opinion of the court
Hancock, Jr., J.The use of hypnosis has not gained general acceptance in the scientific community as a reliable method of restoring a witness’ recollection of an event. We hold, therefore, that the testimony of a witness concerning hypnotically produced recall is, as a matter of law, inadmissible in a criminal trial in this State.
i
Whether such testimony should be admitted, a question apparently not before considered in any appellate court in New York, arises in the context of an appeal from convictions for rape, first degree, burglary, first degree, and assault, second degree. The victim, Nancy Simmons, was unable to recall the details of the incident including the *18identity of her attacker.1 At the suggestion of an investigating police officer, she agreed to be hypnotized. Prior to the first hypnosis session, Mrs. Simmons learned that the police suspected the defendant. During and after this session, which took place in the Public Safety Building and at which, in addition to the hypnotist, the victim’s husband and two investigating police officers were present, the victim named defendant as the rapist and remembered other details. During a second session with the same hypnotist, she recalled nothing additional. Subsequently, Mrs. Simmons on her own initiative contacted a psychiatrist, Dr. Goldfarb, who hypnotized her two more times; he encountered a new memory block which prevented her repetition of the identification previously given under hypnosis. Later, Dr. Goldfarb administered sodium pentothal, and the victim was then able to remember not only that it was defendant who had assaulted her but also that defendant’s brother had been present during part of the incident and had tried to stop him.
At a lengthy pretrial hearing concerning the admissibility of Mrs. Simmons’ testimony, three expert witnesses testified. The prosecutor called Dr. Land. Although he stated that the hypnotic procedures he employed were not biased or suggestive, Dr. Land conceded: that there is no general agreement in the profession as to the reliability of hypnosis as a means of gathering evidence for courtroom purposes; that suggestions made before, during or after hypnosis can affect the subject’s recall; that a subject can create a memory where none previously existed; that a subject will generally be convinced of the accuracy of such false memory; and that his testimony will be given with confidence and have the ring of truth. Neither of the other two expert witnesses — Dr. Goldfarb, called by the People, and Sheldon Malev, professor of clinical psychology, sworn by the defense — disputed Dr. Land’s testimony on these points.
*19In a written decision following the hearing, the trial court, relying on Harding v State (5 Md App 230, cert den 395 US 949) and United States v Adams (581 F2d 193, cert den 439 US 1006), held that “testimony, which is hypnotically induced recollection, is not inadmissible as a matter of law” (People v Hughes, 99 Misc 2d 863, 871) and ruled that Mrs. Simmons was competent to testify as to her posthypnotic recall.
At trial, Mrs. Simmons testified to her recollection of various details recalled after undergoing hypnosis, including her identification of defendant and the presence of defendant’s brother during part of the incident.
ii
County Court’s ruling that hypnotically produced testimony is not legally inadmissible runs counter to the thrust of recent holdings in other jurisdictions that such evidence should not be permitted absent a finding that it satisfies the so-called Frye test (Frye v United States, 293 F 1013) or comparable standard, viz., that before the results of a scientific procedure may be used as evidence, it must be established that the procedure has gained general acceptance in the scientific community (see, e.g., Arizona v Superior Ct.,_Ariz_, 644 P2d 1266; State v Mena, 128 Ariz 226; People v Shirley, 31 Cal 3d 18; Polk v State, 48 Md App 382; People v Tait, 99 Mich App 19; State v Mack,_ Minn_, 292 NW2d 764; Commonwealth v Nazarovitch,_ Pa_, 436 A2d 170). It is significant, we think, that even Maryland’s high court, which decided Harding v State {supra) — for many years the leading case permitting such testimony and holding that the fact of hypnosis of the witness “goes to the weight, not the admissibility” of the evidence2 — has recently repudiated Harding and held *20that admissibility of hypnotically generated testimony must be determined by application of the Frye standard (see Polk v State, supra). New York has adopted a test identical to the Frye rule in considering the admissibility of evidence produced by various scientific procedures, viz., whether the reliability of the results of a procedure is generally acknowledged in the scientific community (see People v Middleton, 54 NY2d 42, admitting bite mark evidence; People v Leone, 25 NY2d 511, excluding results of a polygraph test; People v Forte, 279 NY 204, holding pathometer exam readings not admissible; People v Tarsia, 67 AD2d 210, affd 50 NY2d 1, excluding results of a voice stress test). We hold that the same test should be applied here.
in
Out-of-State courts which have applied the Frye standard to hypnotically produced testimony have generally agreed that such testimony has not been accepted as reliable in the scientific community, citing the following conclusions of recognized authorities in the field.
(1) The subject under hypnosis becomes extremely receptive to suggestions perceived to have come from the hypnotist, even suggestions which were unintentionally or unwittingly communicated. He is also susceptible to suggestions received, whether before or during the hypnosis, from other persons who may have communicated with the witness such as, for example, investigating police officers.3
(2) The subject feels compelled to please the hypnotist and thus, seeking to co-operate, may “fill in” missing details through “confabulation”, or fantasy.4
*21(3) The suggestions adopted by the subject and the details “filled in” by confabulation may create a new memory or “pseudo memory” which the subject accepts as real and which may become so fixed in his mind that effective cross-examination becomes a virtual impossibility.5
(4) Once a subject has been hypnotized, there is no known reliable method enabling anyone — subject, hypnotist, expert witness, Judge or Jury — to distinguish “pseudo memory” from accurate memory which has been enhanced by hypnosis.6
While the authorities are not in complete agreement,7 it is noteworthy, we believe, that even the People’s expert, Dr. Land, generally concurred with the above points including, most significantly, the occurrence of confabulation and the difficulty of distinguishing real from “pseudo” memory.
From our reading of recent decisions in the field and of recognized authorities (see ns 3-6, supra) we are persuaded that hypnotically produced testimony is not generally accepted in the scientific community as reliable and that it should, therefore, be inadmissible (see People v Middleton, 54 NY2d 42, supra; People v Leone, 25 NY2d 511, supra; People v Forte, 279 NY 204, supra; People v Tarsia, 67 AD2d 210, affd 50 NY2d 1, supra).
iv
We must comment briefly on the standards proposed in the dissent for the use of hypnotically produced testimony. Such testimony should be accepted, it is suggested, “when there are facts establishing that the witness did not confabulate.” Corroboration, as the dissenter views it, should be “the linchpin of admissibility.” In other words, what ap*22pears to be true should be admitted; what seems false excluded.
Reliability of hypnotically generated testimony, not its plausibility, should determine whether it is accepted. Polygraph evidence (see People v Leone, supra), voice stress test results (see People v Tarsia, supra), and pathometer exam readings (see People v Forte, supra) are disallowed because the procedures used have not been generally accepted in the scientific community as producing reliable results; such evidence does not become admissible because other evidence corroborates it. For the court to admit what seems to be true and exclude what does not subverts the traditional role of jurors as the exclusive judges of the facts. With corroboration “the linchpin of admissibility”, false hypnotically produced testimony based on “pseudo memory” which happens to coincide with other evidence could come in. Crucial reliable but uncorroborated testimony would be barred. Consequences of such a rule are readily imagined.
If and when hypnosis becomes generally accepted in the scientific community as a reliable method of restoring memory, testimony based on hypnotic recall should in a proper case be admitted. The jury, not the court, should then determine whether such testimony is worthy of belief. Until then, it should be excluded.
Although on a retrial the testimony produced through hypnosis must be excluded, it does not follow that because she has been hypnotized, Mrs. Simmons is incompetent to testify to facts which she was able to recall prior to undergoing hypnosis (see Arizona v Superior Ct., _ Ariz _, supplemental opinion, 644 P2d 1279, 1296; cf. State v Mack,_Minn_, 292 NW2d 764, 771, supra).
We find no error in the other evidentiary rulings of the trial court challenged on appeal.
The judgment should be reversed and a new trial granted.
. Apparently the victim had named the defendant twice prior to hypnosis, stating that she “saw” him or “remembered” him. She did not, however, state that it was he who had attacked her but rather maintained that she could not remember who it was who had done so.
. It should be noted that the Maryland court in Harding v State (5 Md App 230, cert den 395 US 949) apparently accepted a statement of the psychologist who hypnotized the complainant which has subsequently been completely rejected in the scientific community, viz., that hypnosis does not render the subject particularly receptive to suggestion. Thus, the continued soundness of the holdings in Harding and in United States v Awkard (597 F2d 667, 669, cert den 444 US 885) and United States v Adams (581 F2d 193, 198, cert den 439 US 1006), both of which rely on Harding, is subject to question (see *20Diamond, Inherent Problems in the Use of Pretrial Hypnosis on a Prospective Witness, 68 Cal L Rev 313, 322-323).
. See People v Shirley (31 Cal 3d 18, n 46) and authorities cited therein (State v Mack,__ Minn_, 292 NW2d 764, "768; Diamond, op. cit., p 333).
. See State v Mena (128 Ariz 226, 229, quoting Spector and Foster, Admissibility of Hypnotic Statements: Is the Law of Evidence Susceptible?, 38 Ohio St L J 567, 578); People v Shirley (31 Cal 3d 18, _, supra). Diamond, op. cit., p 335 states: “[T]he hypnotically recalled memory is apt to be a mosaic of (1) appropriate actual events, (2) entirely irrelevant actual events, (3) pure fantasy, and (4) fantasized details supplied to make a logical whole.”
. See People v Shirley (31 Cal 3d 18,_, supra); State v Mack (supra, p 769, see excerpt from Dr. Martin Orne’s testimony, n 10); Commonwealth v Nazarovitch (_Pa _, _, 436 A2d 170,174-175); Diamond, Inherent Problems in the Use of Pretrial Hypnosis on a Prospective Witness, 68 Cal L Rev 313, 333-338, 340; Spector and Foster, op. cit., p 593.
. See People v Shirley (supra, pp_,_); State v Mack (supra, p 769); State v Hurd (86 NJ 525, 538-539, quoting Orne, The Use and Misuse of Hypnosis in Court, 27 Int J Clinical & Experimental Hypnosis 311, 317-318); Diamond, op. cit, pp 337, 339-340.
. For example, New Jersey’s Supreme Court, applying New Jersey’s version of the Frye test, held that hypnotically generated evidence was admissible in certain instances, concluding that “[i]f it is conducted properly and used only in appropriate cases, hypnosis is generally accepted as a reasonably reliable method of restoring a person’s memory.” (State v Hurd, supra, p 538; see, also, Spector and Foster, op. cit.)