BRONSON, J.

Defendant and John Duncan Wallach were charged with the first-degree murder of Elmer Evans, in contravention of MCL 750.316; MSA 28.548. Gonzales and Wallach were separately tried, however. Following a jury trial in the Oakland County Circuit Court, defendant was found to be guilty as charged. He was sentenced to a mandatory term of life imprisonment. Defendant now appeals as of right.

On Sunday, January 28, 1979, two bodies were found buried in the snow next to a house located on Silver Down Court in Waterford Township, Michigan. The victims later were identified as Fred Torres and Elmer Evans. Torres had been stabbed to death, and Evans died as the result of a beating.

The case against defendant was based almost solely on the testimony of Rhonna Burns. Burns' testimony had been hypnotically refreshed prior to trial and was admitted over defense objections that the hypnosis may very well have irreparably tainted her true memory so that she was merely parroting suggestions implanted by the police.

According to the testimony of Burns, on Saturday night, January 27, 1979, she was at the Liberty Bar with John Wallach. Late that evening Wallach introduced her to defendant. Wallach then went over to Elmer Evans while Burns and defendant conversed. After talking to Evans, Wallach came back to the booth where Burns and defendant were. He said that Evans needed a ride home. Defendant then went over to Evans who gave defendant his car keys. Defendant then left by way of the front door — apparently to go to Evans' car which Evans had said was parked in front of the bar. Wallach, Burns, and Evans exited through the back door of the bar and went to Wallach's station wagon. The trio drove to a warehouse. Defendant was waiting there, standing in front of a small blue car. Defendant got into the back seat of the station wagon and sat next to Evans. They then drove to a house on a hill and stopped in front of a garage. Stating that he had to go to the bathroom, defendant got out and went out around the garage toward the right. Wallach and Evans then got out of the car and went toward the left, out of Burns' view. After approximately 10 or 20 minutes, defendant returned. Wallach returned immediately thereafter. Evans, however, did not return. Wallach's right hand was cut and blood was on his hand and pants. Burns and Wallach then dropped defendant off downtown.

The question which is dispositive of this appeal is whether the trial court erred in allowing the testimony of Rhonna Burns since her memory had been hypnotically refreshed and her testimony consisted mostly of posthypnosis recollections. If so, defendant's conviction must be reversed. If not, at most, the other claims raised by defendant constitute harmless error.

In People v Tobey, 401 Mich. 141, 145; 257 N.W.2d 537 (1977), the Michigan Supreme Court reaffirmed its adherence to the rule of Frye v United States, 54 US App DC 46, 47; 293 F 1013, 1014 (1923), limiting the admission of scientific evidence to techniques which have gained general acceptance in the particular areas in which they belong. Moreover, the general scientific recognition of a particular technique must be established by disinterested experts and not by those intimately linked to the technique. As will be detailed below, at present we believe that hypnosis as a scientifically accepted technique is akin to the polygraph examination. See People v Barbara, 400 Mich. 352; 255 N.W.2d 171 (1977). While both polygraph examination and the hypnosis technique under consideration here have achieved some degree of acceptance, neither has gained that degree of general acceptance in the scientific community which distinguishes a true science from a pseudoscience.

Two decisions of this Court have spoken to the problem of the admissibility of hypnotically "refreshed" testimony. In People v Hangsleben, 86 Mich. App. 718; 273 N.W.2d 539 (1978), the defense sought to introduce into evidence a tape recording of an interview between a psychiatrist and defendant while the latter was under the effects of hypnosis. The trial court denied the defense request. This Court held that no error had occurred. Hangsleben differs significantly from this case. There the issue was not whether a witness who had been hypnotized could give testimony but, rather, whether a criminal defendant could permissibly inform the jury that his memory was restored by hypnosis to bolster his credibility. The Hangsleben Court noted out-of-state opinions calling posthypnotic testimony suspect and concluded by holding that if error had occurred it was harmless since any reference to a hypnotic memory tended to impeach, rather than enhance, credibility.

The other Michigan decision dealing with this problem, People v Tait, 99 Mich. App. 19; 297 N.W.2d 853 (1980), is much more like the instant matter than Hangsleben. In Tait, a prosecution witness gave hypnotically refreshed testimony. Defense counsel did not learn of the hypnosis until after the first day's testimony and then moved for a mistrial which was denied. In Tait, no evidence concerning the reliability of hypnosis as a method of uncovering a true memory was introduced. Furthermore, the hypnotist was an amateur and not a professional psychiatrist. Tait found that in respect to hypnotically refreshed memory the Tobey test for reliability had not been satisfied.

While Hangsleben was limited in its holding to the lack of adequate foundation presented in that case, Tait is unclear as to the intended scope of its holding. It appears, however, that the Tait decision has a more expansive scope than Hangsleben and stands for the proposition that hypnosis has not achieved sufficient scientific acceptance to allow the admission of testimony affected by the process.

It generally is agreed among those who have studied the hypnotic state that it is a condition of altered consciousness marked by heightened suggestibility. A subject in a hypnotic state may not have accurate recall. Indeed, there is a tendency in the hypnotized subject to relate false memory, distortions of memory caused by the process of age regression induced by hypnosis in which distinct experiences are jumbled and recalled as one, fantasies, and confabulations (the creation of bits of information to fill in the gaps between that which is actually remembered). This problem is exacerbated significantly by the tendency of the subject to respond in a way which he believes the hypnotist desires. See, among others, Diamond, Inherent Problems in the Use of Pretrial Hypnosis on a Prospective Witness, 68 Cal L Rev 313 (1980), Note: Hypnotized Witnesses May Remember Too Much, 64 ABAJ 187 (1978), Spector and Foster, Admissibility of Hypnotic Statements: Is the Law of Evidence Susceptible?, 38 Ohio St LJ 567 (1977), Dilloff, The Admissibility of Hypnotically Influenced Testimony, 4 Ohio N U L Rev 1 (1977), Comment: Refreshing the Memory of a Witness Through Hypnosis, 5 UCLA-Alaska L Rev 266 (1976), Comment: Hypnosis, Truth Drugs, and the Polygraph: An Analysis of Their Use and Acceptance by the Courts, 21 U of Fla L Rev 541 (1969), Note: Hypnotism, Suggestibility and the Law, 31 Neb L Rev 575 (1952), Ladd, Legal Aspects of Hypnotism, 11 Yale LJ 173 (1902).

Bernard Diamond, professor of both law and clinical psychiatry, states in Inherent Problems in the Use of Pretrial Hypnosis on a Prospective Witness, supra, 314-315 (hereinafter Diamond):

"In many cases, presentation to the jury of all facts, whether legally relevant or not, may be desirable. Indeed, I have often expressed the opinion that traditional rules of evidence restricting the information available to the trier of fact may impede a just decision. This principle does not, however, apply to the testimony of witnesses whose memories have been previously enhanced by hypnosis. I believe that once a potential witness has been hypnotized for the purpose of enhancing memory his recollections have been so contaminated that he is rendered effectively incompetent to testify. Hypnotized persons, being extremely suggestible, graft onto their memories fantasies or suggestions deliberately or unwittingly communicated by the hypnotist. After hypnosis the subject cannot differentiate between a true recollection and a fantasy or a suggested detail. Neither can any expert or the trier of fact. This risk is so great, in my view, that the use of hypnosis by police on a potential witness is tantamount to the destruction or fabrication of evidence. Recently, some courts have shown a healthy suspicion of the veracity of this sort of testimony. Yet even under stringent safeguards, including presentation to the trier of fact of the fullest possible information on the effects of hypnosis, the trier will not be able to sort out reality from witness fantasy and weigh this testimony properly." (Footnotes omitted.)

Despite a lack of general scientific acceptance and the problems with hypnotically refreshed or created memory recognized by scholars who have done research in this field, at present the majority of jurisdictions which have considered the question have allowed the testimony of one who has had his memory "refreshed" through hypnosis. In light of our research — which indicates that if such testimony be admitted at all, it should be allowed only with stringent safeguards in the hypnotic process — we were understandably surprised with the widespread acceptance of hypnotically "refreshed" testimony. Upon examination of the case law, however, we find that the lead case on this question, Harding v State, 5 Md. App. 230; 246 A.2d 302 (1968), cert den 252 Md. 731, cert den 395 U.S. 949; 89 S Ct 2030; 23 L Ed 2d 468 (1969), provides an explanation for this phenomenon.

The Harding Court accepted the scientific reliability of hypnosis based solely on the testimony of Ralph P. Oropolo, who was employed as a psychologist by "several law enforcement agencies and state's attorneys' offices in Maryland and other states". Oropolo's educational credentials included a Master's Degree in psychology and some work in pursuit of a doctorate. The court gave great weight to Oropolo's testimony about the effects of hypnosis. Oropolo explicitly discounted the possibility of enhanced suggestibility of the subject while in the hypnotic trance. He further stated that while the witness was under hypnosis he had made no improper suggestions. The Harding Court also found it significant that the witness stated that she was testifying from her own recollection.

The court gave little consideration to authorities other than Oropolo. Near the end of the opinion, the court noted medical authorities which found that hypnosis was recognized as a means for getting a subject to recall painful experiences. The Harding Court then noted in passing that "some authorities warn that fancy can be mingled with fact in some cases". Id., 246.

Upon careful scrutiny, the reasoning of the Harding Court is not persuasive. Indeed, it is riddled with problems. First, Oropolo, the hypnotist in Harding, indicated that hypnosis does not dispose a subject to suggestion. We were unable to locate one article written either before or after Harding which supports this assertion. Diamond, supra, 323, categorically states that this claim directly contradicts all scientific evidence. Second, it must be borne in mind that Oropolo was an employee of various law enforcement agencies. The Court in Harding did not discuss the hard question of whether Oropolo's questionable impartiality should affect the testimony's admissibility. Third, the hypnosis was conducted at a police barracks at a time at which Harding was already the prime suspect. Undoubtedly everybody connected with the case was desirous of confirming in their own minds that Harding was the culprit. Fourth, the victim modified her story while under hypnosis. Dilloff writes in The Admissibility of Hypnotically Influenced Testimony, supra, 19-20 (hereinafter Dilloff), that, at a minimum, the fact of modification should have signaled to the Harding Court the possibility of confabulation or memory distortion. This was particularly true given that the victim not only added new details to the incident but changed other aspects of the original story. Fifth, the court found it significant that the victim believed that she was testifying from her own recollection. However, the scientific studies conducted on hypnosis support the conclusion that hypnotized subjects are unable to sort out accurately fact from distortions after having undergone the process. Dilloff, supra, 20. Diamond, supra, 335-336 writes:

" 6. After the hypnotic subject is awakened, do the distorting effects of the hypnosis disappear?

"This too is a critical issue for the admissibility of testimony where there has been pretrial recall enhancement. For courtroom testimony to be admissible, the witness must have emerged from the trance with his powers of recall intact and presumably enhanced, and must be able to relate from the witness stand his wholly conscious recollections as he now perceives them. The evidence, however, is that the effect of suggestions made during hypnosis endures. Although the more theatrical types of posthypnotic suggestion may last only hours or days, less obvious suggestions may last years or even a lifetime. A highly pertinent example is the so-called `posthypnotic source amnesia.' This occurs when something learned under hypnosis is carried into the wakened state but the fact that the memory or thought was learned under hypnosis is forgotten. A sizable proportion of hypnotic subjects spontaneously do so, and the incidence of source amnesia can be increased by suggestions from the hypnotist. A subject who has lost the memory of the source of his learned information will assume that the memory is spontaneous to his own experience. Such a belief can be unshakeable, last a lifetime, and be immune to all cross-examination. It is especially prone to `freeze' if it is compatible with the subject's prior prejudices, beliefs, or desires. This type of distorted memory is very apt to appear genuine and spontaneous, and will be unlikely to disappear.

"One can only conclude that hypnosis can induce subtle but highly significant distortions of memory that will persist indefinitly, distorting all subsequent related recall of the subject. My own experience has convinced me that even communications and other cues to the subject made in the normal, waking state, both before and shortly after the hypnotic session, may be similarly influenced by the hypnotic experience. Thus the police may tell a witness something just before hypnosis and then hypnotize him. When he awakes, his `source amnesia' may lead him to believe that the police statement was a product of his own memory. Sometimes communications made to the patient after hypnosis may be retroactively integrated into the hypnotic recall. The subject may recall a fact with no awareness that it was not the product of his own mind. Or he may recall being told the fact but insist that he had prior knowledge of it. This often happens when subjects are shown photographs or line-ups for identification just before or just after hypnotic sessions. In my experience, time, rather than weakening the effects of the hypnotic distortion, tends if anything to fix it into a permanent pattern. Therefore, the pretrial hypnosis of a witness appreciably influences all of his subsequent testimony in ways that are outside the consciousness of the witness and difficult, if not impossible, to detect." (Footnotes omitted.)

See also, fn 4, supra.

Dilloff, supra, 19, states that in his opinion the testimony of the complainant in Harding "was probably neither very reliable nor accurate". Dilloff's opinion does not represent a hostility to hypnotically refreshed testimony since he ultimately concludes that such evidence should be deemed admissible despite its potential for mischief.

Obviously, we cannot overrule Harding. We have devoted so much time to a consideration of this case solely because most of the subsequent cases allowing the admission of hypnotically revived or induced testimony rely on Harding or its progeny for reasoning or sources. See, e.g., State v Jorgensen, 8 Or. App. 1; 492 P.2d 312 (1971), State v Brom, 8 Or. App. 598; 494 P.2d 434 (1972), State v McQueen, 295 N.C. 96; 244 S.E.2d 414 (1978), (also advancing the theory that hypnosis is no less acceptable than other memory enhancing procedures), People v Smrekar, 68 Ill. App.3d 379; 24 Ill Dec 707; 385 N.E.2d 848 (1979), (over the dissent of Justice Craven, hypnotically "refreshed" testimony was allowed where, prior to the hypnosis, the witness had indicated that there was only a "50-50 chance" that she could identify the murderer), People v Diggs, 112 Cal.App.3d 522; 169 Cal.Rptr. 386 (1980), (defendant's conviction was reversed on other grounds, however), State v Temoney, 45 Md. App. 569; 414 A.2d 240 (1980), United States v Adams, 581 F.2d 193 (CA 9, 1978), cert den 439 U.S. 1006; 99 S Ct 621; 58 L Ed 2d 683 (1978), United States v Awkard, 597 F.2d 667 (CA 9, 1979), cert den 444 U.S. 885; 100 S Ct 179; 62 L Ed 2d 116 (1979).

The one case which we have uncovered which would allow the admission of hypnotically "refreshed" memory, while recognizing the substantial problems of such testimony's reliability, is State v Hurd, 173 N.J. Super. 333; 414 A.2d 291 (1980). Hurd recognizes the limitations of Harding, supra, and provides an excellent analysis of the hypnotic process. Prior to allowing hypnotically enhanced testimony, the Hurd court required, as a prerequisite to admissibility, meeting a set of six standards proposed by Dr. Orne. See fn 4. In our opinion, Hurd certainly represents a more satisfactory solution to this problem than the indiscriminate admissibility of such evidence as propounded in Harding. Nonetheless, we do not believe that even the Hurd standards provide sufficient safeguards. While the Hurd standards would have the tendency to minimize distortion due to hypnotic suggestibility, our examination of the literature convinces us that, even following these criteria, substantial problems with confabulation, fantasy, and distortion would remain. Consequently, in individual cases, the Hurd standards might have an affirmatively detrimental effect. The standards, themselves, would give the hypnotic process an aura of reliability which, in actuality, it does not possess. It is far too likely that a jury would be even less critical of the testimony because of the indicia of reliability provided by such standards. We find nothing in the literature indicating in what percentage of the cases — under the most rigorous clinical standards — a subject's hypnotic memory later proves to be faulty. Furthermore, as noted previously, there is no way a jury can determine the credibility of the previously hypnotized subject. Since the subject comes to believe the version of the events given while under hypnosis, whether true or not, the jury has no real ability to assess credibility based on demeanor. Indeed, if anything, the body language of the witness will project great confidence in the testimony being given and be persuasive for this reason.

Two recent decisions from the Minnesota and Arizona Supreme Courts are in accord with the position we adopt in this opinion. In State v Mack, 292 N.W.2d 764 (Minn, 1980), a unanimous Supreme Court noted many of the problems with Harding, supra, raised in this opinion. We concur with the conclusion reached in Mack:

"We do not foreclose, by this opinion, the use of hypnosis as an extremely useful investigative tool when a witness is enabled to remember verifiable factual information which provides new leads to the solution of a crime. A witness under hypnosis may, for instance, bring forth information previously unknown to law enforcement authorities, such as a license plate number, which subsequently aids police in identification of a suspect. Experts see no reasonable objection to the use of hypnosis in this manner, provided the witness is willing, as long as the material remembered during hypnosis is not subsequently used in court as part of an eyewitness' testimony. Even where the use of hypnosis truly is to investigate a crime rather than to create a witness, adequate safeguards should be established to assure the utmost freedom from suggestion upon the hypnotized person's memory recall in the event he or she must later be called to testify to recollections recorded before the hypnotic interview." (Footnote omitted.) Mack, supra, 771.

See, also, State v Mena, 128 Ariz. 244; 624 P.2d 1292 (1980).

Although we hold that hypnosis as a technique to enhance memory recall has not received sufficient scientific recognition of reliability to allow the post-hypnotic "recollections" of witnesses to be introduced into evidence, we conclude by briefly discussing the various interviews and hypnotic sessions undergone by Rhonna Burns in this case. Our purpose in so doing is to demonstrate that under any reasonable set of guidelines to ensure the reliability of hypnotic recall, the handling of Burns hopelessly tainted her as a witness.

Prior to any hypnosis sessions, Rhonna Burns was interviewed on four separate occasions by the police. In none of these interviews did Burns indicate that Gonzales was involved in the killings. Indeed, Burns denied having any recollection of the killings. The first two interviews were relatively free of police suggestion at least so far as the record indicates. However, the third and fourth interviews present another situation again.

On March 29, 1979, Detective Sergeant Werner interviewed Burns for the third time. At first Burns did not remember anything new about the incident. Werner then asked Burns if she saw defendant at the Liberty Bar on the night of January 27, 1979. Burns replied she didn't remember seeing him. When asked if she remembered Fred Torres coming into the Liberty Bar, Burns replied affirmatively and also stated that she remembered Wallach saying that he was going to give Chico and Torres a ride to the Wide Track Bar, which was about two blocks away from the Liberty Bar. She and Wallach sat in a booth in the back of the bar. Burns then remembered that Wallach pointed to an old man (perhaps Elmer Evans) sitting at the back of the bar. Wallach stated that they should give him a ride home because he was drunk. She, Wallach, and the old man left the bar together, and they drove the old man home. Burns could not remember if they met anyone at the old man's house. After Wallach and Burns dropped the old man off, they met Chico and drove him home. She and Wallach then returned to the Liberty Bar where Wallach left her alone for a brief time. The interview was interrupted for a short period. Later that same day Werner returned and showed Burns photographs of Evans, but Burns wasn't sure if he was the same old man she had been talking about. Asked if she knew where the old man lived, Burns described a big, new house with dark siding. This description fit the house located at 2610 Silver Down Court, where Evans' body had been found. Burns subsequently identified a picture of the house at 2610 Silver Down Court and said it looked like the house at which the old man had been dropped off. Werner then asked Burns if she had returned to this house at 5 a.m. on Sunday, January 28, and if she got out of the car, saw some bodies, and screamed. Burns replied that she could have, but that she didn't remember. At some point in this interview, Werner showed Burns a photograph of defendant and asked her if she had seen him in the Liberty Bar. Officer Werner, either during or prior to this interview, told Burns that defendant was short, stocky, and of Mexican or Spanish descent. Burns still did not remember defendant being at the Liberty Bar.

Detective Werner held a fourth interview with Rhonna Burns on April 3, 1979. For the first time Burns remembered defendant sitting at the bar on January 27, 1979, with her and Wallach. Burns now also remembered Wallach and defendant walking up to the old man and asking him if they could give him a ride home. The old man gave defendant his car keys. She, Wallach, and Evans left the bar together; they picked up defendant, who got into the back seat of Wallach's car with Evans, and then took Evans to what Burns believed to be his house. Both Wallach and defendant helped the old man out of the car. Burns, defendant, and Wallach then returned to the Liberty Bar. She and Wallach subsequently drove home. Somewhere in town they picked up a man named Chico and drove him home. In response to Werner's question, Burns told him that she remembered Wallach picking up a shovel at Tina Meston's house and that he put it in the back of the car. When asked if she got out of the car and saw Wallach cover the bodies with snow and then screamed, Burns replied that she may have seen the bodies and screamed but she could not remember for sure. At this interview, Burns consented to Officer Werner's request that she undergo hypnosis in order to improve her memory.

On April 13, 1979, Werner and another officer picked up Burns to drive her to Lansing for the hypnosis session. En route, the officers drove Burns to the murder scene, where she stated that this was where she and Wallach had driven the old man. While walking around the grounds, she remembered various details of the night in question, such as paths in the snow going both to the right and left of the garage. The officers then drove Burns to the Crescent Machine Company where Evans' car had been found the day after the bodies were found. Burns was not sure if this was the area where they had picked up defendant on January 27, 1979.

By the time that Rhonna Burns was hypnotized by Dr. Donald Rossi, the various interviews conducted by the police had already suggested to her their theory of the killing. She had been told how defendant looked and was shown a photograph of him. She had been shown a photograph of Elmer Evans. By the police questioning, it was suggested to Burns that she had seen corpses and screamed. After the hypnotic process, Burns recalled seeing the bodies and screaming.

Furthermore, some consideration must be given to Burns as a subject for hypnosis. In Dr. Rossi's opinion, Burns was mildly retarded. This fact may well have exacerbated the already touchy problem of memory distortion while under hypnosis. Rossi acknowledged that a hypnotized person might "recall" events which were merely suggested to them. This can happen particularly where the subject is confused or afraid. It is likely that both words accurately described Burns' mental state. The continual evolution of her version of the events in question demonstrate confusion. At least in the early interviews with the police, Miranda warnings were given to Burns. She might well have believed herself to be a suspect in the killings.

It is clear that Burns' recollections were not desired solely for investigative purposes. See Mack, supra. Basically, what the police sought from her were statements implicating defendant as one of the killers. This was information which could not be independently verified. Even if the police had had Burns hypnotized solely for investigative purposes, the hypnosis session came too late to be free of taint due to the various suggestions made by the police during the various interviews. If for investigative purposes, the hypnosis session should have occurred after the first police interview, if at all.

Reversed and remanded.