BROCK, C.J.

The State appeals the Superior Court's (Groff, J.) ruling that the testimony of two alleged sexual assault victims is not admissible in criminal prosecutions against the defendants, Joel Hungerford and John Morahan. See RSA 606:10, II (1986). We affirm and remand.

For our limited review of the underlying facts, we will rely on the findings that the trial court made for purposes of its ruling on the admissibility of the complainants' testimony. The complainant in State v. Hungerford, Laura, is a woman in her late twenties who had suffered from symptoms of clinical depression and had experienced sexual problems in her marriage prior to entering psychotherapy. Although she had no memory of being abused by her father, defendant Hungerford, she sought therapy in September 1992 after her sister claimed to have recovered memories of being sexually abused by Hungerford. Laura began therapy with Susan Jones, a social worker. According to the trial court, Laura

explained to Ms. Jones that one of her motivations in entering therapy was to explore the possibility that she was sexually abused. Ms. Jones engaged in traditional psychotherapy but did engage specifically in "memory retrieval." In other words, a specific purpose of her psychotherapy with Laura was in part to retrieve or recover memories of possible or suspected sexual abuse.

Laura participated in psychotherapy for approximately nine months, including about one hundred sessions; during this period, she "remembered" several episodes of sexual abuse. She recovered several memories of her father penetrating her, digitally, vaginally, and anally. She recovered these during sessions with Ms. Jones, who instructed her "to close her eyes and focus on the image," and to report "who she was afraid of." In order to facilitate remembering, Laura was instructed to "close her eyes and pretend it was a movie," or to "look around and see what happened."

In March 1993, Laura experienced vaginal pain and a feeling of disgust with her body while taking a shower. A green bar of soap reminded her of a poster above her bed at the family home. She subsequently experienced an "image or flashback" which reminded her that two days before her wedding her father had entered her bedroom, ripped the covers off of her bed, and raped her. Part of this memory seems to have been recalled outside of a particular therapy session, although Ms. Jones did examine the "feeling" with Laura during therapy, and the trial court found that "Laura did recover part or all of the memory of that rape at a therapy session."

Laura reported each of these memories to the Amherst Police Department in March 1993. After the allegations had been made, Hungerford threatened to shoot himself, Laura, and Ms. Jones. Laura and Ms. Jones were aware of this threat. After the defendant made the threat, Laura had a nightmare about black hair, which, after she had drawn a picture of it, Laura recognized as her father's beard. After being instructed to close her eyes, "[l]ook around," and see what was "so terrifying," Laura remembered being tied to a bed with her father beside her, and that something was inside of her vagina. She remembered later, at home, that the object in her vagina had been a gun.

The trial court described Ms. Jones' memory retrieval techniques in some detail:

During these periods when Ms. Jones engaged in the process of memory retrieval, Laura would close her eyes for 15 or 20 minutes, during which the "memory" would be explored. According to Ms. Jones, during these periods, Laura would go into a "self-induced" trance. Ms. Jones indicated that she did not induce the trance with Laura, but rather Laura was able to "enter the traumatic experience by her own access and design." During these episodes, Ms. Jones would ask Laura if she could see or hear anything or anybody, or if anything was happening. These were the only times during therapy that Ms. Jones used this "visualization" technique.

Ms. Jones fashioned or relied on a so[-]called "Repressed Memory Syndrome[,"] which appears unrecognized in the field of psychology. Ms. Jones also believed that dreams are often the first signs of emerging memory, that flashbacks are a sudden reliving of a scene of sexual abuse, and that violent nightmares are a red flag for the existence of sexual abuse. Ms. Jones also described the concept of repression to Laura. Ms. Jones believed that Laura's visualizations were memories of actual abuse. She believed that these incidents of abuse occurred, and by her conduct, communicated this belief to Laura. Laura believed that Ms. Jones indicated that body pains were connected to instances of past sexual abuse. Laura believed that Ms. Jones validated the reality of the remembered sexual abuse. Ms. Jones herself understood that by her actions she had validated the abuse and affirmed the memories for Laura.

The complainant in State v. Morahan, Sarah, presently is in her early twenties. She reported to the trial court that she had negative feelings about her parents, who had divorced, and reported "suffering from depression, narciss[ism] and bulimia at various stages of her life." Sarah began psychological counseling in May 1988. After having been admitted to two private psychiatric hospitals with suicidal ideation, Sarah attended the DeSisto School, a gestalt, "`therapeutic boarding school[,'] where students are required to attend psychotherapy." While at the school, Sarah ceased taking the antidepressant medication that had been prescribed for her previously.

At the DeSisto School, Sarah reported a recurring dream of a man next to her in bed, and suspected she had been abused; at the same time, she expressed conflicting feelings about her grandfather. In 1991, after another hospitalization for suicidal ideation, she reported further suspicions of sexual abuse; the school provided Sarah with "inner child therapy" to "support her and offer her belief and comfort, and to determine what had happened." Sarah reported that her stepfather might have "done something to her," and much of her therapy during this period explored the possibility of sexual abuse.

After several months of therapy, in July 1991, Sarah revealed in therapy that she "now remembered being raped by a teacher in the seventh grade." According to the trial court:

In the next few months, Sarah continued to deal with the alleged rape and develop her memory. She indicated that she had become pregnant as a result of the rape and aborted the pregnancy by an overdose of steroids. In therapy, significant effort was directed to grieving over the abortion. One of Sarah's therapists considered her the classic abused child. Sarah also attended dorm group sessions on a regular basis at which participants talked of their problems, including cases of sexual abuse. On June 8, 1993, Sarah[,] accompanied by DeSisto School personnel, reported the rape to the vice-principal at Hillside Junior High School. Thereafter, she recovered further memories of the rape while talking with her dorm supervisor and a therapist. In mid-August Sarah reported the rape to the Manchester Police.

Sarah was not questioned about the details of the assault at the hearing. The instant prosecutions followed.

Both defendants moved to dismiss the prosecutions, asserting that the complainants' testimony would not be admissible at trial under State v. Cressey, 137 N.H. 402, 628 A.2d 696 (1993), New Hampshire Rule of Evidence 702, State v. Coolidge, 109 N.H. 403, 260 A.2d 547 (1969), rev'd on other grounds, 403 U.S. 443 (1971), or Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). The two cases were consolidated for purposes of addressing the admissibility of the complainants' "repressed memory" testimony. Over the State's objection, the trial court ruled that a preliminary hearing was required to address this question, and that the State would bear the burden of demonstrating that the phenomena of memory repression and recovery are reliable and have gained general acceptance in the psychological community. The court concluded that "[t]estimony that is dependent upon recovery of a repressed memory through therapy cannot be logically disassociated from the underlying scientific technique." The court required the State to demonstrate "that the reasoning or methodology underlying the testimony is scientifically valid; and that it is capable of empirical testing and can properly be applied to the facts in issue."

The court held a two-week admissibility hearing on the issue of repressed memories. The two complainants testified at the hearing, as did seven psychological professionals: Dr. Daniel Brown, Dr. Bessel A. van der Kolk, Dr. Jon Robert Conte, Dr. Elizabeth Loftus, Dr. Paul McHugh, Ms. Susan Jones, and Dr. James Hudson. After the hearing and a review of the materials admitted during the hearing, the trial court defined a repressed memory as "the complete absence of awareness or memory of a traumatic event from the time of its occurrence until a period of years thereafter." See E. LOFTUS K. KETCHAM, THE MYTH OF REPRESSED MEMORY: FALSE MEMORIES AND ALLEGATIONS OF SEXUAL ABUSE 215-17 (1994) (describing repression as "a process of selective amnesia in which the brain snips out certain traumatic events and stores the edited pieces in a special, inaccessible memory `drawer'"). The court ruled that the State failed to meet its burden of proving that there was general acceptance of the phenomenon of repressed memories in the psychological community, and, further, that the State had failed to demonstrate that the phenomenon was reliable. The court accordingly ruled the testimony of the complainants inadmissible. This appeal followed.

On appeal, the State argues (1) that the trial court erred in requiring a preliminary showing of reliability or general acceptance before the witnesses' testimony would be admitted, and (2) that, assuming the preliminary showing was required, the trial court erred in concluding that the State failed to make such a showing. The issues raised in these arguments are intimately related, and we address them together.

We accord the trial court's rulings on evidentiary matters considerable deference, reversing only for an abuse of discretion. See, e.g., State v. Briere, 138 N.H. 617, 620, 644 A.2d 551, 554 (1994). We review the trial court's determination of the reliability of novel scientific evidence with similar deference, see Cressey, 137 N.H. at 405, 628 A.2d at 698, although we review the reliability or general acceptance of novel scientific evidence independently when the determination is not likely to vary according to the circumstances of a particular case, see State v. Vandebogart (DNA), 136 N.H. 365, 376, 616 A.2d 483, 491 (1992). The level of scrutiny we employ in our reliability inquiry will depend upon the complexity of the evidence involved and the impact the evidence likely will have on the trial itself. See, e.g., State v. Murphy, 451 N.W.2d 154, 157 (Iowa 1990).

We agree with the State that lay witnesses are presumed competent to testify, see N.H. R. Ev. 601(a), unless they "lack sufficient capacity to observe, remember and narrate as well as understand the duty to tell the truth," N.H. R. Ev. 601(b). We disagree with the State's assertion, however, that the trial court's decision "was tantamount to an erroneous ruling that [the witnesses] were incompetent" to testify. There seems to be no question that Laura and Sarah believe that they "remember" the events they describe; accordingly, a pure competence inquiry would likely result in a conclusion that their testimony would be admissible. See N.H. R. Ev. 601; Briere, 138 N.H. at 620-21, 644 A.2d at 554. An inconsistency in testimony or the failure to remember aspects of some observed event typically does not disqualify a witness on competence grounds; such gaps in testimony "present questions of credibility for resolution by the trier of fact." Briere, 138 N.H. at 620, 644 A.2d at 554 (quotation omitted).

The present inquiry is in part a question of competence, however, insofar as we are inquiring into the ability of these witnesses to "remember" the events that they seek to describe at trial. See N.H. R. Ev. 601(b); State v. Iwakiri, 682 P.2d 571, 578-79 (Idaho 1984) (treating admissibility of hypnotically-refreshed testimony as competence question); see also State v. Mack, 292 N.W.2d 764, 769 (Minn. 1980). Because these witnesses are not ordinary eyewitnesses with ordinary memories, we must examine the reliability of their "memories," cf. People v. Hughes, 453 N.E.2d 484, 494 (N.Y. 1983) (inquiring into reliability of hypnosis as means of restoring recollection under general acceptance test), just as we inquire into the reliability of scientific evidence, see Cressey, 137 N.H. at 405, 628 A.2d at 698 (expert testimony must reach threshold level of reliability to be admissible). We acknowledge that our inquiry is, in part, into the trial court's determinations concerning the reliability of the victims' memories. Their memories may be actual recollections of actual traumatic events, manufactured narratives of events that never occurred, or some combination of these. See Loftus, The Reality of Repressed Memories, 48 Am. PSYCHOLOGIST 518, 524-25, 533 (1993) [hereinafter The Reality of Repressed Memories]. Laura and Sarah may not know into which category their memories fit. See Wells Murray, Eyewitness Confidence, in EYEWITNESS TESTIMONY: PSYCHOLOGICAL PERSPECTIVES 155, 159-70 (G. Wells E. Loftus eds., 1984) (eyewitness confidence in memory not meaningfully related to accuracy).

Just as our inquiry is not purely one of competence, it is not purely a question of the admissibility of scientific or expert evidence, to be governed solely by reference to New Hampshire Rule of Evidence 702. We recognize that treating the testimony of a percipient witness to a crime as scientific evidence is novel in our law, and that it does not fit precisely within the confines of Rule 702. See Com. v. Kater, 447 N.E.2d 1190, 1195 (Mass. 1983). We agree with the trial court, however, that a recovered memory that previously had been completely absent from a witness's conscious recollection, see E. LOFTUS, MEMORY 41-44 (1980) [hereinafter Memory], cannot be separated from the process, if any, that facilitated the recovery. See People v. Zayas, 546 N.E.2d 513, 518 (Ill. 1989). In this context, "[t]he basic question is not so much whether the process is scientific but rather whether a jury can realistically evaluate the effect of [the process]," Kater, 447 N.E.2d at 1195, on the witness's ability to testify to his or her recollection of an event. See The Reality of Repressed Memories, supra at 523-25. The trial court's gatekeeping power on questions of the admissibility of scientific evidence is the most appropriate procedural tool for evaluating this sort of evidence. See N.H. R. Ev. 104(a); State v. Quattrocchi, 681 A.2d 879, 884 (R.I. 1996). We accordingly conclude that, when challenged, testimony that relies on memories which previously have been partially or fully repressed must satisfy a pretrial reliability determination. See Quattrocchi, 681 A.2d at 884. The trial court correctly ordered the pretrial hearing on admissibility.

The State vigorously argues that the processes of repressing and retrieving memories are normal human functions, common to every person's everyday experience, just as forgetting and remembering are; accordingly, the State contends, such evidence is not beyond the average juror's ability to comprehend, and unique treatment is inappropriate. We disagree. Although there are skeptics, it does seem to be accepted in the psychological community that people are capable of repressing or dissociating conscious recollection of all or part of certain traumatic events. See, e.g., Ernsdorff Loftus, Let Sleeping Memories Lie? Words of Caution About Tolling the Statute of Limitations in Cases of Memory Repression, 84 J. CRIM. L. CRIMINOLOGY 129, 133-34 (1993) [hereinafter Sleeping Memories]; Pope Hudson, Can Memories of Childhood Sexual Abuse Be Repressed?, 25 PSYCHOL. MED. 121, 121 (1995); Taub, The Legal Treatment of Recovered Memories of Child Sexual Abuse, 17 J. LEGAL MED. 183, 187 (1996). There is, however, a vigorous debate on the questions of how the process of repression occurs, how the process of retrieval occurs, and indeed if in fact retrieval is possible at all. See Ault v. Jasko, 637 N.E.2d 870, 875-76 (Ohio 1994) (Wright, J., dissenting); Hough, Recovered Memories of Childhood Sexual Abuse: Applying the Daubert Standard in State Courts, 69 S. CAL. L. REV. 855, 859-63 (1996). A central and divisive question in this debate is whether a person's memory of an event can be accurate or authentic or "true," having been long lost in the person's subconscious mind and subsequently remembered, either spontaneously or by some method seeking to recover the memory. See The Reality of Repressed Memories, supra at 523-29; Taub, supra at 189-91. See generally Pezdek Roe, Memory for Childhood Events: How Suggestible is It?, 3 CONSCIOUSNESS COGNITION 374, 380-83 (1994).

The phenomenon of repressing recollection of a traumatic event, and subsequently "recovering" it, may be familiar to or even accepted by parts of the psychological community, but it is far from being familiar to the average juror. See Com. v. Crawford, 682 A.2d 323, 326 (Pa.Super.Ct. 1996). Some well-publicized accusations may ensure that many people have heard of the concept of repressed memories, see LOFTUS KETCHAM, supra at 79-80 (describing mass media reports of celebrities recovering previously repressed memories); a review of the scientific literature on the subject reveals, however, that ordinary jurors cannot be expected to analyze such claims without the assistance of experts. See Sleeping Memories, supra at 162-63; Crawford, 682 A.2d at 325.

Our case law is clear that if the subject matter in dispute is beyond the general understanding of a jury, the party bearing the burden of proof must adduce expert testimony to explain such evidence. Lemay v. Burnett, 139 N.H. 633, 634-35, 660 A.2d 1116, 1116-17 (1995); see Crawford, 682 A.2d at 324-25. Further, expert testimony is required when the issues in a case are particularly esoteric or when the matter to be determined by the trier of fact is so distinctly related to a particular science, occupation, business, or profession that it is beyond the ability of the average layperson to understand. Wood v. Public Serv. Co., 114 N.H. 182, 186, 317 A.2d 576, 578 (1974); see Lemay, 139 N.H. at 634-35, 660 A.2d at 1116-17.

The offered testimony of Laura and Sarah in the instant cases, even if admitted, could not be understood by the average juror without the assistance of expert testimony. Their memory of the events described above, according to the theory, has undergone a physiological process unlike ordinary memory, with which an average juror would be familiar. Compare Wood, 114 N.H. at 186-88, 317 A.2d at 578-79 (level of insulation required for safe high voltage power lines within understanding of the average juror) with Lemay, 139 N.H. at 635-36, 660 A.2d at 1117-18 (jurors could not assess whether particular diving conditions rendered backyard swimming pool unreasonably dangerous without expert testimony). Further, in the instant cases, the memories are intricately related to the psychological therapy attendant to their recovery. See Note, Recovered Memories of Childhood Abuse: Should Long-Buried Memories Be Admissible Testimony?, 37 B.C. L. REV. 591, 630-35 (1996) [hereinafter Long-Buried Memories]. Even though the General Court has referred to the phenomenon, it remains outside of the understanding of the average juror. See Laws 1990, ch. 213:1 (findings supporting extension of statute of limitations for certain sexual assaults); McCollum v. D'Arcy, 138 N.H. 285, 289, 638 A.2d 797, 800 (1994) (noting that although discovery rule applied to toll statute of limitations, the proponent of recovered memory still bore "the burden . . . to validate the phenomenon of memory repression itself and the admissibility of evidence flowing therefrom"). The trial court properly ordered the State to present expert testimony supporting the reliability of the recovered memories.

We turn to the showing that the proponent must make before evidence of the content of repressed memories will be admissible at trial. New Hampshire Rule of Evidence 702 and the principles we enunciated in Cressey, 137 N.H. 402, 628 A.2d 696, guide our analysis. In Cressey, we evaluated the admissibility of expert psychological testimony under Rule 702, and concluded that such "testimony must rise to a threshold level of reliability to be admissible." Id. at 405, 628 A.2d at 698. We did not define the precise contours of the reliability inquiry for every case, although we did indicate what sorts of concerns ought to guide the inquiry. See id. at 408-10, 628 A.2d at 700-02; State v. Cavaliere, 140 N.H. 108, 110-13, 663 A.2d 96, 98-100 (1995). Specifically, we considered important the presence of objective, quantifiable evaluation results, Cressey, 137 N.H. at 408-09, 628 A.2d at 700-01, the existence of a "logical nexus" between the expert's observations and conclusions, id. at 409, 628 A.2d at 701, the verifiability of any interpretive steps, id. at 409-10, 628 A.2d at 701, and the likely difficulty of effective cross-examination of the expert, id. at 410, 628 A.2d at 701. We apply these principles in the repressed memory context — both to the witness claiming to have recovered memory and to the expert explaining the phenomenon. Also helpful are the considerations enunciated by the United States Supreme Court in Daubert, 509 U.S. at 592-95. In applying Federal Rule of Evidence 702, the Daubert Court discussed four considerations bearing upon the reliability and helpfulness of scientific evidence: (1) whether the theory or technique has been or can be tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) the potential or known error rate; and (4) whether there is general acceptance of the theory or technique in the relevant scientific community. Id. at 593-94.

The extensive case law from other jurisdictions considering the admissibility of various types of refreshed recollection in civil and criminal cases is helpful to our inquiry. In the loosely analogous circumstance of offered testimony relying upon memory that has been enhanced, refreshed, or recovered by hypnosis, courts generally have divided into four groups: those that categorically accept such testimony, those that categorically reject such testimony, those that will admit the testimony only if rigid procedural safeguards have been met, and those that will admit the testimony only after a "totality of the circumstances" review of the reliability of the particular testimony. See, e.g., State v. Brown, 337 N.W.2d 138, 151 (N.D. 1983) (hypnotically refreshed testimony admissible and subject to credibility challenge); People v. Shirley, 723 P.2d 1354, 1383-84 (Cal.) (testimony inadmissible under Frye test), cert. denied, 459 U.S. 860 (1982); State v. Hurd, 432 A.2d 86, 96-97 (N.J. 1981) (admissible if safeguards complied with); Iwakiri, 682 P.2d at 579 (testimony admissible if, under totality of circumstances, it is sufficiently reliable to merit admission). Limitations on the admissibility of eyewitness testimony are generally justified based upon the fact that inaccuracies can be injected into recall during the hypnotic process by suggestion, confabulation, and conflation of true memories with false memories, see Iwakiri, 682 P.2d at 576, and upon the inability of the adversarial process to ferret out such inaccuracies because of memory hardening, see e.g., Hurd, 432 A.2d at 95. See Cressey, 137 N.H. at 410, 628 A.2d at 701 (observing that psychologist's testimony of her interpretation of her evaluations was impenetrable by cross-examination). Outside of the preliminary question of whether to toll the relevant statute of limitations, e.g., McCollum, 138 N.H. at 289, 638 A.2d at 799, few cases involve the more novel question of the admissibility of repressed memories recovered spontaneously, or during or attendant to participation in psychological therapy. See Crawford, 682 A.2d at 327-28; Quattrocchi, 681 A.2d at 881-84.

A review of the psychological literature on the subject of memory repression and recovery convinces us that a case-by-case approach, tempered with skepticism, is most appropriate in this context. See, e.g., The Reality of Repressed Memories, supra at 530-32. See generally Pezdek Roe, supra (reviewing studies of suggestibility of children's memories); Williams, Recall of Childhood Trauma: A Prospective Study of Women's Memories of Child Sexual Abuse, 62 J. CONSULTING CLINICAL PSYCHOL. 1167 (1994) (suggesting loss of memory of sexual abuse may be common).

We are especially concerned with the influence of therapy on the recovery of memory, as in the instant cases. The process of therapy is highly subjective, with its purpose "not the determination of historical facts, but the contemporary treatment and cure of the patient." Tyson v. Tyson, 727 P.2d 226, 229 (Wash. 1986); see Quattrocchi, 681 A.2d at 882. This goal, along with the expectations and predispositions of both therapist and patient, tends to distort the "historical truth" of events in the patient's life. Tyson, 727 P.2d at 229; see Wesson, Historical Truth, Narrative Truth, and Expert Testimony, 60 WASH. L. REV. 331, 337-38 (1985). Within the environment of therapy, a patient may report memories in response to the perceived expectations of the therapist, see, e.g., Taub, supra at 191, or in response to other forces. See Sleeping Memories, supra at 138-39; Nelson Simpson, First Glimpse: An Initial Examination of Subjects Who Have Rejected Their Recovered Visualizations as False Memories, 6 ISSUES IN CHILD ABUSE ACCUSATIONS 123, 126-27 (1994).

Observations like the following are troubling:

[T]he goal of therapy [is to] creat[e] a coherent "narrative truth" that accounts for the events in a patient's life but that does not necessarily make contact with the actual past. The goal is to account for the client's symptoms and allow the client to achieve closure with the past. But the truth of the past is not particularly important; instead, the patient "weaves together" a picture of the past that accounts for his symptoms and allows him to understand his life. Once the past has been reconstructed, however, the past is effectively changed and the original version is lost both for therapy and for all other purposes. The patient's memory will never be the same.

Comment, Repression, Memory, and Suggestibility: A Call for Limitations on the Admissibility of Repressed Memory Testimony in Sexual Abuse Trials, 66 U. COLO. L. REV. 477, 511 (1995) (quotations, footnote, and brackets omitted) [hereinafter Call for Limitations]; see Loftus Ketcham, supra at 265-67.

We do not mean to suggest that all or even a majority of recovered repressed memories are "false." Rather, we merely recognize that the memories are subject to many factors that may affect their reliability, especially, as the trial court found in the instant cases, the uniquely suggestive environment of psychological therapy. See LOFTUS KETCHAM, supra at 265-67. See generally R. GARDNER, TRUE AND FALSE ACCUSATIONS OF CHILD SEXUAL ABUSE 652-66 (1992). As we stated in Cressey:

By this opinion we do not seek to disparage the work being done in psychology and the behavioral sciences, for we can surely see its value; however, we are bound to recognize that the separate fields of behavioral science and criminal justice are different enough in their foundations and goals that what may be considered helpful information in one may not be so valued in the other.

Cressey, 137 N.H. at 407, 628 A.2d at 699. Our approach today reflects our attempt to balance "the legal and emotional needs of survivors of childhood sexual abuse," Roe v. Doe, 28 F.3d 404, 408 (4th Cir. 1994) (Hall, J., concurring), with our duty to ensure that defendants receive a fair trial and that individuals receive a reliable and fair adjudication of their disputes.

Ordinary memory is imperfect. See Hall et al., Postevent Information and Changes in Recollection for a Natural Event, in EYEWITNESS TESTIMONY, supra at 124, 126-27. Studies indicate that memory is not a mechanism that merely reproduces one's perceptions of events; rather,

memory, like perception, is an active, constructive process that often introduces inaccuracies by adding details not present in the initial representation or in the event itself. The mind combines all the information acquired about a particular event into a single storage "bin," making it difficult to distinguish what the witness saw originally from what she learned later.

Note, Did Your Eyes Deceive You? Expert Psychological Testimony on the Unreliability of Eyewitness Identification, 29 STAN. L. REV. 969, 983 (1977); see Shirley, 723 P.2d at 1377-78. See generally Memory, supra at 13-33 (explaining how MEMORY functions).

The law has recognized that an eyewitness's recall of an event or a person's face or features may be irretrievably altered by suggestive identification procedures. See, e.g., State v. Allard, 123 N.H. 209, 213, 459 A.2d 259, 262, cert. denied, 464 U.S. 933 (1983). Indeed, it may be "that the influence of improper suggestion upon identifying witnesses probably accounts for more miscarriages of justice than any other single factor — perhaps it is responsible for more such errors than all other factors combined." United States v. Wade, 388 U.S. 218, 229 (1967) (quotation and brackets omitted); see also Malpass Devine, Research on Suggestion in Lineups and Photospreads, in EYEWITNESS TESTIMONY, supra at 64, 74-86. Courts accordingly exclude in-court identification that has resulted from an out-of-court identification that is unreliable because tainted by improper suggestion. See Allard, 123 N.H. at 213, 459 A.2d at 262.

Our recognition that ordinary memory is subject to suggestion only emphasizes the limitations of eyewitness testimony in any case, see Sleeping Memories, supra at 155; Hall et al., supra at 126-27, and does not conclusively control our evaluation of recovered memories. This point merely establishes the post against which the reliability of recovered memories must be measured. See, e.g., Hurd, 432 A.2d at 95. To establish that a recovered memory is reliable, the proponent of its admission must demonstrate a reasonable likelihood that the recovered memory is as accurate as ordinary human memory. See id.; cf. Shahzade v. Gregory, 923 F. Supp. 286, 290 (D. Mass. 1996) (inquiring only into reliability of phenomenon of repressed memories, not into creditability of particular memory). An inquiry into the reliability of recovered memories generally will comprise the first part of this burden. Further, because of the great possibility of suggestiveness in therapy, see, e.g., The Reality of Repressed Memories, supra at 526-27, if therapy or some other formal technique has been utilized in order to retrieve the memory — or has been engaged in during the time in which the memory was retrieved — then further inquiry is required to determine the effect of that process or technique upon the reliability of the resulting memory. Cf. Call for Limitations, supra at 512-14 (describing similarities between effects of therapy, hypnosis, and interrogation on memory).

In determining the reliability of a recovered memory, — that is, whether the recovered memory is reasonably likely to be as accurate as ordinary memory — the trial court should consider the following factors: (1) the level of peer review and publication on the phenomenon of repression and recovery of memories, see Daubert, 509 U.S. at 593; (2) whether the phenomenon has been generally accepted in the psychological community, see id. at 594; (3) whether the phenomenon may be and has been empirically tested, see id. at 593; (4) the potential or known rate of recovered memories that are false, see id. at 594; (5) the age of the witness at the time the event or events occurred, see Williams, supra at 1168; (6) the length of time between the event and the recovery of the memory, cf. Hall et al., supra at 130; (7) the presence or absence of objective, verifiable corroborative evidence of the event, see Meiers-Post v. Schafer, 427 N.W.2d 606, 610 (Mich.Ct.App. 1988); and (8) the circumstances attendant to the witness's recovery of the memory, i.e., whether the witness was engaged in therapy or some other process seeking to recover memories or likely to result in recovered memories, see BRITISH PSYCHOLOGICAL SOCIETY, EXECUTIVE SUMMARY: RECOVERED MEMORIES § 2.1, at 9 (1995). Cf. Isely v. Capuchin Province, 877 F. Supp. 1055, 1064-67 (E.D. Mich. 1995) (parameters for admissibility of expert testimony).

If the witness was engaged in formal psychological therapy or some other process aimed at, or likely to facilitate, the recovery of memories, then further inquiry into that process is required. See Call for Limitations, supra at 511-12 (describing influence of traditional psychotherapy on memory). In the case of recovery attendant to therapy, this inquiry includes an examination of the therapist's qualifications, the type of therapeutic approach used, whether complaints of false accusations have been filed against the therapist, whether the therapist ordinarily seeks hidden memories or believes that many psychological problems stem from sexual abuse, and whether the therapist remains detached during the process or "validates" allegations of abuse that arise. See Call for Limitations, supra at 521 (suggesting some of these factors); cf. Nelson Simpson, supra at 125-29 (examining effects of various influences on people who develop false memories); Taub, supra at 208-13 (discussing characteristics of individuals who retract claims of recovered memories and their therapists).

Although phrased in different terms, the trial court applied a test of reliability and general acceptance similar to the test we enunciate today. We accordingly defer to its findings insofar as they apply to the facts of these particular cases. See Cressey, 137 N.H. at 405, 628 A.2d at 698; Vandebogart (DNA), 136 N.H. at 376, 616 A.2d at 491.

Considering our first factor, the trial court correctly observed that the phenomenon of memory repression and recovery has received extensive attention in psychological publications. The parties presented photocopies of many articles from medical and psychological publications on the issue, and a review of the literature reveals many more. The level of peer review is high. "[S]ubmission to the scrutiny of the scientific community is a component of `good science,' in part because it increases the likelihood that substantive flaws in methodology will be detected." Daubert, 509 U.S. at 593. In the case of repressed and recovered memories, the level of submission is high, but the debate over methodology and the meaning of results continues. See, e.g., Holmes, The Evidence for Repression: An Examination of Sixty Years of Research, in REPRESSION AND DISSOCIATION 85, 96-98 (J. Singer ed., 1990); Long-Buried Memories, supra at 635.

The psychological community remains deeply divided on the reliability or accuracy of recovered memories. See Ault, 637 N.E.2d at 875 (Wright, J., dissenting); Sleeping Memories, supra at 133-35; Taub, supra at 186-87. Despite common support for the phenomenon in the therapeutic setting, scientists rest their rejection of recovery of repressed memories on the absence of confirming laboratory results. Compare Pope Hudson, supra at 122-25 (criticizing methodology of clinical studies) with Herman Schatzow, Recovery and Verification of Memories of Childhood Sexual Trauma, 1987 PSYCHOANALYTIC PSYCHOL. 4(1), 11-13 (reporting high incidence of repression in group therapy study). Of course, ethically, no complete laboratory study could ever be completed on repression of events as traumatic as sexual abuse. See Sleeping Memories, supra at 133-34; MEMORY, supra at 80-82.

According to the theory of repression, when a person experiences a particularly traumatic event that is unacceptable to the person's conscious existence, the person may repress the memory of the trauma. E.g., Sleeping Memories, supra at 132-33. Although the memory is not permanently "forgotten," it is unavailable to the person's conscious thought process. Sleeping Memories, supra at 132. True repression or traumatic amnesia rendering a person unable to remember any part of a traumatic event are distinguished from ordinary forgetting, see Loftus et al., Memories of Childhood Abuse: Remembering and Repressing, 18 PSYCHOL. WOMEN Q. 67, 68-69, 82 n. 1 (1994) [hereinafter Memories of Childhood Abuse], motivated forgetting, Pope Hudson, supra at 122, incomplete memory, Williams, supra at 1168, and psychogenic amnesia, LOFTUS KETCHAM, supra at 215-16. Cf. Terr, Chowchilla Revisited: The Effects of Psychic Trauma Four Years After a School-Bus Kidnapping, 140 AM. J. PSYCHIATRY 1543, 1545-47 (1983).

Proponents of widespread repression and recovery of memories of sexual abuse consider several facts to support the phenomenon: the existence of psychogenic amnesia and post-traumatic stress disorder, clinical studies in support of the phenomenon, and the prevalence of patients reporting recovery of repressed memories. See Long-Buried Memories, supra at 600-03; Memories of Childhood Abuse, supra at 69-70. Discrete memory repression is a different physiological phenomenon from psychogenic amnesia, where the victim or witness of an extremely traumatic event temporarily may forget ordinary personal details, such as name and address, in addition to the details of the traumatic event. Loftus Ketcham, supra at 215-16. The typical symptoms of post-traumatic stress disorder also do not support fully the notion of complete memory repression. See, e.g., Kinzie, Post-traumatic Effects and Their Treatment among Southeast Asian Refugees, in INTERNATIONAL HANDBOOK OF TRAUMATIC STRESS SYNDROMES 311, 314-15, 318 (J. Wilson B. Raphael eds., 1993) (intrusive memories, recurrent nightmares, and avoidance of memories common symptoms in Southeast Asian refugees with post-traumatic stress disorder); cf. Pynoos Nader, Children's Memory and Proximity to Violence, 28 J. AM. ACAD. CHILD ADOLESCENT PSYCHIATRY 236, 239-41 (1989) (discussing memory disturbance and recall of children with post-traumatic stress disorder differing in severity according to proximity to violence, among other factors).

The clinical studies that support the prevalence of recovery of previously completely repressed memories are subject to some criticism in methodology, as the trial court noted in the instant cases. See, e.g., Pope Hudson, supra at 122-25. For example, in one study, researchers reported that approximately 59% of subjects — patients who were referred by their therapists as having histories of childhood sexual abuse — answered affirmatively when asked: "During the period of time between when the first forced sexual experience happened and your eighteenth birthday was there ever a time when you could not remember the forced sexual experience?" Briere Conte, Self-Reported Amnesia for Abuse in Adults Molested as Children, 6 J. TRAUMATIC STRESS 21, 23-24 (1993). Reviewers raised the following methodological concerns: the subjects were "recruited" by their therapists; it is unclear whether the reported underlying events were confirmed in any way; it is unclear whether the events were "sufficiently traumatic" to have been remembered at every moment; and an affirmative answer to the question conveys insufficient information to conclude that full repression has actually occurred. Pope Hudson, supra at 123. Finally, the reviewers noted the possibility of suggestion in therapy:

All [subjects] were in treatment with therapists who were part of an "informal sexual abuse treatment referral network[,"] and who, therefore, may have communicated to their patients, explicitly or implicitly, that repression of traumatic experiences was to be expected. With this potential degree of expectation, and with therapists choosing which subjects would receive the questionnaire, it would not be surprising if many subjects answered "yes" to a question that asked if there was ever a time when they could not remember an abuse experience.

Pope Hudson, supra at 123; see Sleeping Memories, supra at 134. Other studies are subject to similar complaints. See Pope Hudson, supra at 123-24 (reporting similar methodological limitations in the study reported in Memories of Childhood Abuse, supra); Memories of Childhood Abuse, supra at 71-73 (citing methodological difficulties with previous studies).

Proponents of the phenomenon of recovering repressed memories also rely on the very existence of a large number of patients reporting recovery of repressed memories as validation of the phenomenon. Cf. LOFTUS KETCHAM, supra at 209 ("`Why would anyone invent a story that involved so much anguish and suffering?'").

The stories the patients tell, they argue, are too vivid and too painful to be the product of imagination or fabrication. The emotional troubles these patients have as adults are consistent with the kind of abuse that they remember, and would not be so consistent and so intense in response to a fabricated memory.

Long-Buried Memories, supra at 603 (footnote omitted). Apart from its circularity, the argument lends more support to the concept of suggestibility of memory than to the phenomenon of repression. Cf. The Reality of Repressed Memories, supra at 525; Nelson Simpson, supra at 126-27.

The scientific literature supports the conclusion that, in general, people remember traumatic events well. See, e.g., Malmquist, Children Who Witness Parental Murder: Post-traumatic Aspects, 25 J. AM. ACAD. CHILD PSYCHIATRY 320, 324 (1986) ("Recollection of vivid memories of the event were present in all 16 of the children" studied). In fact, experiencing vivid, intrusive thoughts of the event seems to be a more common memory disturbance resulting from severe trauma than repression. See, e.g., Wilkinson, Aftermath of a Disaster: The Collapse of the Hyatt Regency Hotel Skywalks, 140 AM. J. PSYCHIATRY 1134, 1137 (1983) (repeated recollection of event most frequent symptom among those experiencing collapse). In a study examining the effects on a group of children kidnapped on their school bus, for example, examiners found the children to have intact and detailed memories of the event, although they did observe some memory disturbance. Terr, supra at 1545-46; see also Weine et al., Psychiatric Consequences of "Ethnic Cleansing": Clinical Assessments and Trauma Testimonies of Newly Resettled Bosnian Refugees, 152 AM. J. PSYCHIATRY 536, 540-41 (1995) (many survivors' lives "inundated with traumatic images," while in others, "cognitive overload of the genocidal traumatic experience creates an incomplete perception and registration of the traumatic event").

There have been some cases where repression and later retrieval of a memory of childhood sexual abuse have been claimed to be corroborated from other sources, thus enhancing the credibility of the phenomenon and increasing its acceptance with some therapists. See Sleeping Memories, supra at 134. A recent review of the literature, however, caused the reviewer to note that "despite over sixty years of research involving numerous approaches by many thoughtful and clever investigators, at the present time there is no controlled laboratory evidence supporting the concept of repression." Holmes, supra at 96; see Taub, supra at 188. The scientific community is extremely divided, at best, on the issue of recovery of completely repressed memories.

A degree of scientific divergence of opinion is indeed inevitable, but the degree of divergence surrounding [recovery of repressed memories] is fundamental and goes to the very validity of the process itself. This kind and degree of divergence is notably absent in other areas of scientific evidence generally deemed admissible.

Reed v. State, 391 A.2d 364, 376 (Md. 1978). We cannot say that the phenomenon has gained general acceptance in the psychological community. Cf. Daubert, 509 U.S. at 594.

We turn to the next consideration, whether the phenomenon may be empirically tested. As noted in the foregoing discussion, it would be impossible, ethically, to test repression and recovery of memory of severely traumatic events in a laboratory setting. See Sleeping Memories, supra at 134. Almost all studies of the phenomenon to date, accordingly, involve subjects in the clinical or therapeutic context. See, e.g., Briere Conte, supra at 23-24; Herman Schatzow, supra at 2-4. Further, the studies of memory of childhood sexual abuse involve retrospective self-reporting of prior, typically uncorroborated, sexual abuse. See, e.g., Briere Conte, supra at 23-24 (describing participant description of childhood abuse); Herman Schatzow, supra at 2-3, 10 (reporting that participants were able to obtain confirmation of earlier abuse). One exception is the study by Linda Meyer Williams, who interviewed 129 women who had been treated for sexual abuse in a metropolitan hospital as children approximately seventeen years earlier. Williams, supra at 1169. Of these, forty-nine women, or 38% of the sample, did not report the childhood abuse to the interviewer. Williams, supra at 1170. Williams states that "[a]lthough some of these women may have simply decided not to tell the interviewers about the abuse, additional findings discussed later suggest that the majority of these women actually did not remember the abuse." Williams, supra at 1170. The "findings" to which she refers include the relative openness of the subjects in answering other personal questions, including other incidents of sexual, physical, or emotional abuse. Williams, supra at 1170. Another study has concluded that nonreporting of remembered abuse might be explained by "embarrassment, a wish to protect parents, a sense of having deserved the abuse, a conscious wish to forget the past, and a lack of rapport with the interviewer." Pope Hudson, supra at 124 (quotation omitted). Considering this similar study, Pope and Hudson concluded that it would be "hazardous to conclude that Williams' 49 `non-reporters' actually had amnesia." Pope Hudson, supra at 124. Although empirical testing is difficult, and subject to some methodological complaints, it is possible.

It is difficult to estimate the number or rate of recovered memories that are "false." Cf. Daubert, 509 U.S. at 594. Although some individuals who have recovered memories have since withdrawn their claims, Nelson Simpson, supra at 126-28, there is no way to track the percentage of such false memories, especially when the phenomenon is still subject to such vigorous debate. Cf. Gardner, supra at 661-64.

The remaining factors of the reliability inquiry relate to Laura's and Sarah's memories themselves. The aspects of the memories into which we inquire are, in part, factual; we therefore defer to the facts found by the trial court in this part of the inquiry. See, e.g., State v. Carroll, 138 N.H. 687, 696, 645 A.2d 82, 87 (1994). The charged acts in State v. Hungerford rely on Laura's memories of Hungerford inserting a gun into her vagina in 1990 and of him raping her two days before her wedding in 1991 when she was twenty-two and twenty-three years old, respectively. She began therapy in September 1992 with no memories of the assaults and reported them to the police in March 1993. The length of time between the event and remembering was approximately one and a half years, although the period was longer for the uncharged events. Laura's therapist, Ms. Jones, testified to what she perceived as corroborative evidence: the similar claims of Laura's sister, Amy; the remarks of Laura's mother regarding Hungerford's sexual habits with her and his conduct in the household when the girls were children; and the remarks of staff members at the therapy center where Hungerford was attending therapy.

According to the indictments, the charged acts in State v. Morahan occurred between December 1987 and March 1988, just before Sarah began psychological counseling in May 1988. She was thirteen years old and in the seventh grade at this time. Sarah recovered her memory of the assault in 1991 and reported it to authorities in 1993. There was no testimony at the admissibility hearing about the details of the assault, and we discern no other evidence tending to corroborate or not to corroborate the event.

In State v. Hungerford, Laura's age and the relatively small period of time between the two charged acts and her recovery of memory about them bear in favor of their reliability. Similarly, in State v. Morahan, Sarah's age and the relatively short period of time during which she had no memory of the assault bear favorably on the memory's reliability. Children who are very young are perceived to have incomplete narrative memories even of traumatic events, see Pezdek Roe, supra at 375-76; further, scientists generally agree that individuals are almost completely amnestic for the first few years of their own life, see L. TERR, UNCHAINED MEMORIES: TRUE STORIES OF TRAUMATIC MEMORIES, LOST AND FOUND 226 (1994). See also Briere Conte, supra at 28; Call for Limitations, supra at 498-99. Memory is subject to the influence of innumerable external influences during the "retention" stage of remembering, and thus a shorter period of time between the event and recall offers less opportunity for suggestion. See, e.g., Hall et al., supra at 132-40.

On the presence or absence of objective, verifiable corroborative evidence, the trial court found that "[i]n neither of these cases was there any corroboration or attempt to corroborate the abuse," although the court did find that both complainants demonstrated "serious psychological disturbances." These findings are supported by the record. There was some witness testimony that might have borne on the question of objective corroborative evidence. For example, the corroborative evidence relied upon by Ms. Jones in State v. Hungerford is somewhat convincing. The accusations of Laura's sister are of limited corroborative value. Compare Pope Hudson, supra at 123 (alleged abuse of sibling of questionable corroborative value) with Herman Schatzow, supra at 10-11 (evaluating claim of sibling abuse as corroborative evidence). Ms. Jones' evaluation of the concerns of Hungerford's therapy center staff offer some additional enlightenment, as do Hungerford's wife's reports. They are not, however, directly corroborative. Cf. Meiers-Post, 427 N.W.2d at 609-10 (requiring objective manifested injury and verifiable corroborative evidence, such as an admission by the perpetrator, before tolling statute of limitations). We defer to the trial court's finding that there was no corroboration of the alleged acts.

We next address the circumstances attendant to the recovery of the memories in the two cases, about which the trial court made extensive findings. In State v. Hungerford, Laura was engaged in therapy specifically focused on the issue of sexual abuse. Although on appeal the State attempts to characterize the therapy as being designed to help Laura "understand and cope with her inner feelings," the trial court's conclusion that the focus was recovery or retrieval of memories of sexual abuse is amply supported by the record. Sarah, too, in State v. Morahan, was engaged in therapy. Even if we were to agree with the State that Sarah's therapy was not specifically aimed at recovering memories and that she did not recover her memories during any particular therapy session, we would nonetheless defer to the trial court's conclusion that her memories were recovered attendant to therapy. As the trial court found, "it is difficult to distinguish between Sarah's therapy and real life, because it appears . . . that psychotherapy permeated the everyday structure of her school and social life." In its review of the circumstances attendant to the recovery of memories in both cases, the trial court found

that the psychotherapy utilized by Ms. Jones and by the DeSisto School to "retrieve" Laura's and Sarah's memories of abuse, thoroughly and systematically violated the guidelines and standards of the practice of psychotherapy. Furthermore, the Court finds that the techniques used in the course of psychotherapy in both cases were highly suggestive.

Because the memories in the instant cases were recovered during therapy or while the witness was engaged in therapy, we ordinarily would proceed to examine more closely the circumstances of the therapeutic environment, as discussed earlier. Our review of the memories without regard to the suggestiveness of the therapeutic process, however, convinces us that they do not pass our test of reliability. The phenomenon of recovery of repressed memories has not yet reached the point where we may perceive these particular recovered memories as reliable. "There probably will be a day, as there has been regarding the forensic use of DNA, when courts can be given reliable, competent information on the issue of repressed memory. That day is not here." Ault, 637 N.E.2d at 874-75 (Moyer, C.J., dissenting). The indicia of reliability present in the particular memories in these cases do not rise to such a level that they overcome the divisive state of the scientific debate on the issue.

In a particular case, the court may be satisfied with the state of the scientific debate on the question of recovering repressed memories, and with the general indicators of reliability surrounding a particular recovered memory. If that memory is recovered in the context of therapy, however, we still will be greatly concerned with the suggestiveness of the therapeutic process, and its ability to skew memory and one's confidence in memory. See LOFTUS KETCHAM, supra at 150-73. Because we need not engage in that inquiry in the instant cases, however, we shall not.

Affirmed and remanded.

HORTON, J., did not sit; the others concurred.