dissenting:
,¶ 171. If Curtis Giovanni Flowers is so clearly guilty, why not allow him to defend himself? Conviction of a capital crime and the decision to put a defendant to death should come from a properly instructed jury following a fair trial at which the defendant was afforded every reasonable opportunity to defend against the charges. In this case, the circuit judge erred by refusing to allow the accused to defend the charges by calling a qualified expert to provide perfectly admissible expert opinions, and by instructing the jury on an aggravating factor for which the State provided no evidence.
I. Dr. Neuschatz should have been allowed to testify.
¶ 172. The trial judge categorically prohibited Flowers’s expert, Dr. Jeffrey Neuschatz, from providing expert testimony on eyewitness identification. It was not that the trial judge disagreed with the evidence in Dr. Neuschatz’s expert affidavit — he ignored it. The trial judge felt that any helpful information that could have been provided by the expert could just as easily have been developed through cross-examination. But the information the expert would have provided could never be developed on cross examination, because the witnesses would not know they had made misidentifications.
¶ 173. This Court reviews a trial court’s ruling on the admissibility of evidence under a deferential abuse-of-discretion standard.7 Under Mississippi Rule of Evidence 702 an expert’s testimony must satisfy a two-part test.8 “First, the witness must be qualified by virtue of his or her knowledge, skill, experience!,] or education. Second, the witness’s scientific, technical!,] or other specialized knowledge must assist the trier of fact in understanding or deciding a fact in issue.”9
¶ 174. The trial judge is to act as a gatekeeper and may look to “a non-exhaustive, illustrative list of reliability factors for determining the admissibility of expert witness testimony.”10 The five factors look at whether: (1) the theory or technique can be and has been tested; (2) the theory has been subjected to peer review and publication; (3) there is a high known or potential rate of error; (4) there are standards controlling the technique’s operation; and (5) the theory or technique enjoys general acceptance within a relevant scientific community.11 The presence or absence of any one of these factors is not dispositive.
*1077¶ 175. The issue of allowing experts to testify about the reliability of eyewitness identification is not new. A majority of state jurisdictions allow eyewitness-identification experts to testify, provided they are properly qualified and their opinions satisfy reliability concerns, such as those set forth in Daubert. For instance, the Tennessee Supreme Court — evaluating factors similar to those mentioned in Dau-bert — reversed its prior position seven years ago and now allows properly qualified eyewitness experts to testify, finding that such testimony can be helpful in the jury’s evaluation of an eyewitness’s identification.12 And the vast majority of states and the District of Columbia either give trial courts discretion to allow eyewitness-identification expert testimony, or require the admission of such testimony.13
¶ 176. In this case, the trial judge first found that the trier of fact did not need Dr. Neuschatz’s testimony. He stated: “Given the extensive cross-examination of Mr. Collins and because all other witnesses knew Mr. Flowers on sight, I do not believe an expert on witnesses identification *1078would assist the jury in the least bit in this case.” This conclusion was demonstrably incorrect. The trial judge provided no analysis of how cross-examination of the witnesses could develop the opinions provided by Dr. Neuschatz, or how the witnesses themselves could provide expert testimony. The eyewitnesses certainly could provide the facts, such as distance, lighting, length of time, etc. But Dr. Neuschatz was not presented to testify to the facts alone. His role was to begin with the facts provided by the eyewitnesses, and then provide expert testimony as to the difficulties and reliability associated with identification. The eyewitnesses themselves certainly could not provide the testimony expected from Dr. Neuschatz.
¶ 177. Nor could cross-examination provide an adequate substitute for Dr. Neus-chatz’s testimony. A skillful attorney may utilize cross-examination to expose contradictions in a witness’s testimony. But no attorney — of even the greatest skill — can cross-examine a witness in such a way to expose that the witness did not see what the witness actually believes he saw. And that is exactly the purpose of expert eyewitness-identification testimony. Dr. Neuschatz — if allowed to testify — would have explained scientific principles affecting the accuracy of a witness’s perception of what he saw.
¶ 178. The trial judge’s entire Daubert analysis was as follows:
Further, though, I think aside from the fact that I do not think this would assist the trier of fact, I do not think all the Daubert principles have been shown to the Court to apply to this testimony. The defense has not offered to this Court this idea that the theories the defense would testify to or the area of expertise proffered are generally accepted. I do not know whether the testimony has a potential or known high rate of error. I do not have anything about that. I do not know anything about the principles and methods used to come to conclusions. And basically, the witness did not apply the principles and methods to the facts of this case. He made broad generalizations in the document that was submitted, but he did not specifically apply those principles and methods to the facts of this case. So for all these reasons, I do not think that this witness testimony would be appropriate, and I am going to deny the testimony of this witness.
The following is a detailed analysis of each sentence of the trial judge’s Daubert analysis, and it demonstrates that the trial judge abused his discretion in refusing to allow Dr. Neuschatz to testify as an expert.
I do not think all the Daubert principles have been shown to the Court to apply to this testimony.
¶ 179. This statement suggests the trial court erroneously believed the defendant had a burden to show that “all the Daubert factors” applied. Clearly, this is not the case.
The defense has not offered to this Court this idea that■ the theories the defense would testify to or the area of expertise proffered are generally accepted.
¶ 180. This conclusion was incorrect. Dr. Neuschatz cited thirty-three studies in his affidavit, each of which indicated general acceptance in the field of psychology. The State offered no evidence in rebuttal of Dr. Neuschatz’s affidavit on this point.14
*1079¶ 181. The majority agrees with the trial judge on this point, but with little explanation. The majority notes several times that Dr. Neuschatz did not attach each article to his affidavit, but admits he was not required to do so. Clearly, under our Rules of Evidence, experts are not required to attach the basis of their opinions to their affidavits. Instead, they are allowed to rely on studies that they do not provide to the court, unless requested by the court. Rule 705 provides
The expert may testify in terms of opinion or inferences and give his reasons therefor without prior disclosure of the underlying facts or data, unless the court requires otherwise. The expert may in any event be required to disclose the underlying facts or data on cross-examination.15
¶ 182. In this case, neither the trial judge nor the State requested production of the underlying data or the articles upon which Dr. Neuschatz relied. It was clear error for the trial judge to reject Dr. Neuschatz’s affidavit merely because the articles were not attached. And the majority compounds the error by finding that the trial judge did not abuse his discretion in finding that Flowers presented insufficient evidence on this point. The trial court failed to identify a single piece of evidence that Flowers lacked.
¶ 183. In Watts v. Radiator Specialty Company, this Court considered whether an expert’s opinions were actually supported by the case studies he cited.16 The expert in that case did not provide the studies to the court.17 Though this Court found that the trial court properly excluded the expert’s testimony because the studies did not support his conclusion, the Court never questioned his failure to provide those studies.18
I do not know whether the testimony has a potential or known high rate of error. I do not have anything about that.
¶ 184. Again, on this point, the trial judge was incorrect. In his affidavit, Dr. Neuschatz discussed extensive testing that produced precise error rates. For instance, in discussing the factor of exposure time, Dr. Neuschatz stated:
Exposure duration had a significant impact on identification accuracy. Ninety-five percent of the young adults and 85% of the older adults made correct identifications from the robber-present photo arrays when the robber was exposed for 45 seconds, but only 29% of the young adults and 35% of the older adults made correct identifications when the robber was exposed for 12 seconds.
¶ 185. In discussing the appearance-change disguise factor, he stated: “The average performance levels across the six studies, which involved over 1,300 eyewitness identifications, was 57% correct when *1080uncovered and versus 44% when a hat was worn.”
¶ 186. In discussing cross-race identification, Dr. Neuschatz stated:
The reliability of - the ORB [own-race bias] was examined by Meissner and Brigham (2001) who meta-analyzed the results of 31 separate studies involving 91 separate experimental tests of own versus same-race identifications. The studies included over 5,000 participants. Across all studies, eyewitnesses were 1.4 times more likely to correctly identify members of their own race than members of other races, and they were 1.56 times more likely to falsely identify members of other races than members of their own race.
¶ 187. These and other studies cited in the affidavit speak not only to error rates but also to the reliability of Dr. Neus-chatz’s opinions.
I do not know anything about the principles and methods used to come to conclusions.
¶ 188. Here again, the trial judge was incorrect. The studies cited above, and the balance of Dr. Neuschatz’s affidavit are replete with descriptions of the methods and principles used in the numerous studies.
¶ 189. Dr. Neuschatz’s affidavit contained detailed information about studies on: (1) how memory works; (2) effect of exposure time on reliability of memory; (3) effect of appearance changes and disguises on identification; (4) the phenomenon of unconscious transference where one situation is confused with another; (5) problems with crossracial identifications; (6) problems with post-identification feedback; and (7) problems with suggestive lineup procedures.
And basically, the witness did not apply the principles and methods to the facts of this case. He made broad generalizations in the document that was submitted, but he did not specifically apply those principles and methods to the facts of this case.
¶ 190. Again, the trial judge was incorrect. Dr. Neuschatz indeed applied the principles and methods in his field of expertise to the facts in the Flowers case. For instance, he analyzed in detail the effect of the police’s lineup identification. Also, he addressed issues involving Porky Collins’s identification, including problems with cross-racial identifications.
Refusing to allow Dr. Neuschatz to testify was not harmless error.
¶ 191. Dr. Neuschatz’s testimony would have provided the jury a fair basis upon which to judge the accuracy of Collins’s identification. I also cannot accept the majority’s conclusion that “Collins’s identification was far from the only evidence of guilt in the instant case, and it cannot be label ‘critical.’ ”
¶ 192. It is true that the State had other witnesses, but. none provided the powerful eyewitness testimony that Collins provided. He is the only witness who placed Flowers at the scene of the murders shortly before they were committed. None of the witnesses was more critical to the State’s case, and none was more important for the defense to attack.
¶ 193. The State offered nothing in rebuttal of Dr. Neuschatz’s opinions. No experts, articles, studies, or evidence of any kind that tended to refute his opinions and methodology. Accordingly, Dr. Neus-chatz should have been allowed to testify in this case.
II. The trial judge should have denied the State’s requested instruction on the “commission of a capital crime to avoid arrest” aggravator.
¶ 194. When Flowers’s case proceeded to the sentencing phase, the trial judge *1081instructed the jury to consider three potential aggravating factors:
1. The Defendant knowingly created a great risk of death to many persons.
2. The capital offense was committed while the Defendant was engaged in the commission of the crime of armed robbery for pecuniary gain.
8. The capital offense was committed for the purpose of avoiding or preventing lawful arrest or effecting an escape from custody.
The jury found each aggravating factor and sentenced Flowers to death. Because the trial judge erred when he instructed the jury to consider the third aggravating factor, I would reverse Flowers’s death sentence and remand for resentencing even if this Court affirmed his conviction.
¶ 195. It is elementary that the State bears the burden to prove the defendant’s guilt of a charged offense beyond a reasonable doubt.19 Likewise, our law places upon the State the burden in a capital murder sentencing proceeding to prove the existence of each aggravating factor beyond a reasonable doubt.20 Only when the State satisfies its burden may the jury weigh a particular aggravating factor in favor of a death sentence.21
¶ 196. The majority relies on this Court’s repeated statement that an instruction should be granted on this aggravating factor if the jury may draw an inference from the evidence that avoiding arrest was “a substantial reason for the killing.”22 But that standard — stated in that way with nothing more — cannot be the law. Saying jurors may draw an inference is not enough, unless the inference is so strong that, standing alone, the inference is sufficient for a finding beyond a reasonable doubt. The inference in this case clearly does not fall into that category.
¶ 197. The majority says a reasonable juror could infer that Flowers killed one of the victims to prevent identification and avoid arrest. While. a reasonable juror might very well find that to be a possibility, it would be no more than a guess. And no reasonable juror could make that finding beyond a reasonable doubt. Judges should submit issues to juries through instructions only when a reasonable juror could find the beyond-a-reasonable-doubt burden of proof satisfied on that issue.23
¶ 198. Stated another way, jurors certainly are free to draw reasonable inferences from the facts. But a juror’s ability *1082to draw a reasonable inference is not the same as a reasonable juror’s ability to find that fact beyond a reasonable doubt. For instance, one may draw a reasonable inference that a person who usually goes to lunch from 12:00 to 1:00 p.m. is at lunch on a particular day at 12:30 p.m. But that does not mean one may reach that conclusion beyond a reasonable doubt.
¶ 199. So a trial judge should grant the State’s request to instruct the jury on this aggravating factor only where the State has presented evidence from which a reasonable juror could find beyond a reasonable doubt that “a substantial reason for the killing was to conceal the identity of the killer or killers or to ‘cover their tracks’ so as to avoid apprehension and eventual arrest by authorities.”24 And any case that categorically establishes inferences as sufficient to satisfy the beyond-a-reasonable-doubt standard should be overruled.
¶ 200. The State presented insufficient evidence to meet its burden. The most that may be said in this case is that the jury could speculate that Flowers committed one of the murders to avoid arrest. There was no evidence that he did. In fact, the State argued in the guilt phase that Flowers committed the murders for different purposes: revenge and anger over being fired as well as to effectuate the robbery. From its opening statement, the State’s theory of the case was that Flowers possessed a motive to commit the murders because he had been fired by Tardy’s Furniture. This motive — a revenge killing — is at odds with a conclusion that he killed out of a necessity to avoid arrest. No evidence was presented that Flowers expressed an intent to kill to avoid arrest.
¶ 201. The majority finds that this aggravating-factor instruction is supported by two facts: Flowers knew his victims and Flowers disposed of evidence after the killings. The second can be disposed of easily because it bears no logical connection to the question at hand. Regardless of the reason someone commits murder, that person usually has an incentive to dispose of the evidence. Said differently, if Flowers killed his victims out of pure anger, he would have an incentive to cover up the killing by disposing of evidence. The same could be said if Flowers killed because the victims resisted the robbery. Showing that Flowers disposed of evidence after the killing makes it no more reasonable to conclude that he killed to avoid arrest.
¶ 202. Likewise, the majority’s reliance on the fact that Flowers knew his victims is misplaced. First, by the majority’s own statement, this rationale applies to one victim. The majority states:
The State’s theory was that Flowers intended to kill Bertha Tardy because she fired him and withheld his pay and, in the process of doing so, he shot and killed the others. Flowers knew Carmen Rigby from working at the store, but there is no evidence that he knew Robert Johnson or Derrick Stewart, as the day of the murders was their first day to work at the store.
This statement acknowledges that Flowers killed Tardy — not to avoid arrest — but for revenge. The majority admits that Flowers did not know Johnson or Stewart. So, at best, the majority’s reliance on this evidence relates to Rigby alone. But, even there, it cannot provide sufficient evidence to grant the instruction. It is simply inaccurate to say that a reasonable juror could conclude beyond a reasonable doubt that *1083Flowers killed Rigby to avoid arrest, simply because they had worked together.
¶203. True, this Court has supported an instruction on this aggravating factor in the past based solely on the fact that the victim knew the defendant and could identify him.25 But if we are to enforce the burden of proof required of the State, this was error. Allowing a jury to consider an aggravating factor based on nothing more than pure speculation cannot reach the level of proof beyond a reasonable doubt. Because the State presented insufficient evidence from which a reasonable juror could find this aggravating factor beyond a reasonable doubt, I would reverse Flowers’s sentence of death and remand for a new sentencing hearing.26
KITCHENS AND KING, JJ., JOIN THIS OPINION.
. Miss. Transp. Comm’n v. McLemore, 863 So.2d 31, 34 (Miss.2003) (citing Haggerty v. Foster, 838 So.2d 948, 958 (Miss.2002)).
. Miss. Transp. Comm’n, 863 So.2d at 35.
. Id.
. Id. at 36 (citing Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 592-94, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993)).
. Miss. Transp. Comm'n, 863 So.2d at 36 (citing Daubert, 509 U.S. at 592-94, 113 S.Ct. 2786).
. State v. Copeland, 226 S.W.3d 287, 302-304 (Tenn.2007).
. See Ex Parte Williams, 594 So.2d 1225, 1227 (Ala.1992) ("It is clear from these cases that there is presently a trend in the law to allow expert testimony on the subject of human memory.”); Skamarocius v. State, 731 P.2d 63, 66-67 (Alaska App.1987); State v. Chapple, 135 Ariz. 281, 660 P.2d 1208, 1219 (1983), overruled on different grounds by State v. Benson, 232 Ariz. 452, 307 P.3d 19 (2013); Jones v. State, 314 Ark. 289, 862 S.W.2d 242, 244-45 (1993); People v. Jones, 30 Cal.4th 1084, 135 Cal.Rptr.2d 370, 70 P.3d 359, 388 (2003) ("Exclusion of the expert testimony is justified only if there is other evidence that substantially corroborates the eyewitnesses identification and gives it independent reliability.”); People v. Campbell, 847 P.2d 228, 233 (Colo.App.1992); State v. Guilbert, 306 Conn. 218, 49 A.3d 705, 731-32 (2012); Hager v. U.S., 856 A.2d 1143, 1147 (D.C.2004); McMullen v. State, 714 So.2d 368, 371 (Fla. 1998); Johnson v. State, 272 Ga. 254, 526 S.E.2d 549, 552 (2000); People v. Enis, 139 Ill.2d 264, 151 Ill.Dec. 493, 564 N.E.2d 1155, 1165 (1990); Cook v. State, 734 N.E.2d 563, 570 (Ind.2000); State v. Schutz, 579 N.W.2d 317, 319 (Iowa 1998) ("[Tjhe admissibility of expert testimony relating to the accuracy of eye witness identification rests within the sound discretion of the trial court.”); State v. Carr, 331 P.3d 544, 690 (Kan.2014); Commonwealth v. Christie, 98 S.W.3d 485, 488 (Ky.2002) (noting that the vast majority of jurisdictions have left the admission of eyewitness expert identification testimony to the discretion of trial courts); State v. Rich, 549 A.2d 742, 743 (Me.1988); Bomas v. State, 412 Md. 392, 987 A.2d 98, 114-115 (2010); Commonwealth v. Santoli, 424 Mass. 837, 680 N.E.2d 1116, 1120 (1997); State v. Barlow, 541 N.W.2d 309, 313 (Minn.1995); State v. Du-Bray, 317 Mont. 377, 77 P.3d 247, 255 (2003) ("It shall be an abuse of discretion for a district court to disallow expert testimony on eyewitness testimony when no substantial corroborating evidence exists.”); White v. State, 112 Nev. 1261, 926 P.2d 291, 292 (1996); State v. Henderson, 208 N.J. 208, 27 A.3d 872, 916-17 (2011); People v. LeGrand, 8 N.Y.3d 449, 835 N.Y.S.2d 523, 867 N.E.2d 374, 379 (2007); State v. Fontaine, 382 N.W.2d 374, 378 (N.D.1986); State v. Buell, 22 Ohio St.3d 124, 489 N.E.2d 795, 803 (1986); Bristol v. State, 764 P.2d 887, 890 (Okla.Crim.App.1988); State v. Lawson, 352 Or. 724, 291 P.3d 673, 697 (2012); Commonwealth v. Walker, 92 A.3d 766, 791 (Pa.2014) ("[I]n light of the magnitude of scientific understanding of eyewitness identification and marked developments in case law during the last 30 years, it is no longer advisable to ban the use of expert testimony to aid a jury in understanding eyewitness identification.”); State v. Sabetta, 680 A.2d 927, 933 (R.I.1996); State v. Whaley, 305 S.C. 138, 406 S.E.2d 369, 372 (1991); State v. Hill, 463 N.W.2d 674, 676 (S.D. 1990); Jordan v. State, 928 S.W.2d 550, 555 (Tex.Crim.App.1996); State v. Clopten, 223 P.3d 1103, 1112 (Utah 2009); State v. Percy, 156 Vt. 468, 595 A.2d 248, 253 (1990); State v. Cheatam, 150 Wash.2d 626, 81 P.3d 830, 842 (2003); State v. Utter, No. 13-0479, 2014 WL 1673025, at *3 (W.Va. Apr. 25, 2014); State v. Shomberg, 288 Wis.2d 1, 709 N.W.2d 370, 377 (2006); Engberg v. Meyer, 820 P.2d 70, 79-80 (Wyo.1991).
. Dr. Neuschatz cited studies by: (1) Neisser 1967; (2) Loftus 1979; (3) Loftus and Loftus 1990; (4) Schacter 1995; (5) Ellis, Davis, and Shepard 1977; (6) Maclin, Maclin, and Mal-pass 2001; (7) Shapiro and Penrod 1986; (8) Memon, Hope, and Bull 2003; (9) Culter and Penrod 1988; (10) Culter, Penrod, and Martens 1987 (study a); (11) Culter, Penrod, and *1079Martens 1987 (study b); (12) Culter 1986; (13) O'Rourke 1987; (14) Culter 2006; (15) Tulving 1983; (16) Hunt and Ellis 1974; (17) Read, Tollestrup, Hammersley, McFadden, and Christensen 1990; (18) Ross, Ceci, Dunning, and Toglia 1994; (19) Meissner and Brigham 2001; (20) Gibson 1969; (21) Wells and Bradfield 1998; (22) Neuschatz 2005; (23) Hafstad, Memon, and Logie 2004; (24) Wells, Small, Penrod, Malpass, Fulero, and Brimacombe 1998; (25) Haw and Fisher 2004; (26) Garrioch and Brimacombe 2001; (27) Neuschatz and Gutler (in progress); (28) Douglas, Smith, and Frasier-Hill 2005; (29) Malpass and Devine 1981; (30) Stebaly 1997; (31) Loftus 1993; (32) Johnson 1993; and (33) Pezdek 2007.
. Miss. R. Evid. 705.
. Watts v. Radiator Specialty Co., 990 So.2d 143, 146-48 (Miss.2008).
. Id. at 147, n. 7.
. Id.
. Henley v. State, 136 So.3d 413, 415-16 (Miss.2014).
. Holland v. State, 587 So.2d 848, 874 (Miss.1991) (citing Nixon v. State, 533 So.2d 1078, 1099 (Miss.1987), overruled on other grounds by Wharton v. State, 734 So.2d 985, 991 (Miss.1998)); see also Keller v. State, 138 So.3d 817, 867-68 (Miss.2014).
. Keller, 138 So.3d at 867-68.
. Id. at 868 (quoting Gillett v. State, 56 So.3d 469, 505-06 (Miss.2010) (quoting Leatherwood v. State, 435 So.2d 645, 651 (Miss.1983))); see also Wiley v. State, 750 So.2d 1193, 1206 (Miss.1999); Woodward v. State, 726 So.2d 524, 540 (Miss.1997); Foster v. State, 687 So.2d 1124, 1140 (Miss.1996); Taylor v. State, 672 So.2d 1246, 1275 (Miss.1996); Walker v. State, 671 So.2d 581, 611 (Miss.1995); Carr v. State, 655 So.2d 824, 853-54 (Miss.1995); Chase v. State, 645 So.2d 829, 858 (Miss.1994); Hansen v. State, 592 So.2d 114, 153 (Miss.1991); Lanier v. State, 533 So.2d 473, 490 (Miss.1988).
.Henley, 136 So.3d at 415-16 (quoting Edwards v. State, 469 So.2d 68, 70 (Miss.1985) (citing May v. State, 460 So.2d 778, 781 (Miss.1984))) (a directed verdict should be granted if no reasonable juror could find the defendant guilty beyond a reasonable doubt); Harper v. State, 478 So.2d 1017, 1021 (Miss.1985) (a lesser-offense instruction should be granted unless no reasonable juror could find that the defendant committed the lesser offense).
. Ketler, 138 So.3d at 867-68 (quoting Gillett, 56 So.3d at 505-06 (quoting Leatherwood, 435 So.2d at 651)).
. Wiley, 750 So.2d at 1206.
. Gillett v. State, 148 So.3d 260, 268-69 (Miss.2014) (the Court declines to exercise its authority under Mississippi Code Section 99-19 — 105(3)(d) to re-weigh aggravating and mitigating factors).