center in Denville where he stayed from July 8 to September 26, 1991. From St. Clares’ he was transferred to Bergen Pines, a mental hospital, following which he spent six months at Davis House in Newark where his medication included mellaril and dexedrine, an amphetamine used to treat ADHD (PDR at 2475).
*524It was Dr. Dyer’s belief that at the time of the January 1994 trial, John was suffering from, among other disorders, chronic post-traumatic stress disorder. An individual with this disorder is psychologically overwhelmed by the traumatic event, and cannot cope with it. He experiences:
intrusive, highly emotionally charged recollections of portions of the traumatic event. There is often partial amnesia for important aspects of the traumatic event and this is part of a general pattern of numbing in an attempt to distance one’s self from anything that is reminiscent of the trauma.
“[Functioning is disrupted for a time because of the enormous emotional impact of this trauma,” and perception is affected. With post-traumatic stress disorder, there are typically gaps in memory, and “there is a tendency in individuals with extreme psychological conditions such as organic brain damage and schizoaffective disorder, both of which [John] has been diagnosed with, to fill in gaps in memory by a process that is known as confabulation.”
According to Dr. Dyer, among John’s most serious psychiatric diagnoses is “schizoaffective disorder, bipolar type[,] ... a condition that has both elements of mood disorder, either mania or depression, as well as elements of schizophrenia.” Schizoaffective disorder begins early in life; it “is a disturbance in the biochemical makeup of the brain that is considered to have genetic components as well as being aggravated by environmental factors.” “Schizophrenia ... is considered to be life long.”. Dr. Dyer testified that this severe disorder would have begun in childhood and would have been in existence well before John testified at defendant’s trial at the age of 13.
Dr. Dyer explained the nature and effect of schizoaffective disorder:
The schizophrenic component of the schizoaffective disorder indicates that the individual’s thought processes are confused. That the individual with this diagnosis has eccentric, unrealistic, infantile patterns of thinking that cloud their contact with reality. Individuals with this type of diagnosis tend to employ a process known as confabulation, to fill in gaps in memory. Confabulation refers to a tendency to fabricate some event, ... so that the total memory has some kind of logical consistency. And typically these individuals are not aware that their fabrication is *525not part of the actual memory because their thought processes are so confused and because they are so disturbed psychologically.
Asked to consider John’s changing account of the murders in his statements to police and his testimony at the two trials, of the roles of defendant and Doe in the shooting, and his sometimes inability to recall his prior versions, Dr. Dyer opined that this was consistent with confabulation.
As to John’s attention deficit disorder with hyperactivity (ADHD), first diagnosed in 1987, and its impact upon the child’s cognitive functioning, Dr. Dyer explained that a child with ADHD
has trouble in sustaining or focusing attention. The child’s inner world is disrupted by a physiological pressure for movement, that’s where the hyperactivity comes in. This inability to focus attention and the physiological pressure for motor movement tend to render a child very inefficient in perceiving things.
ADHD “is considered to be a biochemical condition in which there is an actual deficiency in the capacity of the brain to process information to control impulses for motor movement____” In the absence of any indication that John underwent effective treatment for this disorder first identified in 1987, and because he was still being diagnosed with it in 1995, Dr. Dyer was able to conclude “to a reasonable degree of psychological certainty” that John had the condition continuously from 1992 through 1994, encompassing the time of the murders and defendant’s trial. It was also Dr. Dyer’s opinion that the schizoaffective disorder, the chronic post-traumatic stress disorder, and the personality disorders enumerated in Dr. Kaune’s 1995 report “would have predated the 1995 report and would have existed for quite some time earlier than that, most of which conditions had their etiology in childhood.”
Dr. Kaune’s 1995 report also diagnosed John with schizoid, paranoid, borderline, and antisocial personality traits. As to such traits, Dr. Dyer testified that these “are considered to be longstanding maladoptive patterns of adjustment” and thus would have been present long before 1995. And he said, “[sjomebody who ... has paranoid personality traits has a tendency to treat false beliefs or delusions as though they were reality.” “Individuals with schizoid traits tend to have some impairment of cognitive process*526ing. Their processes are somewhat fuzzy.” As a consequence, their recall and perception are mildly affected. Moreover, someone with antisocial personality traits “tend[s] to be manipulative and deceitful, ... may repeatedly lie, ... and [be] extremely irresponsible, ... [and] may not fully appreciate the duty of a witness to tell the truth.”
Dr. Dyer admitted that while psychologists can explain the effects of particular illnesses on an individual’s thought processes and mental functions, in all but the most extreme cases a psychologist would not be able to predict whether a particular individual suffering from specific mental illnesses would be affected by those illnesses on a given day and, thus, he could not say with certainty what John’s precise mental state was at the time of the murders on November 28, 1992, nor the reliability of his testimony at the time of defendant’s trial in January 1994. But he emphasized, with medical certainty, that substantial remission “is almost never observed in someone who suffers from multi-psychiatric problems,” and further, that at the time of trial in January 1994, John was actively symptomatic of the mental illnesses identified between the time of his first psychiatric diagnosis in 1987 and in April 1992, the date of his last recorded treatment prior to his trial testimony. Dr. Dyer also found strong evidence that the “core-aspects” of the additional psychoses not diagnosed until after trial “were present in some form in the intervening period,” and would have affected John’s thought processes. The combination of “all of these various psychological disorders would be expected, to a reasonable degree of psychological certainty, to have a negative impact on the individual’s capacity to tell the truth, to recall things accurately and to perceive things accurately.”
The State’s psychiatrist, Daniel Greenfield, agreed with Dr. Kaune’s numerous diagnoses made in 1995, acknowledged that John’s various mental illnesses “certainly could” affect the truthfulness of his testimony and specifically concurred with Dr. Dyer that John’s “hallucinations, delusions, organic brain syndrome and things like that” might well affect John’s perceptions and memory. *527He concurred that the attention deficit disorder with hyperactivity “goes back to [a] very young” age and affects the ability “[t]o focus, concentrate, [and] interpret events” and that post-traumatic stress disorder if actively symptomatic, would affect perception, memory and veracity. Dr. Greenfield also agreed with Dr. Dyer that no one could say what John’s precise mental state was at the time of defendant’s trial in 1994. Dr. Greenfield added that “[i]f left untreated, the natural history of these things is that, with the exception of certain schizophrenic conditions, they tend to get worse ... if left untreated.” Dr. Greenfield also agreed with the court that, other than an indication of a referral to U.M.D.N.J. as an out-patient upon John’s release from Davis House on April 24, 1992, there was no record of any treatment from the time of John’s release from Davis House 20 months before defendant’s trial and his hospitalization in October 1994, ten months after defendant’s trial.6 However, at the time he saw John in September 1995, Dr. Greenfield found him in remission and asymptomatic.
After reviewing John’s statements to the police, his testimony at the two trials and Dr. Dyer’s report, the only significant inconsistency Dr. Greenfield could detect was the change in John’s account of which intruder ran after the victims, shooting at them. He opined that that kind of memory lapse happens to everybody. But on cross-examination, Dr. Greenfield conceded that John’s accounts contained other substantial contradictions and opined that there was an equal possibility, that it was “50/50,” whether John’s numerous contradictory stories were attributable to his psychiatric conditions or to normal memory lapses. Dr. Greenfield also testified that he did not attribute any psychiatric signifi*528canee to the inconsistencies because “number one, memories do get distorted without any ... particular psychiatric significance ... and, number two, that there was no reason to suspect that it would be something that couldn't be examined or cross-examined or subjected to the scrutiny of the trial process[.]” He also did not find any psychiatric diagnoses made on or prior to defendant’s trial in January 1994, and thus was not certain as to John’s condition at that time, nor did he find any evidence of confabulation regarding the shooting incident. Dr. Greenfield added that “it would be reasonable to assume that his mental state at the time ... in January of 1994 would not have been so bad, so egregious that he wouldn’t have been able to remember things and recount things accurately.” Still, he admitted that a person like John, suffering from a schizoaffective disorder, attention deficit disorder and post traumatic stress disorder, if symptomatic, may have difficulty with the functions essential to memory and making an accurate identification.
Needless to sáy, had a jury been privy to these expert opinions, we are certain it would be Dyer’s analysis that would be accepted. Indeed, in his written opinion, the trial judge on remand concluded that “[tjhere is no question ... that John was and is suffering from a series of mental illnesses,” and found that the evidence of the illnesses was not discovered until after trial and was not reasonably discoverable beforehand. The judge acknowledged that the defense and prosecution experts were in substantial agreement on the possible effects of John’s illnesses on his perception, memory and veracity. But, in reliance on the state’s expert’s observation that individuals with psychotic disorders are not always symptomatic, the judge determined that “it cannot be conclusively stated that he was incapable of understanding his duty to tell the truth” and consequently found John to have been competent to testify at defendant’s trial. Although defendant challenges this conclusion on appeal, we are satisfied it is supported by the remand record.
*529As to the new trial motion, citing R. 3:20-1 and State v. Carter, 85 N.J. 300, 426 A.2d 501 (1981), the judge identified the three-prong new trial test: the evidence must be material, it must actually be newly discovered, and it must be of the sort likely to change the verdict. There was never any dispute that the second prong was satisfied; however, the judge held that the first and third prongs were not met. Deeming the evidence of John’s mental impairments to be “merely” impeaching, and of a nature to “only” affect John’s credibility, the judge concluded that it was not material and thought it possible, but not probable, that the newly discovered evidence would have changed the verdict. As to John’s testimony at the Doe trial, where, as in his police statement but contrary to his sworn testimony in defendant’s trial, he identified defendant as the shooter running into the kitchen, the judge concluded that such “reeant[ed]” testimony was not material in that it would be used “merely” to impeach. He also noted that John had been fully cross-examined on the inconsistencies between his police statement and his testimony during defendant’s trial. Finally, he considered the combination of John’s mental disorders and the recanting testimony and still concluded that the necessary measure of materiality had not been met because the evidence related only to issues of credibility and would probably not have changed the jury’s verdict.
Ill
It goes without saying that a motion for new trial is within the sound discretion of the trial judge which we do not lightly disturb. State v. Conway, 193 N.J.Super. 133, 172, 472 A.2d 588 (App.Div.), certif. denied, 97 N.J. 650, 483 A.2d 174 (1984). And it is also clear that to be entitled to a new trial based upon newly discovered evidence, a defendant must establish not only that the evidence was discovered after trial and not reasonably discoverable before, but that the evidence is material and probably would affect the verdict. See, e.g., State v. Carter, supra, 85 N.J. at 314, 426 A.2d 501. We also recognize the sometimes, perhaps too casually, asserted theme that newly discovered evi*530dence must not be “merely cumulative or impeaching.” Ibid.; State v. Robinson, 253 N.J.Super. 346, 366, 601 A.2d 1162 (App. Div.1992), certif. denied, 130 N.J. 6, 611 A.2d 646 (1992). And we have no doubt that were the only newly discovered evidence that of John’s recantation in the Doe trial, such evidence might well fall within this rubric.
But the evidence of John’s extensive psychiatric disorders and their impact upon his ability to accurately perceive the rapidly unfolding events of November 28,1992, and to accurately, without “confabulation,” recount those events, is much more than “merely” impeaching in the sense of some inconsistencies. That evidence goes directly to the heart of the State’s case — the identification by John of defendant as one of the two shooters. That identification was the linchpin upon which the verdict rested for, not only was it only John who identified defendant, but there was no other evidence tying him to the murders. As the trial judge acknowledged during the January 1994 trial, “I don’t think there’s any question that, one of the crucial issues here is the question of identification,” and instructed the jury to consider each witness’s ability to perceive events and make accurate identifications, emphasizing “[i]t is particularly appropriate that you consider the capacity or the ability of the witnesses to make observations or perceptions____”
The State asserts, however, that “[mjateriality concerns the relation between the propositions for which the evidence is offered and the issues in the case.” State v. Hutchins, 241 N.J.Super. 353, 359, 575 A.2d 35 (App.Div.1990); State v. Allison, 208 N.J.Super. 9,17, 504 A.2d 1184 (App.Div.), certif. denied, 102 N.J. 370, 508 A.2d 235 (1985). Since, argues the State, John’s “mental health records ... clearly did not have a direct bearing on whether the murders actually took place or not,” it claims they are not material. But, as we have pointed out, the issue at trial was not whether the victims were murdered. Clearly they were. The crucial issue was whether defendant was one of the perpetrators. *531The entire answer to that issue was John’s identification of defendant.
Thus, John’s mental capacity and ability to make accurate perceptions and to accurately and reliably recall and recount his perceptions was the focal point of the trial. “Material facts are those that have some bearing on the claims being advanced.” Korostynski v. Division of Gaming Enforcement, 266 N.J.Super. 549, 555, 630 A.2d 342 (App.Div.1993). See State v. Robinson, supra, 253 N.J.Super. at 366, 601 A.2d 1162 (an exculpatory statement by a codefendant is material to the issue of defendant’s guilt). Most certainly John’s mental history and psychiatric disorders relate not only “with some bearing” but directly to the focal issue of the trial and must be considered material. See State v. Bunyan, 299 N.J.Super 467, 473, 691 A.2d 417 (App.Div.), certif. granted, 151 N.J. 74, 697 A.2d 546 (1997) (newly discovered eyewitness “patently” material where identification was “the only real issue at trial.”).
Indeed, had the evidence been discovered prior to defendant’s trial in January 1994, there could be no question but that he would have had an absolute right to present such evidence and that a refusal to so permit would have required a reversal. State v. Franklin, 52 N.J. 386, 399, 245 A.2d 356 (1968); State v. Vigliano, 50 N.J. 51, 58-59, 232 A.2d 129 (1967); State v. Wormley, 305 N.J.Super 57, 66-67, 701 A.2d 944 (App.Div.1997); State v. Johnson, 216 N.J.Super 588, 603, 524 A.2d 826 (App.Div.1987). Cf. State v. Butler, 27 N.J. 560, 603-05, 143 A.2d 530 (1958). There is no reason to consider such evidence as any less relevant and material in the context of a newly discovered evidence motion simply because it may be cast in terms of impeachment evidence.
To be sure, “merely” impeaching or cumulative evidence obtained after trial will not suffice for a new trial. See, e.g., State v. Carter, supra, 85 N.J. at 314, 426 A.2d 501. This is so because evidence of that quality would not ordinarily make a difference in the jury’s verdict. State v. Carter, 91 N.J. 86, 114, 449 A.2d 1280 (1982) (“[e]vidence that is merely cumulative does not create a *532reasonable possibility that the verdict would have been affected.”). State v. Coburn, 221 N.J.Super. 586, 600-01, 535 A.2d 531 (App. Div.1987), certif. denied, 110 N.J. 300, 540 A.2d 1281 (1988), is illustrative. There, following defendant’s conviction of murder and other related charges, defendant had moved for a new trial based upon a newly discovered toxicological report. At trial, the defense essentially was that the killing was an accident caused by the victim who, defendant claimed, was “a wild woman.” During trial, pointing to fresh needle marks discovered on the victim’s wrist during her autopsy, defendant argued that she was under the influence of drugs at the time, causing her wild, threatening behavior. The toxicological report received after trial confirmed that the victim was under the influence of drugs at the time she died. In affirming the trial judge’s denial of his motion for a new trial, we acknowledged the “potentially material” nature of the report, but, noted that the trial evidence of fresh needle marks on the victim “indicated recent injections of drugs, probably heroin.” The defendant, thus, was able to make that claim based upon evidence then before the jury. We concluded, further, that the verdict was “most likely” based on the jury’s rejection of the defendant’s theory of accident because of its implausibility in light of the undisputed evidence that the gun had to have been cocked to fire and required four and one-half pounds of pressure to pull the trigger. We also noted that defendant’s trial defense was undermined by the evidence of his admissions to his ex-wife and the police shortly after the murder. In light of these considerations, we agreed with the trial judge’s assessment that the evidence was cumulative as to defendant’s credibility and that is was “highly improbable” that it would have affected the verdict. But we did not do so simply because the evidence might have been “merely cumulative” and thus not material. The critical factor was that it would not have affected the verdict.
On the other hand, in State v. Gookins, 135 N.J. 42, 45-48, 637 A.2d 1255 (1994), the Supreme Court ordered new trials for the convicted defendants based, in part, on newly discovered “other crimes,” that is evidence that the State’s main witness, a state *533trooper, falsified breathalyzer tests. The “other crimes” evidence was impeachment evidence. Nonetheless, the court recognized that such “impeachment” evidence related to the heart of the case, the credibility of the State’s key witness. That evidence was considered material, requiring a new trial. And see Keser v. State, 737 P.2d 756, 760 (Wyo.1987) (asserting “[t]here is a difference between evidence which merely goes to credibility or impeaches a witness by calling his credibility into question and [impeaching] evidence which is offered to show an ‘eyewitness’ to a crime gave false identification testimony,” the latter being material for the purposes of a new trial motion); State v. Stone, 869 S.W.2d 785, 789 (Mo.Ct.App.1994) (finding “[t]he newly discovered fingerprint and handwriting evidence goes directly to the principal issue in this case: the identity of the person who forged check number 1061. It is not, therefore, merely impeachment evidence and can justify a new trial”).
We consider the analysis of materiality within the context of Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), and its progeny as analogous. Under Brady, a defendant’s due process rights are violated if the prosecution withholds evidence which is material to either the defendant’s guilt or punishment. 373 U.S. at 87, 83 S.Ct. at 1196-97, 10 L.Ed.2d at 218. “The jury’s estimate of the truthfulness and reliability of a given witness may well be determinative of guilt or innocence, and it is upon such subtle factors as ... [credibility] ... that a defendant’s life or liberty may depend.” State v. Engel, 249 N.J.Super. 336, 399, 592 A.2d 572 (App.Div.), certif. denied, 130 N.J. 393, 614 A.2d 616 (1991) (quoting Napue v. Illinois, 360 U.S. 264, 269, 79 S.Ct. 1173, 1177, 3 L.Ed.2d 1217, 1221 (1959)). Thus, “[e]videnee impeaching the testimony of a government witness falls within the Brady rule when the reliability of the witness may be determinative of a criminal defendant’s guilt or innocence.” State v. Carter, supra, 91 N.J. at 111, 449 A.2d 1280. And see State v. Knight, 145 N.J. 233, 245-48, 678 A.2d 642 (1996); State v. Landano, 271 N.J.Super. 1, 637 A.2d 1270 (App.Div.), certif. denied, 137 N.J. 164, 644 A.2d 612 (1994). It is, therefore, clear that under the *534Brady analysis withheld evidence that is material may be that which impeaches a witness where the issue of the witness’ reliability and credibility is crucial. See State v. Knight, supra, 145 N.J. at 247-48, 678 A.2d 642.
We recognize that a new trial analysis premised upon a Brady violation and a new trial analysis premised upon newly discovered evidence, are different and that materiality for the latter has been characterized as “more stringent.” State v. Carter, supra, 85 N.J. at 314, 426 A.2d 501. In Carter, the court commented that “[wjhereas the test of materiality for the granting of a new trial under a Brady analysis is simply whether the suppressed evidence might have affected the outcome of the trial, ... the test to be satisfied under a newly discovered evidence approach is more stringent.” Ibid. (Emphasis added).
To begin with, defendant must establish that the evidence must not only have not been discovered until after trial, but could not have been with due diligence. So that, in that respect, there is clearly a more stringent requirement. But we are not sure that the standard of materiality is any different. See State v. Carter, supra, 91 N.J. at 121, 449 A.2d 1280 (“[f|or the reasons outlined in our discussion of the Brady violation, we hold that [as to the motion for new trial based on newly discovered evidence] the evidence ... is neither material nor of the sort that would lead to a change in the jury’s verdict.”).
We think to understand the court’s comment in Carter, supra, 85 N.J. at 314, 426 A.2d 501, one must recognize the context in which it was made. That context was a prosecutorial failure to provide defendant with specifically requested exculpatory evidence. In that setting, the court’s consideration of materiality under Brady was perceived to be dependent upon whether the evidence “might” have affected the outcome of the verdict, as opposed to whether it “probably” would have affected the outcome. Id. at 312, 426 A.2d 501. And see State v. Marshall, 123 N.J. 1,199-200, 586 A.2d 85 (1991), cert. denied, 507 U.S. 929,113 S.Ct. 1306, 122 L.Ed.2d 694 (1993). But see State v. Knight, *535supra, 145 N.J. at 247, 678 A.2d 642 (criticizing Marshall’s reference to two separate standards in a Brady analysis dependent upon whether the suppressed evidence was or was not specifically requested.).
Of course, a possibility of impact upon a verdict is not the type of impact required under a newly discovered evidence motion. But where a Brady analysis has been conducted within the setting of withheld exculpatory evidence not specifically requested, materiality is coupled with the requirement of a reasonable probability that the verdict would have been different had the evidence been disclosed. See, e.g., State v. Knight, supra, 145 N.J. at 247, 678 A.2d 642. Yet, even in this context, evidence that is impeaching may be considered material where it probably would have changed the verdict. Id. at 247-48, 678 A.2d 642.
The critical issue, then, we are convinced, is whether the additional evidence probably would have affected the outcome, regardless of whether it is characterized as impeachment evidence. We simply cannot fathom the trial judge’s conclusion here that it would not. John’s trial testimony already was replete with inconsistencies. Its reliability was, to put it mildly, suspect. The jury must have struggled to reach its verdict. But had the jury heard the nature and extent of his mental disorders, the substantial medication he was subjected to, and the probable effects of such disorders upon his cognitive abilities, we are absolutely certain that what may have appeared to be only troublesome questions as to the reliability of John’s observations and recall would have become insurmountable hurdles.
In this respect, the jury would have heard that his substantial psychiatric disorders were such that he “has poor reality testing when stressed,” that his chronic post-traumatic stress disorder triggered by the murders affects his perception of the traumatic event and causes memory gaps, that his schizophrenia causes “confused ... eccentric, unrealistic, infantile patterns of thinking that cloud ... contact with reality” and cause “confabulation, to fill in gaps in memory” without being “aware ... [of] their *536fabrication ...,” that his ADHD creates “an actual deficiency in the capacity of the brain to process information ...” and that his antisocial, paranoid and schizoid personality traits tend to cause a person “to be manipulative and deceitful, ... [to] repeatedly lie, ... and [be] extremely irresponsible, ... [and] ... not fully appreciate the duty of a witness to tell the truth,” “to treat false beliefs or delusions as though they were reality” and to “have some impairment of cognitive processing affecting recall and perception.” Had the jury heard this information, we are almost certain defendant, like his two codefendants, would have been acquitted. At the least, that result would have been most probable. A new trial, therefore, should have been granted.
IV
Finally, we address defendant’s contentions in his second supplemental brief that his guilty plea to the November 21, 1992 charges should be vacated because he was not advised of the consequences in the event an appeal of the convictions for the November 28,1992 murders was successful and because there was an inadequate factual basis for the November 21,1992 convictions. We agree as to the latter and need not, therefore, consider the former.
It is well established that “if an appellate court subsequently determines that a plea has been accepted without an adequate factual basis, the plea, the judgment of conviction, and the sentence must be vacated, the dismissed charges reinstated, and defendant allowed to re-plead or to proceed to trial.” State v. Barboza, 115 N.J. 415, 420, 558 A.2d 1303 (1989). In order to accept a guilty plea in non-capital cases, a judge must be convinced that the defendant, in his own words and in light of the surrounding circumstances, has committed the acts which constitute the crime. Id. at 422, 558 A.2d 1303. “Even if a defendant wished to plead guilty to a crime he or she did not commit, he or she may not do so. No court may accept such a plea.” State v. Smullen, 118 N.J. 408, 415, 571 A.2d 1305 (1990).
*537The following factual basis was provided by defendant in connection with the November 21,1992 charges:
THE COURT: All right, now, Mr. Henries, now, Count 1 and Count 3, Count 1 says that, on November, the 21st, of 1992, in Newark, that you — well, it’s being downgraded — basically you pointed a weapon, a handgun, at James Johnson, and likewise, in Count 3 of this indictment, it also states, on the same date, that you did the same thing with regard to Alonzo Heath.
Can you tell me, in your own words, what happened, sir? Speak up.
THE DEFENDANT: The day of the incident, November 21st, I would say I had a handgun and pointed at the defendants, Alonzo Heath and James Johnson. THE COURT: For the record, not the defendants, the victims?
THE DEFENDANT: Victim.
THE COURT: This handgun you had, what kind of handgun?
THE DEFENDANT: I don’t remember.
THE COURT: Was it a real gun?
THE DEFENDANT: Yes, your Honor.
THE COURT: And can we assume that you did hot have the proper permits to have this gun?
THE DEFENDANT: Yes, your Honor.
THE COURT: [Prosecutor] are there any questions you wish to ask him concerning the co-defendant?
[PROSECUTOR ]: Yes, Judge.
Mr. Henries, at the time you pointed the weapon at Mr. Heath and Mr. Johnson, were you alone, or was somebody with you?
THE COURT: What is your answer? I’m sorry.
THE DEFENDANT: Yes.
THE COURT: Yes, you were alone, or yes, someone was with you?
THE DEFENDANT: People was with me.
[PROSECUTOR ]: All right, without going into names, how many people were with you?
THE DEFENDANT: I don’t recall.
[PROSECUTOR ]: Was it more than one or more than two?
THE DEFENDANT: I can’t recall either; probably two or three.
THE COURT: The answer was two or three.
[PROSECUTOR ]: And were those people armed also?
THE DEFENDANT: Yes.
[PROSECUTOR ] And was the weapon that you had with you a 9 millimeter type of weapon?
THE DEFENDANT: I don’t recall.
[PROSECUTOR ]: Was it an automatic weapon?
*538THE DEFENDANT: I don’t recall.
[.PROSECUTOR ]: Was it a revolver?
THE DEFENDANT: I don’t recall.
This testimony was intended to provide a factual basis for a conviction of N.J.SA 2C:12-lb(4), a fourth-degree aggravated assault where the actor “[k]nowingly under circumstances manifesting an extreme indifference to the value of human life points a firearm, as defined in section 2C:39-lf., at or in the direction of another, whether or not the actor believes it to be loaded[ ],” and for a conviction of third degree possession of a handgun without a permit pursuant to N.J.SA 2C:39-5b. As to the former, the Plea Form and the Request to Recommend Disposition both state that defendant would plead guilty to N.J.S.A 2C:12-lb(3), which is a fourth-degree aggravated assault requiring that the actor “[r]ecklessly causes bodily injury to another with a deadly weapon.” When the prosecutor made his presentation to the court at the plea hearing, he stated that the State would move to amend the second-degree aggravated assault counts to fourth-degree counts under N.J.SA 2C:12-lb(3). Later at the hearing, when the judge first addressed defendant, he said to him that the assault charges were being downgraded to “pointing” (N.J.SA 2C:12-lb(4)). The judgment of conviction reflects the “final charges” as two counts of aggravated assault under “NJS 2C:12-lb(3)(4),” seemingly encompassing both sections.
Clearly, however, there was not a factual predicate for a conviction of N.J.SA 2C:12-lb(3). Neither do we think the necessary predicate existed for a N.J.S.A 2C:12-lb(4) or 2C:39-5b offense. The most defendant said as to being in possession of a gun was “I would say” he had a handgun. As to the attendant circumstances, “he would say” he pointed it at the victims and that there were other people with him who had guns, though he could not recall how many or who. But he did not recall the type of gun he had, or whether it was an automatic or revolver. Other than the pointing of a gun at the victims, nothing more is known. For instance, nothing is known about the type of weapon the handgun was or whether it was operable and loaded. As to the latter *539factor, while an offense under N.J.S.A. 2C:12-lb(4) does not require belief on the part of a defendant that the weapon is loaded, certainly that would be an important circumstance manifesting extreme indifference to human life, particularly where there is a dearth of any other such circumstances.
Moreover, although the statement that he “would say” he had a gun and that he had no permit might support a conviction of the third-degree offense, we think the overall circumstances warrant a reinstatement of the underlying indictment. As defendant points out:
In determining the adequacy and reliability of the factual basis offered by defendant in support of these charges, the circumstances under which the plea was entered must be recalled. Before defendant agreed to the instant plea on Counts 1 through 8, he had twice rejected very favorable offers from the state to plead to Counts 9 through 17, which alleged more serious offenses, including an offer to dismiss both counts of murder in exchange for a plea to two counts of assault with a maximum term of 10 years with no more than one-third- without parole. Defendant felt strongly that if a jury heard the evidence, he would be acquitted of the offenses charged in Counts 9 through 17, and, in fact, both his codefendants were acquitted. Mr. Henries, however, was convicted.
Mr. Henries did not agree to plead to Counts 1 through 8 until after he was convicted of Counts 9 through 17 and sentenced to spend the rest of his life in prison for acts to which he maintained his innocence. At the point, defendant accepted a plea to a minimal concurrent term of five years, which did not add any time to the two consecutive life terms with an aggregate 60 years without parole which had been imposed following conviction on Counts 9 through 17.
Under these circumstances, though it seems defendant’s plea may, at the time, have been voluntary and knowing, it certainly was influenced by his murder convictions and life sentences. Had those convictions not occurred there is every reason to believe he would not have willingly and knowingly pled guilty. We are convinced, therefore, that defendant should be allowed to withdraw his plea and defend the original charges.
V
Reversed and remanded for further proceedings consistent with this opinion.