(dissenting). In my view a rule of per se inadmissibility is “unnecessarily broad” and results in the exclusion of worthwhile evidence (State v Hurd, 86 NJ 525). Hypnosis is neither supernatural nor magical but “a *23reduced state of consciousness” (Dillof, Admissibility of Hypnotically Influenced Testimony, 4 Ohio North L Rev 1, 3) in which the subject is able to relax and thereby, in some cases, recall events that the conscious mind has repressed. The past decade has seen a hypnosis “boom” (Diamond, Inherent Problems in the Use of Pretrial Hypnosis on a Prospective Witness, 68 Cal L Rev 313,341) in law enforcement, resulting in much abuse of the technique. For example, some witnesses have been hypnotized by law enforcement personnel rather than by impartial professionals, and convictions have been obtained with no corroborating evidence (see Diamond, op. cit., pp 324-325). Hypnosis has also been used to refresh the recall of a person inebriated at the time of the crime (People v Shirley, 31 Cal 3d 18). These and other inappropriate uses of hypnosis have led many courts to adopt a rule of per se inadmissibility. Other courts have rejected this approach by holding that the evidence is admissible if the hypnotic session was conducted in accord with various procedural safeguards (State v Hurd, supra; State v Greer, 609 SW2d 423, vacated on other grounds 450 US 1027; People v Smrekar, 68 Ill App 3d 379; United States v Adams, 581 F2d 193, cert den 439 US 1006; see, also, People v Lucas, 107 Misc 2d 231; People v McDowell, 103 Misc 2d 831, 834-836). Although I recognize the danger of confabulation and the limitations of procedural safeguards, I would not rule out hypnotically refreshed testimony in those cases where there is independent verification that the subject has not confabulated but has, in fact, recalled accurately who the assailant was.
The majority apply the Frye rule (Frye v United States, 293 F 1013) to hypnotically refreshed testimony and conclude that there is no general acceptance of the reliability of such evidence. I have some doubts as to whether that rule should apply, since the evidence being presented is not an expert testifying as to the results of a scientific test but rather the witness testifying as to her own recollection, albeit a refreshed one. Assuming, arguendo, that the Frye rule should apply, my application of it to the facts at hand results in a different conclusion, for there is general acceptance in the scientific community that hypnosis can restore the subject’s memory, and there is general agreement that *24the key to determining whether the subject confabulated or recalled the event accurately is independent verification. A leading expert in the field, Dr. Martin Orne, has summed up this view by stating that: “hypnosis may be useful in some instances to help bring back forgotten memories following an accident or a crime while in others a witness might, with the same conviction, produce information that is totally inaccurate * * * As long as this material is subject to independent verification, its utility is considerable and the risk attached to the procedure minimal” (State v Hurd, supra, p 539, citing Orne, The Use and Misuse of Hypnosis in Court, 27 Int J Clinical & Experimental Hypnosis, 311, 317-318). Even the leading proponent of the per se inadmissibility rule, Dr. Bernard L. Diamond, concedes that hypnosis can, in some cases, restore memory (Diamond, op. cit., p 340). A review of the many cases discussing this topic reveals that most of the courts place undue emphasis on the opinions of Dr. Diamond to the exclusion of other experts. However, “the test is not whether a particular procedure is unanimously indorsed by the scientific community, but whether it is generally acceptable as reliable” (People v Middleton, 54 NY2d 42, 49). Hypnosis to refresh a memory is generally accepted as reliable when there is independent verification of the fact recalled, and thus I would admit hypnotically refreshed testimony when there are facts establishing that the witness did not confabulate. While corroboration has sometimes been listed as a procedural safeguard (see, e.g., People v Lucas, 107 Misc 2d 231, 236, supra), I would make corroboration the linchpin of admissibility.
Turning to the facts at hand, there was ample corroboration that the victim accurately identified her attacker and not some innocent party. A neighbor saw defendant enter the apartment building where he and the victim both lived at 10:10 p.m. About five minutes later, two other neighbors heard screams coming from the victim’s apartment. One of the neighbors heard the victim’s four-year-old son screaming and heard the victim cry out, “What are you doing here? What do you want? Get out.” She then heard a thumping noise, like something falling down the back stairs. About 45 minutes later, a fourth neighbor who lived *25across the street was called outside by her husband who was leaving for work at 10:55 p.m. when he heard the victim’s son crying. This neighbor and her nephew stood outside and saw the child come outside, look around, and go back inside. A few minutes later, at approximately 11:05 p.m., they saw defendant emerge from the victim’s back yard. Defendant, wearing no shirt, looked up and down the street and then cut through the shrubbery and entered the apartment house. The next day, when police questioned defendant, he had scratches on his back, dirt on his elbows, and bruises on his right knee and hands. There were grass stains on the pants and semen on the underwear that he had been wearing the day before. While the victim was in the hospital she told the investigating officer that she kept remembering defendant, but did not know why. In view of the evidence placing defendant in the victim’s back yard at the precise time she was being raped and coupled with the physical evidence showing that defendant had been in a scuffle in the outdoors, there is ample independent verification that the victim did not confabulate but did, in fact, recall who her attacker was. In view of the corroborating evidence, the victim’s identification should have been admissible because the hypnotic session was conducted in a fair and impartial manner* and was reliable under the totality of the circumstances (Manson v Brathwaite, 432 US 98, 114; People v Graham, 67 AD2d 172, 177).
The majority feel that the admissibility of a hypnotically refreshed identification should not hinge on corroboration. They note that corroboration does not render admissible other evidence, such as polygraph tests. However, polygraphs, voice stress tests and the like are genetically different from hypnotically refreshed testimony. The former are “tests”; therefore, they can have no probative worth unless the validity of the tests has been established, and it is of no consequence that the test produced a correct *26result in a particular case. By contrast, the evidence at issue is a witness’ identification which is presumed reliable unless some “suggestive” procedure has tainted it. Hypnosis often helps a witness recall facts accurately, but the only way to guard against the danger of confabulation is to have independent verification of the facts recalled. Because hypnosis is conceptually different from “tests” it would be appropriate to establish a different rule of law as to the admissibility of such evidence.
In enumerating the various reasons, why hypnotically refreshed testimony is subject to question, the majority seems to suggest that the admissible testimony of any other witness will produce evidence clear from taint, prejudice, or simple mistaken recall. We know that such is not the case. Nor does the rule of law I propose usurp the function of the jury. The ultimate determination of weight and credibility in either case remains for the trier of the fact. The problem which I envision in the “per se” rule of the majority is that it rejects, out of hand, testimony which is reasonably likely to be as reliable as that of any other witness.
Assuming, arguendo, that the majority position is correct, I must further disagree with the conclusion that the subject is not incompetent as a witness to testify to her recollections prior to hypnosis. A major justification for rejecting hypnotically refreshed testimony is that the subject cannot separate pre- and posthypnotic recollections and, once hypnotized, the subject acquires great confidence in the memory thus rendering cross-examination ineffective (see State v Mena, 128 Ariz 226). These considerations have prompted California to declare the hypnotized witness incompetent to testify at all as to matters discussed under hypnosis (People v Shirley, 31 Cal 3d 18,_, supra).
Finally, I make several concluding observations. Hypnosis should not be used in a great many cases, particularly those in which the witness was inebriated at the time or unable to remember clearly for some other reason (see, e.g., People v Shirley, supra). Moreover, hypnotically refreshed testimony should not be used merely to elicit details where the victim has substantial memory of the event. The greater the detail required of the witness, the more chance *27there is for confabulation and the more difficult it is to verify the recall. Lastly, the need for hypnosis most often arises in the context of a rape or other sexual assault where the victim has repressed the event because of its uniquely traumatic nature (State v Mack,__ Minn_, 292 NW2d 764; State v La Mountain, 125 Ariz 547; Polk v State, 48 Md App 382; Harding v State, 5 Md App 230, cert den 395 US 949). Generally, there are no other witnesses, and thus the need for the victim’s testimony is all the more pronounced. I believe that admitting the hypnotically refreshed identification only when there is “clear and convincing” verification that the witness did not misidentify the attacker would best accommodate the competing interests of the need for the evidence with the rights of an accused.
For the reasons stated herein, I would affirm defendant’s conviction.
Dillon, P. J., Callahan and Schnepp, JJ., concur with Hancock, Jr., J.; Doerr, J., dissents and votes to affirm in an opinion.
Judgment reversed, on the law and the facts, and a new trial granted.
In proposing a rule of admissibility I note that various procedural safeguards should be followed to minimize the possibility of confabulation. In this regard, I observe that the procedure in this case was tainted because the victim inadvertently learned, prior to hypnosis, that defendant was a suspect. However, in view of the very strong corroborating evidence, I do not view this fact as dispositive.