INTRODUCTION

T1 Defendant, Deon Lomax Clopten, appeals his conviction for murder on grounds that the trial court abused its discretion when it excluded expert testimony regarding the reliability of eyewitness identification. Following existing Utah precedent, the court of appeals affirmed Clopten's conviction while inviting this court to revisit our position on the admissibility of such expert testimony. We reverse the decision of the court of appeals, vacate the conviction, and remand for a new trial.

BACKGROUND

1 2 In February 2006, Clopten was convict, ed of first-degree murder for the shooting of Tony Fuailemaa outside a Salt Lake City nightclub. At trial, Clopten maintained that someone else-a man named Freddie White-was responsible for the shooting. The testimony of several individuals who witnessed the murder and who identified Clopten as the perpetrator countered this assertion. In the absence of strong physical or forensic evidence against Clopten, the State leaned heavily on the eyewitness testimony to secure a conviction.

18 As part of his defense, Clopten sought to introduce the testimony of Dr. David Dodd, an expert on eyewitness identification. Clopten intended to elicit testimony from Dr. Dodd regarding various factors that can af-feet the accuracy of eyewitness identifications, including cross-racial identification, the impact of violence and stress during an event, the tendency to focus on a weapon rather than an individual's facial features, and the suggestive nature of certain identification procedures used by police.

T4 At Clopten's first trial, the district court initially allowed the expert testimony, but later reversed itself and ruled that Dr. Dodd could not testify. The district court changed course again and decided to permit the testimony, but this ruling was nullified *1106when, in May 2005, a mistrial was declared because of a conflict of interest unrelated to the issue before us. At the second trial, the court excluded the expert testimony. The trial court reasoned that the testimony was unnecessary since potential problems with eyewitness identification could be explained using a jury instruction, as has been the common practice in Utah since this court's decision in State v. Long, 721 P.2d 488 (Utah 1986). The trial court concluded that the jury instruction (hereinafter a "Long instruction") "does an adequate job" and that Dr. Dodd's testimony would be "superfluous" and "would only confuse the issue."

15 Clopten appealed the trial court's ruling. The court of appeals held that trial judges are afforded "significant deference to exclude expert testimony on this topic" and upheld the conviction. State v. Clopten, 2008 UT App 205, 119-21, 186 P.3d 1004. However, the court also cited numerous studies concluding "that jury instructions and eross-examinations do not adequately address the vagaries of eyewitness identification." Id. 119. Judge Thorne wrote a separate concurrence, in which he urged this court to "revisit the boundaries of trial court discretion in excluding expert testimony on the subject." Id. 182 (Thorne, J., concurring). We granted certiorari review, and we have jurisdiction under Utah Code section T8A-3-102(5) (2008).

ISSUE AND STANDARD OF REVIEW

T6 We granted certiorari review on whether expert testimony regarding the reliability of eyewitness identification should be presumed admissible when timely requested. "On certiorari, we review de novo the decision of the court of appeals, not that of the trial court." State v. Gardner, 2007 UT 70, T20, 167 P.3d 1074. A trial court's exclusion of expert testimony is reviewed for an abuse of discretion and is reversed if it "'exceeds the limits of reasonability.'" State v. Hollen, 2002 UT 35, 166, 44 P.3d 794 (quoting State v. Larsen, 865 P.2d 1855, 1861 (Utah 1998)).

ANALYSIS

T 7 Our analysis proceeds in four parts. In Part One, we summarize the evolution of Utah law as it pertains to the introduction of expert testimony regarding eyewitness identifications. Part Two examines the wealth of empirical research that, since our decision in Long, has solidly established the importance of expert testimony to explain factors contributing to eyewitness fallibility and the resulting possibility of mistaken identifications. In Part Three, we provide new guidance for the introduction of expert testimony on this subject. Finally, Part Four addresses our holding in this case.

I. UTAH LAW HAS BEEN INTER PRETED AS ESTABLISHING A DE FACTO PRESUMPTION AGAINST THE ADMISSION OF EXPERT TESTIMONY ON EYEWITNESS IDENTL-FICATION

[ 8 When we decided State v. Long in 1986, it was already apparent that "(allthough research has convincingly demonstrated the weaknesses inherent in eyewitness identification, jurors are, for the most part, unaware of these problems." 721 P.2d 488, 490 (Utah 1986). Thus we confronted a troubling quandary: while eyewitness identifications are frequently crucial to the State's case against a criminal defendant, the human ability to perceive and remember accurately is subject to numerous limitations. See id. at 488. In addition, it appears that jury members are frequently unaware of these limitations and thus give eyewitness identifications a disproportionate weight. Id. at 490.

T9 In Long, we considered the appropriateness of jury instructions as a way of familiarizing the fact-finder with these issues. Id. at 492. There, the defendant was convicted of aggravated assault based on an identification made by the victim, who had been wounded by a shotgun blast and acknowledged that his vision was "glossy" when he saw the shooter. Id. at 484. Counsel for the defendant requested a cautionary instruction regarding the accuracy of the identification, which the trial court declined to give. Id. at 487.

1 10 Prior to Long, the decision to issue a cautionary instruction regarding the infirmities of eyewitness testimony was left entirely *1107to the trial court's discretion. State v. Tucker, 709 P.2d 318, 316 (Utah 1985); State v. Reedy, 681 P2d 1251, 1252 (Utah 1984); State v. Newton, 681 P.2d 883, 834 (Utah 1984). Although this court cautioned that refusing to provide a requested instruction could constitute an abuse of discretion, see e.g., Reedy, 681 P.2d at 1252-53, until Long we had never reversed a single conviction on the grounds of such a refusal. Long, 721 P.2d at 487. As a result, trial judges in Utah rarely used the instruction, even in cases where there was serious doubt as to the reliability of the identification. Id. We therefore faced a choice between abandoning any pretext of requiring a cautionary instruction or giving the requirement teeth. We chose the latter course, reversed Long's conviction, and remanded the case for a new trial. Id. at 495. In addition, we directed trial courts to provide instructions "whenever eyewitness identification is a central issue in a case and such an instruction is requested by the defense." Id. at 492.

111 We also acknowledged that, because of doubts regarding its effectiveness in educating the jury, "[al cautionary instruction plainly is not a panacea." Id. at 492 n. 5. Despite that warning, Long left undisturbed previous holdings that discouraged the use of expert testimony as an alternative to jury instructions. These disincentives first appeared in State v. Griffin, which dismissed eyewitness expert testimony as a mere "lee-ture" that could invade the jury's role as sole evaluator of witness credibility. 626 P.2d 478, 481 (Utah 1981); accord State v. Maim-rose, 649 P.2d 56, 61 (Utah 1982). The Malmrose decision provoked a dissent by Justice Stewart, who argued that because the "inherent dangers of good faith error in eyewitness identification are widely recognized," it was error for the trial judge to neither admit expert testimony nor issue a cautionary instruction. Id. at 62, 65-66 (Stewart, J., dissenting). These misgivings later became the core of the majority opinion in Long.

12 It was never the intent of this court to establish cautionary instructions as the sole means for educating juries about eyewitness fallibility. Indeed, we carefully acknowledged that "[fjull evaluation of the efficacy of cautionary instructions must await further experience." Long, 721 P.2d at 492, n. 5. With the benefit of hindsight, however, it is clear that Long actually discouraged the inclusion of eyewitness expert testimony by failing to dispel earlier notions that such testimony would constitute a "lecture to the jury about how they should perform their duties." Malmrose, 649 P.2d at 61. As a result, trial judges reached two logical conclusions: (1) when in doubt, issuing cautionary instructions was a safe option; and (2) allowing expert testimony was hazardous if the expert "lectured the jury" about the credibility of a witness.

113 Subsequent decisions reinforced this bias. In State v. Hubbard, we held that the substance of expert testimony "can be just as adequately conveyed to the jury through the judge in a jury instruction." 2002 UT 45, 1 17, 48 P.3d 958. Further, we affirmed trial court rulings that "such evidence could cause confusion of the issues and could cause undue delay." State v. Butterfield, 2001 UT 59, 4 44, 27 P.8d 1183. Proponents of eyewitness expert testimony also found themselves in a dilemma regarding the specificity of the proffered testimony. On one hand, eyewitness expert testimony that was too specific was excluded as having "a significant tendency to cause the jury to abdicate its role as fact finder." Hubbard, 2002 UT 45, 1 20, 48 P.3d 953. If, on the other hand, the eyewitness expert only gave general testimony about memory phenomena, then it could be exelud-ed because it "did not deal with the specific facts from this case but rather would constitute a lecture to the jury about how it should judge the evidence." Buiterfield, 2001 UT 59, 144, 27 P.3d 1183. In addition, we held that a Long instruction is enough to render an erroneous exclusion harmless, even if the instruction failed to mention significant portions of the proffered expert testimony. FHubbard, 2002 UT 45, €20, 48 P.8d 953. Finally, in a continuation of our history prior to Long, neither this court nor the court of appeals has ever reversed a conviction for failure to admit eyewitness expert testimony. Given this history, it is not surprising that there is a de facto presumption against eyewitness expert testimony in Utab's trial courts.

*1108{14 This trend, acknowledged by both parties, is troubling in light of strong empirical research suggesting that cautionary instructions are a poor substitute for expert testimony. We turn now to a review of that research.

II. EMPIRICAL RESEARCH HAS CONVINCINGLY ESTABLISHED THAT EXPERT TESTIMONY IS NECESSARY IN MANY CASES TO EXPLAIN THE POSSIBILITY OF MISTAKEN EYEWITNESS IDENTIFICATION

1 15 " vagaries of eyewitness identification are well known; the annals of criminal law are rife with instances of mistaken identification" State v. Long, T21 P.2d 483, 491 (Utah 1986) (quoting Umited States v. Wade, 388 U.S. 218, 228, 87 S.Ct. 1926, 18 LEd.2d 1149 (1967)). Decades of study, both before and particularly after Long, have established that eyewitnesses are prone to identifying the wrong person as the perpetrator of a crime, particularly when certain factors are present.1 For example, people identify members of their own race with greater accuracy than they do members of a different race.2 In addition, accuracy is significantly affected by factors such as the amount of time the culprit was in view, lighting conditions, use of a disguise, distinetiveness of the culprit's appearance, and the presence of a weapon or other distractions.3 Moreover, there is little doubt that juries are generally unaware of these deficiencies in human perception and memory and thus give great weight to eyewitness identifications.4 Indeed, juries seemed to be swayed the most by the confidence of an eyewitness, even though such confidence correlates only weakly with accuracy.5 That the empirical data is conclusive on these matters is not disputed by either party in this case and has not been questioned by this court in the decisions that followed Long. See State v. Hubbard, 2002 UT 45, 116, 48 P.3d 958 (noting the "inherent deficiencies" in eyewitness testimony); State v. Butterfield, 2001 UT 59, \ 42, 27 P.3d 1183 (recognizing "the vagaries of eyewitness identification").

'I 16 The remaining issue is whether expert testimony is generally necessary to adequately educate a jury regarding these inherent deficiencies. As discussed below, we are now convinced that it is.6 In the absence of expert testimony, a defendant is left with two tools-eross-examination and cautionary instructions-with which to convey the possibility of mistaken identification to the jury. Both of these tools suffer from serious shortcomings when it comes to addressing the merits of eyewitness identifications.7 Addi*1109tionally, the admission of eyewitness expert testimony is gaining support in courts throughout the country.

A. Expert Testimony Has Been Shown to Be the Best Method for Educating the Jury About Factors That Can Contribute to Mistaken Eyewitness Identifications

{17 The most troubling dilemma regarding eyewitnesses stems from the possibility that an inaccurate identification may be just as convincing to a jury as an accurate one. In one study, subjects watched a mock trial of a defendant accused of armed robbery. One group of subjects heard only cireumstan-tial evidence against the defendant-they convicted at a rate of only 18 percent. The conviction rate jumped to 72 percent for a second group, which heard an eyewitness identify the defendant. A third group heard the same evidence and the same eyewitness, but was also told that the eyewitness was legally blind and had not been wearing glasses at the time of the crime. Despite the obvious unreliability of the eyewitness, 68 percent of this group still voted to conviet.8 As one leading researcher said: "[There is almost nothing more convincing than a live human being who takes the stand, points a finger at the defendant, and says 'That's the onel'" Elizabeth F. Loftus, Eyewitness Testimony 19 (1979). Because of this overreliance on questionable eyewitnesses, juries will often benefit from assistance as they sort reliable testimony from unreliable testimony.

€18 The challenge arises in determining how best to provide that assistance in cases where mistaken identification is a possibility. It is apparent from the research that the inclusion of expert testimony carries significant advantages over the alternatives, namely eross-examination and jury instructions.9

1. Expert Testimony Effectively Educates the Jury About the Possibility of Mistaken Identification Without Unfairly Favoring the Defendant

{19 Typically, an expert is called by a criminal defendant to explain how certain factors relevant to the identification in question could have produced a mistake. The expert may or may not be familiar with the facts of the case prior to the testimony, and in any case will not offer an opinion on whether the specific eyewitness identification is accurate or not. Instead, the relevant research is discussed in more general terms, thus allowing the jury to apply the information to whatever degree it sees fit.

20 Such testimony performs two beneficial functions. First, it teaches jurors about certain factors-such as "weapon focus" and the weak correlation between confidence and accuracy-that have a strong but counterin-tuitive impact on the reliability of an eyewitness. In other words, the testimony enables jurors to avoid certain common pitfalls, such as believing that a witness's statement of certainty is a reliable indicator of accuracy.10 Second, it assists jurors by quantifying what most people already know. An expert may discuss, for example, the degree to which accuracy is affected by a disguise or a long lapse between the crime and the identification. Importantly, expert testimony does not unfairly favor the defendant by making the jury skeptical of all eyewitnesses. In fact, when a witness sees the perpetrator under favorable conditions, expert testimony actually makes jurors more likely to convict.11 When expert testimony is used correctly, the end result is a jury that is better able to reach a just decision.

*11102. Cross-examination May Be Ineffective When the Eyewitness has Made a Mistaken Identification in Good Faith

T21 In the absence of expert testimony, the method most commonly used to challenge the veracity of eyewitnesses is cross-examination. But because eyewitnesses may express almost absolute certainty about identifications that are inaccurate, research shows the effectiveness of cross-examination is badly hampered. Cross-examination will often expose a lie or half-truth, but may be far less effective when witnesses, although mistaken, believe that what they say is true.12 In addition, as we recognized in Long, eyewitnesses are likely to use their "expectations, personal experience, biases, and prejudices" to fill in the gaps created by imperfect memory. 721 P.2d at 489. Because it is unlikely that witnesses will be aware that this process has occurred, they may express far more confidence in the identification than is warranted.13

122 Even if cross-examination reveals flaws in the identification, expert testimony may still be needed to assist the jury. Crossg-examination might show, for example, that the perpetrator was a different race than the eyewitness and was also wearing a disguise. Without the assistance of expert testimony, a jury may have difficulty assessing the import of those factors in gauging the reliability of the identification.14 For these reasons, we cannot rely on cross-examination as a surefire way to uncover the possibility of mistaken identification.

3. Research has Shown That Cautionary Instructions are Ineffective at Educating the Jury

123 Trial courts in Utah and around the nation have often tried to remedy the possibility of mistaken identification by giving cautionary instructions to the jury. This approach was pioneered in Umited States v. Telfaire, in which the D.C. Cireuit proposed a model instruction touching on common problems with identifications. 469 F.2d 552, 558-59 (D.C.Cir. 1972). In Long, we pointed out several deficiencies with the Telfuire approach and suggested a longer, more detailed instruction that incorporated more recent research,. Long, 721 P.2d at 494 n. 8. Variations of this model were thereafter used regularly by Utah trial courts and became collectively known as the "Long instruction." The standard Long instruction consists of general cautions about many factors known to contribute to mistaken identifications, such as brief exposure time, lack of light, presence of disguises and distractions, and effects of stress and cross-racial identification. Id. At the time, it seemed logical that this measure would substantially enhance a jury's ability to evaluate eyewitness accuracy.

124 Subsequent research, however, has shown that a cautionary instruction does little to help a jury spot a mistaken identification.15 While this result seems counterin-tuitive, commentators and social scientists advance a number of convincing explanations. First, instructions "given at the end of what might be a long and fatiguing trial, and buried in an overall charge by the court" are unlikely to have much effect on the minds of a jury. Peter J. Cohen, How Shall They Be Known? Daubert v. Merrell Dow Pharmaceuticals and Eyewitness Identification, 16 Pace L.Rev. 287, 272 (1996). Second, instructions may come too late to alter the jury's opinion of a witness whose testimony might have been heard *1111days before.16 Third, even the best cautionary instructions tend to touch only generally on the empirical evidence. The judge may explain that certain factors are known to influence perception and memory, but will not explain how this occurs or to what extent. As a result, instructions have been shown to be less effective than expert testimony.17

25 In conclusion, there is little reason to be confident that cross-examination and cautionary instructions alone provide a sufficient safeguard against mistaken identifications. In contrast, expert testimony has been shown to substantially enhance the ability of juries to recognize potential problems with eyewitness testimony.

B. The Need to Admit Eyewitness Expert Testimony is Recognized by Courts Throughout the Nation

126 The admissibility of eyewitness expert testimony was first considered by the nation's courts starting in the 1970s. In general, these early decisions excluded the testimony on grounds that have since been undercut by the research cited above.18 The majority of courts that have considered the issue since then have held that admission or exclusion of the evidence is within the broad discretion of the trial court.19 Starting in the 1980s, however, numerous state and federal courts recognized that the statistical evidence on eyewitness inaccuracy was too substantial to ignore. Many of these appellate courts instructed trial judges that, under certain cireumstances, it would be an abuse of discretion not to allow expert testimony on the subject.

1 27 The first such decision came from the Arizona Supreme Court in State v. Chapple, 185 Ariz. 281, 660 P.2d 1208 (1983). There, the trial court exeluded an eyewitness expert on grounds that the testimony would not assist the jury and that cross-examination was sufficient to reveal problems with the identification. Id. at 1228-24. The Arizona Supreme Court reversed and held that, in cases where the expert would provide information about eyewitness factors relevant to the case, it was error to exclude the testimony as unhelpful. Id. Over the last two decades, numerous other state courts have either reversed decisions to exclude or encouraged the inclusion of eyewitness expert testimony.20 Calling on courts to "face up to the reliability problems of eyewitness identification," the Supreme Court of California held in People v. McDonald that "it will ordinarily be error" to exclude qualified expert testimony when an eyewitness identification is key to the prosecution's case and is not "substantially corroborated" by independent evidence. 37 Cal.8d 351, 208 Cal.Rptr. 236, 690 P.2d 709, 717, 727 (1984), overruled on other grounds by People v. Mendoza, 28 Cal.4th 896, 98 Cal.Rptr.2d 481, 4 P.3d 265 (2000).

[ 28 Similar positions have been adopted in the federal courts. The Sixth Circuit was the first to hold that eyewitness expert testimony is sufficiently reliable to assist the jury. United States v. Smith, 786 F.2d 1108, 1107 (6th Cir. 1984). The next year, the Third Cireuit held that exclusion of expert testimony was an abuse of discretion when the conviction was based solely on an eyewitness identification. United States v. Down*1112ing, 758 F.2d 1224, 1226 (3rd Cir. 1985). The Third Circuit further held that the discretion of trial judges was limited in this area, and that rule 702 of the Federal Rules of Evi-denee "requires ... that expert testimony on eyewitness perception and memory be admitted at least in some circumstances." Id. at 1282. The Fifth Circuit reached a similar conclusion in Umited States v. Moore and held that "in a case in which the sole testimony is casual eyewitness identification, expert testimony regarding the accuracy of that identification is admissible and properly may be encouraged." 786 F.2d 1308, 1818 (5th Cir. 1986).

29 In short, a growing number of courts have recognized that eyewitness expert testimony is both reliable and helpful to the jury. See United States v. Smithers, 212 F.3d 306, 311-12 (6th Cir. 2000). Numerous courts have also rejected the idea that such testimony is impermissible because it is misleading or because it "invades the province" of the jury. See, eg., McDonald, 208 Cal.Rptr. 236, 690 P.2d at 722 (calling the latter argument no "more than a shibboleth which ... would deprive the jury of important information") (quoting Patrick M. Wall, Hye-Witness Identification in Criminal Cases 2183 (1965). With this in mind, we now provide new guidance regarding the admissibility of such expert testimony in Utah trial courts.

III. EXPERT TESTIMONY REGARDING EYEWITNESS IDENTIFICATION SHOULD BE ADMITTED IF IT MEETS THE STANDARDS SET OUT IN UTAH RULE OF EVIDENCE 702

130 Our previous holdings have cere-ated a de facto presumption against the admission of eyewitness expert testimony, despite persuasive research that such testimony is the most effective way to educate juries about the possibility of mistaken identification. Clopten urges us to remedy this situation by creating a new rule establishing eyewitness expert testimony as presumptively admissible. We decline to adopt an outright presumption. Instead, we clarify how eyewitness expert testimony fits into the Utah Rules of Evidence. We ultimately hold that the testimony of a qualified expert regarding factors that have been shown to contribute to inaccurate eyewitness identifications should be admitted whenever it meets the requirements of rule 702 of the Utah Rules of Evidence. We expect this application of rule 702 will result in the liberal and routine admission of eyewitness expert testimony, particularly in cases where, as here, eyewitnesses are identifying a defendant not well known to them.

131 We conclude that this approach best conforms to the intent of rule 702. Under the rule, trial judges perform a gatekeeper function to sereen out unreliable expert testimony and are advised to view proposed experts with "rational skepticism." Utah R. Evid. 702, advisory comm. note (2007). It should be noted that the current rule 702 was amended in 2007 and was not in effect when Clopten was convicted. We therefore analyze the admission of eyewitness expert testimony under both the current version of the rule and the version that was in place at the time of Clopten's trial.21 Analysis of the current version of rule 702 consists of two basic parts. First, the trial judge must find that the expert testimony will "assist the trier of fact." Utah R. Evid. 702(a). Second, the testimony must "meet a threshold show*1113ing" of reliability. Id. 702(b). Focusing on this analysis allows us to dispose of several arguments that have been used in the past to exclude eyewitness experts.

A. Expert Eyewitness Testimony Is Helpful in Cases Involving Identification of a Stranger and One or More of the Factors Known to Affect Eyewitness Aceu-racy

As explained in Part One, trial judges have often exeluded eyewitness experts on grounds that the testimony will not be helpful to the jury. We now hold that, in cases where eyewitnesses are identifying a stranger and where one or more established factors affecting accuracy are present,22 the testimony of an eyewitness expert will meet rule 702's requirement to "assist the trier of fact." As the research makes clear, the top-ies covered by eyewitness experts are often beyond the common knowledge of ordinary jurors and usually cannot be effectively elicited through cross-examination alone. It is therefore inappropriate for a trial judge to exclude an eyewitness expert merely on grounds that the jurors' life experiences already provide enough information, that cross-examination will suffice to reveal weaknesses with the identification, or that the testimony presented will be misleading or confusing.

133 We are not mandating the admission of eyewitness expert testimony in every case. Trial judges must still analyze whether the testimony will assist the jury, and in some cases it will not. The research on eyewitness identifications, for example, almost exelusively focuses on individuals who are attempting to identify a stranger. If the eyewitness is identifying someone with whom he or she has been aequainted over a substantial period of time (for example, a family member, longtime business associate, neighbor, or friend), then expert testimony is not likely to assist the jury in evaluating the accuracy of a witness's testimony. Similarly, there may be cases in which a witness viewed the perpetrator under such ideal conditions that an expert would not be able to identify factors that could have contributed to a misidentification. In such cases, the trial judge retains the discretion to exclude the testimony. We expect, however, that in cases involving eyewitness identification of strangers or near-strangers, trial courts will routinely admit expert testimony.

134 In addition, the testimony of an eyewitness expert should not be considered cumulative or duplicative of cautionary instructions to the jury. To reconcile this holding with our previous decision in Long, we modify our guidance regarding cautionary instructions. In cases where the defense does not call an eyewitness expert, the holding in Long still applies. In other words, the trial judge must provide a cautionary instruction if one is requested by the defense and eyewitness identification is a "central issue." State v. Long, 721 P.2d 488, 492 (Utah 1986). Where eyewitness expert testimony is heard, however, Long no longer applies and the inclusion of a cautionary instruction, if requested, is a matter for the trial judge's discretion.

*1114B. Eyewitness Expert Testimony Is Sufficiently Reliable to Be Admitted Under Rule 702 and Should Not Be Excluded as a Lecture to the Jury

T35 Though it disputes whether judicial notice of reliability is appropriate, the State concedes that the testimony of eyewitness experts is based on sufficiently reliable principles to merit admission. Under rule 702, there are two ways to establish reliability. First, the proponent may show that the "principles or methods underlying the testimony meet a threshold showing that they (1) are reliable, (M) are based upon sufficient facts or data, and (iii) have been reliably applied to the facts of the case." Utah R. Evid. 702(b) (2007). Second, the proponent may show that the underlying principles or methods "are generally accepted by the relevant expert community." Id. We hold that the testimony of eyewitness experts satisfies both tests. The phenomena that eyewitness experts seek to explain have been reviewed and replicated many times in recent decades. In addition, this court recognized in State v. Rimmasch that it was appropriate to take judicial notice of "general acceptance" of those principles in the community of researchers that specialize in the study of eyewitness identification. 775 P.2d 888, 398 (Utah 1989).

136 Finally, we hold that eyewitness expert testimony should not be excluded as intruding on the province of the jury or as an impermissible "lecture." The current version of rule 702 resolves the Catch-22 of the old rule, in which proponents of eyewitness expert testimony risked being too specific on the one hand and too general on the other. The rule as now written recognizes that "lilt might be important in some cases for an expert to educate the factfinder about general principles, without attempting to apply these principles to the specific facts of the case." Utah R. Evid. 702, advisory comm. note (2007). It is therefore acceptable for an eyewitness expert to "give a dissertation or exposition" of factors found in the case that are understood to contribute to eyewitness inaccuracy. Id. As long as the expert does not attempt to tell the jury that a specific eyewitness identification either is or is not accurate, then the expert has not impinged on the jury's duty as the sole evaluator of witness credibility.

C. Eyewitness Expert Testimony Was Also Admissible Under the Version of Rule 702 That Was in Effect When Clopten Was Tried

137 Our analysis has up to this point relied on the current version of rule 702. As the State correctly points out, this version was not in effect when Clopten was tried. Thus, we must evaluate the decision to exclude the expert testimony in light of the law that existed in 2005. Under the previous version of rule 702, expert testimony could be admitted if the witness was properly qualified as an expert and the testimony would "assist the trier of fact." Utah R. Evid. 702 (2006). There was no language in the older version of rule 702 to require a threshold showing of reliability. However, we previously held that unreliable expert testimony could not assist the jury and therefore applied the three-prong test developed in Rim-masch to test reliability. State v. Crosby, 927 P.2d 688, 640-41 (Utah 1996) (citing Rimmasch, TTS P.2d at 400).

(38 Because the current version of rule 702 incorporates an updated reliability analysis for expert testimony, the Rimmasch test has been subsumed in the new rule. However, the old "Rule 702 plus Rimmasch" test yields the same result as the current rule when applied to eyewitness expert testimony. The Rimmasch opinion itself states in regards to Long that judicial notice may be taken "of the fact that eyewitness identification is unreliable on [the] basis of general acceptance of that view." 775 P.2d at 898. The Rimmasch reliability standard is therefore satisfied. Since there has never been a dispute over Dr. Dodd's qualifications, the only remaining issue is helpfulness. The language in rule 702 has not changed regarding helpfulness; in both the old and new versions, expert testimony may be admitted if it "will assist the trier of fact to understand the evidence or to determine a fact in issue." The application of the older version of rule 702 therefore does not change the analysis for this case. In cases where an eyewitness *1115is identifying a stranger and in which various factors that can affect accuracy are present, eyewitness expert testimony is helpful to the jury and thus admissible. Having determined that Dr. Dodd's testimony could have been admitted, we now consider whether its exclusion was an abuse of discretion.

IV. THE TRIAL COURT ABUSED ITS DISCRETION BY EXCLUDING THE EXPERT TESTIMONY

189 Before deciding the ultimate outcome of this case, we analyze whether the decision to exclude Dr. Dodd's testimony constituted an abuse of discretion by "'exceed[ing] the limits of reasonability.'" State v. Hollen, 2002 UT 35, 166, 44 P.3d 794 (quoting State v. Larsen, 865 P.2d 1355, 1861 (Utah 1998)). We are less likely to find an abuse of discretion "'where there has been no showing that the excluded evidence would probably have had a substantial influence in bringing about a different verdict.'" State v. 2002 UT 45, 120, 48 P.8d 953 (quoting State v. Butterfield, 2001 UT 59, 4 43, 27 P.8d 1188). Similarly, we find errors by the trial court harmful only if there is a "reasonable likelihood" that the verdict would have been different had the expert testimony been included. Steffensen v. Smith's Mgmt. Corp., 862 P.2d 1842, 1847 (Utah 1998). If no such reasonable likelihood exists, then the error is harmless and reversal is not warranted. In order to analyze these issues, we review the facts presented in the record.

€40 Tony Fuailemaa, the victim in this case, was shot and killed outside a nightclub following a rap concert. An undercover police officer responded and was told by the victim's girlfriend, Shannon Pantoja, that the shooter was "the guy in the red.23 The officer gave chase and saw several men jump into a Ford Explorer and drive away at high speed. A police pursuit ensued and resulted in the capture of Clopten and three other men. Clopten was in the driver's seat of the Explorer at the time of the arrest. Freddie White, the individual identified by Clopten as the shooter, was in the rear passenger seat. Both Clopten and White are African-American. Clopten was wearing both a red hooded sweatshirt and red pants at the time of arrest, while White was wearing a red T-shirt. Another red hooded sweatshirt was later found in the Explorer near where White had been sitting; the evidence suggested that White had been wearing it earlier in the evening. The handgun was found on the side of a road, having been thrown from the Explorer during the pursuit.

141 The State was unable to link Clopten to the handgun using fingerprints or other forensic evidence. Instead, the State relied heavily on eyewitness testimony. Pantoja testified that she was standing some fifteen feet away when she saw the man in red shoot Fuailemaa. She was taken by police to the scene of the arrest and there identified Clop-ten as the shooter. Pantoja also picked Clopten out of a police lineup some thirteen months later and identified him as the shooter at trial. Clopten was also identified by Melissa Valdez, another concertgoer who witnessed the shooting. Valdez testified that she had spoken briefly to a man dressed in a red sweatshirt both before and after the concert at the nightelub. Valdez said that just after speaking to the man in red for the second time, she saw him shoot Fuailemaa in the back of the head. She identified Clopten as the man in red both in a photo array and at trial. The State also introduced testimony given by Christopher Hamby, who accompanied Clopten to the concert and was riding in the Explorer at the time of the arrest. Hamby testified at a preliminary hearing that Clopten was the shooter. Finally, the State called Robert Land, who had been in prison with Clopten following the arrest. Land claimed that Clopten confessed to killing Fuailemaa.

T 42 There were significant problems, however, with the State's case. Hamby initially told police that he had not witnessed the *1116shooting. Hamby changed his story after being told by police that Clopten would inevitably be convicted and that he could either "be a witness" or "go to jail for many years." Hamby disappeared prior to trial, leaving the State no alternative but to read his preliminary hearing testimony into evidence. Land's credibility was also attacked since he received a substantially reduced prison sentence in exchange for his testimony. There were also questions about whether Clopten would have been able to have the in-depth discussion that Land described, since the two inmates were housed in different parts of the prison. Given these issues, the testimony of the other eyewitnesses-particularly that of Pantoja-were of primary importance.

143 Pantoja had never met or seen Clop-ten prior to the night of the murder. She testified at trial that, upon arriving at the nightclub, Fuailemaa recognized Clopten as someone with whom he had previously been incarcerated. Fuailemaa pointed Clopten out to Pantoja, and told her that Clopten "had some problems with the homies out in the prison." The murder occurred some two hours later, as Fuailemaa and Pantoja were returning to their parked car. Pantoja spoke to the police on numerous occasions, and she testified at a preliminary hearing and at both trials.

1 44 Pantoja's testimony at trial contained numerous inconsistencies when compared to her previous statements. At trial, for example, Pantoja testified that she looked closely at Clopten as she entered the nightclub prior to the shooting. In an interview with police just two days after the shooting, however, she stated that "I didn't really look at [Clop-ten] when I entered the club." In addition, the circumstances under which she saw the shooting and later identified the perpetrator indicate that inaccuracy could have occurred at several points. Just after the shooting, for example, a frantic Pantoja told police that the shooter was "the guy in red." She did not identify Clopten by name, even though Fuailemaa had pointed him out by name earlier in the evening. Less than an hour later, a still-distraught Pantoja was told by police that "they have some suspects pulled over" and that they needed her "to go identify these guys." When she demurred, an officer told her to "[dJo it for Tony." After being transported to the seene of the arrest, she viewed the four men one at a time at a distance of some thirty-five feet. All four men were handcuffed. Although there was evidence that White had been wearing a red hooded sweatshirt earlier in the night, at the time of the identification only Clopten was wearing such a garment. Pantoja identified Clopten as the shooter, and a short time later picked Clopten out of a photo array. She also identified Clopten in a police lineup approximately thirteen months later and again at trial. When asked by defense counsel how she could identify Clopten in a lineup after so much time had passed, Pantoja testified that she picked Clopten out due to his distinctive hairline. Every other eyewitness, however, testified that the perpetrator had a hood pulled up over the top of his head at the time of the shooting.

4 45 The testimony of Melissa Valdez presented similar difficulties. Valdez had also never seen Clopten prior to the night of the murder. At trial, she said that she had spoken twice to a man wearing red about getting tickets to the concert. Moments after speaking to the man the second time, Valdez heard shots and turned to see him standing over Fuailemaa with a gun. She later picked Clopten out of a photo array. In a police interview shortly after the shooting, however, Valdez stated that the shooter had not been wearing red pants. This weakened her own identification, contradicted Pantoja's testimony, and suggested that someone other than Clopten (who was wearing red pants when arrested) was responsible. In addition, Valdez's description of the red sweatshirt worn by the shooter varied from the sweatshirt actually worn by Clopten, and may have better matched the sweatshirt later found near White's seat in the Explorer.

1 46 Both Valdez and Pantoja witnessed a brutal crime committed by a stranger. They each saw the shooter for no more than a few seconds, from some distance away, at night, and while in extreme fear for their own lives. The shooter's facial features were likely disguised by a hood. The shooter was of a different race than either eyewitness, and the presence of a weapon may have served as a significant distractor. Pantoja's identifica*1117tion may have been biased by her expectations, since Fuailemaa had told her just before the murder that he and Clopten were enemies. Her identification may also have been affected by cireumstances that occurred later, such as the fact that Clopten was the only individual wearing a red sweatshirt at the time of the initial "show up" identification. Pantoja's statement that she was urged by police to go identify a perpetrator for the sake of her murdered boyfriend, at a time when she was still extremely distraught, also creates doubts as to her accuracy. Finally, the fact that Pantoja insisted that she remembered the shooter's distinctive hairline, when others testified that the shooter's head was covered, raises a fair question as to whether Pantoja actually recalled the shooter's hairline, or if she later incorporated that feature into her memory after seeing pictures of Clopten.

¶47 In short, the cireumstances found in the Clopten trial are exactly those under which the testimony of an eyewitness expert is most helpful to a jury. Dr. Dodd, the proffered expert in this case, could have testified about research into how eyewitness identification of a stranger is affected by stress, disguises, darkness and length of exposure. He could have quantified the impact of factors such as weapon focus and eross-racial identification. Dr. Dodd could also have testified as to the impact that comments made by police officers may have on an eyewitness making an identification. Additionally, he could have discussed a common phenomenon in which witnesses fill gaps in their memory with information obtained later and thus, over time, become more and more certain of identifications that may be inaccurate. All of these factors were present here, and thorough testimony by a qualified expert as to their nature would have significantly assisted the jury in evaluating the accuracy of the State's most important witnesses. In addition, the critical importance of the eyewitnesses here forces the conclusion that the proffered testimony might have had a "substantial influence in bringing about a different verdict." It was therefore unreasonable for the trial court to rule that such expert testimony would be superfluous. While we acknowledge that the trial court followed established precedent, we hold that the court of appeals erred in concluding that the exclusion of Dr. Dodd's testimony was not an abuse of discretion.

148 The court of appeals acknowledged the body of research that supports the admission of eyewitness expert testimony. State v. Clopten, 2008 UT App 205, 19, 186 P.3d 1004. It then noted that "[the precise situation in this case is somewhat different" because some of the eyewitnesses "were not complete strangers to Clopten." Id. 120. The court of appeals therefore concluded that any error made by the trial court was not prejudicial, Id. We disagree. While Valdez and Pantoja each had brief encounters with Clopten prior to the shooting, both were essentially strangers to him.24 To say that either eyewitness was "acquainted" with Clopten, and thus not subject to the vagaries of identifying a stranger, stretches the definition of acquaintance too far. In addition, there is little doubt that eyewitness testimony was of paramount importance in the State's case against Clopten, as there was no other independent corroboration of the shooter's identity. Thus, we cannot conclude that the jury would have convicted solely on the basis of other evidence or testimony from other witnesses. There is therefore a reasonable likelihood that, if allowed to hear Dr. Dodd's testimony, the jury would have questioned the accuracy of the eyewitnesses more rigorously and would not have convicted Clopten. While the trial judge did give a Long instruction, there is no reason to believe this measure was effective in mitigating the error. This is particularly true because the instruction discussed neither the phenomenon of "weapon focus" nor the lack of correlation between eyewitness confidence and accuracy. Dr. Dodd's testimony would have reached both issues. We therefore conclude that it is necessary to reverse the court of appeals and vacate Clopten's conviction.

CONCLUSION

[ 49 We are always reluctant to reverse a jury's decision to convict, particularly when the crime in question is as serious as this one. The seriousness of the crime, however, *1118makes it only more imperative that the jury's decisionmaking abilities are supported by the best information available. If unreliable identifications are not addressed properly at trial, then there exists an unacceptable risk of the innocent being punished and dangerous criminals remaining at large. We therefore hold that, in cases where eyewitnesses are identifying a stranger and one or more established factors affecting accuracy are present, the testimony of a qualified expert is both reliable and helpful, as required by rule 702. Such eyewitness expert testimony should therefore be routinely admitted, regardless of whether the trial judge decides to issue a cautionary instruction. Given the cireumstances present in this case, we hold that the court of appeals erred; the trial court's decision to exclude eyewitness expert testimony was an abuse of discretion that cannot be considered harmless. Accordingly, we reverse the decision of the court of appeals, vacate Clopten's conviction and remand for a new trial in accordance with our decision today.

150 Justice PARRISH and Justice NEHRING concur in Chief Justice DURHAM'S opinion.

concurring in part and dissenting in part:

151 Our case law has consistently recognized that the decision to admit or exclude expert testimony is within the discretion of the district court.1 Although this discretion is not unconstrained, and is obviously limited by the parameters set out in the Utah Rules of Evidence, it is nonetheless real and important. It is fundamentally a product of the structure of our judicial system, in which district court judges are placed in a superior position to evaluate the proffered testimony in light of the principles set out in the rules of evidence.2

4 52 Accordingly, I agree with the majority's rejection of the presumption against admissibility of eyewitness expert testimony that seems to have been precipitated by our decision in State v. Long3 as well as the majority's clarification that the presence of a Long instruction does not necessarily render eyewitness expert testimony improper or superfluous. I also concur in the majority's express rejection of a presumption in favor of eyewitness expert testimony and agree with its acknowledgment that decisions regarding the admissibility of eyewitness expert testimony should be made by district courts based on a standard application of Utah Rule of Evidence 702.

€58 But after declaring its view that all requests for eyewitness expert testimony are to be evaluated according to the principles of rule 702, the majority then goes on to perform this evaluation for the district courts, concluding that these requirements are satisfied in every case involving an eyewitness identification of a stranger.4 This effectively dictates to district courts how they should conduct the rule 702 analysis in all cases involving an eyewitness identification of a stranger, leaving them no discretion in an area where we have emphasized the importance and propriety of their discretion in the past.

T54 I would take a different approach. Rather than categorically decide that eyewitness expert testimony satisfies the elements *1119of rule 702, I would simply instruct the district courts that they are to treat eyewitness expert testimony like any other type of expert testimony and determine its admissibility based on the requirements of the rule. I would neither create a presumption in favor, nor one against, the admission of eyewitness expert testimony, and district court rulings on the admissibility of such expert testimony would be entitled to the same deference we have traditionally accorded rulings on the admissibility of other types of expert testimony.

(55 Because a presumption against the admission of eyewitness expert testimony seems to have developed based on our decision in Long, and because this presumption may have influenced the district court's decision to deny admission of Clopten's proffered eyewitness expert testimony, I would remand this case for a new trial to allow the district court to exercise its discretion in light of the principles I have set forth above. Accordingly, I respectfully dissent from Parts III and IV of the majority opinion.

1] 56 Justice WILKINS concurs in Associate Chief Justice DURRANTS opinion.