dissenting:
I respectfully dissent because the majority opinion affirms a Texas Court of Crimi*351nal Appeals (TCCA) death penalty judgment that is contrary to the federal law clearly established by Atkins v. Virginia, 536 U.S. 304, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002): When Atkins banned the execution of mentally retarded offenders, it defined mental retardation as generally conforming to the three-part clinical definitions set forth by the American Association on Mental Retardation (AAMR) and the American Psychiatric Association (APA), which were quoted by the Court in Atkins. In the present case, the TCCA, instead of applying the “adaptive skill areas” part of that definition, id. at 308 n. 3, 122 S.Ct. 2242, applied its own substantively contrary state law, known as the “Briseno factors,” in erroneously deciding that Elroy Chester failed to prove that he is mentally retarded. For this reason, I would vacate the federal district court’s judgment affirming the TCCA’s judgment and remand the case to the federal district court for further proceedings applying the entire correct clinical definition of mental retardation as required by Atkins.
In Atkins, the Supreme Court held that the Eighth Amendment’s prohibition of “cruel and unusual punishments” bars the execution of mentally retarded offenders. The Court reasoned that: (1) there is a national consensus among state legislatures and Congress that the execution of mentally retarded offenders is excessive punishment, id. at 314-16, 122 S.Ct. 2242; (2) the statutory definitions of “mental retardation” used by states in that national consensus are not identical, but generally conform to, the clinical definitions of “mental retardation” set forth by the AAMR and APA, id. at 317 n. 22, 122 S.Ct. 2242; and (3) the Supreme Court’s independent evaluation of the issue revealed no reason for the Court to disagree with the national legislative consensus, id. at 321, 122 S.Ct. 2242. Based on this rationale, the Atkins Court concluded that the Eighth Amendment “ ‘places a substantive restriction on the State’s power to take the life’ of a mentally retarded offender,” id. (quoting Ford v. Wainwright, 477 U.S. 399, 405, 106 S.Ct. 2595, 91 L.Ed.2d 335 (1986)), and that this constitutional restriction protects individuals who “fall within the range of mentally retarded offenders about whom there is a national consensus,” id. at 317, 122 S.Ct. 2242. Further, Atkins made plain that the AAMR and APA “clinical definitions of mental retardation require not only subaverage intellectual functioning, but also significant limitations in adaptive skills such as communication, self-care, and self-direction that became manifest before age 18.” Id. at 318, 122 S.Ct. 2242.1 Thus, in deciding whether a person is mentally retarded and therefore exempt from execution under the Eighth Amendment, a state court must apply each of the three prongs — subaverage intellectual functioning; adaptive functioning limita*352tions; and onset prior to age eighteen — of a definition that generally conforms to the AAMR and APA clinical definitions of “mental retardation.”
In Atkins, the Court also concluded that “[n]ot all people who claim to be mentally retarded will be so impaired as to fall within the range of mentally retarded offenders about whom there is a national consensus. As was our approach in Ford v. Wainwright with regard to insanity, “we leave to the State[s] the task of developing appropriate ways to enforce the constitutional restriction upon [their] execution of sentences.’” Id. at 317, 122 S.Ct. 2242 (alterations in original) (citation omitted) (quoting Ford, 477 U.S. at 416-17, 106 S.Ct. 2595). In so doing, the Ford Court made clear that when “the Eighth Amendment bars execution of a category of defendants defined by their mental state[,] [t]he bounds of that category are necessarily governed by federal constitutional law.” Ford, 477 U.S. at 419, 106 S.Ct. 2595 (Powell, J., concurring in part and concurring in the judgment, but speaking for the majority on this point).2 Thus, by closely following Ford, Atkins signals clearly that federal constitutional law governs the bounds of the category of mentally retarded individuals who are exempt from execution, although the states, within the confines of due process, may devise procedures to govern mental retardation determinations.
Our en banc court and panels have adopted this understanding of Atkins. See Moore v. Quarterman, 533 F.3d 338, 341 (5th Cir.2008) (en banc) (“Atkins specifically reserved to the states the adoption of procedures to implement its new constitutional rule....” (emphasis added)); Wiley v. Epps, 625 F.3d 199, 207 (5th Cir.2010) (“[E]ven though Atkins left to the states the job of implementing procedures for determining who is mentally retarded, ‘it was decided against the backdrop of the Supreme Court’s and lower court’s due process jurisprudence.’ ” (emphasis added) (quoting Rivera v. Quarterman, 505 F.3d 349, 358 (5th Cir.2007))). Accordingly, the states retain substantial discretion to create appropriate procedures, but they may not substantively redefine mental retardation so as to permit the execution of those who “fall within the range of mentally retarded offenders about whom there is a national consensus.” Atkins, 536 U.S. at 317, 122 S.Ct. 2242.
In Atkins, the Supreme Court quoted and referred to the AAMR definition of mental retardation as follows: “Mental retardation refers to substantial limitations in present functioning. It is characterized by [ (1) ] significantly subaverage intellectual functioning, [ (2) ] existing concurrently with related limitations in two or more of the following applicable adaptive skill areas: communication, self-care, home living, social skills, community use, self-direction, health and safety, functional academics, leisure, and work[; and (3) ] Mental retardation manifests before age 18.” 536 U.S. at 309, 122 S.Ct. 2242.3
The Texas habeas trial court, in considering Chester’s habeas petition, did not apply the second element of the AAMR definition to determine if Chester had “re*353lated limitations in two or more of the following applicable adaptive skill areas: communication, self-care, home living, social skills, community use, self-direction, health and safety, functional academics, leisure, and work.” Instead, the trial court applied the “Briseno factors,” a series of questions originally suggested by the TCCA in Ex parte Briseno, 135 S.W.3d 1 (Tex.Crim.App.2004), which the TCCA itself has described as being “non-diagnostic criteria.” Ex parte Van Alstyne, 239 S.W.3d 815, 820 (Tex.Crim.App.2007). Although the factors were initially conceived of as “evidentiary factors which factfinders ... might also focus upon in weighing evidence as indicative of mental retardation or of a personality disorder,” Briseno, 135 S.W.3d at 8, the TCCA in the present case used the Briseno factors as a substantive part of its mental retardation definition, instead of the second prong of the AAMR definition (the AAMR clinical adaptive skills criteria) to determine that Chester had not proved he had adaptive skills deficits, and, therefore, that Chester failed to prove he is mentally retarded. See Ex parte Chester, No. AP-75037, 2007 WL 602607 (Tex.Crim.App. Feb. 28, 2007) (unpublished).
Chester introduced evidence that he scored under 70 in 4 of the 5 full-scale IQ tests that he has taken since the age of seven years, and scored 57 on the Vineland Adaptive Behavior Survey (VABS), a clinical adaptive functioning test. His evidence showed he has been unable to live or work independently, and his experts and lay witnesses related that he suffered from mild mental retardation and was deficient in several adaptive behavioral areas. The state habeas trial court, which gave no reason for applying the Briseno factors to the exclusion of the AAMR adaptive skills criteria and the VABS test results, found that Chester failed to disprove any of the Briseno factors, was therefore not limited in adaptive functioning, and thus not mentally retarded for this reason alone. See Chester, 2007 WL 602607, at *4. The TCCA affirmed, approving of the trial court’s use of the Briseno factors to the exclusion of clinical adaptive skills criteria to define mental retardation. Id. at *9. The U.S. District Court denied federal habeas relief, on the ground that Chester’s application failed under the seventh Briseno factor alone. Chester v. Quarterman, No. 5:05cv29, 2008 WL 1924245, at *7 (E.D.Tex. Apr. 29, 2008) (unpublished).
The TCCA’s decision was contrary to the federal law that was clearly established by the Supreme Court in Atkins. Under Atkins, state courts must apply a mental retardation definition that generally conforms to all three parts of the clinical AAMR or APA definitions. The TCCA’s unique nondiagnostic Briseno factors are more constricted than, unrelated to, and substantively contrary to the adaptive deficits criteria identified in the second prongs of the AAMR and APA clinical definitions of mental retardation. Exclusively applying the Briseno factors to assess the substantive adaptive skills prong of the mental retardation definition inevitably leads to anomalous and unreliable results, including the execution of offenders who should be classified as mentally retarded and shielded from execution under Atkins and the comprehensive clinical definitions quoted therein. In other words, by affirming the Texas courts’ erroneous use of the Briseno factors in place of the adaptive skills prong of the substantive three-part rule defining mental retardation, the majority allows those state courts to circumvent the constitutional rule of Atkins and to use their more constricted definition of mental retardation to exclude substantial numbers of mentally retarded offenders from protection from execution under the Eighth *354Amendment.4 Because the decisions of the TCCA and the federal district court are based on the same error of clear constitutional law, the judgment of the federal district court should be set aside and the case should be remanded to it for further proceedings generally conforming to the clinical definitions of mental retardation as required by Atkins.
I.
The petitioner, Elroy Chester, robbed and raped two young women at gunpoint in their home. When the women’s uncle unexpectedly arrived at the house, Chester shot and killed him. He pleaded guilty to the crime of capital murder and was sentenced to death by a Texas jury in 1998. His conviction and sentence were affirmed on direct appeal. Chester then sought post-conviction relief at the state and federal levels on the grounds that he could not be executed because he is mentally retarded.
Chester’s federal habeas petition was pending in 2002 when the Supreme Court held in Atkins that the Eighth Amendment forbids the execution of mentally retarded offenders. Consequently, Chester’s petition was dismissed without prejudice to allow him to renew his claim for state post-conviction relief on the basis of Atkins. Chester v. Cockrell, 62 Fed.Appx. 556 (5th Cir.2003).
The TCCA granted Chester leave to file a successive state habeas petition and remanded his case to the state trial court for further proceedings. The trial court held an evidentiary hearing on the Atkins claim. In order to prove that he was mentally retarded, under Atkins, 536 U.S. 304, 122 S.Ct. 2242, and Briseno, 135 S.W.3d l,5 Chester was required to show *355by a preponderance of the evidence that he had (1) significantly subaverage general intellectual functioning and (2) related limitations in adaptive functioning, (3) the onset of which had occurred before Chester reached age 18. Briseno, 135 S.W.3d at 7. Chester introduced substantial evidence of his mental retardation, including full-scale IQ scores of 69 at age seven, 59 at age twelve, 77 at age thirteen, 69 at age eighteen, and 66 at age twenty-nine and a score of 57 on VABS, a standardized test of adaptive functioning. The Texas Department of Criminal Justice administered one of these IQ tests — on which Chester scored a 69 — and the VABS test in 1987, when Chester was eighteen years old. The State’s own expert testified that a person with those test scores would be correctly diagnosed as mentally retarded. In addition, an expert retained by Chester testified that he was indeed mentally retarded. Chester also supplied other evidence tending to show that he suffers from limitations in adaptive functioning. He presented evidence that he was placed in special education early in schooling and admitted into the prison Mentally Retarded Offenders Program (MROP) at approximately age eighteen; achievement testing in prison placed him at third grade levels or below. Two of his sisters testified regarding his adaptive behavior deficits, including his inability to five or work independently. A special education teacher testified regarding his limited abilities at school. One expert diagnosed Chester as mentally retarded based on a review of his records, interviews with Chester, and observation of the State expert’s interview with him. Another expert classified him as mildly mentally retarded based on a review of his MROP records.
Chester asserted that he demonstrated deficits regarding the broader conceptual and practical adaptive skill areas, as well as the specific skill areas of communication, work, functional academics, self-direction, and community use. The trial court, adopting the entirety of the prosecution’s proposed findings of fact and conclusions of law, concluded that Chester had failed to carry his burden of proof as to either of the first two elements of mental retardation: significantly subaverage general intellectual functioning and related limitations in adaptive functioning. The trial court found that Chester failed to prove that he had adaptive skill deficits but it did not apply the clinical AAMR criteria to make this finding. Instead, it applied the Briseno “evidentiary factors” and concluded that, under the Briseno factors, Chester was not limited in adaptive functioning and therefore failed to satisfy the adaptive functioning prong of the tripartite clinical definitions of mental retardation. He therefore was found to be not mentally retarded. If the trial court had applied the clinical adaptive functioning criteria instead, Chester’s evidence arguably would have shown he satisfied this prong as well as the other two and that he is mentally retarded.
On appeal, the TCCA partly overruled the trial court’s findings and conclusions. Chester, 2007 WL 602607. The TCCA held that Chester “has met his burden in regard to demonstrating significant limitations in intellectual functioning,” id. at *3, and that “there is no dispute as to the third part of the test, that the evidence in *356favor of a finding of mental retardation occurred and was recorded before the applicant reached the age of eighteen,” id. at *2. Thus, the TCCA held that Chester had satisfied the first and third prongs of the definition of mental retardation. However, the TCCA denied habeas relief because it concluded that Chester had not “[s]hown [significant [deficits in [a]daptive [b]ehavior” as required by the second prong, id. at *4, and he had therefore not carried the burden of showing that he was mentally retarded. In reaching this conclusion, the TCCA did not consider the AAMR’s or APA’s clinical definitions of mental retardation, but relied exclusively on the factors that had previously been presented in Briseno as merely “some other evidentiary factors which factfinders in the criminal trial context might also focus upon in weighing evidence as indicative of mental retardation or of a personality disorder,” Briseno, 135 5.W.3d at 8. See Chester, 2007 WL 602607, at *4-5.
Chester then filed another federal habeas petition. In that proceeding, the state did not dispute that the evidence of his mental deficits developed before he was eighteen years old or “that Chester suffers from significantly sub-average intellectual functioning.” Chester, 2008 WL 1924245, at *2. But the district court departed even further than the state courts from the Atkins clinical definitions of mental retardation, expressly holding that a finding as to just one of the seven Briseno factors was a sufficient basis for denying an Atkins claim: “[A]n affirmative finding as to the seventh and final Briseno evidentiary factor is sufficient by itself to uphold a denial of relief in a habeas corpus proceeding.” Id. at *7.6
Thus, the district court held that the TCCA’s “rejection of [Chester’s] mental retardation claim is neither inconsistent with, nor the result of an unreasonable application of, clearly established federal law,” and denied habeas relief. Id7
Chester now appeals from the district court’s order denying habeas relief.
II.
“In a habeas corpus appeal, we review the district court’s findings of fact for clear error and review its conclusions of law de novo, applying the same standard of review to the state court’s decision as the district court.” Beazley v. Johnson, 242 F.3d 248, 255 (5th Cir.2001) (quoting Thompson v. Cain, 161 F.3d 802, 805 (5th Cir.1998)) (internal quotation marks omitted).8 I conclude that Chester has shown *357that his habeas petition should be granted under the “contrary to” clause of AEDPA, 28 U.S.C. § 2254(d)(1). Therefore, I need not consider whether he also satisfies the “unreasonable application” clause of AED-PA, 28 U.S.C. § 2254(d)(1), and need not review the district court’s findings of fact for clear error.
“Under the Antiterrorism and Effective Death Penalty Act (AEDPA), a federal court may not grant § 2254 habeas relief on any ground previously disposed of on the merits by a state court unless the state decision ‘was contrary to, or involved an unreasonable application of, clearly established [fjederal law, as determined by the Supreme Court of the United States,’ or ‘was based on an unreasonable determination of the facts in light of the evidence presented in the [s]tate court proceeding.’ ” Geiger v. Cain, 540 F.3d 303, 306-07 (5th Cir.2008) (alterations in original) (quoting 28 U.S.C. § 2254(d)(l)-(2)); see also Woods v. Quarterman, 493 F.3d 580, 584 (5th Cir.2007). “A state court decision is contrary to clearly established federal law if it ‘applies a rule that contradicts the governing law set forth in [Supreme Court] cases,’ or if the state court confronts a set of facts that are materially indistinguishable from a decision of [the Supreme Court] and nevertheless arrives at a result different from [the Court’s] precedent.’ A state court decision involves an unreasonable application of clearly established federal law if the state court ‘correctly identifies the governing legal rule but applies it unreasonably to the facts of a particular prisoner’s case....’” Williams v. Quarterman, 551 F.3d 352, 358 (5th Cir.2008) (alterations in original) (citations omitted) (quoting Williams v. Taylor, 529 U.S. 362, 405-08, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000)).
III.
In Atkins, the Supreme Court early in its opinion set forth in full the generally accepted clinical definitions of mental retardation:
The American Association on Mental Retardation (AAMR) defines mental retardation as follows: “Mental retardation refers to substantial limitations in present functioning. It is characterized by significantly subaverage intellectual functioning, existing concurrently with related limitations in two or more of the following applicable adaptive skill areas: communication, self-care, home living, social skills, community use, self-direction, health and safety, functional academics, leisure, and work. Mental retardation manifests before age 18.” Mental Retardation: Definition, Classification, and Systems of Supports 5 (9th ed.1992).
The American Psychiatric Association’s [ (APA) ] definition is similar: “The essential feature of Mental Retardation is significantly subaverage general intellectual functioning (Criterion A) that is accompanied by significant limitations in adaptive functioning in at least two of the following skill areas: *358communication, self-care, home living, social/interpersonal skills, use of community resources, self-direction, functional academic skills, work, leisure, health, and safety (Criterion B). The onset must occur before age 18 years (Criterion C). Mental Retardation has many different etiologies and may be seen as a final common pathway of various pathological processes that affect the functioning of the central nervous system.” Diagnostic and Statistical Manual of Mental Disorders 41 (4th ed.2000). “Mild” mental retardation is typically used to describe people with an IQ level of 50-55 to approximately 70. Id., at 42-43.
Atkins, 536 U.S. at 308 n. 3, 122 S.Ct. 2242. The Court then surveyed developments in state and federal law that showed a growing nationwide consensus that mentally retarded defendants should not be executed. See id. at 313-16, 122 S.Ct. 2242. Beginning with a Georgia law enacted in 1988, the Court counted eighteen states, along with the federal government, which had enacted legislation prohibiting the execution of mentally retarded defendants. See id. at 313-15, 122 S.Ct. 2242. Furthermore, the Court observed that “even in those States that allow the execution of mentally retarded offenders, the practice is uncommon[:] ... only five [states] have executed offenders possessing a known IQ less than 70” since 1989. Id. at 316, 122 S.Ct. 2242.
The Court therefore concluded that “a national consensus has developed against” the execution of mentally retarded defendants. Id. “To the extent there is serious disagreement about the execution of mentally retarded offenders, it is in determining which offenders are in fact retarded.... Not all people who claim to be mentally retarded will be so impaired as to fall within the range of mentally retarded offenders about whom there is a national consensus.” Id. at 317, 122 S.Ct. 2242. The Court noted that “[t]he [states’] statutory definitions of mental retardation are not identical, but generally conform to the clinical definitions” promulgated by the AAMR and the APA. Id. at 317 n. 22, 122 S.Ct. 2242. Thus, the Court’s holding that “a national consensus has developed,” id. at 316, 122 S.Ct. 2242, was based on statutes which employed “definitions of mental retardation” that “generally conform to the clinical definitions” quoted above, id. at 317 n. 22, 122 S.Ct. 2242.
As an additional step in its reasoning, the Court undertook an “independent evaluation of the issue.” Id. at 321, 122 S.Ct. 2242. In this step, the Court reasoned that “by definition,” people with mental retardation have certain characteristics that make their execution cruel and unusual. Id. at 318, 122 S.Ct. 2242. These characteristics include “diminished capacities to understand and process information, to communicate, to abstract from mistakes and learn from experience, to engage in logical reasoning, to control impulses, and to understand the reactions of others.” Id. Because of these characteristics, “executing the mentally retarded will not measurably further the goal of deterrence,” id. at 320, 122 S.Ct. 2242, and “the lesser culpability of the mentally retarded offender ... does not merit that form of retribution.” Id. at 319, 122 S.Ct. 2242. Moreover, mentally retarded defendants “face a special risk of wrongful execution” because of “the possibility of false confessions, [and] the lesser ability of mentally retarded defendants to make a persuasive showing of mitigation.” Id. at 320-21, 122 S.Ct. 2242. Therefore, the Court concluded, there was “no reason to disagree with the judgment of ‘the legislatures that have recently addressed the matter’ and concluded that death is not a suitable punishment for a mentally retarded criminal.” *359Id. (quoting Enmund v. Florida, 458 U.S. 782, 801, 102 S.Ct. 3368, 73 L.Ed.2d 1140 (1982)).
In short, the holding of Atkins is that the Eighth Amendment prohibits the execution of individuals who are “so impaired as to fall within the range of mentally retarded offenders about whom there is a national consensus.” Id. at 317, 122 S.Ct. 2242. And the Court described the parameters of that “national consensus” as “generally conforming] to the clinical definitions” of mental retardation used by the AAMR and APA. Id. at 317 & n. 22, 122 S.Ct. 2242.
The Court refrained from setting forth procedures for determining whether a particular defendant is “so impaired as to fall within the range of mentally retarded offenders about whom there is a national consensus.” Id. at 317, 122 S.Ct. 2242. Rather, it declared, “we leave to the States the task of developing appropriate ways to enforce the constitutional restriction upon their execution of sentences.” Id. (quoting Ford, 477 U.S. at 405, 106 S.Ct. 2595) (brackets and internal quotation marks omitted).
The Atkins Court noted that this was the same “approach” that the Court had previously followed “in Ford v. Wainwright with regard to insanity.” 536 U.S. at 317, 122 S.Ct. 2242. We have previously observed that “Ford is instructive” in Atkins cases “because of the similarity of the competency and mental retardation issues: both decisions affirmatively limit the class of persons who are death penalty eligible.” Rivera, 505 F.3d at 358. The Court’s opinions in Atkins and Ford both expressly announced “that the Constitution ‘places a substantive restriction on the State’s power to take the life’ ” of certain offenders. Atkins, 536 U.S. at 321, 122 S.Ct. 2242 (quoting Ford, 477 U.S. at 405, 106 S.Ct. 2595); see also Panetti, 551 U.S. at 957, 127 S.Ct. 2842 (same).
In Ford, Justice Powell’s controlling concurring opinion9 explained that when “the Eighth Amendment bars execution of a category of defendants defined by their mental state[,] [t]he bounds of that category are necessarily governed by federal constitutional law.” 477 U.S. at 419, 106 S.Ct. 2595. Further, the Panetti Court explained that Ford “broadly identified]” a “substantive standard for incompetency”; applying this substantive standard, the Panetti Court “rejected] the standard followed by the Court of Appeals” as being inconsistent with Ford. 551 U.S. at 960, 127 S.Ct. 2842. Thus, it is clear under both Ford and Panetti that the definition of incompetency to be executed is a matter of federal substantive constitutional law. Because Atkins expressly adopted Ford’s approach by announcing “a substantive restriction,” 536 U.S. at 321, 122 S.Ct. 2242, while giving states procedural room to “develop[ ] appropriate ways to enforce” that restriction, id. at 317, 122 S.Ct. 2242, it follows that under Atkins, too, the substantive definition of mental retardation for Eighth Amendment purposes is a matter of federal constitutional law.
Atkins identified a substantive standard for mental retardation by announcing that states may not execute offenders who are “so impaired as to fall within the range of mentally retarded offenders about whom there is a national consensus.” Atkins, 536 U.S. at 317, 122 S.Ct. 2242. The Court explained that the “national consensus” was based on state-law definitions of mental retardation that “generally conform to the clinical definitions” in the AAMR-9 and DSM-IV-TR, which the Court quoted in full in its opinion. Id. at 308 n. 3, 317 & n. 22, 122 S.Ct. 2242. Therefore, in order *360to “appropriately] ... enforce the constitutional restriction” imposed by Atkins, id. at 317, 122 S.Ct. 2242 (quoting Ford, 477 U.S. at 416, 106 S.Ct. 2595), states must apply definitions of mental retardation that “generally conform to the clinical definitions,” id. at 317 n. 22, 122 S.Ct. 2242.10
IV.
In 2004, the Texas Court of Criminal Appeals issued an opinion declaring how the Texas courts would “enforce the constitutional restriction” under Atkins. Briseno, 135 S.W.3d at 5 (quoting Atkins, 536 U.S. at 317, 122 S.Ct. 2242). The court observed that it had “previously employed the definitions of ‘mental retardation’ set out by the AAMR, and that contained in section 591.003(13) of the Texas Health and Safety Code. Under the AAMR definition, mental retardation is a disability characterized by: (1) ‘significantly subaverage’ general intellectual functioning; (2) accompanied by ‘related’ limitations in adaptive functioning; (3) the onset of which occurs prior to the age of 18. As noted above, the definition under the Texas Health and Safety Code is similar.... ” Id. at 7 (footnotes omitted) (citing, inter alia, AAMR-9, supra note 1, at 5).11 Accordingly, the court concluded that “[u]ntil the Texas Legislature provides an alternate statutory definition of ‘mental retardation’ for use in capital sentencing, we will follow the AAMR or section 591.003(13) criteria in addressing Atkins mental retardation claims.” Id. at 8.
However, the Briseno court expressed some concern with the “adaptive behavior criteria” in the AAMR definition,12 which it considered to be “exceedingly subjective.” Id. at 8. For this reason, the court identified “some other evidentiary factors which factfinders in the criminal trial context might also focus upon in weighing evidence as indicative of mental retardation or of a personality disorder.” Id.13 Those additional factors are as follows:
• Did those who knew the person best during the developmental stage — his family, friends, teachers, employers, *361authorities — think he was mentally retarded at that time, and, if so, act in accordance with that determination?
• Has the person formulated plans and carried them through or is his conduct impulsive?
• Does his conduct show leadership or does it show that he is led around by others?
• Is his conduct in response to external stimuli rational and appropriate, regardless of whether it is socially acceptable?
• Does he respond coherently, rationally, and on point to oral or written questions or do his responses wander from subject to subject?
• Can the person hide facts or lie effectively in his own or others’ interests?
• Putting aside any heinousness or gruesomeness surrounding the capital offense, did the commission of that offense require forethought, planning, and complex execution of purpose?
Id. at 8-9. Although the TCCA identifies the Briseno factors as “adaptive functioning” criteria, the factors bear no resemblance to the AAMR or APA adaptive functioning criteria. The Briseno court cited no sources or authorities for the proposition that these “other evidentiary factors” are an accurate or useful way to tell whether a person is genuinely mentally retarded; thus, the factors appear to be wholly the TCCA’s own creation. These factors are “non-diagnostic,” Van Alstyne, 239 S.W.3d at 820, and based on the judges’ impressions and assumptions about mental retardation. See Blume, supra note 13, at 712. The court gave no explanation of how the factors are any less “subjective” than the criteria for adaptive functioning in the AAMR’s definition of mental retardation, and they are not based on or drawn from the adaptive functioning skill areas identified by the AAMR or the APA. See Atkins, 536 U.S. at 308 n. 3, 122 S.Ct. 2242; see also supra note l.14
After setting forth these evidentiary factors, the TCCA went on to determine the *362procedural means by which Atkins claims in state habeas petitions would be resolved. The court decided that “Atkins does not require a jury determination of mental retardation in a post-conviction proceeding,” id. at 9, and that “[t]he defendant bears the burden of proof, by a preponderance of the evidence, to establish that he is mentally retarded,” id. at 12 (bold face removed). The court then concluded that the petitioner in that case, Jose Garcia Briseño, had failed to prove by a preponderance of the evidence that he was mentally retarded. See id. at 14-18.
In summary, as relevant here, the Briseno court first determined that Texas courts “w[ould] follow the AAMR or section 591.003(13) criteria in addressing Atkins mental retardation claims,” and then mentioned seven “other evidentiary factors which factfinders ... might also focus upon” in determining whether a defendant has the limitations in adaptive behavior that are required for a finding of mental retardation. Id. at 8.
V.
In the present case, the TCCA held that Chester fulfilled two of the three requirements of mental retardation: significantly subaverage intellectual functioning and onset before age 18.15 It is undisputed in this appeal that Chester meets those requirements. Whether Chester is mentally retarded therefore depends on whether he meets the one remaining requirement: significant limitations in adaptive functioning. In the state habeas proceedings, Chester presented a substantial amount of evidence that tended to prove that he does have significant limitations in adaptive functioning, under the standard clinical definitions of mental retardation to which the national consensus generally conforms, see Atkins, 536 U.S. at 317 & n. 22, 122 S.Ct. 2242. However, the TCCA concluded that Chester is not mentally retarded — a conclusion which was based entirely and exclusively on the TCCA’s application of the Briseno evidentiary factors. Chester, 2007 WL 602607, at *4-5, *9.
Chester provided the state courts with factual evidence and expert testimony indicating that he has significant limitations in adaptive functioning as defined by the AAMR and APA criteria.16 Notably, he received a score of 57 on the VABS test, *363which is a clinical method of assessing a person’s level of adaptive functioning for the purpose of diagnosing mental retardation.17 The TCCA observed that “even the State’s expert witness acknowledged that a person with a Vineland score of 57, combined with an IQ score of 69 as measured at the same time, would be correctly diagnosed as mildly mentally retarded.” Chester, 2007 WL 602607, at *3. These two scores were obtained in 1987 by the Texas Department of Criminal Justice, when Chester was eighteen years old. He was admitted to the department’s Mentally Retarded Offender Program at that time. Id. That was eleven years before Chester committed the murder for which he was sentenced to death, and fifteen years before the Supreme Court decided Atkins; thus, Chester could not have been faking mental retardation to avoid the death penalty. The TCCA’s opinion gave no specific reason for disbelieving or disregarding Chester’s VABS score of 57. Neither has the State of Texas given us any reason why the score would not be reliable.
Chester also presented substantial evidence of limitations in at least four out of the AAMR’s ten “adaptive skill areas”: communication, home living, functional academics, and work. Atkins, 536 U.S. at 308 n. 3, 122 S.Ct. 2242 (quoting AAMR-9, supra note 1, at 5).18 For instance, there was evidence that while growing up, Chester had severe communication difficulties: he spoke in jumbled sentences with misused words and a small vocabulary. His school diagnosed him with a “serious communicative handicap” and provided therapy. Other children called him “stupid” and “retarded.” He attended special education classes from third through twelfth grades. All his classes were taught to him at his own academic level, which was never higher than third grade. As an adult, when out of jail, Chester never lived independently; never had a driver’s license or a bank account; held only menial jobs; could not read or write well enough to fill out a job application; and could not shop or cook by himself. At eighteen years of age he could read “only small words” and could not name the months of the year, according to the Texas Department of Criminal Justice’s assessment which resulted in his being admitted to the MROP.19
*364Despite Chester’s VABS test score of 57 — -which the state’s own expert witness testified was low enough to require a diagnosis of mental retardation — and despite his substantial evidence of adaptive functioning limitations in at least four of the AAMR skill areas, the state courts relied exclusively on the Briseno factors to determine that Chester did not meet this requirement of mental retardation. Thus, the state courts did not use the Briseno factors in the limited way in which the TCCA’s Briseno opinion announced that they would be used — as “other evidentiary factors” that a court “might also focus upon” in addition to the AAMR’s clinical definition. 135 S.W.3d at 8 (emphases added).
The TCCA in Briseno stated that it would apply the AAMR’s 1992 definition of mental retardation, which includes adaptive functioning criteria, in Atkins cases. Id. The adaptive functioning prong of the AAMR’s 1992 definition of mental retardation, as quoted in Atkins, is satisfied by a finding of “limitations in two or more of the following applicable
adaptive skill areas: communication, self-care, home living, social skills, community use, self-direction, health and safety, functional academics, leisure, and work.” Atkins, 536 U.S. at 308 n. 3, 122 S.Ct. 2242 (quoting AAMR-9, supra note 1, at 5) (quotation marks omitted). Similarly, the APA has stated that mentally retarded individuals will have “significant limitations in adaptive functioning in at least two of the following skill areas: communication, self-care, home living, social/interpersonal skills, use of community resources, self-direction, functional academic skills, work, leisure, health, and safety.” Id. (quoting DSM-IV-TR, supra note 1, at 41 (quotation marks omitted)). Yet the TCCA in this case made no attempt whatsoever to determine whether Chester has significant limitations in two or more of those areas of adaptive functioning.20 The state trial court likewise made no such attempt.
Instead, the TCCA and the state trial court both treated their findings under the Briseno factors as if they were a sufficient basis, standing alone, for determining whether a person has significant limita*365tions in adaptive functioning. The courts did not employ — nor did they even mention — any of the clinical definitions or criteria for adaptive functioning limitations. The trial court stated: “This Court’s review of evidentiary factors relating to adaptive behavior suggested relevant by the Court of Criminal Appeals in Ex parte Briseno weigh conclusively against Applicant’s argument that he is mentally retarded.” State v. Chester, No. 76044-B, slip op. at 25 (Crim. Dist. Ct. Jefferson Cnty., Tex. July 26, 2004). The TCCA stated: “The trial court’s findings addressed all seven evidentiary factors listed in Briseno, and noted carefully how the applicant had failed to persuade the trial court on each one.” Chester, 2007 WL 602607, at *4. The TCCA, in effect, dispensed with any measurement of “adaptive functioning” as it is conceived of by the clinical mental retardation definitions and the adaptive functioning criteria contained therein.
The federal district court, in denying Chester habeas relief, likewise made no attempt to determine whether Chester has significant limitations in adaptive functioning according to a rule generally conforming to the clinical definition of mental retardation. The federal district court relied entirely on the seventh Briseno factor, and nothing else, as the adaptive functioning test: “an affirmative finding as to the seventh and final Briseno evidentiary factor is sufficient by itself to uphold a denial of relief in a habeas corpus proceeding.” Chester, 2008 WL 1924245, at *7. Thus, not only did the federal district court disregard the AAMR and APA clinical criteria that were used by the Atkins Court to define mental retardation, but it disregarded six of the seven Briseno factors as well.
VI.
A.
The TCCA’s decision in the present case is contrary to the clear holding of the Court in Atkins. “A state court decision is contrary to clearly established federal law if it ‘applies a rule that contradicts the governing law set forth in [Supreme Court] cases....’” Williams v. Quarter-man, 551 F.3d at 358 (first alteration in original) (quoting Williams v. Taylor, 529 U.S. at 405, 120 S.Ct. 1495). The TCCA applied a rule of decision using the Briseno factors as its exclusive basis for holding that Chester does not have significant limitations in adaptive functioning and that he is therefore not mentally retarded for Eighth Amendment purposes. By so holding, the TCCA adopted and applied a rule of decision that the Briseno “other evidentiary factors,” standing alone, can be used to determine that an offender does not have significant limitations in adaptive functioning and is therefore not mentally retarded, even when the offender has presented strong evidence that he satisfies all the AAMR or APA clinical criteria for mental retardation.
The TCCA’s rule of decision is contrary to Atkins because, as explained above, Atkins held that the Eighth Amendment imposes a “substantive restriction,” 536 U.S. at 321, 122 S.Ct. 2242, which prevents states from executing offenders who are “so impaired as to fall within the range of mentally retarded offenders about whom there is a national consensus,” id. at 317, 122 S.Ct. 2242. In order to “appropriately] ... enforce th[is] constitutional restriction,” states must apply definitions of mental retardation for Eighth Amendment purposes that reflect the “national consensus” by “generally conforming] to the [AAMR and APA] clinical definitions” of mental retardation. Id. at 317 & n. 22, 122 S.Ct. 2242. The rule applied by the TCCA in this case involved a radical departure from general conformity with those clinical definitions of mental retardation. It rede*366fined the adaptive functioning element in such a way that it clearly contradicts and fails to carry out Atkins’ s mandate to protect from execution all offenders who fall within the national consensus’s understanding of mental retardation that generally conforms to the AAMR and APA definitions.
The Briseno evidentiary factors, standing alone, cannot be used to accurately determine that a person does not have significant limitations in adaptive functioning according to the national consensus’s understanding of mental retardation that the Atkins Court identified. This is because the Briseno evidentiary factors are substantively very different from, and ask different questions than, the AAMR and APA criteria. For instance, the Briseno factors do not consider, and therefore cannot determine, whether a person has significant limitations in the adaptive skill areas of self-care, home living, community use, health and safety, functional academics, leisure, or work. The AAMR-9 and APA criteria or precepts generally conforming to them require only that a person must have significant limitations in two out of ten or eleven adaptive skill areas (as well as showing significantly sub-average intellectual functioning and onset before age 18). The AAMR and APA have noted that mentally retarded individuals have strength areas and that the most appropriate way to identify adaptive functioning deficits is to focus on the individual’s limitations. See AAMR-10, supra note 1, at 8 (“[P]eople with mental retardation are complex human beings who likely have certain gifts as well as limitations .... These may include strengths in social or physical capabilities, strengths in some adaptive skill areas, or strengths in one aspect of an adaptive skill in which they otherwise show an overall limitation.”); Blume et al., supra note 13, at 706, 710 (quoting AAMR-10, supra note 1, at 8); see also Holladay v. Allen, 555 F.3d 1346, 1363 (11th Cir.2009) (“Individuals with mental retardation have strengths and weaknesses, like all individuals. Indeed, the criteria for diagnosis recognizes this by requiring a showing of deficits in only two of ten identified areas of adaptive functioning.”); Lambert, 126 P.3d at 651 (“Unless a defendant’s evidence of particular limitations is specifically contradicted by evidence that he does not have those limitations, then the defendant’s burden is met no matter what evidence the State might offer that he has no deficits in other skill areas.”).21
*367The Briseno factors, used exclusive of the clinical definitions, operate in a significantly different manner. As in this case, if the factfinder concludes that the petitioner met one of the Briseno factors even in a limited period of time or situation, the factfinder may then overlook the petitioner’s limitations and conclude that the petitioner is not mentally retarded. Thus, a person could easily have significant limitations in two or more of these areas and yet be determined not to be mentally retarded by a court that relied exclusively on the Briseno evidentiary factors. Exclusive reliance on the Briseno factors renders evidence of significant limitations meaningless, and is directly contrary to the clinical definitions.
Moreover, at least some of the Briseno factors are concerned with aspects of a person’s behavior that bear no relation to mental retardation as defined by the AAMR and APA. The second Briseno evidentiary factor (“Has the person formulated plans and carried them through or is his conduct impulsive?”) describes impulsivity as a sign of mental retardation, but the DSM-IV-TR flatly contradicts this: “Some individuals with Mental Retardation are passive, placid, and dependent, whereas others can be aggressive and impulsive.” DSM-IV-TR, supra note 1, at 44. The fourth and fifth Briseno evidentiary factors (“Is his conduct in response to external stimuli rational and appropriate, regardless of whether it is socially acceptable? Does he respond coherently, rationally, and on point to oral and written questions or do his responses wander from subject to subject?”) treat rationality and coherent communication as being inconsistent with mental retardation, but the clinical definitions do not indicate that mentally retarded people cannot communicate or behave rationally. On the contrary, the DSM-IV-TR states that people with mild mental retardation — who are “the largest segment (about 85%) of those with the disorder” — “typically develop social and communication skills” and “usually achieve social and vocational skills adequate for minimum self-support.” DSM-IV-TR, supra note 1, at 43.
Finally, the Briseno evidentiary factors also substantially depart from clinical definitions of mental retardation because they focus on isolated instances of a person’s behavior rather than on a person’s typical, day-to-day level of adaptive functioning. The seventh Briseno factor especially focuses on how the underlying capital crime was committed. This focus on isolated events — and especially on extraordinary events, like a situation in which a person commits murder — runs directly contrary to the clinical definitions’ emphasis on day-to-day functioning and on the fact that mentally retarded people, like anyone else, have strengths as well as limitations, in particular circumstances as well as in different skill areas. See Holladay, 555 F.3d at 1363 (criticizing the state’s expert for her “predominant focus on [the offender’s] actions surrounding the crime,” which “suggested] that she did not recognize [that people with mental retardation have strengths as well as weaknesses]”); see also Caroline Everington & J. Gregory Olley, Implications of Atkins v. Virginia: Issues in Defining and Diagnosing Mental Retardation, 8 J. Forensic Psychol. Prac., no. 1, 2008, at 1,11 (“[P]erhaps most important, adaptive behavior is the individual’s typical performance in his/her community setting. The details of the crime *368cannot be considered to be a sample of typical behavior.”).
As the Supreme Court explained in Williams v. Taylor, 529 U.S. at 405, 120 S.Ct. 1495 (O’Connor, J., concurring in part and concurring in the judgment, and speaking for the court on this point), “a state-court decision is contrary to this Court’s precedent if the state court arrives at a conclusion opposite to that reached by this Court on a question of law” or “if the state court confronts facts that are materially indistinguishable from a relevant Supreme Court precedent and arrives at a result opposite to ours.” Further elaborating, the Court stated:
The word “contrary” is commonly understood to mean “diametrically different,” “opposite in character or nature,” or “mutually opposed.” Webster’s Third New International Dictionary 495 (1976). The text of § 2254(d)(1) therefore suggests that the state court’s decision must be substantially different from the relevant precedent of this Court.... A state-court decision will certainly be contrary to our clearly established precedent if the state court applies a rule that contradicts the governing law set forth in our cases.
Id. (emphasis added).
Then, the Court gave two examples to clarify its meaning. First, it explained that:
If a state court were to reject a prisoner’s claim of ineffective assistance of counsel on the grounds that the prisoner had not established by a preponderance of the evidence that the result of his criminal proceeding would have been different, that decision would be “diametrically different,” “opposite in character or nature,” and “mutually opposed” to our clearly established precedent because we held in Strickland that the prisoner need only demonstrate a “reasonable probability that ... the result of the proceeding would have been different.”
Id. at 405-06, 120 S.Ct. 1495 (quoting Strickland v. Washington, 466 U.S. 668, 694, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984)). The Court then provided an example of when a state law is not “contrary to” a clearly established federal law:
[A] run-of-the-mill state-court decision applying the correct legal rule from our cases to the facts of a prisoner’s case would not fit comfortably within § 2254(d)(l)’s “contrary to” clause. Assume, for example, that a state-court decision on a prisoner’s ineffective-assistance claim correctly identifies Strickland as the controlling legal authority and, applying that framework, rejects the prisoner’s claim. Quite clearly, the state-court decision would be in accord with our decision in Strickland as to the legal prerequisites for establishing an ineffective-assistance claim, even assuming the federal court considering the prisoner’s habeas application might reach a different result applying the Strickland framework itself. It is difficult, however, to describe such a run-of-the-mill state-court decision as “diametrically different” from, “opposite in character or nature” from, or “mutually opposed” to Strickland, our clearly established precedent.
Id. at 406, 120 S.Ct. 1495.
Applying the Court’s precepts to the present case, it is clear that the TCCA has arrived at a decision that is contrary to Atkins because it applied a rule that contradicts the governing law set forth in Atkins. Similarly to the first example given in Williams v. Taylor, the TCCA rejected Chester’s Atkins claim on the grounds that he had not established his limitations in adaptive functioning by proving he was so limited or deficient accord*369ing to the state-law Briseno factors. That decision is “diametrically different,” “opposite in character or nature,” and “mutually opposed” to the Court’s clearly established precedent because the Court held in Atkins that Chester need only demonstrate that he has adaptive functioning deficits according to a definition that generally conforms to the AAMR or APA clinical definitions. The Briseno factors do not generally conform to, and indeed are “substantially different” from, either the AAMR or APA adaptive functioning criteria. See Williams v. Taylor, 529 U.S. at 405, 120 S.Ct. 1495.
In other words, this is not a run-of-the-mill state-court decision in which the state court correctly identified the clinical adaptive functioning prong of Atkins’ s clinical definition of mental retardation and, applying that framework, rejected the prisoner’s claim. Instead, the TCCA in this case disregarded the adaptive functioning framework recognized by the Court in Atkins and applied the state-court-created Briseno factors, which are “diametrically different,” “opposite in character or nature,” and “mutually opposed” to a definition that generally conforms to the clinical criteria recognized by the Court in Atkins for defining mental retardation. Specifically, the TCCA’s decision was contrary to Atkins itself, where the Court held that a defendant may demonstrate his adaptive functioning deficits by showing limitations in specific skill areas that generally conform to the AAMR or APA clinical mental retardation definitions.
Thus, because the Briseno evidentiary factors are substantively different from and contrary to the clinical and diagnostic approaches to determining deficits in adaptive functioning, when they are used as the sole measure of a person’s adaptive functioning, rather than concordantly with clinically standardized test results and professionally accepted criteria, they cannot accurately determine whether a person is mentally retarded in accordance with the national consensus’s understanding of mental retardation that the Supreme Court identified in Atkins.
B.
In addition to being contrary to Atkins, the TCCA decision in this case did not follow its own earlier opinion in Briseno in which it had stated that it would apply the AAMR definition of mental retardation or the similar Texas Health Code § 591.003(13) definition. Briseno, 135 S.W.3d at 7-8. The Briseno court mentioned some “other evidentiary factors” that a court “might also focus upon,” id. at 8 (emphases added), but it did not suggest that those factors could be used as a substitute for substantive clinical criteria in determining whether an offender has significant limitations in adaptive functioning. In this case, the TCCA disregarded the Atkins second prong clinical criteria altogether and used Briseno’s “other evidentiary factors” as its sole basis for determining that Chester does not have significant limitations in adaptive functioning. The district court took this substantive definition change a step further and held that an affirmative finding of the seventh Briseno factor alone can serve as the sole basis for affirming the TCCA.
The three-judge dissent in Lizcano v. State, No. AP-75879, 2010 WL 1817772 (Tex.Crim.App. May 5, 2010) (Price, J., concurring and dissenting) (unpublished), made the point forcefully that the TCCA majority, in using or allowing the use of the Briseno factors to the exclusion of clinical diagnostic criteria to determine whether a petitioner has satisfied the second prong of the tripartite definition of mental retardation, contradicts both At*370kins and Briseno. In Lizcano, as in Chester, the factfinder used the Briseno factors to the exclusion of clinical diagnostic criteria to reject the petitioner’s mental retardation claim and to affirm his death sentence. The dissent acknowledged that there may be “fodder” in the Briseno decision to support the Lizcano majority’s argument that the jury is not “bound by the diagnostic criteria,” id. at *34, but then correctly argued that such a belief runs contrary to established federal law: “Atkins adopted a categorical prohibition. It was founded upon the Supreme Court’s ratification of the prevalent legislative judgment that it is inappropriate to execute mentally retarded offenders. That legislative judgment comprehended mental retardation in essentially the same ‘clinical’ terms as the AAMR’s and APA’s diagnostic criteria.” Id. at *35.
The Lizcano dissent argued that the TCCA was not “justif[ied] [in its] apparent grant of latitude to fact-finders in Texas to adjust the clinical criteria for adaptive deficits to conform to their own normative judgments with respect to which mentally retarded offenders are deserving of the death penalty and which are not.” Id. In failing to “anchor the fact-finder’s decision on the specific diagnostic criteria,” id., the majority acted unconstitutionally: “Even if the Supreme Court in Atkins ‘did not mandate the application of a particular mental health standard for mental retardation, ... it did recognize the significance of professional standards and framed the constitutional prohibition in medical rather than legal terms.’ It would be anomalous to allow the fiat of a fact-finder to undermine the essentially diagnostic character of the inquiry. We should not ... permit a [fact-finder] capriciously to deviate from the specific diagnostic criteria in order to conform to its own normative, unnecessarily subjective, and certainly unscientific judgment regarding who deserves the death penalty.” Id. (first alteration in original) (footnotes omitted).
In summary, in order to determine whether Chester is mentally retarded and protected from execution under Atkins, the state courts were constitutionally obligated to employ a definition of mental retardation that would identify and protect the class of offenders covered by Atkins’s “substantive restriction on the State’s power to take [a] life” — namely, those offenders who are “so impaired as to fall within the range of offenders about whom there is a national consensus.” Atkins, 536 U.S. at 317, 321, 122 S.Ct. 2242. However, in their adjudication of Chester’s claim, the state courts transformed the Briseno evidentiary factors into a stand-alone substantive test for the adaptive functioning prong of the mental retardation definition, a test that does not “generally conform to the clinical definitions,” id. at 317 n. 22, 122 S.Ct. 2242, and hence cannot accurately determine whether an offender falls within the class that is protected by Atkins. The state courts did this even though Chester presented standardized test scores and other evidence tending to show that he satisfied the clinical criteria for adaptive functioning limitations (as well as the other two prongs of mental retardation, which he has undisputedly established). Therefore, the state courts’ decision that Chester is not mentally retarded was “contrary to” Atkins, under 28 U.S.C. § 2254(d)(1), because the state court applied a rule that contradicted the governing law clearly set forth in Atkins.
VII.
The issue presented by this case is not whether a state or federal court must strictly apply the AAMR or APA clinical definitions of mental retardation in deciding Atkins claims — as the majority sug*371gests — but rather whether a court must apply a definition that generally conforms to those clinical definitions, or whether a court can disregard or depart freely from them and make up its own unscientific and non-clinical definition of mental retardation that contradicts the definitions to which the national consensus generally conforms. Atkins requires that states apply a definition that “generally conforms” to the AAMR and APA clinical definitions and diagnostic assessments of mental retardation, which the national consensus has embraced. In this case, the TCCA utilized parts of the tripartite definition of mental retardation, but its definition of the adaptive functioning prong does not generally conform to the national consensus’s definition of this prong and in fact departs substantially from the nationally accepted criteria for determining whether a petitioner has adaptive functioning deficits.
The majority erroneously concludes that the Briseno factors provide a constitutionally acceptable means of limiting the class of defendants who are death eligible, and that Chester’s claim must fail. There is a vast distance between a holding requiring strict adherence to a clinical definition and a holding that would allow states to develop their own definitions of mental retardation without regard for the clinical definitions or the national consensus. While the former is not required by Atkins, the latter clearly falls outside Atkins’s constitutional bounds because Atkins requires that the state’s definition “generally conform” to the clinical definitions that the national consensus relied upon in narrowing the class of death eligible defendants to exclude mentally retarded defendants.
The Atkins holding clearly prohibits the execution of mentally retarded defendants. Although as in Ford, the Supreme Court left to the states the task of enforcing this restriction, Atkins, 536 U.S. at 317, 122 S.Ct. 2242, “[t]he bounds of that category are necessarily governed by federal constitutional law.” Ford, 477 U.S. at 419, 106 S.Ct. 2595 (Powell, J., concurring in part and concurring in the judgment, and speaking for the majority on this point). The prohibition becomes meaningless unless it is moored to a generally agreed upon definition of “mental retardation.” Yet this is what the majority does: it releases the definition from its moorings. The TCCA should not be permitted to circumvent Atkins’s constitutional prohibition by totally supplanting the definition of adaptive functioning that, prior to Briseno, had been utilized by Texas courts and which “generally conform[ed]” both with the AAMR clinical definition and with the national consensus that had developed around the AAMR and APA definitions.
The majority does not attempt to argue that the Briseno factors, standing alone, fall within the national consensus. Indeed, the opinion seems to suggest that states are empowered to ignore the national consensus. This national consensus tracks the “ ‘evolving standards of decency that mark the progress of a maturing society,’ ” which underlies Eighth Amendment jurisprudence. Atkins, 536 U.S. at 312, 122 S.Ct. 2242 (quoting Trop v. Dulles, 356 U.S. 86, 101, 78 S.Ct. 590, 2 L.Ed.2d 630 (1958)). States may not ignore it in favor of their own restrictions that would allow for the execution of individuals that the national consensus has decided should not be executed. This is clear from the text of Atkins, which left to the states only “ ‘the task of developing appropriate ways to enforce the constitutional restriction upon [their] execution of sentences.’ ” Atkins, 536 U.S. at 317, 122 S.Ct. 2242 (alteration in original) (emphasis added) (quoting Ford, 477 U.S. at 416-17, 106 S.Ct. 2595). It does not give states the power to define what the constitutional restriction is; that had already been determined by the na*372tional consensus upon which the Atkins Court based its holding.
In this case, the TCCA used the Briseno factors as a substitute for the clinical definition that Texas had previously pledged to follow and to which the national consensus generally conforms. It presents a “scientifically unsound and Atkinsviolative assessment!] of adaptive functioning.” Blume, supra note 13, at 706. Moreover, standing alone, the Briseno factors turn on its head the consensus’s approach to determining whether the petitioner has significant limitations in adaptive functioning. The AAMR and APA correctly assess adaptive functioning by analyzing the petitioner’s limitations. According to the AAMR, a person who is deficient in two out of the ten AAMR adaptive skill areas may be categorized as having significant limitations in adaptive functioning. The Briseno factors function in the opposite manner. According to the district court below, which the majority today affirms, the fact finder may find that a petitioner is not mentally retarded merely because he meets one of the seven Briseno factors. Moreover, several of the Briseno factors are markedly different from the clinical adaptive skill areas, and several are based on a person’s actions in a single moment instead of over a person’s lifetime. In other words, if the Briseno factors, standing alone, are allowed to replace an analysis that generally conforms to the clinical definitions, a single area or moment of strength can discount substantive evidence of significant limitations in numerous areas of adaptive functioning. See Blume, supra note 13, at 717-18. Furthermore, the TCCA, in developing the Briseno factors, did not conduct an assessment of the national consensus or draw from the clinical definitions of mental retardation around which the national consensus has coalesced. The factors are unmoored from the national consensus’s general understanding of what constitutes mental retardation. Used alone, these factors may determine that a subclass of persons protected by Atkins’s holding are, indeed, death eligible in Texas. The use of the Briseno factors in the present case therefore is clearly “contrary to ... clearly established Federal law, as determined by the Supreme Court of the United States.” 28 U.S.C. § 2254(d)(1).
VIII.
Chester’s claim that he is mentally retarded must be adjudicated in a manner that is consistent with Atkins. The TCCA’s adjudication of Chester’s claim was contrary to the law that Atkins clearly established. See Williams v. Quarterman, 551 F.3d at 358 (5th Cir.2008) (“A state court decision is contrary to clearly established federal law if it ‘applies a rule that contradicts the governing law set forth in [Supreme Court] cases....’” (first alteration in original) (quoting Williams v. Taylor, 529 U.S. at 405, 120 S.Ct. 1495)). Accordingly, I would vacate the judgment of the district court and remand this case to allow the district court to determine, not inconsistently with this opinion and pursuant to the federal law clearly established by the Supreme Court in Atkins, whether Chester is mentally retarded under a definition of mental retardation that “generally conform[s] to the clinical definitions” set forth in Atkins, 536 U.S. at 317 n. 22, 122 S.Ct. 2242, and thus “fall[s] within the range of mentally retarded offenders about whom there is a national consensus,” id. at 317, 122 S.Ct. 2242, and hence is protected from execution by the Eighth Amendment.
. Authorities have used the terms "adaptive functioning,” "adaptive behavior,” and "adaptive skills” to refer to this element of mental retardation. See Am. Ass’n on Mental Retardation, Mental Retardation: Definition, Classification, and Systems of Supports 1 (10th ed.2002) [hereinafter AAMR-10] (referring to “limitations ... in adaptive behavior as expressed in conceptual, social, and practical adaptive skills"); Am. Psychiatric Ass’n, Diagnostic and Statistical Manual of Mental Disorders 41 (4th ed., text rev.2000) [hereinafter DSM-IV-TR] (referring to “limitations in adaptive functioning in at least two of [eleven] skill areas”); Am. Ass’n on Mental Retardation, Mental Retardation: Definition, Classification, and Systems of Supports 5 (9th ed.1992) [hereinafter AAMR-9] (referring to "limitations in two or more of [ten] adaptive skill areas”). Neither the state nor Chester argues that these terminological differences have any effect on the issues in this appeal. Accordingly, this opinion refers to adaptive functioning, adaptive behavior, and adaptive skills interchangeably.
. See Panetti v. Quarterman, 551 U.S. 930, 949, 127 S.Ct. 2842, 168 L.Ed.2d 662 (2007) (“Justice Powell’s opinion constitutes ‘clearly established’ law for purposes of [28 U.S.C.] § 2254 and sets the minimum procedures a State must provide to a prisoner raising a Ford-based competency claim.’’)
. In Atkins, the Court also quoted and referred to the similar APA clinical definition of mental retardation. Because the Texas courts have adverted only to the AAMR clinical definition, and the two clinical definitions are substantively the same, I set forth and refer here only to the AAMR clinical definition of mental retardation.
. At least one study suggests that Texas may be favorably resolving far fewer Atkins claims than the national average. As of 2011, of the 81 Texas petitioners whose Atkins collateral review claims have been resolved, only 14 (17% of the total) have been successful. This "is significantly lower than the 'national average’ success rate of thirty-eight percent identified in a 2008 study of states that had resolved Atkins claims." Peggy M. Tobolowsky, A Different Path Taken: Texas Capital Offenders' Post-Atkins Claims of Mental Retardation, 39 Hastings Const. L.Q. 1, 71 & nn. 203-04, 373-74 (2011) (citing John H. Blume et al., An Empirical Look at Atkins v. Virginia and its Application in Capital Cases, 76 Tenn. L.Rev. 625, 628-29, 637 (2009)).
. In Ex parte Briseno, the TCCA determined how, in the absence of guidance from the state legislature, Texas courts would procedurally " ’enforce the constitutional restriction' ” imposed by Atkins. Briseno, 135 S.W.3d at 5 (quoting Atkins, 536 U.S. at 317, 122 S.Ct. 2242). The Briseno court held that Atkins claims in habeas corpus applications would be decided by judges rather than juries, id. at 11, and that the defendant would bear the burden of proof to establish mental retardation by a preponderance of the evidence, id. at 12.
In addition to establishing these procedural specifications, the Briseno court described "some other evidentiary factors which factfinders in the criminal trial context might also focus upon in weighing evidence as indicative of mental retardation or of a personality disorder.” Id. at 8. They include: (1) “Did those who knew the person best during the developmental stage ... think he was mentally retarded at that time, and, if so, act in accordance with that determination"; (2) "Has the person formulated plans and carried them through or is his conduct impulsive?”; (3) "Does his conduct show leadership or does it show that he is led around by others?"; (4) "Is his conduct in response to external stimuli rational and appropriate, regardless of whether it is socially acceptable?”; (5) "Does he respond coherently, rationally, and on point to oral or written questions or do his responses wander from subject to subject?"; (6) "Can the person hide facts or lie effectively in his own or others’ interests?”; (7) "Putting aside any heinousness or gruesomeness surrounding the capital offense, did the commission of that offense require forethought, planning, and complex execution of purpose?” Id. at 8-9. *355As will be discussed more fully below, the Briseno court did not initially present these factors as a substitute or alternative definition of mental retardation, but only as additional factors which a court “might also” take into consideration. The Briseno factors differ substantially from the adaptive skill areas identified by the AAMR and APA in the clinical definitions of mental retardation. See infra Sections IV, VI.A.
. The seventh Briseno evidentiary factor is as follows: "Putting aside any heinousness or gruesomeness surrounding the capital offense, did the commission of that offense require forethought, planning, and complex execution of purpose?” Briseno, 135 S.W.3d at 8-9.
. However, the district court granted Chester a certificate of appealability as to three issues, the first two of which are as follows:
1. The state courts’ determination that Mr. Chester is not mentally retarded is unreasonable in light of the evidence presented in the state court and violates the Eighth Amendment by permitting the execution of a mentally retarded defendant. 2. As a way to assess significantly subaverage adaptive functioning, the Briseno factors have no basis in the scientific literature and conflict with the accepted definitions of mental retardation recognized and relied on in Atkins v. Virginia, making the state courts’ exclusive reliance on the Briseno factors an unreasonable application of federal law clearly established in Atkins and violating the Eighth Amendment by permitting the execution of a mentally retarded defendant.
.Because the TCCA was the last state court to address Chester’s arguments, we review the decision of that court. See Wood v. Quarterman, 491 F.3d 196, 202 (5th Cir.2007) (fo*357cusing on whether “the last reasoned state court decision” was “contrary to, [or] an unreasonable application of, federal law”); see also Watson v. Anglin, 560 F.3d 687, 690 (7th Cir.2009) (“We review the decision of the last state court to address [the petitioner's] arguments.”); Mark v. Ault, 498 F.3d 775, 783 (8th Cir.2007) (“[W]e apply the AEDPA standard to ... the ‘last reasoned decision' of the state courts.” (quoting Ylst v. Nunnemaker, 501 U.S. 797, 804, 111 S.Ct. 2590, 115 L.Ed.2d 706 (1991))); Joseph v. Coyle, 469 F.3d 441, 450 (6th Cir.2006) ("[T]he decision we review is that of 'the last state court to issue a reasoned opinion on the issue.’" (quoting Payne v. Bell, 418 F.3d 644, 660 (6th Cir.2005))); Barker v. Fleming, 423 F.3d 1085, 1091 (9th Cir.2005) ("When more than one state court has adjudicated a claim, we analyze the last reasoned decision.”).
. See Panetti, 551 U.S. at 949, 127 S.Ct. 2842 (“Justice Powell's opinion [in Ford] constitutes 'clearly established law’ for purposes of [28 U.S.C.]§ 2254....”).
. See Pruitt v. State, 834 N.E.2d 90, 108 (Ind.2005) (plurality opinion) ("[T]he Eighth Amendment must have the same content in all United States jurisdictions.... Because Atkins explains that state statutes that provided the ‘national consensus' against the execution of the mentally retarded 'generally conform' to the AAMR or DSM-IV definitions, we conclude that Atkins requires at least general conformity with those definitions, but allows considerable latitude within that range.... [W]e think that the prohibition of the execution of the mentally retarded must have some content. There may be some flexibility in determining mental retardation, but we think that if a state's definition of mental retardation were completely at odds with definitions accepted by those with expertise in the field the definition would not satisfy the prohibition.”).
. Section 591.003(13) of the Texas Health and Safety Code states: " 'mental retardation' means significantly subaverage general intellectual functioning that is concurrent with deficits in adaptive behavior and originates during the developmental period.”
. As previously noted, the AAMR’s 1992 definition of mental retardation, as quoted by the Supreme Court in Atkins, includes an "adaptive functioning” prong, which requires "limitations in two or more of the following applicable adaptive skill areas: communication, self-care, home living, social skills, community use, self-direction, health and safety, functional academics, leisure, and work.” Atkins, 536 U.S. at 308 n. 3, 122 S.Ct. 2242 (quoting AAMR-9, supra note 1, at 5). The AAMR issued a revised definition of mental retardation in 2002, see AAMR-10, supra note 1, but the differences between the AAMR’s 1992 and 2002 criteria for adaptive functioning or adaptive behavior are not relevant to this appeal.
. The Briseno court may have mistakenly assumed that a defendant cannot suffer from both mental retardation and a personality disorder, when in fact individuals often suffer *361both from mental retardation and personality disorders. "|T|he scientific and clinical definitions emphasize that individuals with mental retardation often have mental disorders as well. No reasonable clinician would have determined that [a defendant] did not have mental retardation merely because the evidence also supported a diagnosis of antisocial personality disorder.” John H. Blume et al., Of Atkins and Men: Deviations from Clinical Definitions of Mental Retardation in Death Penalty Cases, 18 Cornell J.L. & Pub. Pol'y 689, 692 (2009). The AAMR has authoritatively explained that ”[t]he types of mental health disorders are the same in people with and people without mental retardation.” AAMR-10, supra note 1, at 172. "The prevalence of mental health disorders among individuals with mental retardation is difficult to estimate due to problems in sampling and diagnosis. In general, mental health disorders are much more prevalent in this population compared to the general population.” Id. Thus, an offender might well be correctly diagnosed with both mental retardation and a personality disorder. See Lambert v. State, 126 P.3d 646, 659 (Okla.Crim.App.2005) ("Mental retardation and mental illness are separate issues. It is possible to be mentally retarded and mentally ill.”).
. It is also unclear how the Briseno factors could be used as an independent, conclusive way to determine whether a person has significant limitations in adaptive functioning. The AAMR-9, DSM-IV-TR, and AAMR-10 definitions of mental retardation clearly specify how their adaptive functioning criteria are to be used: a person fulfills the criteria if he or she has significant limitations in two of ten, or two of eleven, or one of three, areas. By contrast, the Briseno factors do not indicate whether a defendant must negate one factor, or a majority of them, or all seven, or if some factors have more weight than others.
. The state trial court, on habeas, adopted the prosecution’s proposed findings of fact and conclusions of law in their entirety and thereby held that Chester did not have significantly subaverage intellectual functioning. However, the TCCA observed that "the trial court's findings on the applicant’s [IQ] test scores were in conflict with the record” and concluded that Chester “has met his burden in regard to demonstrating significant limitations in intellectual functioning.” Chester, 2007 WL 602607, at *2-3.
. Chester also argues that the evidence shows that he is mentally retarded under the revised definition promulgated by the AAMR’s tenth edition in 2002. See generally AAMR10, supra note 1. The questions raised in this appeal do not turn on any differences between the AAMR’s 2002 definition and the earlier definitions that the Supreme Court quoted in Atkins. Thus, we have no occasion to consider what the result should be if a defendant were to show that he was retarded under one clinical definition but not under another one. Cf., e.g., Wiley v, 625 F.3d at 216 n. 13 (noting that expert witnesses in that case "indicated that their diagnoses of [the defendant] were the same under both” the AAMR’s 1992 and 2002 definitions of the adaptive behavior prong); In re Hearn, 418 F.3d 444, 445 (5th Cir.2005) (citing the AAMR-10 in support of a general definition of mental retardation: "Mental retardation is a disability characterized by three criteria: significant limitation in intellectual functioning, significant limitation in adaptive behavior and functioning, and onset of these limitations before the age of 18”); United States v. Hardy, 762 F.Supp.2d 849, 879 (E.D.La.2010) ("While these differences in definition deserve *363note, they are ultimately of no consequence to the Court’s task. Just as in Wiley, ... the court need not decide which is preferable or correct, because the differences between them are mostly theoretical. Both the APA and AAMR direct clinicians to the same standardized measures of adaptive behavior, such as the Vineland Adaptive Behavior Scales-II....”).
.Courts, including the TCCA, have often considered VABS test scores to be a valid way of assessing the adaptive functioning prong of mental retardation. See, e.g., Van Alstyne, 239 S.W.3d at 820 n. 12 ("The Vineland Adaptive Behavior Test is one of the recognized standardized scales for measuring adaptive deficits.” (citing Ex parte Blue, 230 S.W.3d 151, 165 n. 55 (Tex.Crim.App.2007))).
Moreover, the AAMR favors the use of standardized tests, such as the VABS, as a way to diagnose mental retardation: "For the diagnosis of mental retardation, significant limitations in adaptive behavior should be established through the use of standardized measures normed on the general population, including people with disabilities and people without disabilities. On these standardized measures, significant limitations in adaptive behavior are operationally defined as performance that is at least two standard deviations below the mean....” AAMR-10, supra note 1, at 76 (emphasis added).
. The AAMR and APA definitions of mental retardation require that a person have limitations in only two adaptive skill areas. Atkins, 536 U.S. at 308 n. 3, 122 S.Ct. 2242.
. There are disputes about the significance of some of the evidence in the record. For instance, Dr. Edward Gripon, the state’s mental retardation expert, testified that the cor*364rect diagnosis for someone with an IQ score of 69 and a VABS score of 57 is mental retardation. Yet he also testified that he did not think Chester was mentally retarded.
The state emphasizes that although Chester was placed in the MROP, he was officially classified as having "borderline intellectual functioning" rather than mental retardation. However, Dr. Henry Orloff, a psychologist who was the MROP’s director at the time, testified that there was no basis for the diagnosis of borderline intellectual functioning, and that the erroneous diagnosis had been allowed to stand because at that time, it made no practical difference.
The state also emphasizes that in school, Chester was categorized as being learning disabled rather than mentally retarded. However, Chester counters that a diagnosis of learning disability would be simply invalid in light of his consistently low IQ test results, since a person with a learning disability by definition has a discrepancy between IQ and academic achievement. The DSM-IV-TR explains: "Learning Disorders are diagnosed when the individual’s achievement on individually administered, standardized tests in reading, mathematics, or written expression is substantially below that expected for age, schooling, and level of intelligence.” DSM-IV-TR, supra note 1, at 49 (emphasis added). By contrast, "In Mental Retardation, learning difficulties are commensurate with general impairment in intellectual functioning.” Id. at 51 (emphasis added).
. Nor did the TCCA make any attempt to apply any other clinical criteria for assessing adaptive functioning, such as the AAMR’s 2002 definition, which requires "performance that is at least two standard deviations below the mean of either (a) one of the following three types of adaptive behavior: conceptual, social, or practical, or (b) an overall score on a standardized measure of conceptual, social, and practical skills.” AAMR-10, supra note 1, at 14.
. The AAMR’s 2002 definition of mental retardation incorporates the following explanation:
Assumption 3: "Within an individual, limitations often coexist with strengths." This means that people with mental retardation are complex human beings who likely have certain gifts as well as limitations. Like all people, they often do some things better than other things. Individuals may have capabilities and strengths that are independent of their mental retardation. These may include strengths in social or physical capabilities, strengths in some adaptive skill areas, or strengths in one aspect of an adaptive skill in which they otherwise show an overall limitation.
AAMR-10, supra note 1, at 8. In accordance with this principle, the Oklahoma Court of Criminal Appeals, has explained that evidence regarding an offender's criminal activity is relevant for Atkins purposes only insofar as it tends to show that the offender does or does not have limitations in the areas specified by the clinical definitions: "Lambert was required to show ... that he had limitations in adaptive functioning in two of nine areas.... None of the evidence of criminal activity went to any of Lambert’s claims of adaptive function limitations. Thus, strictly speaking, none of it was relevant to disprove those claims.” Lambert, 126 P.3d at 656.
The Briseno evidentiary factors, because they focus heavily on isolated instances of a person's behavior, by design are not meant to *367indicate whether a person meets the standard clinical criteria for mental retardation, which assess an individual’s limitations in adaptive functioning based on his or her typical behavior.