Roger Lee McQueen v. Samuel P. Garrison Attorney General of North Carolina, Rufus Edmisten

SPROUSE, Circuit Judge: The State of North Carolina 1 appeals from the district court’s judgment granting a writ of habeas corpus to Roger Lee McQueen. McQueen was convicted of multiple murders in a North Carolina state court in 1977. 2 In this habeas corpus action, he contends that his sixth and fourteenth amendment rights were violated in his 1977 trial by the admission of testimony from a key witness who previously had been hypnotized to assist her recall of the details of the murders. The district court dismissed the fourteenth amendment claim, 3 but agreed with McQueen that the use of the hypnotically enhanced testimony violated his sixth amendment right to confront that witness. 4 We disagree and reverse. I. In 1964 McQueen was convicted of second-degree murder in Missouri. He escaped from a penitentiary of that state in 1972. After wandering through several areas, he traveled to Bennettsville, South Carolina, and stayed with relatives who operated a house of prostitution. There he met two prostitutes: Jacqueline Christine Stanton, known as “J.C.” or “Chris,” and Barbara Kiser. He moved into a house with both women, maintaining a friendly relationship with Kiser and an amorous relationship with Stanton. On April 25, 1972, McQueen married Stanton. In early June, Stanton and Kiser applied for positions as prostitutes in a Fayetteville, North Carolina, house operated by Wilma Norris. Linda Lingle was then working as a prostitute at the Norris establishment. Wilma Norris and Linda Lingle were the two victims murdered in the June 23, 1972, crimes *953 for which McQueen was convicted five years later. Wilma Norris, the “madam,” employed Barbara Kiser to work as a prostitute starting about June 6, 1972, but refused Stanton’s application because she was too old. Stanton, now married to McQueen, found employment as a prostitute at a nearby location, but was soon arrested and jailed — accused of á previously committed burglary. Her bond was set at $1,000. According to trial testimony, it was McQueen’s efforts to obtain money for his wife’s bond that led to the murders of Wilma Norris and Linda Lingle. Although his wife was refused employment at the Norris establishment, McQueen remained friendly with Kiser and became friends with Norris and Lingle. He visited the Wilma Norris house frequently during the month of June 1972— driving there and parking his wife’s Pinto automobile by the house. According to Kiser’s trial testimony, McQueen drove there on June 23, entered the house and initiated the events ending in the murders of Norris and Lingle. McQueen admitted that he was there on the morning of June 23, but swore that he left at noon — well before the women were murdered. Kiser testified as an eyewitness to the murders. She had been hypnotized two weeks prior to trial to help her more clearly recall the details of the crime. After hypnosis, however, she related details differing in one crucial respect from prior statements. 5 In her pre-hypnosis statements, Kiser had related that she heard a shot, entered Norris’ bedroom and saw Norris lying on her bed, wounded in the chest. Kiser also told how McQueen had ordered her to pack her clothes and get in the car. Going to the car, she heard four more shots while she was outside the house. She then reentered the house and encountered McQueen in a hallway. However, while under hypnosis and in her testimony in McQueen’s state trial, she related that she had witnessed McQueen shoot the two women with the last four shots. According to this version, McQueen had threatened and abused Norris and Lingle in an attempt to get $1,000 for his wife’s bond. Kiser heard one shot, entered Norris’ bedroom and found her lying on a bed, wounded. After further abuse, McQueen ordered Norris and Lingle to lie face down on the bed. He then proceeded to fire two bullets into the head of each woman (first Norris, then Lingle) while Kiser stood there and witnessed the executions. Although this central portion of Kiser’s story thus differed, most of the details recalled in her pre-hypnosis statements were identical to her hypnotically enhanced testimony at trial. In both versions, she told of traveling to Bennettsville, South Carolina, with Stanton; of meeting McQueen; of staying in South Carolina and North Carolina; and of her friendly, but not sexual, relationship with McQueen. Moreover, in both accounts she related that McQueen parked his wife’s car by the Norris house in Fayetteville, North Carolina, on June 23, entered, and demanded money and jewelry that Norris and Lingle kept in the house. McQueen shot Norris in the chest after she resisted his efforts to rob the house, but Lingle, seeing Norris shot, succumbed to his demands. He slapped the women several times with his hand and, when Norris sat up on the bed after being shot initially, he hit her with a pistol. Lingle, following his frenzied orders and threats, gathered money and jewelry and gave it to him. McQueen told the women he would not harm them further if they cooperated with him. Kiser’s recollection of events after the murders also showed a consistency between her pre-hypnosis statements and her hypnotically enhanced trial testimony. In both accounts, McQueen had ordered her to pack her clothes and go to the car. Leaving the house, Kiser saw four young boys *954 working in a field behind the house and warned them to retreat. McQueen joined her and they drove away. He handed her a pistol and, following his instructions, Kiser threw it out the window in the vicinity of a stream. 6 The two then left the Fayetteville area and eventually traveled through a number of states. They first stopped in Roanoke, Virginia, where McQueen ordered Kiser to wire his wife $1,000 so she could post bond. They stopped in Chicago where McQueen visited Vera Hignight and renewed his acquaintance with her. 7 They later traveled to Hot Springs, Arkansas, where they met another woman, Vendell Kelly, and took her with them. After arguing with Kelly, McQueen took her to an abandoned field outside Texarkana, Arkansas. Kiser, later summoned by McQueen into the field, found Kelly laying on the ground, naked from the waist down. According to Kiser, Kelly said to McQueen, “You promised that you would not shoot me,” to which McQueen replied, “I lied, Bitch,” and shot her twice in the head while she lay on the ground. During the entire odyssey, according to all of Kiser’s statements, McQueen continually threatened to kill Kiser’s infant child and her parents if she did not cooperate. Kiser finally left McQueen in Indiana. She returned to her parents’ home in Gary, Indiana, and surrendered to the FBI. Later, the FBI fugitive warrant then outstanding against her was dropped and she was given immunity by North Carolina prosecutors, with whom she cooperated both during investigation and trial. McQueen’s defense consisted principally of his own testimony. He stated he was not at the Norris house at the time of the murders, but admitted with variations most of the circumstantial evidence produced by the state against him. Both McQueen and Vera Hignight, his prison visitor and St. Louis lover, testified that Kiser told them on separate occasions that she had killed the two women. 8 There can be no doubt that the trial evidence was sufficient to convict McQueen. In fact, either version of Kiser’s story, supported by the other evidence produced at trial, would have been sufficient to sustain McQueen’s murder convictions. The trial, however, was conducted in 1977, five years after the crimes, because McQueen had been captured and returned to the Missouri prison after the murders of the two prostitutes. In the meantime, Kiser had given statements in 1972 to North Carolina police and in 1977 to Arkansas police 9 which, as we have stated, contained details of the shootings that differed from her 1977 statement while under hypnosis and from her trial testimony given after hypnosis. The state produced considerable other evidence probative of McQueen’s guilt that *955 corroborated much of Kiser’s testimony. Young boys digging potatoes in a field behind Wilma Norris’ house on the day of the murders saw a car matching the description of McQueen’s auto park beside the house, a man exit and enter the house. They later heard women’s screams and gunshots. The state proved that McQueen’s wife was arrested, jailed and placed under $1,000 bond, and that the bail money was wired to her from Roanoke, Virginia, after the murders. Witnesses testified that on the evening of the murders McQueen tried to sell them jewelry matching the description of the jewelry taken from the Norris establishment. The uncontradicted testimony of investigators and pathologists was that both murder victims were “stomach down” on the bed with their hands behind them in a position similar to that of having been tied. Although the hands of the dead women were not tied when viewed by investigators, a coil of rope lay on the bed beside their bodies. 10 Lingle’s face bore a discoloration mark that was consistent with a conclusion that she could have been struck. The pre-murder and post-murder movements and associations of McQueen and Kiser were traced. Pennsylvania state police related the details of McQueen’s capture on the Pennsylvania Turnpike six weeks after the North Carolina murders. He was traveling as a hitchhiker in a truck and was apprehended after jumping from the vehicle and firing a pistol at police. A search revealed that he carried several weapons and ammunition for a .32 caliber pistol. The “slugs” removed from the two victims were .32 caliber but they could not be matched because the murder weapon was never found. After his 1972 capture in Pennsylvania, McQueen was returned to the Missouri prison from which he had previously escaped. In 1977 he was released and brought to North Carolina to stand trial for the 1972 murders. Barbara Kiser also returned to Fayetteville at the request of North Carolina investigators to assist them in preparation for trial and to testify against McQueen. They again searched for the murder weapon but could not find it. Kiser suggested to Major Richard Washburn of the Cumberland County, North Carolina, Police Department that she be hypnotized. She speculated that under hypnosis she might better remember where she threw the gun and other details of the killings. Major Washburn readily agreed to Kiser’s request and she was hypnotized on September 10, 1977, by Joe Raynor, an amateur hypnotist. Raynor was not associated with the Cumberland County police, although he was a North Carolina state senator who chaired a Senate committee on law enforcement and crime control. Ray-nor had no formal training as a hypnotist, nor had he ever hypnotized a person at the request of any police department. At McQueen’s habeas corpus hearing, however, Raynor claimed to have hypnotized thousands of persons in the past twenty-five to thirty years, and to have read widely in the field of hypnosis. The hypnosis of Kiser was recorded on audio tape, but the initial phase of the session, when Kiser was placed in the hypnotic state, was not included. In addition to Raynor, present during the hypnosis were Major Washburn, Ed Grannis (the prosecuting attorney) and Kiser’s mother. While Kiser was under hypnosis, Raynor, Washburn, and Grannis all questioned her about the events of June 23, 1972. To reiterate, although Kiser’s version of the murders, as recollected under hypnosis, was similar in most details to her prior statements, it differed in several respects. Most importantly, under hypnosis she stated that, instead of being outside the house, she was present in the bedroom when the final four shots were fired and she observed McQueen shoot each woman twice in the back of the head as they lay face down on the bed. 11 *956 At trial, Kiser testified about the events of June 23, 1972, in accordance with her memory as refreshed by hypnosis, i.e., that she was an eyewitness to McQueen’s shooting of both women. McQueen’s counsel was told of the hypnosis and given a tape of the session on the opening day of trial, two weeks after Kiser had been hypnotized. His counsel, however, made a tactical decision not to make the hypnosis an issue in the case, fearing that it might result in the jury giving additional weight to Kiser’s testimony. Instead, defense counsel merely cross-examined Kiser in an attempt to impeach her credibility — by pointing to the inconsistencies between her pre-hypnosis and post-hypnosis accounts. Defense counsel, however, objected to the admission of any testimony by Kiser surrounding the events of June 23, 1972, on the grounds that the hypnosis prevented effective cross-examination. 12 After the habeas corpus proceedings were concluded, the district court agreed with McQueen that the state trial court committed error of constitutional magnitude in allowing Kiser’s testimony. The focus of this appeal by the State of North Carolina is, therefore, whether the admission of Kiser’s hypnotically enhanced testimony violated McQueen’s constitutionally guaranteed right to confront witnesses. II. Courts have differed markedly over the trial use of post-hypnosis testimony. Some courts have held that it is per se inadmissible. See, e.g., State v. Peoples, 311 N.C. 515, 319 S.E.2d 177 (1984); State v. Collins, 296 Md. 670, 464 A.2d 1028 (1983); People v. Shirley, 31 Cal.3d 18, 181 Cal. Rptr. 243, 723 P.2d 1354, cert. denied, 459 U.S. 860, 103 S.Ct. 133, 74 L.Ed.2d 114 (1982). Others have held that it is always admissible. Under this view, the fact that a witness has been hypnotized is a question of credibility for the trier-of-fact. See, e.g., United States v. Awkard, 597 F.2d 667 (9th Cir.), cert. denied, 444 U.S. 885, 100 S.Ct. 179, 62 L.Ed.2d 116 and 444 U.S. 969, 100 S.Ct. 460, 62 L.Ed.2d 383 (1979); United States v. Adams, 581 F.2d 193 (9th Cir.), cert. denied, 439 U.S. 1006, 99 S.Ct. 621, 58 L.Ed.2d 683 (1978); United States v. Narcisco, 446 F.Supp. 252 (E.D.Mich.1977). Other courts have held that hypnotically enhanced testimony may be admissible in particular circumstances if the proponent of the evidence can demonstrate that sufficient procedural safeguards were employed to ensure the reliability of the testimony. See, e.g., Wicker v. McCotter, 783 F.2d 487 (5th Cir.), cert. denied, — U.S.-, 106 S.Ct. 3310, 92 L.Ed.2d 723 (1986); Clay v. Vose, 771 F.2d 1 (1st Cir.1985), cert. denied, —U.S.-, 106 S.Ct. 1212, 89 L.Ed.2d 324 (1986); Sprynczynatyk v. General Motors Corp., 771 F.2d 1112 (8th Cir.1985), cert. denied, —U.S.-, 106 S.Ct. 1263, 89 L.Ed.2d 572 (1986); State v. Hurd, 86 N.J. 525, 432 A.2d 86 (1981). In our recently decided case of Harker v. State of Maryland, 800 F.2d 437 (4th Cir. 1986), we adopted a middle ground, holding that hypnotically enhanced testimony may be admissible in appropriate circumstances. We held, however, that to protect a defendant from the dangers of hypnotically induced testimony, the in-court testimony must be shown to be independent of the dangers associated with hypnosis. 13 *957 In Harker, a man shot the victim, Mervyn Thompson, outside Thompson’s home, following a conversation between the two men. Thompson worked with the police to develop a composite sketch of his assailant. Later, the police hypnotized Thompson so that he could better visualize the events surrounding his shooting. Under hypnosis, Thompson gave a description of his assailant that closely matched the description he had given police shortly after the shooting. Several days after the hypnosis, the police showed Thompson ten photographs, from which he selected a picture of David Hark-er and said it resembled his attacker. It later was learned that Harker was to be at the Baltimore County Circuit Court at a particular time. Thompson went to the courthouse, where it was estimated that several hundred people were milling around. Thompson, upon observing Hark-er entering a courtroom, nevertheless identified him as resembling his assailant. With a better view of Harker’s exit, Thompson positively identified him as his assailant. These identifications were admitted at trial over the objection that they had been tainted impermissibly by the hypnosis session. We affirmed the district court’s denial of Harker’s habeas corpus petition. In rejecting Harker’s fourteenth amendment claim, the majority recognized the deficiencies in the hypnosis procedures. It found, however, that Thompson’s post-hypnosis identification of Harker had a basis that was independent of the dangers associated with hypnosis because, prior to hypnosis, Thompson had helped police develop a composite sketch that even Harker’s father admitted closely resembled his son. In rejecting Harker’s sixth amendment claim, the majority reasoned that the jury had been apprised of the uses and dangers of hypnosis, the trial court had permitted full exploration of the hypnosis session, and Thompson had been cross-examined on his opportunity for pre-hypnosis observation. In his concurring opinion, Judge Murnaghan referred favorably to the standards adopted by the Supreme Court of New Jersey in State v. Hurd, 86 N.J. 525, 432 A.2d 86 (1981). 14 When evaluated under these guidelines, he felt the police procedures used in hypnotizing Thompson were inadequate whether considered under either the sixth or fourteenth amendments. Judge Mumaghan concurred, however, because in his opinion Thompson’s identification of Harker was sufficiently reliable, apart from any hypnotic influence, so that the admission of Thompson’s hypnotically enhanced testimony was harmless error. Harker, like most reported “hypnosis” cases, involved a witness’ identification of a suspect after the witness had been hypnotized. Such cases are complicated if for no other reason than they apply judicial techniques to unproven scientific theories relating to the memory and other components of the human mind. Kiser’s testimony that McQueen shot the two prostitutes presents even greater complications. Identification normally involves few mnemonic exercises. 15 Kiser’s testimony pertained to highly *958 traumatizing events that occurred over a period of forty-three days. She was faced with recalling hundreds of details. In Harker we said that, in considering hypnotically enhanced testimony, a court must conduct a balanced inquiry to determine if the testimony had a basis that was independent of the dangers associated with hypnosis — in other words, a balanced inquiry to determine whether a witness’ memory and ability to testify from it was distorted by the earlier hypnosis. The balanced inquiry contemplated in Harker cannot be circumscribed by narrow considerations, and Judge Wilkinson’s well-written opinion demonstrates that these appeals must be determined by a detailed factual analysis on a case-by-case basis. Our consideration of this appeal confirms that conclusion. In Harker, we did not adopt the Hurd guidelines as a litmus test for determining the reliability of pre-trial hypnosis — nor do we here. Hurd, however, certainly represents the type of general reliability inquiry that must be made and, in our view, each of the six potential safeguards suggested in Hurd is relevant to the hypnosis procedures now under review. 16 As we held in Harker, however, a court cannot necessarily rest solely on the reliability vel non of the hypnosis procedures in ruling on the admissibility of the proffered testimony. Even though all of the Hurd safeguards might be employed, the defendant may still be able to demonstrate by expert testimony that a witness’ memory has been irreparably distorted by hypnosis. See Diamond, Inherent Problems in the Use of Pretrial Hypnosis on a Prospective Witness, 68 Calif.L.Rev. 313, 332-42 (1980). On the other hand, even if the hypnosis procedures are flawed, a trial or appellate court might discern that a witness’ testimony was nonetheless independent of the dangers associated with hypnosis. Harker, 800 F.2d at 843. As in Harker, the procedures used in Barbara Kiser’s hypnosis were far from ideal. Kiser was hypnotized by an amateur hypnotist who possessed no formal training and no prior experience in conducting hypnosis sessions in conjunction with police investigations. Washburn, the chief investigating police officer, and Grannis, the prosecuting attorney, were present and questioned Kiser while she was under hypnosis — a factor hardly assuring the independence of the hypnotist. None of the information regarding Kiser’s prior recollection of June 23, 1972, was recorded before the session. The session was not videotaped, and the audio tape (the only recording of the session) failed to include the initial time period in which the hypnotist put Kiser in a hypnotic state. Our inquiry thus begins with a strong inference that Kiser’s testimony was tainted by hypnosis because of the faulty procedures used by the hypnotist. 17 *959 We, of course, strive to achieve an objective standard in determining the kind and quantum of evidence necessary to rebut an inference of prejudice arising from faulty hypnosis procedures. Many courts wrestling with the dangers of hypnotically enhanced testimony do not face this task. Obviously, it is not a problem for those courts that automatically permit or deny the use of such testimony. Likewise, courts following the Hurd rationale reject the testimony once it is demonstrated that the procedures used were faulty. In our view, however, the reliability of the hypnosis procedure does not assure admissibility, nor does demonstrated unreliability necessarily mean the hypnotically enhanced testimony is not admissible. Harker, 800 F.2d at 443. We have, then, the task of determining what will rebut inferences flowing from evidence that a particular hypnosis session is good or bad. As we have stated, some courts take the position that this is simply a question of credibility for the jury to determine. In Harker, the fact that the jury had considered such reliability evidence weighed in our equation that found Thompson’s testimony to be independent of the dangers associated with hypnosis. There were other factors, however. Prehypnosis and post-hypnosis evidence corroborated Thompson’s post-hypnosis identification. For example, he assisted in the drawing of a composite sketch shortly after the incident and, after hypnosis, he identified Harker in a crowd of people under circumstances suggesting that the identification was independent of any dangers associated with hypnosis. Here, we do not have the advantage of the jury’s credibility determination. Moreover, there is the unassailable fact that Kiser changed a critical part of her story after being hypnotized. On the other side of the ledger is considerable evidence corroborating the altered version of Kiser’s recall — evidence probative of a finding that the crime happened as she related in her eyewitness account. If this evidence is believed (and some of it is irrefutable physical evidence), it not only corroborates evidence pointing to McQueen’s guilt, 18 but it is direct evidence that if Kiser’s memory was affected by hypnosis, it was positively affected. If Kiser is to be believed, she could not remember the actual shooting until placed under hypnosis. Of course, evidence that events occurred as she related them does not prove that hypnosis did not affect her memory. To the contrary, it may be probative of a finding that her memory was correctly restored by hypnosis, rather than distorted by it. It is also probative, however, of a finding that she was not testifying from suggestions made to her during hypnosis, that she had not engaged in confabulation, and that her memory had not hardened. Looking then to evidence corroborating Kiser’s post-hypnosis testimony, we first note that her trial testimony is generally consistent with her pre-hypnosis statements, varying only in the description of the actual shooting. Her descriptions of events before and after the murders are identical. Her descriptions of McQueen’s behavior on June 23, the day of the murders, are identical except as to the actual shooting. In all versions, she related McQueen’s brutal treatment of Norris and Lingle, his robbing them to obtain bond money for his wife, his announcement of his intention not to harm them if they cooperated, and his initial wounding of Norris. Importantly, considerable circumstantial evidence corroborated Kiser’s testimony. She testified that McQueen ordered Lingle to obtain rope so he could tie the hands of *960 the two women while they lay on the bed, that Lingle brought the rope, that both women lay on their stomachs on the bed, and that McQueen shot them in the back of the head. Investigators testified that a coil of rope lay on the bed next to the bodies and the women’s wrists were positioned as if they had been tied behind their backs. Face marks indicated that Lingle had been struck — as Kiser testified. The pathologist testified that the fatal shots had been fired from about twenty-four inches from the women’s heads. Kiser testified that McQueen ordered her to place a telephone in a sink of water. Investigators found it there. She described McQueen’s ordering Lingle to search for Norris’ jewelry and her bringing it to him. Independent witnesses testified that McQueen attempted to sell them the same jewelry later that day. Pennsylvania state police testified that McQueen possessed some of that jewelry when captured. All of this evidence is probative of the factual conclusion that Kiser was testifying from observed facts, as distinguished from facts either artificially implanted in her memory by hypnosis or created by her in an effort to bolster her story. No appellate court has gone, nor would we go, so far as to judge whether hypnotically enhanced testimony was free of the dangers associated with hypnosis solely from a review of the witness’ testimony. Such tests would too heavily involve the subjective judgment of reviewing courts. However, a review of Kiser’s testimony against the background of evidence corroborating the veracity of her recall is illuminating. For example, responding to questions concerning her reason for requesting hypnosis, she replied: I asked [Major Washburn] if he would have — make arrangements to have me hypnotized so that when this thing all came to court that I could tell the honest to god truth, not something that I believed and maybe had been a mistake. Sometimes I knew I saw him kill them and sometimes I really knew that I didn’t. I just — and each time I was sure, you know. I couldn’t — I couldn’t remember, it was driving me crazy. I had bad dreams for a long — for like a.year, year and a half after the murders happened and drank a lot so I could sleep. In explaining why the details previously had been blocked from her mental processes, she said: I always blame myself for them girls getting killed because when he came in, I always though [sic] in my mind if I would have packed my clothes when he told me to pack my clothes that he wouldn’t have killed anybody. Her responses generally were not the automatic responses of a preconditioned mental process. For example, she appeared to exhibit sincere uncertainty in both her pre-hypnosis and post-hypnosis recall of the initial shooting of Wilma Norris. Kiser testified that she and Lingle went to Norris’ bedroom, but she was unsure which one entered first. (“I believe Linda, but I don’t really know____ It couldn’t have been more than a couple of seconds either way.”) Kiser testified on direct examination that, after shooting Norris, McQueen then slapped Norris; but Kiser was unsure whether he slapped Norris when she reached across the bed toward the nightstand, or at some later point. Kiser also was unable to remember whether McQueen slapped Norris with his hand or a gun. Under cross-examination, however, she testified that McQueen first slapped Norris “by the bedroom” with his open hand and that he later hit her with the pistol while she lay on the bed. In her testimony concerning McQueen’s robbery of the jewelry, Kiser exhibited the same mental fluctuations. She stated that McQueen told Lingle to bring in jewelry and the keys to Norris’ strongbox, although she did not remember whether Lingle ever found the keys. Later, she testified that she saw McQueen counting the money from the strongbox. She did not testify, however, on how the box was opened. On another matter, Kiser testified that she put a phone in the kitchen sink, but she could not remember which phone. On cross-examination, McQueen’s counsel asked Kiser whether she remembered *961 what happened under hypnosis. Kiser replied, “I don’t remember every exact word for word that I said,” but went on to add, “I recall being able to relive in my mind seeing him stand there and go bang and kill them.” Kiser recognized that she had testified previously to seeing a gun in McQueen’s hand when she first entered Norris’ bedroom. She admitted on cross-examination, however, that she did not know whether her prior statement to Major Washburn had mentioned this fact. Kiser did not know when Norris and Lingle were positioned on their stomachs, only that she had observed them in that posture. (“I don’t remember if they rolled or what. I remember they were on their stomachs____ Everything was happening so fast. It’s not like this was a two-day drawn out thing in slow motion. It all happened very, very rapidly and I’m trying my best to remember exactly how it happened.”) She was also unsure how much time passed between the time Norris was initially shot in the chest and the time Norris and Lingle were shot in the head. (“Not too long, I don’t know, as I said, everything was happening so fast, I wasn’t looking at a clock. I wasn’t trying to tell the time. I was trying to figure what the hell was going on around me. I never seen anybody get shot before, not in the chest, not in the head, not anywhere.”) Throughout her testimony, Kiser exhibited the characteristics normally expected from a witness recalling details of facts five years after their occurrence. Significantly, she was uncertain about some of the minor details that she had clearly remembered in 1972. Kiser’s testimony, in general, however, portrays a witness in possession of her critical faculties. There is no indication of a memory unshakably frozen by hypnosis. She exhibited reasoned judgment in these responses and her disagreements with questioning counsel appear to be from independent reasoning rather than from photographic images. In short, her general testimony presents the appearance that she testified independent of possible suggestion from the hypnotist and that her memory had not “hardened.” There also remains the danger of “confabulation” — the hypnosis-related phenomenon whereby a subject falsely adds details to present events in a manner favorable to her self interest. McQueen testified that Kiser admitted shooting the women. If that were true, it would have given Kiser urgent reason for confabulation or simple prevarication. The largely uncontroverted evidence corroborating her version of the shootings, however, effectively negates the presence of “confabulation.” In considering the probative nature of Kiser’s testimony, as well as that of the corroborative evidence, it is important to keep our chore in the proper perspective. Our task is to determine if the hypnosis could have affected her ability to testify and respond freely to cross-examination. In making that determination, we assume Kiser sincerely was trying to reconstruct the crime. It is always possible, of course, that a witness could act from improper motivation. As the Supreme Court of North Carolina intimated, Kiser, in asking to be hypnotized, and in changing her recall to include the actual witnessing of the murders, possibly could have acted from some improper motivation. McQueen, 244 S.E.2d at 428. As the North Carolina court noted, she had been immunized from prosecution. Id. It is always possible that if Kiser had been indicted, a jury might have inferred that she and McQueen both participated in the robbery and murders. The jury, of course, was told by McQueen and his lover, Hignight, that Kiser admitted to them that she killed the two women. See supra footnote 8. Whether she was protecting herself, however, is not the question. That was a credibility issue resolved by the jury. As we have stated, our principal task on this appeal is to determine whether hypnosis could have affected Kiser’s ability to testify and respond freely to cross-examination. Despite the faulty character of the hypnosis, the record demonstrates that she testified from a basis that was independent of the dangers associated with it. We conclude, therefore, that the admission of Kis *962 er’s hypnotically enhanced testimony did not violate McQueen’s sixth amendment right to confront witnesses. In view of the above, the judgment of the district court is reversed and remanded for further proceedings consistent with this opinion. REVERSED AND REMANDED. Footnotes [1] The named appellants in this action are Samuel P. Garrison and Rufus L. Edmisten. When this action was commenced, Garrison was the Warden at Central Prison, Raleigh, North Carolina, and Edmisten was the Attorney General of North Carolina. [2] McQueen’s conviction was affirmed by the Supreme Court of North Carolina in State v. McQueen, 295 N.C. 96, 244 S.E.2d 414 (1978). [3] McQueen v. Garrison, 619 F.Supp. 116 (D.C.N. C.1985). The district court adopted the magistrate’s recommendation that the claim be dismissed because McQueen did not receive a fundamentally unfair trial. Id. at 117. [4] McQueen v. Garrison, 617 F.Supp. 633, 638 (D.C.N.C.1985). [5] Kiser initially gave a statement to North Carolina police in August 1972. Later, on September 6, 1977, she gave a statement to Arkansas police investigating a murder McQueen was alleged to have committed in that state. At trial, she testified to the version of events as recollected while under hypnosis. [6] She later searched for the weapon with the police but it was never found. [7] Mrs. Hignight testified that she learned of McQueen through a mutual acquaintance while he was incarcerated in Missouri in 1969. She began corresponding with him by mail in late 1970 and first visited him in prison in 1971. She then began visiting McQueen regularly in an attempt to convert him to Christianity. When McQueen escaped in February 1972, Hignight met him in St. Louis and spent the night with McQueen in a local motel. (According to Hignight, she had sexual relations with McQueen on a dozen occasions.) Two days later, she drove him to Chicago and, shortly thereafter, to Kentucky. In the month following McQueen’s escape, Hignight saw him in Kentucky on one or two other occasions. She did not see him again, however, until the. end of June. McQueen called her on June 29 and asked her to come to Chicago. She went to Chicago, met McQueen and Kiser, and traveled with them until July 3. She finally left McQueen and Kiser in Little Rock, Arkansas, but, before leaving, she rented them a car in her name. [8] McQueen testified that Kiser told him about the murders several days after they left Fayetteville. During a conversation about the pair’s finances, Kiser said "[djon’t worry about money. We have plenty____” She said she "killed the bitches” and got a large sum of money from Norris’ place. Hignight testified that Kiser told her that she [Kiser] had shot the women and later threatened Hignight that Til blow your brains out just like I did them" if Hignight bothered McQueen. [9] See supra footnote 5. McQueen was sought in Arkansas because of his alleged connection with the previously described murder of Kelly. [10] Kiser testified that McQueen had ordered Lingle to bring him some rope so that he could tie up Lingle and Norris. [11] In addition to remembering that she witnessed the murders, while under hypnosis Kiser also recalled (for the first time) being slapped around by McQueen prior to the shootings and being told by McQueen to place a telephone in a *956 sink filled with water. Officials investigating the murders found a telephone in a sink filled with water. [12] This objection was overruled by the trial court. [13] Judge Wilkinson's well-considered opinion in Harker discusses three testimonial dangers associated with hypnosis: suggestibility, confabulation, and memory hardening. Suggestibility refers to the possibility that the hypnotized person may be susceptible to suggestions from the hypnotist or questioner. "[T]he subject may give the response he or she thinks the hypnotist [or questioner] wants to hear.” Harker, 800 F.2d at 440. Confabulation involves a subject’s tendency to fill in missing details in an attempt to make his account more plausible and coherent. "Memory hardening” is what the term suggests. There is a danger that the hypnotist may so firmly implant an impression that a witness is incapable of reexamining it under cross-examination or otherwise. As both the majority and concurring opinions in Harker observe, it would be nearly impossible to elicit independent testimony from a witness impaired by all three phenomena. [14] In Hurd, the Supreme Court of New Jersey established the following guidelines for evaluating the admissibility of hypnotically enhanced testimony. First, the hypnosis session must be conducted by a trained psychiatrist or psychologist. Second, the person conducting the session must be independent of the police investigation, the prosecution and the defense. Third, the hypnotist's knowledge of the facts in the case should be recorded. Fourth, before hypnosis, the hypnotist should obtain a detailed description of the subject’s current memory of the event. Fifth, the hypnosis session should be recorded — videotaping is strongly recommended — so a permanent record exists to ensure against suggestive procedures. Finally, only the hypnotist may interact with the subject just prior to, during, and just after the session, and no one else should be in the room during the session unless their presence is essential. Hurd, 86 N.J. at 545-47, 432 A.2d at 96-97. [15] "Hypnosis” cases dealing only with the identification of a defendant can involve similar due process issues as those confronted in identification cases generally, as well as issues involving sixth amendment confrontation rights. Here, however, where a witness’ recall concerning all or most of the details of a crime has been hypnotically enhanced, possibly injected into the trial is a question of whether the defendant’s due process rights were violated because the use of hypnotically enhanced testimony may have resulted in a fundamentally unfair trial. See Grundler v. North Carolina, 283 F.2d 798 (4th *958 Cir.). The same sixth amendment problem exists in both testimonial circumstances, however. [16] Kiser testified on direct examination that she had been hypnotized and that her testimony was based on her memory of the events of June 23, 1972, as refreshed through the hypnosis. In Harker, Thompson was cross-examined on the hypnosis procedures and other evidence established the complete details surrounding the hypnosis session, so that the jury could weigh the possible affect hypnosis might have had on Thompson’s testimony. Although McQueen’s counsel mentioned the hypnosis, it was only in the context of cross-examining Kiser on the inconsistencies between her pre-hypnosis and post-hypnosis versions of events. As a result, the jury had only a sketchy account of what transpired during Kiser’s hypnosis. However, having deliberately waived the right to present evidence on the inadequacies of hypnosis in general, and Kiser's testimony in particular, McQueen cannot now be heard to complain of this deficiency. There is, however, still presented here the core issue present in every “hypnosis” case— should the jury have been permitted to hear a witness after she had been hypnotized. Once a court properly determines such evidence to be admissible, however, a defendant can, as here, waive his right to present testimony upon which a jury could determine the hypnotically enhanced testimony to be incredible because it was affected by hypnotic influences. A thin but necessary line thus exists between the court’s examination of the hypnosis procedures from a legal sufficiency standpoint and the credibility determinations by the jury. [17] As we noted in footnote 16, supra, McQueen waived his opportunity to present to the jury details of the hypnosis session and expert opinion concerning the effects of hypnosis. We *959 consider, however, whether the jury should have been permitted to hear Kiser’s testimony. [18] It again should be noted that the evidence we consider is in the context of our inquiry as to whether Kiser’s testimony was independent of the dangers associated with hypnosis. Quite a different matter is other evidence generally corroborative of McQueen’s guilt but not connected directly to Kiser’s testimony. Within our inquiry to determine the basis of Kiser’s testimony, the only helpful corroborative evidence is that which bears directly on the accuracy of her testimony. Other generally corroborative evidence, however, might well go to the issue of harmless error.