OPINION

delivered the opinion of the Court, in which

This case comes before us on petition for discretionary review by the State of Texas to decide whether Zani v. State, 758 *782S.W.2d 233 (Tex.Crim.App. 1988), was overruled by Kelly v. State, 824 S.W.2d 568 (Tex.Crim.App. 1992). In Zani, a case based on the general acceptance standard of Frye v. United States, 293 F. 1013 (D.C.Cir. 1923), we held that hypnotically enhanced testimony is admissible in Texas.1 Because Kelly overruled the general acceptance standard and Texas courts’ reliance on Frye, the State argues that Zani is implicitly overruled by Kelly. We conclude that, although based on the general acceptance standard expressed in Frye, the Zani case, in developing a highly specific framework to ensure that hypnotically enhanced testimony offered in a particular case is reliable, is not overruled, either explicitly or implicitly, by Kelly or its progeny.

This appeal originally arose from the trial court’s exclusion of hypnotically enhanced testimony from the only eyewitness in a capital murder case. In the initial appeal, the court of appeals, citing State v. Roberts, 940 S.W.2d 655, 658 (Tex.Crim.App. 1996), held that it did not have jurisdiction to consider the trial court’s exclusion of the evidence. State v. Medrano, 987 S.W.2d 600, 603-04 (Tex.App.-El Paso 1999). We granted the State’s petition for discretionary review, reversed the court of appeals’ decision, overruled Roberts, and remanded the appeal for consideration on the merits. State v. Medrano, 67 S.W.3d 892, 902-03 (Tex.Crim.App. 2002). On remand, the court of appeals affirmed the trial court’s exclusion of the evidence holding that the trial court correctly applied Zani in ruling on the admissibility of hypnotically enhanced testimony. State v. Medrano, 86 S.W.3d 369, 373 (Tex.App.-El Paso 2002). We granted the State’s petition to determine whether Zani was overruled by Kelly. We affirm.

As aptly expressed by the Eighth Court of Appeals, our opinion in Zani dealt with a “highly specific, detailed discussion of a narrowly defined question.” State v. Medrano, 86 S.W.3d 369, 372 (Tex.App.-El Paso 2002). That question was “whether hypnotically induced testimony is admissible in Texas.” Zani, 758 S.W.2d at 234 (internal citations omitted). In analyzing the admissibility of hypnotically enhanced testimony, we worked within the framework of the applicable law at the time: Frye v. United States.2 Judge Clinton, writing the opinion for this Court, stated: “we consider Frye ‘applicable because lay testimony that is dependent upon hypnosis cannot be logically dissociated from the underlying scientific technique.’” Zani, 758 S.W.2d at 241 (quoting Contreras v. State, 718 P.2d 129, 134 (Alaska 1986)). A discussion of hypnosis and hypnotically enhanced testimony included a discussion of Frye, the case that governed the admissibility of scientific evidence.

In Zani, while applying Frye and discussing the “general acceptance” of hypnosis and hypnotically enhanced testimony in the medical and legal community, we recognized that the accuracy of hypnotically enhanced testimony was questionable. We stated, “To the extent that ordinary disabilities of perception, memory, and articulation may be amplified by hypnosis, it will result in recall less accurate than would *783ordinarily be expected.” Id. at 242 (internal quotation marks omitted). We reasoned that because hypnosis may amplify perception, memory and articulation in the witness’s testimony, and because the amplification may go undetected, the best solution would be to adopt a per se exclusionary rule. Ibid.

Despite the lack of general acceptance among the scientific community, we did not impose a per se rule with Zani. Rather, citing the United States Supreme Court decision in Rock v. Arkansas, 483 U.S. 44, 107 S.Ct. 2704, 97 L.Ed.2d 37 (1987), we “declinefd] to adopt an exclusionary rule prohibiting hypnotic recall in every case.” Zani, 758 S.W.2d at 243. Although the law governing scientific evidence, the general acceptance standard of Frye, militated in favor of a per se exclusionary rale, the Supreme Court’s decision in Rock “rendered such a position untenable.” Id. at 242. See Rock, 483 U.S. 44, 107 S.Ct. 2704, 97 L.Ed.2d 37 (holding a per se exclusionary rule banning hypnotically refreshed testimony of defendants testifying on their own behalf unconstitutional under the Fifth, Sixth, and Fourteenth Amendments of the United States Constitution). Yet the question of the reliability of hypnotically enhanced testimony remained.

With Zani, the Court provided a mechanism to allow for the admission of hypnotically enhanced testimony and at the same time to ensure that this admitted testimony was reliable. The solution was the Zani standard. Under the Zani standard, the Court instituted procedural safeguards to protect against “the four-prong dangers of hypnosis: hypersuggestibility, loss of critical judgment, confabulation, and memory cementing.” Id. at 244 (internal quotation marks omitted). The Court adopted a non-exclusive list of factors from People v. Romero, 745 P.2d 1003, 1017 (Colo. 1987),3 which a trial court is to consider in deciding whether hypnotically enhanced testimony is admissible in a particular ease. Zani, 758 S.W.2d at 244. The Zani standard permits admission of hypnotically enhanced testimony “[i]f, after consideration of the totality of the circumstances, the trial court should find by clear and convincing evidence that hypnosis neither rendered the witness’s posthypnotic memory untrustworthy nor substantially impaired the ability of the opponent fairly to test the witness’s recall by cross[-]examination.” Ibid. By requiring clear and convincing evidence of the trustworthiness and the lack of impairment of a witness’s hypnotically enhanced testimony, the aim of this Court was to ensure the reliability of the evidence admitted.

Since Zani was decided in 1988, the law governing the admissibility of scientific evidence in Texas criminal courts has evolved from the “general acceptance” standard enunciated in Frye v. United *784States, 293 F. 1013 (D.C.Cir. 1923), to the current standard presented in Kelly v. State, 824 S.W.2d 568 (Tex.Crim.App. 1992). In Kelly, we stated that the general acceptance standard was no longer valid for two reasons. First, the Frye standard was no longer valid because, with the 1986 adoption of Texas Rule of Criminal Evidence 702 (stating “If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.”), there was “no textual basis ... for a special admissibility standard for novel scientific evidence.” Kelly, 824 S.W.2d at 572. Second, the Frye standard was no longer valid because “scientific evidence may be shown reliable even though not yet generally accepted.... ” Ibid. Because the Frye-based general acceptance standard was no longer valid, the Court presented a new standard in Kelly.

In determining the admissibility of novel scientific evidence, the threshold question asked by both Rule 702 and Kelly was, “whether the testimony will help the trier of fact understand the evidence or determine a fact in issue.” Ibid, (citing Duckett v. State, 797 S.W.2d 906, 910 (Tex.Crim.App. 1990)); see also Texas Rule of Evidence 702 (stating that evidence should be admitted, “[i]f scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue.”). In Kelly, we reasoned that “the trial court’s first task is to determine whether the testimony is sufficiently reliable and relevant to help the jury in reaching accurate results,” because “ ‘[ujnreliable ... scientific evidence simply will not assist the [jury] to understand the evidence or accurately determine a fact in issue.’” Kelly, 824 S.W.2d at 572 (quoting K. Kreiling, Scientific Evidence: Toward Providing the Lay Trier With the Comprehensible and Reliable Evidence Necessary to Meet the Goals of the Rules of Evidence, 32 Ariz. L.Rev. 915, 941-42 (1990)). Kelly states that “before novel scientific evidence may be admitted under Rule 702, the proponent must persuade the trial court, by clear and convincing evidence, that the evidence is reliable and therefore relevant.” Id. at 573 (internal citations omitted). The reliability of novel scientific evidence is the cornerstone of our opinion in Kelly.

Proving reliability requires that the proponent establish: “(a) [that] the underlying scientific technique [is] valid; (b) [that] the technique applying the theory [is] valid; and (c) [that] the technique [has] been properly applied on the occasion in question.” Ibid, (internal citations omitted). Additionally, the Court cited seven factors that a trial court may consider which could affect reliability.4 This standard was expanded by the Court through Hartman v. State, 946 S.W.2d 60, 63 (Tex.Crim.App. 1997), in which the Court held that the standard established in Kelly applies to all scientific evidence offered under Rule 702. *785See also Nenno v. State, 970 S.W.2d 549, 560 (Tex.Crim.App. 1998). However, we soon recognized that the criterion used to prove reliability by clear and convincing evidence under Kelly may become cumbersome under certain circumstances.

In Nenno v. State, the Court warned that the “general principles announced in Kelly (and Daubert [v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993)]) apply [to nonscientific expert testimony], but the specific factors outlined in those cases may or may not apply depending on the context.” Nenno, 970 S.W.2d at 560. Although Kelly remains the law after Nenno, “[w]hen addressing fields of study aside from the hard sciences, such as the social sciences or fields that are based primarily upon experience and training as opposed to the scientific method, Kelly’s requirement of reliability applies but with less rigor than to the hard sciences.” Id. at 561. Through Nenno, the Court refined the Kelly framework to provide a more adequate means of analyzing the reliability of scientific evidence based on soft sciences.

In Nenno, we were not attempting ‘To develop a rigid distinction between ‘hard’ and ‘soft’ sciences, or nonscientific testimony.” Id. at 560-61. However, Nenno offered three inquiries more appropriately tailored to analyze soft sciences where “the validity of a ‘theory’ or ‘technique’ ... may be roughly accurate but somewhat misleading.” Ibid. When soft sciences are at issue, the trial court should inquire: “(1) whether the field of the expert’s testimony is within the scope of that field, (2) whether the subject matter of the expert’s testimony is within the scope of that field, and (3) whether the expert’s testimony’ properly relies upon and/or utilizes the principles involved in the field.” Id. at 561. These questions may be more appropriate when dealing with soft sciences because “hard science methods of validation, such as assessing the potential rate of error or subjecting a theory to peer review, may often be inappropriate for testing the reliability of fields of expertise outside the hard sciences.” Ibid. However, because the objective of both Kelly and Nenno was to ensure the reliability of expert testimony and scientific evidence, Nenno “[did] not categorically rule out employing [the Kelly] factors in an appropriate case.” Id. at 561 n. 9. Rather, the objective in Nenno was to provide a mechanism to ensure the reliability in the more amorphous sciences.

Reliability remains the cornerstone, of not just Kelly, but also of all of the scientific evidence cases. By requiring a threshold of general acceptance in the field before scientific evidence would be admissible, the court in Frye was seeking to ensure the reliability of novel scientific evidence. Frye, 293 F. at 1014. Moving beyond the general acceptance test in Frye, our objective in Kelly was to provide a three-prong framework that would ensure that evidence derived from a scientific source was reliable despite lacking in general acceptance in the relevant scientific community. Kelly, 824 S.W.2d at 573.

Yet the Kelly test, while readily applicable to hard sciences based primarily on the scientific method, did not provide an adequate test of reliability to soft sciences “based primarily upon experience and training.” Nenno, 970 S.W.2d at 561. In recognition of this limitation, we offered a similar three-prong test but made it an “appropriately tailored translation of the Kelly test” to areas of soft science and non-traditional sources of expert testimony. Ibid. We noted that the unique nature of scientific evidence derived from soft sciences does not vitiate the Kelly requirement of reliability. Ibid. In Weatherred v. State, 15 S.W.3d 540 (Tex.*786Crim.App. 2000), we made clear that despite the apparently relaxed standard created for “soft sciences” in Nenno, “[u]nder Rule 702, the proponent of scientific evidence must show, by clear and convincing proof, that the evidence he is proffering is sufficiently relevant and reliable to assist the jury in accurately understanding other evidence or in determining a fact in issue.” Weatherred, 15 S.W.3d at 542 (citing Nenno v. State, 970 S.W.2d 549, 560-61 (Tex.Crim.App. 1998); Hartman v. State, 946 S.W.2d 60, 62-63 (Tex.Crim.App. 1997); Jordan v. State, 928 S.W.2d 550, 553-55 (Tex.Crim.App. 1996); Kelly v. State, 824 S.W.2d 568, 572-73 (Tex.Crim.App. 1992)). Regardless, of the type of scientific evidence, reliability remains crucial to admissibility.

The primacy of reliability is as paramount to the admission of scientific evidence today as it was when we decided Zani. In 1988, when the Court adopted the Zani standard, it sought to ensure that all hypnotically enhanced testimony that was admitted by the Texas criminal courts was rehable. The Zani standard, although relying on Frye, followed the 1986 promulgation of the Texas Rules of Criminal Evidence and, more specifically, Rule 702. The Zani case was tried in 1981 and as such was a pre-Rules case. By the time the appeal reached this Court the new Rules of Criminal Evidence were already in effect. Zani v. State, 679 S.W.2d 144, 149 (Tex.App.-Texarkana 1984). Consequently, when the appeal was decided, although not pursuant to the new Rules, the Court was no doubt aware of the importance of reliability under the newly enacted Rule 702 governing the admissibility of expert testimony and scientific evidence. Reliability was and remains a central theme of Rule 7025 because the rule requires that any expert be able to “assist the trier of fact to understand the evidence or to determine a fact in issue,” Texas Rule of Evidence 702, and because “ ‘[u]n-reliable ... scientific evidence simply will not assist the [jury] to understand the evidence or accurately determine a fact in issue.’” Kelly, 824 S.W.2d at 572 (quoting K. Kreiling, Scientific Evidence: Toward Providing the Lay Trier With the Comprehensible and Reliable Evidence Necessary to Meet the Goals of the Rules of Evidence, 32 Ariz. L.Rev. 915, 941-42 (1990)). Although, we stated in Nenno that, with regard to the soft sciences, “Kelly’s requirement of reliability applies but with less rigor than to the hard sciences,” this does not mean that scientific evidence derived from the soft sciences is always afforded less rigorous scrutiny. Zani is illustrative.

With Zani, the Court created a standard for the trial courts to apply in determining the admissibility of hypnotically enhanced testimony — evidence. based on a soft science. The Court recognized in Zani that the reliability of hypnotically enhanced testimony is especially in question due to the undetected amplification of disabilities in perception, memory and articulation in a witness’s testimony. Although this standard was created in 1988, predating both Kelly and Nenno, the objective of reliability remains constant in all three opinions. In Kelly and Nenno, frameworks were developed to ensure the reliability of novel scientific evidence in the broad categories of hard and soft sciences, respectively. Zani, on the other hand, is a standard that applies in a soft science situation where a narrowly tailored framework was created *787to ensure the reliability of a particular scientific technique.

The Zani standard provided the Texas trial courts with the appropriate framework to protect against the four-prong dangers of hypnosis. These four dangers, hypersuggestibility, loss of critical judgment, confabulation, and memory cementing, are dangers that directly undercut the reliability of a witness’s hypnotically enhanced testimony. The Zani standard minimizes these dangers and, consequently, ensures the reliability of the testimony. Certainly, the Zani standard is more rigorous than the standards applied in Kelly or Nenno. However, our opinion in Zani exhaustively analyzed both the dangers and solutions inherent in hypnotically enhanced testimony.

The standard created by the Court is a highly specific framework applicable specifically to the admissibility of hypnotically enhanced testimony. Turning to a broader standard under Kelly and Nenno, while seeking the same objective of reliability realized by the current standard in Zani, would do a disservice to the trial courts. To adopt a broader standard that was created without hypnotically enhanced testimony in mind in place of a narrowly defined standard specifically designed to ensure the reliability of hypnotically enhanced testimony would be imprudent. Because this highly specific framework is a naiTowly defined standard that seeks to ensure the reliability of a specific type of scientific evidence, and because Zani is faithful to the primary objective of ensuring reliability in the admission of scientific evidence, we conclude that Zani remains the standard to be applied by Texas trial courts in assessing the reliability and determining the admissibility of hypnotically enhanced testimony. The decision of the court of appeals is affirmed.

WOMACK, J., filed a concurring opinion.

filed a concurring opinion.

I agree with the Court’s decision to affirm the courts below in excluding the evidence of the hypnotized witness. I also agree that we should adhere to the Court’s decision in Zani,1 even though it assigns a higher burden to the proponent of such evidence than the burden on the proponent of scientific evidence.

The reason that the burdens are different is that the question in the hypnosis cases is different from the questions that were presented in Frye,2 Kelly,3 or Daubert.4 Those cases involve the admissibility of scientific evidence, specifically experimental data or expert opinion about the results of scientific tests. Cases like the one that is now before us do not involve the admissibility of such evidence. The issue in these cases is the competency of a lay witness to testify about his or her memory of events that he or she perceived. The question is the extent to which hypnosis has affected the reliability of the witness’s memory. The standards of the scientific-evidence cases can do no more than *788inform a court’s decision about the admissibility of the witness’s testimony.5

For these reasons, I agree with Judge Cochran that it was inappropriate to import into our scientific-evidence decisions the “clear-and-eonvincing” burden of Zani6 I would conform to Daubert the burden of Kelly, but not the burden of Zani

Zani was correct not only in imposing a higher burden, but also in adopting a flexible standard that can take into view “the totality of the circumstances” of the case to consider whether “that hypnosis neither rendered the witness’ posthypnotic memory untrustworthy nor substantially impaired the ability of the opponent fairly to test the witness recall by cross[-]examination.”7 As another court has put it:

The district court must then determine whether in view of all the circumstances, the proposed testimony' is sufficiently reliable and whether its probative value outweighs [the danger of] its prejudicial effect, if any, to warrant admission. Ultimately the district court must decide whether the risk that the testimony reflects a distorted memory is so great that the probative value of the testimony is destroyed.8

In Zani we quoted a long, but non-exclusive, list of circumstances.9 Others could always be added, such as a consideration of the purpose for the hypnosis10 and evidence of the witness’s “hypnotizability.”11 On these circumstances, expert opinion may be relevant.12

filed a concurring opinion, joined by KELLER, P.J., and HOLCOMB, J.

I concur in the Court’s judgment and agree that the reliability approach concerning post-hypnotic testimony that is set out in Zani v. State,1 was carried forward *789under Kelly v. State2 and its progeny.3 However, I respectfully disagree with Zani’s creation of a “clear and convincing” standard of proof for admitting post-hypnotic testimony and this Court’s continued use of a “clear and convincing” standard for admitting all expert testimony. This standard might have made some sense in Zani itself, but it is not appropriate for determining the admissibility of all expert testimony under Rule 702. It is an error that this Court should rectify.

In Zani, we addressed the question of “whether hypnotically-induced testimony is admissible in Texas.”4 Although the defendant had argued that hypnotically enhanced identification testimony was per se inadmissible under Frye v. United States,5 this Court held that a per se rule of exclusion was untenable under United States Supreme Court precedent.6 Instead, in Zani, we adopted the reliability approach set out in an earlier Colorado Supreme Court case, People v. Romero,7 and we quoted Romero’s non-exclusive factors for trial judges to use in assessing the trustworthiness of a witness’ memory recall after hypnosis.8 This Court also stated:

This is not to say that the proponent of hypnotically refreshed testimony ought not bear a heavy burden of showing it should go to the jury. We conclude that because of the uncertainties inherent in posthypnotic testimony it is appropriate to require the proponent of such testimony to demonstrate to the satisfaction of the trial court, outside the jury’s presence, by clear and convincing evidence, that such testimony is trustworthy.9

*790We did not cite any authority for imposing this heightened burden of proof for determining the admissibility of evidence. In fact, in Romero, the very case that we quoted and relied upon for the factors to assist in determining the reliability of the witness’ memory recall, the Colorado Supreme Court explicitly used the normal “preponderance of the evidence” standard. It stated:

While some courts have imposed a clear and convincing standard of proof on the issue of reliability, we believe the preponderance of evidence is .the suitable standard for resolving this issue. The preponderance of evidence standard has often been referred to as the “orthodox view,” since it is in keeping with the traditional burden applicable to the resolution of most preliminary questions of admissibility. Indeed, it has been cogently observed that the preponderance of evidence standard “is appropriate for resolving most preliminary fact questions, even in a criminal case, and even when the reliability of the ultimate verdict is arguably affected by the decision on the preliminary issue.” We accordingly adopt the preponderance of evidence standard for resolving challenges to the reliability of testimony from a previously hypnotized witness.10

Our “clear and convincing” standard of proof in determining the admissibility of evidence apparently snuck into Texas law without reliance upon any Texas statute, rule, or case authority.11

This “clear and convincing” standard for assessing the reliability, hence admissibility, of certain testimony next appeared in Kelly v. State,12 in which this Court simply cited Zani for the proposition that this was an appropriate standard in determining the reliability of any “novel” scientific evidence under Tex.R. Evid. 702.13 Sure *791enough, when we later applied the Kelly reliability factors to all “hard science” expert evidence in Hartman v. State,14 and to all “soft science” expert testimony in Nenno v. State,15 we simply dragged Zani’s “clear and convincing” standard of proof forward. We did this in the same breath that we quoted approvingly the Supreme Court’s statement in Daubert that “a rigid [Frye ] ‘general acceptance’ requirement would be at odds with the ‘liberal thrust’ of the Federal Rules and their ‘general approach of relaxing the traditional barriers to “opinion” testimony.’ ”16 Adopting a “clear and convincing” standard for the admission of all expert opinion testimony is not “liberal,” it is not logical, and it is not appropriate under the Texas Rules of Evidence.

Rule 104(a) of our Texas Rules of Evidence governs the trial court’s authority to decide questions about the admissibility of evidence.17 Under that rule, the trial judge is the “gatekeeper” of evidence. It is his responsibility to ensure that all evidence admitted at trial is relevant, rehable, and admissible under the pertinent constitutions, statutes, rules of evidence and common law principles.

Although Rule 104(a) does not explicitly specify a burden of proof, the United States Supreme Court has always held that the proponent of evidence (whether it is expert testimony offered under Rules 702-703, lay opinion testimony under Rule 701, documents offered under Rule 902, or any evidence offered under any other Federal Rule of Evidence) must demonstrate by a preponderance of proof that the proffered item or testimony is admissible.18 We have expressly adopted that standard and cited to the seminal Supreme Court cases, Huddleston and Bourjaily, in other Rule 104(a) contexts.19 The federal courts have no difficulty in using the normal Rule 104(a) “preponderance of proof’ or “preponderance of evidence” standard set out in Daubert for their gatekeeping determi*792nations on the admissibility of expert testimony.20

I fail to understand why we employ an entirely different standard in determining the admissibility of expert evidence than the federal courts do under exactly the same rules -of evidence. I also fail to understand why we employ an- entirely different standard in determining the admissibility of expert evidence than we ourselves do under any other rule of evidence except Rule 702.

Furthermore, use of a “clear and convincing” standard to determine the admissibility of expert evidence leads to anomalous results in criminal and civil cases. For example, suppose Officer Obie proposes to testify as an expert in a criminal trial on the cause of a traffic accident. The trial judge finds that his testimony is not sufficiently rehable under this “clear and convincing” standard to allow the jury to hear it. However, in a civil case, Officer Obie’s same testimony on the same cause of the same traffic accident would be sufficient to support a final verdict and a large monetary recovery based on a mere preponderance of the evidence. Under our current case law, the standard for admitting expert evidence in a criminal trial is higher than the ultimate standard of proof for upholding a verdict in a civil trial. I cannot conclude that this dichotomy is logically or legally justifiable.

Therefore, while I agree with much of the Court’s reasoning in this case, I respectfully disagree with its continued use of an illogical and inappropriate “clear and convincing” standard of proof for admitting expert testimony or identification testimony by a lay witness who has been previously hypnotized.. •