728 A.2d 231 (1999)
321 N.J. Super. 5
STATE of New Jersey, Plaintiff-Respondent,
v.
Ralph A. RASO, Defendant-Appellant.
Superior Court of New Jersey, Appellate Division.
Submitted February 24, 1999.
Decided March 17, 1999.
Ivelisse Torres, Public Defender, attorney for the defendant-appellant (Claire Drugach, Assistant Deputy Public Defender, of counsel and on the brief).
Peter Verniero, Attorney General of New Jersey, attorney for plaintiff-respondent (Jordana Jakubovic, Deputy Attorney General, of counsel and on the brief).
Before Judges CONLEY, KIMMELMAN and LEFELT.
The opinion of the court was delivered by
CONLEY, J.A.D.
Following a jury trial, defendant was convicted of knowing or purposeful murder, N.J.S.A. 2C:11-3a(1), (2) (count one), and possession of a weapon for an unlawful purpose, N.J.S.A. 2C:39-4d (count two). Count two was merged into count one and a life sentence with a thirty-year disqualifier was imposed, along with the necessary fines and penalties. On appeal defendant raises a number of issues including the contention that "introduction into evidence of unsubstantiated and unreliable expert testimony by Dr. Alvin Krass was prejudicial error and requires reversal of defendant's conviction." We agree and conclude that this error alone requires a reversal. We do not, therefore, consider defendant's other contentions.
I.
The convictions arose from the killing of Joanne Turek, with whom defendant had had a long term relationship. Defendant, who has a history of psychiatric problems, admitted to the killing, but asserted diminished capacity as a defense. Here are the critical facts.
Defendant lived with the victim for approximately ten years prior to September 8, 1996 when he killed her. Two weeks before that date, the victim had asked him to leave. Defendant was upset by this rejection and made several attempts to reconcile with her. There was also an apparent suicide attempt which the State asserted was feigned. During the two weeks before September 8, defendant's behavior deteriorated as he became increasingly agitated and angry about the ending of his relationship with the victim. By September 7, 1997, the day before the murder, defendant was described as "distraught" and "in a daze." There was evidence to suggest that at that time he was taking substantial amounts of Prozac.
On September 8, 1997, defendant went to the victim's residence at or around 8:30 a.m. for the purpose of retrieving some of his belongings. In a statement given shortly after the killing, defendant related that after unsuccessfully pleading with her to take him back, he "just went bananas." He reached under a nearby small table and grabbed a knife that had been stored there, repeatedly stabbed her, inflicting a total of eighty-one stab wounds to her back, torso and arms. Still wearing his bloody clothes, he went to the police station and confessed, saying "I went berserk" and "I lost it. I did it." A taped formal interrogation then occurred.
As we have indicated, defendant's defense at trial was diminished capacity. In support of this, he presented two expert witnesses. The first expert was Dr. Azariah Eshkenazi, a forensic psychologist. It was his opinion, within a reasonable degree of medical certainty, that prior to the killing, defendant's mental state was "one of depression, severe with agitation, in addition to obsessive compulsive disorder." That conclusion was based on the doctor's interview with defendant and his review of, among other materials, defendant's taped confession, reports from his former psychologist and discussions with his former wife. The doctor also discussed the effects of the medication Prozac, large doses of which had been prescribed to defendant for his depression. Dr. Eshkenazi opined that Prozac, which acts as a stimulant and can cause agitation and psychosis,[1] had the effect of increasing defendant's already extreme emotional disturbance. The doctor's ultimate conclusion was that, based on his review of the case and attendant diagnosis, defendant was in a highly distressed state of mind at the time of the killing and his ability to act knowingly and purposely was seriously affected.
Defendant's next expert, Dr. Edward Dougherty, was an expert in the field of forensic neuropsychology. Dr. Dougherty saw defendant a total of eleven hours during three sessions on September 16, 1996, November 25, 1996 and in January 1997. The doctor recounted as significant facts that defendant, then fifty-one years old, was a twelfth grade dropout, his mother died when he was eleven and he had a failed marriage. At the time of the first interview, a little more than two weeks after the killing, defendant was agitated, disheveled, not focused and was confused as to which medications he had been taking, at first indicating Zanax and then both Zanax and Prozac. During the first interview, defendant recalled going to see the victim and having a brief conversation about wanting to come back. He recalled going into the bathroom for a while and then recalled getting into his van, driving down the street "seeing blood, going back, thinking he had a dream, going back to the apartment of the victim, going in and seeing he killedor that she was dead and covered with blood ... he kissed her ... leftclosed the bedroom door, saw the victim's son, said he was going to get bagels and drove to the police station...." During the two other interviews "he started putting the pieces of the puzzle together." During the second interview, defendant recalled picking up a knife, but could not then recall what happened. By the time of the third interview, defendant knew that he had stabbed the victim, but did not know how many times. Dr. Dougherty then explained:
The critical thing [at that point] for me to do is try to say is he playing games with me, just selectively recalling certain events to see what my reaction would be, and I ruled that out because there was just too much specificity in certain parts of the events. And my clinical opinion, dealing with people that have memory problems, that's what I had specialized in, we have a thing called confabulation where basically youlike if you were really drunk, you got home, you weren't sure how you got there, you assumed you got in your car, your car is in the driveway, sleeping in your own bed. So many times you have periods you don't remember exactly what you did, you said I got in the car and drove home and got into bed. You tell yourselves lies to confabulate what had happened. I tried to figure out what is going on with this? What is this personality? Who is Ralph Raso? I had to do a number of psychological tests to try to pin down who the aspect of this person was.
The doctor first performed two intelligence tests and found a "big" discrepancy between the verbal and the "matrices" or abstract portions of the first. He observed "there's a first red flag, there's a discrepancy between verbal and intelligence. There could be something else going on, an initial sign of, perhaps, a neurological problem." He performed another, similar test, and still found discrepancies "so [the doctor was] left with this dilemma of why, with the history of a person that at times worked for himself, had such discrepancies and scores and [he] had to do some further searching."
An important test that he performed was the Bender Gestalt Test, first developed in 1938. As described by the doctor, "[i]t's a series of drawings. Basically, there's been a scoring technique developed over the last ten years based upon research, the basis of the testing and screening for organic brain damage." The test meets American Psychological Testing standards.
As to defendant's drawings, the doctor said:
By 11 years of age, you should be able to complete them all without errors. I found that Mr. Raso could not complete the drawings as well as he should if everything was intact. He received approximately six errors which is another sign of organic brain dysfunction. Doesn't mean he's totally out of it but something is not functioning totally correctly in terms of his neurological functioning.
In explaining this to the jury, the doctor demonstrated from the Bender Gestalt Screening for Brain Dysfunction literature the drawings that a neurologically normal person should be able to do by age 11 and compared defendant's drawings to the standard drawings. He related:
The brain behavior relationship is how you perceive things, how you hold them in your mind. These are presented on cards, no time limit. He can see the cards the entire time. We also have a history of Mr. Raso being rather compulsive and being rather obsessive. I expected him to be a very careful drawing. These are less actual drawings. I'll show you what the specifics are. The problems with his drawings, if you'll note, these are dots, not circles, not loops. This was an angulation problem, angled this way, not angled. He repeatedly used loops and circles instead of dots. After he was introduced to circles on the second drawing, that's a problem called perseveration. Perseveration, your stimulus bound, you cue into something you cannot get away from it.
With young children, they start drawing a circle and drawing a circle with adults, it's morethey can repeat the same thing when the stimulus is changed from circle to a dot. The closure problem is these are suppose to touch. This is suppose to be discreet, angle that is not there. This overlapping is not ... There are nine drawings. Total placement is not that critical, however, this is all in a very disorganized manner. Someone very obsessive would put them in order, number them many times. That wasn't there. That's just a clinical assessment of his behavior at the time. This drawing basically should be a diamond and touching, very important they touch....
It's showing lack of closure, showing he didn't perceive that's what's suppose to be done. The most significant ones are the perseveration, repeating during the loops and circles rather than the dots. This is disconnect angle, no angle here, called angulation. You could see the Gestalt, the whole figure. When you have different types of minimal brain damage, you don't see the whole picture, you see parts. That's basically what this is measuring.
After performing a few more tests, the doctor concluded that they showed evidence of neurological damage and evidence of a serious personality problem and further concluded that defendant suffered from organic brain impairment and a borderline personality disorder as defined in the American Psychiatric Association diagnostic manual. In response to a question about the significance of defendant's conduct immediately after the killing in going to the police station in his blood covered clothes, the doctor observed:
Two things: Number one, he went to the police directly. He was in this taped interview. He was there. I took into consideration how he talked. I took into consideration the facts of the case as presented through discovery materials, when he came to police station he was covered in blood.
Q. What significance does that have?
A. Quite a bit out of the context as we saw the history of Mr. Raso, that basically he was a very meticulous man, very obsessive. He organized his clothing in certain ways, very concerned about looking good. Totally out of character. He was described by the couple, they thought it was paint on him, red specks throughout what he was wearing, all over his clothing and on his ... I think they said on his body in the police report. So he was covered with blood, and that's quite out of character for somebody that has obsessive and compulsive activities about their cleanliness.
Q. How would you expect a compulsive obsessive person to act?
A. When their faculties are intact and modus operandi, I would expect them to change their clothes, clean themselves up. They're always concerned about how they present themselves. That's one of Ralph's things. We know that he spent a tremendous amount of time to maintain the perfect tan and look just right, this was totally out of character.
Dr. Dougherty ultimately concluded that at the time of the killing, defendant's mental state had deteriorated to a temporary psychotic state and that he had "totally broke[n] down."
The State presented three experts in response. Two of them, Dr. Kenneth Weiss and Dr. John R. Motley, reviewed the medical evidence and examined defendant, reaching the conclusion that at the time of the killing defendant was acting purposefully and knowingly. Doctor Motley, however, agreed that defendant suffered from an obsessive compulsive disorder and a borderline personality disorder.
The third doctor, Dr. Krass, specifically addressed Dr. Dougherty's test evaluations, disagreeing with some of the conclusions. In this respect, the doctor testified:
Q. Now, have you evaluated the testing that was provided to you by Dr. Dougherty?
A. Yes. I looked through his findings and then tried to make my own judgment.
Q. Did you review the Bender Gestalt drawings?
A. Yes.
Q. Of Mr. Raso?
Q. Did you review the figure that he drew?
Q. Anything else that he noted or responded in either questions or in drawings?
A. Yes. I looked through all of the material.
Q. Do you have an opinion based on that as to whether there is any neurological impairment of any kind as reflected in those particular tests?
A. In my opinion Mr. Raso's Bender Gestalt drawings are normal. They are not in my opinion in any way indicative of perceptual impairment similar to that obtained from persons known to be brain-damaged.
The doctor did not limit his opinion to a comparison of defendant's drawings against the accepted Bender Gestalt standards. In response to the prosecutor's question "[c]an you expand on that," the doctor continued:
A. Yes. I think everybody must have seen these drawings that Raso gave. I brought the results of somebody that I saw who is known to be mildly brain-damaged who is administered this test to show the difference.
[Defense counsel]: Judge, I object to that. We have no idea. This could be from a person with total neurological impairment. For him to make that
THE COURT: He said "mildly neurological impaired," and I will allow counsel to cross-examine him in that regard. He's attempting to compare before the jury what was drawn by Mr. Raso from which Dr. Dougherty drew certain conclusions to others which he has. I'm going to allow it.
[Defense counsel] Judge, just for the record, the prosecutor didn't furnish any of this to me.
A. This is the drawing of a young man who attended school, could read, had problems in perception and was known as a little child to have suffered high fevers with resulting neurological impairment. As you can see, the drawings have certain characteristic changes in their output compared to the designs themselves and compared to Mr. Raso's performance.
[Prosecutor] Can we mark this [drawings of unknown individual] for identification?
THE COURT: S-61.
(Document described above marked State's Exhibit S-61 for identification.)
Q. Would you be able to take that and just walk in front of the jury, so that they can get a little clearer picture of what it is that you're talking about?
A. Let me show you.
Q. Just slowly walk a section off the jury and thenokay, sir. Thank you.
A. Could I make a comment?
Q. Yes, sir.
A. If somebody were profoundly neurologically impaired, they couldn't draw the designs at all. They would be profoundly screwed up.
THE COURT: I like that, layman's term, the only thing I have understood.
THE WITNESS: Sorry, Your Honor.
A. Some of the characteristic defects in this, which are common to neurologically impaired people, is the fact that even though there are eleven dots in the presented card, [the unknown person] just draws a whole series of dots right across the paper, That's called perservation. Once [the unknown person] starts doing something he loses sight of what the limits of the job ought to be. That's a very characteristic problem with people who are neurological impaired. [The unknown person] can't get the connection between the design in the right order. This little sought [sic] of curved doohickey should be at the corner of the rectangle, and he gets it up in the middle. When [the unknown person] has to draw little tiny circles, he starts by making dots and then gets, you know, changes and is really unable to correct himself.
One of the characteristics of people who are not brain-damaged or neurologically impaired, that is, when they make an error they see it and correct it. People who are neurologically impaired know that they've make an error, but then can't fix it. They don't know what to do to fix it, and that's one of the characteristic symptoms of neurological impairment. The other last comment I'd make is in terms of direction. Neurologically impaired people can't smoothly go from left to right. They sometimes don't know their left from their right. If you tell them, kick something with their left foot, they have to think for a moment, what is their left foot. They can't just do it automatically. They have to change directions. They may move to the right, think about it, try to move to the left, but keep moving to the right and then make a move to the left. We do see that in this [unknown] person's drawings, but don't see that in the drawings performed by the defendant.
[Emphasis added.]
After demonstrating the drawings of the unknown allegedly "mildly brain damaged" person, the doctor compared defendant's Bender Gestalt test, not against the recognized approved standards contained in the Bender Gestalt literature, but rather against the unknown person and then against "your general average functioning adult," and was asked by the prosecutor "[w]ould you expect any of us to perform any better on that test than [defendant] did?" The doctor responded "no."
II.
At the outset, we recognize that the entire focus of the jury's inquiry was the diminished capacity issue upon which the jury had the benefit of expert evidence from both sides. The witnesses did not necessarily disagree that defendant had some psychological deficits, but sharply disagreed as to whether those deficits impaired his ability, at the time of the killing and under the attendant circumstances, to act purposefully or knowingly. The battle, then, was a choice of the experts, the primary choice, we think, being Dr. Dougherty or Dr. Krass. At the least, the testimony of these two witnesses could have been dispositive, depending upon which one the jury accepted. There were, of course, a number of factors, including defendant's own conduct and statements at the time of the killing and shortly before, that were important. But critical to the choice of the experts was the jury's consideration of the evidence each of these doctors presented on defendant's performance of the neurological tests. Significant among the various tests was the Bender Gestalt Test.
Dr. Dougherty and Dr. Krass, using the results of the tests performed by Dougherty, opined at some length on those results. Dr. Dougherty was convinced that defendant's results were abnormal and Dr. Krass was convinced they were not. Were their differing views based solely upon each one's analysis of defendant's drawings compared to the accepted Bender Gestalt standards, there would be no issue here.
They were not. As we have set forth in our recitation of the evidence, the State's expert, Dr. Krass, over the objection of defendant, brought to court and demonstrated to the jury the drawings, and test results, of an unknown individual who, he asserted, was "mildly brain-damaged." Using not the Bender Gestalt standards but these drawings, Krass offered his opinion that defendant's drawings were normal as compared to those of the unknown individual. Indeed, not only did Dr. Krass utilize this unknown person as a standard of comparison, but he, through the questions of counsel, asserted that defendant's drawings were as good as either the prosecutor or the jurors could produce.
Thus viewed, we think the conclusion that this aspect of Krass' testimony was improperly admitted scarcely needs citation. To begin with, the use of the unknown individual as a standard sharply compromises defendant's sixth amendment ability to cross-examine the doctor. See generally State v. Harvey, 151 N.J. 117, 187-88, 699 A.2d 596 (1997). While N.J.R.E. 703 does permit an expert to utilize hearsay data where it is the type reasonably relied upon by others in the field, it is not a rule of wholesale admissibility of otherwise inadmissible evidence. See State v. Rose, 112 N.J. 454, 499-501, 517, 548 A.2d 1058 (1988); State v. Burris, 298 N.J.Super. 505, 512, 689 A.2d 860 (App.Div.), certif. denied, 152 N.J. 187, 704 A.2d 17 (1997); Day v. Lorenc, 296 N.J.Super. 262, 267, 686 A.2d 1193 (App.Div.1996); State v. Pasterick, 285 N.J.Super. 607, 620-21, 667 A.2d 1103 (App.Div.1995). Dr. Krass proffered no evidence whatsoever that in the field of forensic psychology, an unknown individual's Bender Gestalt test results is data reasonably relied upon to determine whether a particular defendant's results are normal.
Perhaps more importantly, one of the criteria under N.J.R.E. 702 for the admissibility of expert testimony is that the testimony be based on "reasonably reliable scientific premises." State v. Cavallo, 88 N.J. 508, 516-526, 443 A.2d 1020 (1982) (rapist profile inadmissible as scientifically unreliable). See State v. W.L., 278 N.J.Super. 295, 304, 650 A.2d 1035 (App.Div.1995); State v. Michaels, 264 N.J.Super. 579, 616-619, 625 A.2d 489 (App.Div.1993), aff'd, 136 N.J. 299, 642 A.2d 1372 (1994). The reliability of a scientific methodology may be established by 1) expert testimony as to its general acceptance and reliability, 2) authoritative scientific and legal writings, or 3) judicial opinions. E.g. State v. Kelly, 97 N.J. 178, 210, 478 A.2d 364 (1984). And, we take note that admissibility of all evidence, particularly scientific evidence being used in a criminal proceeding, must be "clearly established." State v. Haskins, 131 N.J. 643, 649, 622 A.2d 867 (1993); State v. Cary, 99 N.J.Super. 323, 333, 239 A.2d 680 (Law Div.1968), aff'd, 56 N.J. 16, 264 A.2d 209 (1970) ("[t]he need for a high degree of scientific acceptance, and particularly reliability, is vital when a criminal case is involved where the individual's freedom or, in fact, his life may be at stake."). See State v. Fortin, 318 N.J.Super. 577, 609, 724 A.2d 818 (App. Div.1999) (in murder trial, linkage-analysis methodology of identifying defendant as the murderer based upon a comparison of one other crime committed by him not scientifically reliable). Contrast State v. Clowney, 299 N.J.Super. 1, 19-20, 690 A.2d 612 (App. Div.), certif. denied, 151 N.J. 77, 697 A.2d 549 (1997) (on cross-examination of defendant's diminished capacity expert witness, prosecutor's questions incorporating rapist profile traits, not disagreed with by the expert, was proper cross-examination and not substantive proffer of rapist profile methodology).
Here, to be sure, the Bender Gestalt Test and standard of comparison used by Dr. Dougherty is an established reliable methodology for determining whether a particular individual exhibits signs of neurological impairment. But the critical question is whether Dr. Krass' use of the unknown individual's results of that test as a standard to measure defendant's results against, and as an indication of whether defendant suffered from a neurological impairment, was established as reliable. Clearly it was not; no effort whatsoever was made to do so. Unlike Dr. Dougherty's testimony as to the use of the Bender Gestalt Test and comparison standards established by the accepted literature, Dr. Krass offered no evidence that his quite different comparison of the unknown individual's results against defendant's results was a scientifically accepted method of determining whether the defendant suffered from some neurological disorder. Neither has the State proffered any independent literature or case law to establish the reliability of such method.
We are not persuaded by the State's argument that Dr. Krass "was not promoting some sophisticated and untested scientific methodology" that would require "a special inquiry as to its reliability." Whether characterized as sophisticated or complex, the comparison of one's Bender Gestalt drawings against another's is just as much beyond the ordinary knowledge of jurors as is the comparison of fingerprints. The difference, of course, is that the methodology involved in the latter has long been established. And compare State v. Haskins, supra, 131 N.J. at 645, 622 A.2d 867 (scientific evidence in the form of a measuring tape used to measure distance of drug transaction from a school was admissible even though not shown to be accurate because the "subject matter of such measurements is within the common knowledge and experience of jurors"); State v. Carroll, 256 N.J.Super. 575, 597-98, 607 A.2d 1003 (App.Div.), certif. denied, 130 N.J. 18, 611 A.2d 656 (1992) (jury may compare a known sample of defendant's handwriting with the handwriting on a contested document without expert testimony).
Alternatively, the State contends that the use of the unknown person's test results was not as substantive evidence, but simply as "illustrative" or "demonstrative" evidence. And it has been observed that "[t]here is nothing inherently improper in the use of demonstrative or illustrative evidence." State v. Scherzer, 301 N.J.Super. 363, 434, 694 A.2d 196 (App.Div.), certif. denied, 151 N.J. 466, 700 A.2d 878 (1997). See State v. Feaster, 156 N.J. 1, 82, 716 A.2d 395 (1998). Demonstrative or illustrative evidence may be evidence that replicates or illustrates certain aspects of a crime scene. E.g. id. at 83, 716 A.2d 395 (use of mannequin with a knitting needle inserted in head to show trajectory of bullets through victim's head); State v. Mayberry, 52 N.J. 413, 435-36, 245 A.2d 481 (1968), cert. denied, 393 U.S. 1043, 89 S.Ct. 673, 21 L.Ed.2d 593 (1969) (admission of a gun "very similar" to unrecovered murder weapon); State v. Scherzer, supra, 301 N.J.Super. at 434, 694 A.2d 196 (admission of a bat and broom the victim identified as similar to bat and broom used during the sexual assault upon her). We do not view the evidence here as merely demonstrative or illustrative. It was substantive evidence to establish that defendant was not neurologically impaired.
Finally, we reject the State's contention that the evidence was relevant only insofar as it concerned Dr. Krass' credibility. The quick answer to that is, were that so, defendant clearly was substantially hampered in his ability to cross-examine any aspect of the unknown nature of the person who took the test. All that was offered was the doctor's hearsay assertions of the characteristics of this person. Since the comparison was not contained in the doctor's report, defendant had no ability to obtain any factual data to effectively cross-examine the doctor as to this. Moreover, as we have said, this was not just credibility evidence. It is quite clear that the unknown person's test results were offered as independent proof that defendant was not neurologically damaged.
The admission of this aspect of Dr. Krass' expert testimony, then, was erroneous. We are convinced, moreover, that the evidence was such as to have been clearly capable of producing an unjust result, that is that there is a reasonable possibility that the error contributed to the verdict. R. 2:10-2; State v. Macon, 57 N.J. 325, 337-39, 273 A.2d 1 (1971).
In this respect, we certainly cannot say that the State's experts were any more or less persuasive than defendant's experts. Neither can we ignore some of the factual circumstances of the incident itself that would suggest defendant was not in his right mind at the time. The infliction of eighty-one stab wounds alone, coupled with the rather bizarre conduct of walking away from the body in bloody clothes and telling the victim's son he was going for bagels, would raise questions. There was, also, the evidence of defendant's agitated state just before the incident and, a juror could conclude, heavy intake of Prozac. Moreover, as we have said there was no real disagreement that defendant did suffer from a psychiatric or psychological condition. To be sure, defendant's immediate actions and statements, partially relied on by the State's witnesses, were indicative of some awareness and consciousness. But we cannot say that a jury would have accepted the evidence supporting the opinion that defendant was acting knowingly or purposefully were Dr. Krass' objectionable evidence to be disregarded. We have no idea what role that evidence actually played in the jury's evaluation, but it was a substantial part of his testimony and, as we have said, the battle of the experts was largely waged between Dr. Dougherty and Dr. Krass. Under these circumstances, we are convinced the error was not harmless.
Reversed and remanded for a new trial consistent with this opinion.
NOTES
[1] The doctor described psychosis as becoming "totally detached from reality ... hearing voices, seeing things ... thought processes are racing, the thoughts are coming in your mind faster than you can express them, you cannot think very clearly...."