The state’s appeal of the district court’s grant of Eric Lynn Moore’s petition for writ of habeas corpus was returned to this panel from the en banc court. We affirm.
I
Moore was convicted and sentenced to death for the murder of Helen Ayers in 1991. The Texas Court of Criminal Appeals (“TCCA”) affirmed Moore’s conviction and death sentence in 1994, and rejected Moore’s first habeas petition. The federal district court denied Moore’s habe-as petition and this Court affirmed in 2002. See Moore v. Cockrell, 54 Fed.Appx. 591 (5th Cir.2002).
In 2002, the United States Supreme Court decided Atkins v. Virginia, 536 U.S. 304, 122 S.Ct. 2242, holding that the execution of a mentally retarded inmate violates the Eighth Amendment. Moore filed a second state habeas petition with the TCCA, claiming that he was mentally retarded and thus ineligible for the death penalty under Atkins. The TCCA denied Moore’s claim. Ex Parte Moore, No. 38,670-02 (Tex.Crim. App 2003). Moore then filed a federal habeas petition, again making his Atkins claim. This Court authorized him to file a successive § 2254 petition. In re Moore, 67 Fed.Appx. 252 (5th Cir.2003). Further procedural activity ensued in this Court and in the district court, eventually resulting in Moore being granted an evidentiary hearing by the district court. After considering the evidence, the district court found Moore to be mentally retarded and enjoined the State from executing him.
The State appealed the district court’s ruling. This panel reversed the ruling on *67procedural grounds. Moore v. Quarterman, 454 F.3d 484 (5th Cir.2006). Moore petitioned for rehearing en banc, and the panel withdrew its opinion, replacing it with a second opinion that also reversed the district court on procedural grounds. Moore v. Quarterman, 491 F.3d 213 (5th Cir.2007) Moore filed a second petition for rehearing en banc, which was granted. The en banc Court reversed the panel, finding that Moore had established cause and prejudice for failure to exhaust his Atkins claim in state court. The case was returned to the panel for review of the district court’s determination on the merits of Moore’s Atkins claim. Moore v. Quarterman, 533 F.3d 338, 342 (5th Cir.2008) (“We return this case to the panel for review of the ultimate Atkins determination of mental retardation under the clear error standard.”)
II
To succeed on an Atkins claim, a defendant must prove by a preponderance of the evidence that he is mentally retarded. Lewis v. Quarterman, 541 F.3d 280, 283 (5th Cir.2008). We review a district court’s findings of fact for clear error and its conclusions of law de novo. 28 U.S.C. § 2254(d); e.g. Lewis, 541 F.3d at 283. The question of whether a defendant suffers from mental retardation is a question of fact, and thus subject to clear error review. See Clark v. Quarterman, 457 F.3d 441, 444 (5th Cir.2006).
Factual findings are clearly erroneous only if they are implausible in light of the record as a whole. Rivera v. Quarterman, 505 F.3d 349, 360 (5th Cir.2007). “If the district court’s account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently. Where there are two permissible views of the evidence, the factfinder’s choice between them cannot be clearly erroneous.” Anderson v. City of Bessemer City, N.C., 470 U.S. 564, 573-74, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985) (citing United States v. Yellow Cab Co., 338 U.S. 338, 342, 70 S.Ct. 177, 94 L.Ed. 150 (1949)). A reviewing court accords deference to determinations of credibility made by the factfinder; this deference also extends to findings based on “physical or documentary evidence or inferences from other facts.” Id. at 574, 105 S.Ct. 1504.
Ill
The district court conducted its assessment of the facts under the framework developed by the Texas Criminal Court of Appeals in Ex Parte Briseno, 135 S.W.3d 1 (Tex.Ct.Crim.App.2004), which provides the guidelines for determining mental retardation for the purposes of Atkins claims.1 Under Briseno, a capital defendant must prove by the preponderance of the evidence that he is mentally retarded. This burden of proof applies to three showings: (1) the defendant had significant subaverage intellectual functioning, (2) with related limits in adaptive functioning, and (3) the onset of which occurred prior to the defendant turning eighteen. This “three-pronged test” is an adaptation of the definition of mental retardation provided by the American Association for Mental Retardation (AAMR) and set out in section 591.003(13) of the Texas Health and Safety Code. Briseno, 135 S.W.3d at 7. *68The district court focused the bulk of its analysis on intellectual and adaptive functioning.2
A
Significantly subaverage intellectual functioning is defined as an IQ of about 70 or below (approximately 2 standard deviations below the mean). Briseno, 135 S.W.3d at 7 n. 24 (citing to AmerioaN Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders (Text Revision, 4th ed.2000) (hereinafter “DSM-IV”); AmeRican Association on Mental Deficiency, Classifioation in Mental Retardation N1 (Grossman ed.l983)(hereinaf-ter “AAMD”)). IQ tests may overstate or understate the subject’s actual level of intellectual functioning. See, id., 135 S.W.3d at 8 (“Psychologists and other mental health professionals are flexible in their assessment of mental retardation; thus, sometimes a person whose IQ has tested above 70 may be diagnosed as mentally retarded while a person whose IQ tests below 70 may not be mentally retarded.”).
The standard instrument for measuring intellectual functioning is the Wechsler Adult Intelligence Scales test (WAIS III). Atkins, 536 U.S. at 309 n. 5, 122 S.Ct. 2242. Moore presented the results of two Wechsler-based IQ tests before the district court. His scores were, respectively, 76 (on a Wechsler Adult Intelligent Scale-Revised, or WAIS-R, test,) and 66 (on the WAIS III test). He also presented the results of a 1973 Primary Mental Abilities (PMA) Test that was administered by his school; the result was 74. The most recent test, the WAIS III, was conducted in conjunction with a test of non-verbal abilities (TONI-2), which reduces the impact of the test taker’s educational background on his score. Moore’s expert, Dr. Antonin Llórente (“Dr.Llorente”), testified that the TONI-2 results placed him in the bottom eight-tenths percentile of the entire population.
The district court based its holding that Moore had proved by preponderance of the evidence that he exhibited significant subaverage intellectual functioning on two facts: First, the court found that the State’s expert, Dr. Gary Mears (“Dr. Mears”), agreed under cross-examination that Moore satisfied the intellectual functioning prong of the Briseno test; second, the court averaged the three IQ test scores, for a score of 72, and applied a five-point standard error of measurement. The district court found that the result satisfied the AAMR criterion of subaver-age intellectual functioning. The State argues that both findings were clearly erroneous because (1) its expert did not “concede” that Moore satisfied the intellectual functioning prong; and (2) all three of the test scores are so unreliable as to be invalid.
The district court relied on the following exchange between Moore’s lawyer and Dr. Mears, in reference to the intellectual functioning prong of the Briseno test:
Q: So you agree with Dr. Llórente that whether we use the DSM[-IV] definition or the AAMR definition, the first prong is satisfied in your professional opinion?
A: The prong in terms of — I have some questions about his accuracy of score, but I would still, nevertheless, because of my opinion about the adaptive functioning, I will accept that.
Q: You do accept that?
*69A: I will accept it.
Q: Because I don’t want to spend half a day talking about it?
A: Right. I will accept it — regardless of the scoring errors, I still will accept it.
The discussion then moved on to the adaptive functioning prong of the Briseno test.
Though the State contends that this exchange does not constitute a “concession” by Dr. Mears that Moore satisfies the intellectual functioning prong, we find that it is at best ambiguous and defer to the district court’s view of the testimony. It is clear that Dr. Mears did not consider Moore to be mentally retarded because (in his assessment) the strength of Moore’s adaptive functioning outweighed any limitations in intellectual functioning. However, it is also a reasonable interpretation of Dr. Mears’ testimony that he believed Moore to have significant limitations in intellectual functioning based solely on his IQ test scores.
The dissent characterizes Dr. Mears’ prior testimony about the IQ tests as breaking dramatically with his later agreement that Moore satisfied the intellectual functioning prong. On the contrary, the report submitted by Dr. Mears one day before he testified, which the dissent references, clearly states, “I accepted the results of the administration, testing, and scoring of the intelligence testing done by both Drs. Llórente and Fulbright. Psy-chometrically, the scores are rather consistent considering a more than a decade lag.” Though Dr. Mears discusses the IQ tests at length in his testimony, he never states that any scoring error in the administration of the WAIS-III renders it so unreliable or inaccurate such that it could not be used to determine intellectual functioning. Rather, he maintains reservations about potential scoring errors (though he does not state that the test should be thrown out altogether) and disagrees about the degree to which reliance on the scores alone, without analysis of adaptive functioning, can be the basis for making an overall determination of mental retardation. Dr. Mears’ views on Moore’s adaptive functioning, and his overall view that Moore is not mentally retarded, is not incompatible with agreement that Moore’s intellectual functioning as measured by the IQ tests satisfies Briseno. The point is that the district court’s interpretation of the exchange, in light of the other documentary and evidentiary testimony, was reasonable. We are not left with the “definite and firm conviction that a mistake was committed.” Anderson, 470 U.S. at 574-75, 105 S.Ct. 1504.
The State argues next that the unreliability of all three test scores nullifies any basis upon which to base the intellectual functioning finding. The State also points to evidence it says indicates that Moore deliberately scored poorly on the WAIS-III test.
While there was conflicting testimony as to the presence of scoring errors on all three tests, the district court had a clear basis in the record to find that the scores were consistent after accounting for their margins of error, and were thus collectively a sufficiently reliable indicator of Moore’s IQ. For example, Dr. Mears stated at the hearing that he did not administer his own independent test because he “had sufficient number of tests of the quantify intelligence tests [sic] to at least get a view of how [Moore] compares with the people that take those tests.” Dr. Llórente testified that “we are not trying to reach here a conclusion about a patient’s intellectual level on the basis of one test alone. That is not the right thing to do, and I agree with Dr. Mears on that. So it should be on the basis of multiple, for *70example procedures, which is what we did.” In averaging the test scores and relying on the five-point margin of error that Dr. Llórente testified was applicable to each score, the district court attempted to find a way to reconcile all three test scores.3
The district court also heard expert testimony that supported the averaging of the three scores and the application of the confidence interval. Dr. Llórente extensively testified as to each of the testing protocols and noted that, despite potential scoring errors and various other considerations, all three tests demonstrated consistency in their findings; he also testified that a five point margin of error was applicable to each test score. Relying on the AAMR definition and how the AAMR accounts for the standard error of measurement, Dr. Llórente testified that a score of “75 and below would be considered mental retardation.” In the expert report he submitted to the district court, Dr. Mears also noted that the AAMR associates mild mental retardation with IQs ranging from 50 to 75.4
Thus, there were several bases to support the district court’s finding that Moore satisfied the intellectual functioning prong of the Briseno test: the existence of the WAIS III score below 70; Dr. Llorente’s testimony that the other two adjusted test scores were consistent with the WAIS III score considering the standard errors of measurement; and Dr. Llorente’s testimony that a score of 75 or less constitutes mental retardation.5 Based on this evidence, we find no clear error.
*71Finally, the district court also rejected the evidence proffered by the State of “malingering” by Moore. See id. (“Moore’s scores on [] tests [other than the PMA] showed no response bias, which is an attempt by Moore to perform poorly.”).
We therefore find that, because the record supports the district court’s assessment of Moore’s intellectual functioning, the district court did not clearly err in holding that Moore satisfies that prong of the Briseno test.
B
The district court spent the bulk of its opinion discussing evidence under the adaptive functioning prong of the Briseno test. Under this prong, the factfinder looks for “significant limitations in an individual’s effectiveness in meeting the standards of maturation, learning, personal independence, and/or social responsibility that are expected for his or her age level and cultural group, as determined by clinical assessment and, usually, standardized scales.” Briseno, 135 S.W.3d at 7 n. 25 (citing American Association on Mental DEFICIENCY (AAMD), CLASSIFICATION IN Mental Retardation 11).
In arguing that the district court clearly erred in its evaluation of the adaptive functioning evidence, the State primarily argues that the “great weight” of the evidence cuts the other way, and that the district court abused its discretion in how it conducted its factfinding. Again, however, we must be mindful of the deferential lens through which we view the lower court’s credibility determinations and conflicting testimony and evidence. The district court relied exclusively on the AAMR’s guidelines for determining mental retardation, and followed the AAMR in examination of three areas: Moore’s conceptual skills, Moore’s social skills, and Moore’s practical skills.6 Moore, 2005 WL 1606437 at *5. The district court consid*72ered “experts’ clinical judgments and other evidence to determine whether Moore [had] significant deficits that place him approximately two standard deviations below the mean in adaptive functioning.” Id. at *6. As part of assessing the “clinical judgment” of the experts, the court looked to their training, experience with others who are mentally retarded, and familiarity with the individual and his environment. Id. The court also found Dr. Llorente’s testimony to be more credible than Dr. Mears, especially with respect to evaluation of adaptive functioning, based on the fact that Dr. Llórente spent a great deal more time than Dr. Mears evaluating Moore and interviewing people who knew him:
On all three criteria, but especially Moore’s adaptive functioning, the Court finds Dr. Llorente’s testimony and assessment more credible than Dr. Mears’s. Dr. Llórente spent seven to eight hours interviewing and evaluating Moore, while Dr. Mears spent only two to two and half hours. Dr. Llórente contacted and interviewed Moore’s family members to learn about Moore’s childhood, while Dr. Mears did not contact or interview anyone because he thought their opinions were not useful. Instead, Dr. Mears relied exclusively on his comparatively brief interview with Moore and his review of Moore’s records. Dr. Llórente administered a large battery of tests to Moore. Dr. Mears also performed some tests on Moore, but not nearly as many.
Id. at *13.
We first address the State’s contention that the district court abused its discretion in not considering “other factors” suggested in Briseno to help the factfinder distinguish mental retardation from antisocial personality disorder. These factors are: (1) whether those who knew the person best during the developmental stage thought he was mentally retarded at the time, and if so, acted according to that determination; (2) whether the person formulated plans and carried them through; (3) whether the person’s conduct showed leadership; (4) whether the person’s conduct in response to external stimuli is rational and appropriate, regardless of whether it is socially acceptable; (5) whether the person responds coherently, rationally, and on point to oral or written questions; (6) whether the person hides facts or lies effectively in his own or others’ interests; (7) whether the commission of the offense required forethought, planning, and complex execution of purpose. Briseno, 135 S.W.3d at 8. The district court declined to apply these factors for the following reasons: it found that the State did not argue or present evidence that Moore had an antisocial personality disorder that accounted for his adaptive functioning deficiencies; that, even if the argument that Moore suffered from antisocial personality disorder had been raised, there was no evidence to support a finding that such a disorder caused Moore’s cognitive and adaptive deficits; that the factors are not part of the AAMR’s definition of mental retardation; and that application of the factors is purely discretionary. Moore, 2005 WL 1606437 at *5 n. 6.
The district court did not abuse its discretion in not explicitly considering the Briseno factors, as it found that the State did not present evidence to support a finding that a personality disorder is responsible for Moore’s cognitive and adaptive deficits. Id. Though the State does point to testimony provided by its experts that Moore displayed symptoms of antisocial personality disorder, we must defer to the district court’s determination that any such symptoms did not cause his subaver-*73age intellectual and limited adaptive functioning.7
Notwithstanding the district court’s statement that it would not consider the Briseno factors, the transcript of the evi-dentiary hearing indicates that the court did consider evidence implicating the factors, including the factor that is of particular importance to the State — Moore’s role in the crime for which he was convicted. The district court specifically asked Dr. Llórente whether his evaluation of Moore’s mental retardation would be changed by evidence that Moore was the leader of the crime, and Dr. Llórente responded that it would not. Moore’s counsel then followed up by pointing out that in the 1991 assessment of Moore, the report stated that Moore was more likely a follower than a leader, and Dr. Llórente agreed that Moore’s mental capacity is such that he would be easily manipulated and influenced by others. Though this discussion is absent in the district court’s memorandum opinion, it is clear from the transcript of the hearing that the court considered it.
Evidence pertaining to the other Brise-no factors also appears throughout the evidentiary hearing. In the hundreds of pages of testimony, the district judge heal'd: opinions from people who knew Moore during his developmental stage about whether they thought he was mentally retarded; evidence about Moore’s functioning at school and at work showing his planning skills and response to external stimuli; and evidence about how he responds to oral and written questions. For example, the district court looked at evidence from Moore’s school records, and heard from numerous witnesses including teachers, family members, classmates and acquaintances regarding his inability to complete schoolwork and perform other basic tasks until the age of eighteen. The district court also heard testimony that Moore was unable to properly dress himself, could not tie his shoes, consistently scored well below his grade level, and had difficulties learning to speak in sentences and paragraphs, as well as testimony that Moore had difficulties performing basic assignments, and was a follower who was easily tricked as a child. Moore, 2005 WL 1606437 at *7-*10. The district court also heard from former co-workers and employers as to deficits in his ability to complete work-related tasks. Id. at *11. In addition to testimony and documentary evidence regarding Moore’s school and work performance, the district court considered Dr. Llorente’s and Dr. Mears’ examinations of Moore, which included both oral and written evaluations. Id. at *13.
The district court relied on this evidence for its finding that Moore’s deficits in adaptive functioning satisfy Briseno. The State argues that the district court incor*74rectly viewed the evidence, arguing, for example, that substantial evidence showed that Moore’s academic performance would have improved if he applied himself, that Moore’s teachers did not remember his inability to tie his shoes, that he was employed at several jobs and was commended for good performance, and that he communicated and interacted well with his classmates. Where evidence conflicts, however, we must defer to the fact-finder’s decision as to which evidence to credit, and we may not simply re-weigh the facts to come to a contrary conclusion. Anderson, 470 U.S. at 574, 105 S.Ct. 1504. The State also argues that a determination of mental retardation for the purposes of Atkins cannot rest on such subjective factors. However, the Briseno court acknowledged that the adaptive behavior criteria are indeed “exceedingly subjective,” Briseno, 135 S.W.3d at 8, and we are bound to apply the law as set out by the Texas courts.
IV
In arguing that the district court clearly erred, the State essentially re-litigates the evidentiary ease it presented below. However, the district court was in a better position than this court to judge the credibility of the witnesses who testified on the extent, duration, and causes of Moore’s intellectual and adaptive functioning limitations. See Rivera, 505 F.3d at 363. Because its findings were not “implausible,” they “survive[] clear error review.” Id. Accordingly, we AFFIRM.
Pursuant to 5th Cir. R. 47.5, the court has determined that this opinion should not be published and is not precedent except under the limited circumstances set forth in 5th Cir R. 47.5.4.
. The Texas legislature has yet to enact legislation for determining mental retardation under Atkins. See, e.g. Neal v. State, 256 S.W.3d 264, 270 (Tex.Ct.Crim.App.2008) ("Over four years later [following Briseno], the Texas legislature still has not enacted any legislation on this matter.”)
. The state does not argue that the district court erred in finding that Moore fulfilled the third criterion of the Briseno test: manifestation of significantly subaverage intellectual functioning and related adaptive deficits before the age of eighteen. See Moore v. Dretke, 2005 WL 1606437, at *15 (E.D.Tex. July 1, 2005).
. The district court could have relied solely on Dr. Llorente's testimony and concluded that Moore demonstrated significant subaver-age intellectual functioning, especially as it expressly found Dr. Llorente's testimony to be more credible than Dr. Mears as to all three Briseno criteria, given the greater comprehensiveness of Dr. Llorente's examination. Moore, 2005 WL 1606437, at *13.
. Dr. Mears never addressed in his testimony whether a score of 75 or below was compatible with a finding of mental retardation. As previously discussed, Dr. Mears testified that IQ test scores alone, without consideration of educational and cultural factors, could not be used to determine whether an individual was mentally retarded.
. Judge Smith would hold that, under Texas law, there is a bright-line IQ score cutoff of 70 and apply it in this case. Notwithstanding the fact that the State never made this argument before the Court, and thus we found no reason to address it, we note that neither Briseno nor its progeny establish that such a bright-line cutoff exists. The TCCA decisions which apply the Briseno criteria do not change the standard outlined in Briseno, which defined significantly subaverage functioning as occurring in persons with an "IQ of about 70 or below (approximately 2 standard deviations below the mean).'' 135 S.W.3d at 7 n. 24 (citing to DSM-IV and American Association on Mental Deficiency, Classification in Mental Retardation N1 (AAMD)) (emphasis added). The Briseno court declined to create any presumption that mental retardation must be established by IQ scores lower than 70. See Briseno, 135 S.W.3d at 7 n. 24 (declining to adopt Ohio's rebuttable presumption that a defendant is not mentally retarded if his or her IQ is above 70). Briseno did not alter the standards outlined in Atkins, which states that mental retardation can be found in range of 70-75. 536 U.S. at 309, 122 S.Ct. 2242 ("It is estimated that between 1 and 3 percent of the population has an IQ between 70 and 75 or lower, which is typically considered the cutoff IQ score for the intellectual function prong of the mental retardation definition.”) TCCA cases are not in conflict with Briseno or Atkins. See e.g. Ex parte Modden, 147 S.W.3d 293, 298 (Tex.Crim.App.2004) ("the applicant’s IQ scores of 58 and 64 are well below the 70-75 score that generally indicates subaverage general intellectual functioning.”). Briseno makes clear the accounting for the standard error of measurement is an inquiry conducted by the factfinder based on expert testimony. See Briseno, 135 S.W.3d at 7 n. 24 (recognizing that “[p]sychologists and other mental health professionals are flexible in their assessment of mental retardation” and that a *71person tested with an IQ above 70 may be mentally retarded); id. at 14 n. 53 (discussing expert disagreement as to application of the confidence interval and ultimately rejecting application due to lack of evidence in the record).
The criticism that the majority fails to take the opportunity to "make sense of Texas's jurisprudence regarding retardation ... and to provide guidance” is misplaced. First, the ultimate responsibility for providing guidance and clarifying the relevant standards, especially to the degree that such "clarification” actually calls into question the language articulated in Atkins and in Briseno, rests with the Texas courts, not with us. See Atkins, 536 U.S. at 317, 122 S.Ct. 2242 (”[W]e leave to the State[s] the task of developing appropriate ways to enforce the constitutional restriction upon [their] execution of sentences.”) (internal citations omitted). Second, we are reluctant to create new law in this area where the parlies fail to identify or address these novel issues.
. The criticism that the district court erred as a matter of law in “inconsistently” applying the AAMR 9th and 10th editions was never made by the State, likely because no rule on AAMR editions exists in either Texas or the 5th Circuit. Briseno itself cites both the AAMR 9th and AAMR 10th almost interchangeably, 135 S.W.3d. at 7-8, as Judge Smith acknowledges. The court in Briseno also does not cite either the AAMR 9th or the AAMR 10th for a definition of limitations in adaptive functioning, instead citing the American Association on Mental (AAMD) definition. Id. at 7 n. 25 (citing AAMD at 11). This definition, while similar, is not the same as either the AAMR 9th or the AAMR 10th. Further, this Court has never distinguished between the AAMR 9th and the AAMR 10th.
Given the absence of any briefing by the parties on this issue, we decline to create a new mandatory methodology in applying the AAMR. Adoption of such a rule would again contravene Atkins’ mandate to leave the development of standards for weighing mental retardation to the states. Atkins, 536 U.S. at 317, 122 S.Ct. 2242.
. Further, Briseno makes clear that the application of the factors is discretionary. Brise-no, 135 S.W.3d at 8 ("There are, however, some other evidentiary factors which factfin-ders in the criminal trial context might also focus upon in weighing evidence as indicative of mental retardation or of a personality disorder ... ”) (emphasis added). Subsequent TCCA cases do not alter this discretionary language. See Hunter v. State, 243 S.W.3d 664, 666 (Tex.Crim.App.2007) ("Other eviden-tiary factors that fact finders in the criminal-trial context might also focus upon in weighing evidence as indicative of mental retardation include ..."); Gallo v. State, 239 S.W.3d 757, 769-70 (Tex.Crim.App.2007) (same) (emphasis added); Ex Parte Modden, 147 S.W.3d 293, 296 (Tex.Crim.App.2004) ("In Ex parte Briseno, we concluded that the criteria in these definitions are subjective, and thus, we set out some additional factors that factfin-ders may use.")', Rosales v. Quarterman, 291 Fed.Appx. 558, 562 (5th Cir.2008) (per cu-riam) ("In Briseno, the TCCA also laid out a number of guidelines that courts can consider when making mental retardation determinations.”) (emphasis added).