Full opinion text
MEMORANDUM & ORDER NICHOLAS G. GARAUFIS, District Judge. The execution of those who are mentally retarded violates both the Federal Death Penalty Act and the Eighth Amendment. See 18 U.S.C. § 3596(c); Atkins v. Virginia, 536 U.S. 304, 321, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002). Defendant Earl Ronell Wilson, a convicted murderer of two undercover police officers, claims that he is mentally retarded and therefore ineligible to receive the death penalty. For the reasons that follow, he is incorrect. I. BACKGROUND In 2003, Wilson murdered two undercover NYPD detectives who were posing as gun purchasers. (See Second Superseding Indictment (Dkt. 179) ¶¶ 7, 9.) He was tried in this court for capital-eligible crimes. (Trial Tr. (Dkts. 362-404).) The jury convicted Wilson and voted to impose the death penalty. (Jury Verdict (Dkt. 351).) The court accordingly sentenced Wilson to death. (Judgment (Dkt. 407).) Wilson appealed, and the Second Circuit affirmed Wilson’s convictions but vacated his death sentence on constitutional grounds and remanded to this court for retrial of his penalty phase. United States v. Whitten, 610 F.3d 168 (2d Cir.2010). After the Second Circuit’s mandate issued, Wilson requested “a pretrial hearing to determine whether he is a person with mental retardation” and thus ineligible for the death penalty under the Eighth Amendment and the Federal Death Penalty Act (the “Atkins claim”). (Dkt. 614.) The court granted this request and set a schedule for exchange of expert information, motions related to the Atkins claim, and an evidentiary hearing (the “Atkins hearing”). (See Feb. 2, 2012, Order (Dkt. 618).) Wilson provided notice of his intent to call four mental health experts at the Atkins hearing: (1) John Olley, Ph.D., a psychologist; (2) Bruce Shapiro, M.D., a developmental pediatrician; (3) Joette James, Ph.D., a neuropsychologist; and (4) George Woods, Jr., a licensed physician. (Mar. 7, 2012, Def. Ltr. (Dkt. 637).) The Government stated that it would call three experts at the hearing: (1) Robert Denney, Psy.D., a neuropsychologist; (2) Robert Mapou, Ph.D., a neuropsychologist; and (3) Raymond Patterson, M.D., a psychiatrist. (Mar. 7, 2012, Gov’t Ltr. (Dkt. 638); Apr. 6, 2012, Gov’t Ltr. (Dkt. 676); May 2, 2012, Gov’t Ltr. (Dkt. 697).) Since then, the parties and their experts have conducted extensive discovery and testing in preparation for the Atkins hearing. The court has also issued two opinions in response to motions filed by the parties regarding the scope of discovery. See United States v. Wilson, No. 04-CR-1016 (NGG), 2012 WL 3890951, at *4-8 (E.D.N.Y. Sept. 7, 2012); United States v. Wilson, No. 04-CR-1016 (NGG), 920 F.Supp.2d 287, 295-306, 2012 WL 6962982, at *6-16 (E.D.N.Y. June 22, 2012). On September 7, 2012, the parties exchanged expert reports. (See Sept. 7, 2012, Def. Ltr. (Dkt. 868).) Each of the Government’s experts opined that Wilson is not mentally retarded. (Denney Rep. (Dkt. 956) at 48; Patterson Rep. (Dkt. 957) at 18; Mapou Rep. (Dkt. 958) at 35.) Each of Wilson’s experts opined that he is mentally retarded. (James Rep. (Dkt. 959) at 1, 17; Olley Rep. (Dkt. 960) at 28; Shapiro Rep. (Dkt. 961) at 2, 22; Woods Rep. (Dkt. 962) at 29.) The court held the Atkins hearing over nine days in November and December 2012. (See Minute Entries (Dkts. 950-55, 976-78).) It heard testimony from all seven of the experts mentioned above and four other witnesses. (See Atkins Hr’g Tr. (“Tr.”).) The parties submitted briefing on the Atkins claim after the hearing. (Def. Mem. (Dkt. 982); Gov’t Mem. (Dkt. 983); Def. Reply (Dkt. 999).) II. STANDARD FOR MENTAL RETARDATION Two provisions of law forbid federal courts from imposing a death sentence upon a person who is mentally retarded. First, the Federal Death Penalty Act (“FDPA”), originally enacted by Congress in 1988 and amended in 1994, provides that a “sentence of death shall not be carried out upon a person who is mentally retarded.” 18 U.S.C. § 3596(c). Second, the execution of mentally retarded individuals violates the Eighth Amendment’s ban on “cruel and unusual punishments.” Atkins v. Virginia, 536 U.S. 304, 321, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002); see U.S. Const, amend. VIII (“Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted”). In Atkins, the Supreme Court found that, due to the relatively recent legislative efforts of several states, “a national consensus ha[d] developed against” the execution of mentally retarded offenders. Atkins, 536 U.S. at 316, 122 S.Ct. 2242. Such executions were therefore inconsistent with “ ‘the evolving standards of decency that mark the progress of a maturing society’ ” — the guiding principle of the Eighth Amendment. Id. at 311-12, 126 S.Ct. 941 (quoting Trop v. Dulles, 356 U.S. 86, 101, 78 S.Ct. 590, 2 L.Ed.2d 630 (1958)). The Court concluded that although the intellectual deficiencies of mentally retarded criminals did “not warrant an exemption from criminal sanctions ”— including life imprisonment — such criminals “should be categorically excluded from execution ” ■ for two main reasons. Id. at 318, 126 S.Ct. 941 (emphases added). First, there was a “serious question” as to whether the execution of mentally retarded offenders would serve the deterrence or retribution justifications of the death penalty. Id. at 318-319, 126 S.Ct. 941. Second, there was an enhanced risk in the case of mentally retarded offenders “that the death penalty w[ould] be imposed in spite of factors which may call for a less severe penalty,” both because of “the possibility of false confessions” by mentally retarded defendants and because of the “lesser ability of mentally retarded defendants to make a persuasive showing of mitigation.” Id. at 320, 126 S.Ct. 941. It is therefore clear that this court may not sentence a mentally retarded criminal to death, but that is where most of the clarity ends. The difficult task is deciding which persons qualify as “mentally retarded” under the FDPA and Atkins — an issue of first impression in this Circuit. A. Sources of the Definition Neither the FDPA nor Atkins mandates a particular definition of mentally retardation. The FDPA' provides simply that “mentally retarded” persons may not be executed. 18 U.S.C. § 3596(c); see also Garcia Briseno v. Dretke, No. 05-CV-08, 2007 WL 998743, at *10 n. 8 (S.D.Tex. Mar. 29, 2007). And Atkins expressly left “to the States the task of developing appropriate ways to enforce the constitutional restriction upon their execution of sentences.” 536 U.S. at 317, 122 S.Ct. 2242 (alterations omitted); see also Bobby v. Bies, 556 U.S. 825, 129 S.Ct. 2145, 2150, 173 L.Ed.2d 1173 (2009) (“Our opinion [in Atkins ] did not provide definitive procedural or substantive guides for determining when a person' who claims mental retardation ‘will be so impaired as to fall [within Atkins ’ compass].’ ” (second alteration in original) (quoting Atkins, 536 U.S. at 317, 122 S.Ct. 2242)). Atkins noted, however,' that although state “statutory definitions of mental retardation [we]re not identical, [they] generally conform[ed] to [ ] clinical' definitions” promulgated by two groups: (1) the American Association on Mental Retardation (“AAMR”), which has since changed its name to the American Association on Intellectual and Developmental Disabilities (“AAIDD”); and (2) the American Psychiatric Association (“APA”). Id. at 308 n. 3, 317 n. 22, 122 S.Ct. 2242. The Court further noted that these clinical definitions “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. In short, Atkins declined to mandate a definition of mental retardation but left it to the states to define the term, while noting that existing state definitions generally conformed to the clinical definitions set forth by the AAMR and the APA. Atkins did not hold that federal courts are bound to apply the mental retardation definitions of the particular states in which they are located, nor does the FDPA contain any such mandate. Federal courts that have decided cases involving both Atkins and FDPA claims have taken inconsistent approaches in this regard: some have applied their forum state’s law, see, e.g., United States v. Cisneros, 385 F.Supp.2d 567, 571-72 (E.D.Va.2005) (applying Virginia law), while others have made no mention of their state’s law and applied only clinical definitions of mental retardation, see, e.g., United States v. Hardy, 762 F.Supp.2d 849 (E.D.La.2010) (no mention of Louisiana law); United States v. Davis, 611 F.Supp.2d 472 (D.Md.2009) (no mention of Maryland law). This court will consider New York law in determining the definition of mental retardation, while noting that: (1) Atkins does not explicitly require that the court be bound by New York law; and (2) even if it did, an application of New York law would, as discussed below, ultimately lead the court to rely primarily upon clinical definitions of the term. New York has been without the death penalty since 2004, when the New York Court of Appeals held that the State’s capital sentencing statute violated its Constitution. See People v. LaValle, 3 N.Y.3d 88, 783 N.Y.S.2d 485, 817 N.E.2d 341 (2004). This statute is, however, still on the books for the most part. See New York Criminal Procedure Law (“C.P.L.”) § 400.27. As before LaValle, the statute requires a court to side aside a defendant’s capital sentence' if - it finds that the defendant is mentally retarded, with certain exceptions. See id. § 400.27(12). The statute provides further that “ ‘mental retardation’ means significantly subaverage general intellectual functioning existing concurrently with deficits in adaptive behavior which were manifested before the age of eighteen.” Id. § 400.27(12)(e). Atkins cited this statute in support of its finding that a national consensus had developed against the execution of mentally retarded individuals. See 536 U.S. at 314 & n. 13, 122 S.Ct. 2242; see generally People v. Smith, 193 Misc.2d 462, 751 N.Y.S.2d 356, 357 (N.Y.Sup.Ct.2002). New York’s statute, however, provides little guidance as to the definition of mental retardation for three reasons. First, although the statute is still technically in force, it has been effectively rendered a nullity by the invalidation of New York’s death penalty scheme, and thus can be considered at most only a weak expression by the State of the definition of mental retardation for Atkins purposes. Second, even if the statute could inform the definition of mental retardation under Atkins, it likely would not affect the definition under the FDPA, which independently forbids the execution of mentally retarded offenders. See 18 U.S.C. § 3596(c). Third, the definition in New York’s statute is essentially identical to the clinical definitions discussed below, and neither the statute itself nor any New York case law provides guidance beyond the statute’s definitional statement. Indeed, the language of the statute tracks very closely with a 1983 definition propounded by the AAIDD (then known as the American Association on Mental Deficiency). See American Association on Mental Deficiency, Classification in Mental Retardation 1 (8th ed. 1983) (“Mental retardation refers to significantly subaverage general intellectual functioning existing concurrently with deficits in adaptive behavior, and manifested during the developmental period.”). For these reasons, the court relies largely on the clinical definitions of mental retardation promulgated by the AAIDD and the APA, the two leading authorities on the subject. These authorities were cited favorably in Atkins, and nothing in either the FDPA or New York law prevents the court from relying upon them. Most federal courts have taken the same approach when deciding Atkins cases. See, e.g., United States v. Smith, 790 F.Supp.2d 482, 485-86 (E.D.La.2011); United States v. Lewis, No. 08-CR-404 (SO), 2010 WL 5418901, at *5, *23 (N.D.Ohio Dec. 23, 2010); Hardy, 762 F.Supp.2d at 854; Davis, 611 F.Supp.2d at 474. The court emphasizes, however, that “psychology informs, but does not determinatively decide, whether an inmate is exempt from execution.” Ortiz v. United States, 664 F.3d 1151, 1168 (8th Cir.2011). Atkins “did not delegate to psychologists the determination of whether an inmate should not face execution.” United States v. Bourgeois, No. 02-CR-216, 2011 WL 1930684, at *24 (S.D.Tex. May 19, 2011); see also Hooks v. Workman, 689 F.3d 1148, 1172 (10th Cir.2012) (“Atkins could have adopted the clinical standard, but explicitly declined to do so.”); Clark v. Quarterman, 457 F.3d 441, 445 (5th Cir.2006) {Atkins “did not dictate that the approach” to defining mental retardation “must track the approach of the [AAIDD] or the APA exactly”); United States v. Candelario-Santana, No. 09-CV-427 (JAF), 916 F.Supp.2d 191, 194, 2013 WL 101615, at *2 (D.P.R. Jan. 8, 2013) (“Though the clinical standards have informed our analysis, we emphasize that a clinical standard is not a constitutional command.” (internal quotation marks omitted)). Instead, while noting the leading clinical definitions of mental retardation, Atkins expressly permitted state legislatures and courts to exercise their own judgments as to the definition of mental retardation, even if those judgments diverged from those of leading psychologists. See Atkins, 536 U.S. at 317, 122 S.Ct. 2242 (“[W]e leave to the States the task of developing appropriate ways to enforce the constitutional restriction upon their execution of sentences.” (alterations omitted)); see also id. at 317 n. 22, 122 S.Ct. 2242 • (“The statutory definitions of mental retardation ... generally conform to the clinical definitions .... ” (emphasis added)). This must logically be true as well in situations like this one in which a federal court must define the term in the absence of significant state legislative or state judicial guidance. See Bourgeois, 2011 WL 1930684, at *24 {Atkins “left the contours of the constitutional protection to the courts”). The court will thus rely heavily upon clinical definitions and expert testimony to determine the definition of mental retardation for capital punishment purposes, but, particularly where these definitions and testimony are ambiguous or conflicting (as they often are in this case), it will apply its own judgment as to the “appropriate ways” to enforce the ultimately legal prohibition on executing mentally retarded offenders. Atkins, 536 U.S. at 317, 122 S.Ct. 2242. The court must also decide whether it should rely upon current clinical definitions of mental retardation or those that were in place at the time of Atkins. Although Atkins cited the APA definition that is used today, see 536 U.S. at 308 n. 3, 122 S.Ct. 2242 (citing APA, Diagnostic and Statistical Manual of Mental Disorders 41 (4th ed. 2000) (“DSM-IV-TR”)), it cited the 1992 version of the AAIDD’s definition, see id (citing AAMR, Mental Retardation: Definition, Classification, and Systems of Supports 5 (9th ed. 1992)), which the AAIDD has since supplanted with two recent publications, see AAIDD, Intellectual Disability: Definition, Classification, and Systems of Supports (11th ed. 2010) (“AAIDD 2010 Manual”); AAIDD, User’s Guide: Intellectual Disability: Definition, Classification, and Systems of Supports (11th ed. 2012) (“AAIDD 2012 User’s Guide”). The Government argues that because “these later AAIDD materials were not contemplated by the Atkins Court” and “fall outside the scope of the ‘national consensus’ upon which the Supreme Court relied in Atkins,” the court should not adopt them as part of the AAIDD’s definition of mental retardation. (Gov’t Mem. at 18-19; see also id at 24-26.) The court disagrees. Contrary to the Government’s argument, Atkins did not conclude that there was a national consensus as to the definition of mental retardation; it suggested just the opposite. See 536 U.S. at 317, 122 S.Ct. 2242 (“To the extent there is serious disagreement about the execution of mentally retarded offenders, it is in determining which offenders are in fact retarded.”). What the Court concluded was that there was a national consensus against execution of those offenders that fit within a given state’s definition of mental retardation, while permitting the states to continue to define the contours of the definition in their own— and differing — ways. See id. The Government cannot seriously dispute that a state would be permitted to define mental retardation according to current clinical definitions as opposed to those existing at the time of Atkins. It logically follows that, in the absence of binding law to the contrary, this court is also permitted to exercise its judgment as to the best interpretation of “mental retardation,” even if that interpretation diverges from the understanding of the term at the time Atkins was decided. In any event, surely nothing in the FDPA prevents the court from doing so. It is also important to note that the Government’s approach would be very difficult (if not impossible) to apply in practice. For example, if Atkins requires the court to apply only the clinical standards in place at the time it was decided, does that mean the court prohibited from considering intelligence tests developed after Atkins? The Government apparently does not think so, because its own expert, Dr. Denney, administered an intelligence test on Wilson that was published in 2008, six years after Atkins. (See Denney Rep. at 40.) Moreover, as will be discussed in Part III, clinical judgment is essential to the interpretation of intelligence testing. The Government’s approach may require clinicians to set aside much of their training in post-Atkins psychological standards and to train themselves (for Atkins purposes alone) in the outdated standards existing in 2002. Atkins should not be read to require this result. Thus, because the AAIDD 2010 Manual reflects the AAIDD’s view of the current best practices in the field, the court will rely upon this edition. Such reliance is permissible under Atkins and the FDPA, sensible as a practical matter; and consistent with the approach of other federal courts. See, e.g., United States v. Northington, No. 07-CR-550-05, 2012 WL 4024944, at *3 (E.D.Pa. Sept. 12, 2012) (“Northington II ”); Smith, 790 F.Supp.2d at 484; Bourgeois, 2011 WL 1930684, at *23 n. 27; Lewis, 2010 WL 5418901, at *8; Hardy, 762 F.Supp.2d at 854 n. 5. B. Clinical Definitions of Mental Retardation The definitions of mental retardation set forth by the AAIDD and the APA are “essentially identical.” Davis, 611 F.Supp.2d at 475; see also Ortiz, 664 F.3d at 1158; United States v. Northington, No. 07-CR-550-05, 2012 WL 2873360, at *2 n. 6 (E.D.Pa. July 12, 2012) (“Northington I ”); Lewis, 2010 WL 5418901, at *5; see generally United States v. Nelson, 419 F.Supp.2d 891, 894-95 (E.D.La.2006) (explaining the minor differences between the definitions and noting that they “do not appear to conflict”). According to the APA, a diagnosis of mental retardation requires: A. Significantly subaverage intellectual functioning: an IQ of approximately 70 or below on an individually administered IQ test____ B. Concurrent deficits or impairments in present adaptive functioning (i.e., a person’s effectiveness in meeting the standards expected for his or her age by his or her cultural group) in at least two of the following areas: communication, self-care, home living, social/interpersonal skills, use of community resources, self-direction, functional academic skills, work, leisure, health and safety. C. The onset is before 18 years of age. DSM-IV-TR at 49. The AAIDD defines mental retardation (which it now calls “intellectual disability” or “ID”) as follows: “Intellectual disability is characterized by significant limitations both in intellectual functioning and in adaptive behavior as expressed in conceptual, social, and practical adaptive skills. This disability originates before 18.” AAIDD 2010 Manual at 1. Putting these two clinical standards together, the definition of mental retardation has three “uniformly accepted” requirements, Bourgeois, 2011 WL 1930684, at *24, which the court will at times refer to as “prongs one, two, and three.” These requirements are: (1) significantly subaverage intellectual functioning; (2) significant deficits in adaptive behavioral skills; and (3) onset of the condition before age eighteen. See AAIDD 2010 Manual at 7, 27, 41; DSM-IV-TR at 49; Taylor v. Quarterman, 498 F.3d 306, 307 (5th Cir. 2007); Northington II, 2012 WL 4024944, at *3; Davis, 611 F.Supp.2d at 475; cf. Atkins, 536 U.S. at 318, 122 S.Ct. 2242. The three prongs are each “indispensable” to a finding of mental retardation. Blue v. Thaler, No. 05-CV-2726 (H), 2010 WL 8742423, at *9 (S.D.Tex. Aug. 19, 2010), aff'd, 665 F.3d 647 (5th Cir.2011); see also AAIDD 2010 Manual at 7 (the three elements are each “essential”). The third prong — onset before the age of eighteen — -bears clarification because it is essentially a prerequisite to satisfying the first two prongs. To be mentally retarded, both an individual’s significantly subaverage intellectual functioning and his significant deficits in adaptive behavioral skills must become manifest before the age of 18. See Blue, 2010 WL 8742423, at *9 (definition requires “(1) substantial limitations in intellectual functioning; (2) significant limitations in adaptive area skills; and (3) manifestation of those limitations before age 18” (emphasis added)); see also Atkins, 536 U.S. at 318, 122 S.Ct. 2242 (“[C]linical definitions 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.” (emphasis added)). Moreover, in deciding an Atkins claim, the court must determine whether the defendant “was mentally retarded at the time of the crime.” Hardy, 762 F.Supp.2d at 881 (emphasis added); see also Holladay v. Allen, 555 F.3d 1346, 1353 (11th Cir.2009) (“Though the factors state that the problems had to have manifested themselves before the defendant reached the age of eighteen, it is implicit that the problems also existed at the time of the crime.” (internal quotation marks omitted)). Thus, mental retardation must “be diagnosed, if it is to be diagnosed at all, retrospectively in every sense of the word.” Hardy, 762 F.Supp.2d at 881. C. Additional Legal Principles As noted above, clinical definitions of mental retardation do not provide the full picture for an Atkins case. Two important general legal principles govern the court’s analysis. First, whether an individual is mentally retarded “is a question of fact, and not a mixed question of law and fact.” Clark, 457 F.3d at 444; see also Ortiz, 664 F.3d at 1164; Walker v. Kelly, 593 F.3d 319, 323 (4th Cir.2010). The standard for whether someone is mentally retarded and ineligible for the death penalty under the Eighth Amendment and FDPA is a legal matter (as discussed above), but “the ultimate issue of whether [Wilson] is, in fact, mentally retarded” is for the court to decide as a factual matter, “based upon all of the evidence and determinations of credibility.” In re Briseno, 135 S.W.3d 1, 9 (Tex.Crim.App.2004). Second, the court must decide the burden of proof for Atkins /FDPA claims. Although neither Atkins nor the FDPA addresses this issue, the vast majority of courts to address it have held that the defendant bears the burden of proof by a preponderance of the evidence. See, e.g., Northington II, 2012 WL 4024944, at *3; Smith, 790 F.Supp.2d at 484; Bourgeois, 2011 WL 1930684, at *46; Lewis, 2010 WL 5418901, at *4; Hardy, 762 F.Supp.2d at 851; Davis, 611 F.Supp.2d at 474; Thomas v. Allen, 614 F.Supp.2d 1257, 1296 (N.D.Ala.2009); Nelson, 419 F.Supp.2d at 894. But see, e.g., People v. Vasquez, 84 P.3d 1019, 1023 (Colo.2004) (upholding a Colorado statute requiring proof by the defendant by clear and convincing evidence); Head v. Hill, 277 Ga. 255, 261, 587 S.E.2d 613 (2003) (upholding a Georgia statute requiring proof by the defendant beyond a reasonable doubt). This is also the standard set forth in New York’s capital statute. See C.P.L. § 400.27(12)(a) (“[T]he defendant has the burden of proof by a preponderance of the evidence that he or she is mentally retarded.”). Both parties agree that this burden applies here (see Def. Mem. at 1; Gov’t Mem. at 49), and the court finds no compelling reason to apply a more stringent burden of proof. It therefore adheres to the majority and New York position: Wilson will have the burden of proving that he is more likely than not mentally retarded. * * * * * * In sum, both the FDPA and the Eighth Amendment forbid the court from imposing a death sentence upon a person who is mentally retarded. A person is mentally retarded only if he satisfies three necessary requirements: (1) significantly subaverage intellectual functioning; (2) significant deficits in adaptive behavioral skills; and (3) onset of those limitations before the age of 18. In developing the nuances of these requirements, the court will rely heavily upon modern clinical definitions of mental retardation, particularly those of the APA and the AAIDD. At the same time, the definition of mental retardation is ultimately a legal matter, and so the court •may — and will — exercise its own judgment as to the appropriate definition of mental retardation in the Atkins /FDPA context. Once the court expounds upon the legal definition of mental retardation, it must decide as a factual matter whether Wilson himself is mentally retarded, an issue on which Wilson bears the burden of proof by a preponderance of the evidence. III. INTELLECTUAL FUNCTIONING With these principles in mind, the court turns to the first prong of the mental retardation definition: significantly subaverage intellectual functioning. The court will begin by discussing intellectual functioning generally, with a focus on some of the especially difficult and disputed issues involved in analyzing it. (See Part III.A.) Resolution of these issues will then guide the court’s analysis of Wilson’s own intellectual functioning. (See Part III.B.) A. Intellectual Functioning in General The AAIDD has stated that “intellectual functioning is currently best conceptualized and captured by a general factor of intelligence,” and defines “intelligence” as “a general mental ability” that “includes reasoning, planning, solving problems, thinking abstractly, comprehending complex ideas, learning quickly, and learning from experience.” AAIDD 2010 Manual at 31. Intellectual functioning is primarily evaluated using standardized tests that measure a person’s “Intelligence Quotient,” or “IQ.” See id. (“Although far from perfect, intellectual functioning is currently best represented by IQ scores when they are obtained from appropriate, standardized and individually administered assessment instruments.”); Hardy, 762 F.Supp.2d at 875 (“Both the APA and AAMR/AAIDD indicate that a diagnosis of mental retardation should be made based on IQ test results where it is possible to perform such a test.”); Thomas, 614 F.Supp.2d at 1264. At the same time, the AAIDD makes clear that IQ scores themselves do not tell the whole story about someone’s intelligence; rather, “one needs to use clinical judgment” to interpret those scores and other relevant information. AAIDD 2010 Manual at 35. The most widely-accepted IQ tests in the United States are the Wechsler Intelligence Scales, which include the Wechsler Intelligence Scale for Children (“WISC”) and the Wechsler Adult Intelligence Scale (“WAIS”). Each Wechsler test is composed of several subtests, some of which evaluate a person’s “verbal” abilities and some a person’s “performance” abilities. A person’s IQ is calculated by adding together the number of points earned on all of the subtests and then using a mathematical formula to convert this raw score into an overall score, called the “full scale IQ.” In addition to the full scale IQ, the tests also produce a “verbal IQ” and a “performance IQ,” which are based solely on the subtests relating to those particular skills. The Wechsler tests are “standardized” instruments, meaning that during their design phases, they were administered to a large, representative sample of the population in order to predict the distribution of results that the general population would likely obtain. See Thomas, 614 F.Supp.2d at 1264. The results obtained by the representative sample were then analyzed for the purposes of creating scaled test scores, a process called “norming.” All of the Wechsler tests are normalized so that their average scaled score is 100. The “standard deviation” indicates how far a score is away from that average. It is the baseline against which a person’s intellectual deficits (or strengths) are measured, and can be translated into a percentile that indicates a person’s relative intelligence within the population. The Wechsler tests are normalized to have a standard deviation of 15 points. These concepts are often depicted with a bell-shaped curve. Although their definitions are worded somewhat differently, both the AAIDD and the APA define significantly subaverage intellectual functioning by reference to an IQ score approximately two standard deviations below the mean, or 70. The AAIDD defines it as “an IQ score that is approximately two standard deviations below the mean, considering the standard error of measurement for the specific assessment instruments used and the instruments’ strengths and limitations.” AAIDD 2010 Manual at 27. The APA similarly defines it as “an IQ of approximately 70 or below on an individually administered IQ test.” DSM-IV-TR at 49. These definitions are simple enough to state, but they raise several challenging issues. 1. The Standard Error of Measurement and Confidence Intervals Although both the APA and the AAIDD refer to an IQ score of approximately 70 in their definitions of significantly subaverage intellectual functioning, neither advocates the use of a fixed “cutoff score” for a finding of mental retardation, and the AAIDD explicitly advises against it. See AAIDD 2010 Manual at 40 (“A fixed point cutoff score for ID is not psyehometrieally justifiable.”). Instead, “[b]oth the APA and [AAIDD] direct that [an IQ] test’s measurement error must be taken into account when interpreting its result.” Hardy, 762 F.Supp.2d at 856; see DSM-IV-TR at 41-42; AAIDD 2010 Manual at 35. The concept of measurement error is grounded in the idea that each person has a “true” IQ score — the hypothetical score that person would obtain if no error influenced the results of the IQ test. Thomas, 614 F.Supp.2d at 1269. All IQ tests, however, contain at least some possibility of error, making it impossible to state a test subject’s “true” IQ score with certainty. Id. “An IQ score is subject to variability as a function of a number of potential sources of error, including variations in test performance, examiner’s behavior, cooperation of the test taker, and other personal and environmental factors.” AAIDD 2010 Manual at 36. The Wechsler tests take measurement errors into account through the use of a mathematical concept known as the “Standard Error of Measurement” (“SEM”). See id.; Thomas, 614 F.Supp.2d at 1270. The SEM is an index showing the variability of test scores produced by the people forming the normative sample for a particular test. Thomas, 614 F.Supp.2d at 1270. It is used to calculate a “confidence interval,” or a range of scores within which we can be confident to a certain degree that an individual’s “true” IQ lies. Id. The confidence interval refers to a percentage corresponding to our degree of confidence that an interval around the obtained IQ score contains the true IQ score. See Wiley v. Epps, 668 F.Supp.2d 848, 893-94 (N.D.Miss.2009). For example, the 95% confidence interval for a given IQ score would show the range of scores within which we can be 95% confident that a person’s true IQ score falls. This means that “if we could administer the test to that person 100 times (as if new each time), 95 times out of the 100 we would observe a score that is within those confidence bounds.” (James Rep. at 4.) The AAIDD discusses two particular confidence intervals bearing a relatively simple relationship to the SEM: (1) the 66% confidence interval is the range from one SEM below the observed score to one SEM above the observed score; and (2) the 95% confidence interval is the range from two SEMs below the observed score to two SEMs above the observed score. See AAIDD 2010 Manual at 36. For example, the most recent edition of the Wechsler test has an average SEM of 2.3 points. {See Shapiro Rep. at 8.) Thus, if an individual scored a 70 on this test, the 66% confidence interval would be from 67.7 to 72.3 (i.e., 70 + 2.3), and the 95% confidence interval would be from 65.6 to 74.6 (i.e., 70 ± 4.6). Courts and state legislatures generally have not focused on the precise SEM of particular IQ tests or on the choice of a particular percentage confidence interval. Those courts that have accounted for measurement error generally have found that the SEM for well-standardized IQ tests is approximately 5 points (without distinguishing based on the test edition), and, using a range from one SEM below to one SEM above the observed score (the 66% confidence interval), have set an upper bound of 75 for a finding of mental retardation. See, e.g., Smith, 790 F.Supp.2d at 490 (“There is [] general agreement ... that a score of 75 should be used as the upper bound of the IQ range describing mild mental retardation.”); Bourgeois, 2011 WL 1930684, at *25 (“Because IQ tests typically have a standard error of measurement (also called a ‘confidence interval’ or ‘confidence band’), a base IQ score actually represents a range that could be five points higher or lower. Thus, the psychological profession accepts 75 as a qualifying score for a diagnosis of mental retardation.”); Lewis, 2010 WL 5418901, at *8 (“Taking into consideration the SEM of 5 points on either side of 70, an IQ score for intellectual disability falls within the range of 65 to 75.”); Hardy, 762 F.Supp.2d at 857 (“All the experts in this case agree that a score of 75 should be used as the upper bound of the IQ range describing mild mental retardation. Indeed, there is almost universal agreement on this point.”); Blue, 2010 WL 8742423, at *9 (“The psychological profession [] sets 75 as the base score that may qualify for a diagnosis of mental retardation....”); Wiley, 668 F.Supp.2d at 893 (“[A]n obtained score of 70 on a Wechsler intelligence test actually represents a range of possible scores between 65 and 75.”); Davis, 611 F.Supp.2d at 475 (“[T]he SEM in IQ assessments is approximately 5 points, therefore raising the operational definition of mental retardation to 75.”); see also Atkins, 536 U.S. at 309 n. 5, 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 score for the intellectual function prong of the mental retardation definition.”). In addition to any applicable state statutes, these courts have relied upon parts of the DSM-TV-TR and previous publications by the AAIDD that appear to advise the use of a 5-point SEM and an approximate upper bound of 75. See DSM-IV-TR at 41-42 (“It should be noted that there is a measurement error of approximately 5 points in assessing IQ.... Thus, it is possible to diagnose Mental Retardation in individuals with IQs between 70 and 75 who exhibit significant deficits in adaptive behavior.”); AAMR, Mental Retardation: Definition, Classification, and Systems of Supports 59 (10th ed. 2002) (“AAMR 2002 Manual”) (noting that, “[i]n effect, [the SEM] expands the operational definition of mental retardation to 75”); see, e.g. Hardy, 762 F.Supp.2d at 856-57. The AAIDD’s more recent publications do not make reference to any approximate cutoff score of 75. See AAIDD 2010 Manual; AAIDD 2012 User’s Guide. And rightly so. The court can see no particular reason to apply a blanket SEM of 5 points to every kind of IQ test, when the precise SEMs for the various tests are readily available. See AAIDD 2010 Manual at 27 (significantly subaverage intellectual functioning is “an IQ score that is approximately two standard deviations below the mean, considering the standard error of measurement for the specific assessment instruments used ” (emphasis added)). Moreover, the courts that have used a range from 5 points below to 5 points above the observed score implicitly assumed without analysis that a 66% confidence interval is appropriate for the interpretation of IQ scores in Atkins and FDPA cases. That may be the case (and indeed, the court concludes below that it is the case), but it is certainly not self-evident. Cf. AAIDD 2012 User’s Guide at 22 (discussing both 66% and 95% confidence intervals); AAIDD 2010 Manual at 36 (same). Thus, like the AAIDD, the court will depart from the practice of using an approximate upper bound of 75 for a finding of mental retardation. It will apply the precise SEMs applicable to the IQ tests Wilson has taken, and will consider— as a matter of first impression — which percentage confidence interval is appropriate in light of modern clinical literature, the expert opinions, and the nature of these proceedings. Contrary to Wilson’s suggestion (see, e.g., Def. Mem. at 35), the court concludes that a 66% confidence interval — i.e., a range of one SEM below to one SEM above the obtained score — is appropriate in this context for three reasons. First, the court is aware of no clinical authority (outside of some of the expert opinions in this case) that has expressly advocated for the use of more than one SEM to set the confidence interval. Although the AAIDD’s most recent publications mention both a 66% and a 95% percent confidence interval, they do not clarify which one is appropriate. See AAIDD 2010 Manual at 36 (“[A] range of confidence can be established with parameters of at least one standard error of measurement (i.e., . -.. 66% probability) or parameters of two standard error[s] of measurement (i.e., ... 95% probability).”); AAIDD 2012 User’s Guide at 22 (same). And once again, the overwhelming practice prior to the AAIDD’s most recent (and inconclusive) statements on the subject has been to set a range from one SEM below to one SEM above the observed IQ score, which corresponds to using a 66% confidence interval. See, e.g., AAMR 2002 Manual at 49; DSM-IV-TR at 41-42; Smith, 790 F.Supp.2d at 490; Bourgeois, 2011 WL 1930684, at *25; Lewis, 2010 WL 5418901, at *8; Hardy, 762 F.Supp.2d at 857; Wiley, 668 F.Supp.2d at 893; Davis, 611 F.Supp.2d at 475; Thomas, 614 F.Supp.2d at 1271; see also Atkins, 536 U.S. at 309 n. 5, 122 S.Ct. 2242. (See also Tr. at 623 (Dr. Olley’s testimony).) In other words, it appears that the use of a range from one SEM below to one SEM above the observed score remains the best practice in both the psychological and legal communities. Absent any definitive contradictory statement by the AAIDD, the court has no particular reason to depart from this practice (only reasons to the contrary, as explained below). Second, and on a similar note, the application of a 95% confidence interval would permit diagnoses of mental retardation well above what has previously been considered the approximate upper bound for a finding of subaverage intellectual functioning. As the AAIDD notes, SEMs for IQ scores typically range from “approximately 3 to 5 points.” If the court were to apply a 95% confidence interval to an IQ test with an SEM of 5, then a person could conceivably be diagnosed with mental retardation if his observed IQ score were 80 — i.e., two SEMs above 70. So far as the court is aware, no court or clinician has made a finding of mental retardation based on such a high IQ score, and neither the AAIDD nor the APA has ever suggested that such an IQ score would be an indication of significantly subaverage intellectual functioning. Third, the court considers the use of a 95% percent confidence interval to be particularly inappropriate in the context of an Atkins claim. In the clinical context, there may be good reason to resolve any doubts in favor of a finding of mental retardation and — as a consequence — the provision of treatment for the disability; in Atkins cases, by contrast, the law places the burden on the defendant to prove his mental retardation by a preponderance of the evidence. Cf. Bourgeois, 2011 WL 1930684, at *26 (“[W]ith the main purpose of most IQ testing being to qualify an individual for needed supports or services, the mental health community has little reason to question the results of IQ testing. Courts, however, endeavor to determine [as a matter of fact] whether a borderline score represents an intelligence capacity above or below the mental-retardation threshold.”). Wilson’s contention — that the court must ascertain the range of scores in which we are 95% confident that his true score falls — is in serious tension with his burden of showing that he is more likely than not mentally retarded. A final point on confidence intervals. Although the AAIDD maintains that there is no fixed cutoff score for a finding of mental retardation and defines significantly subaverage intellectual functioning as an IQ score of “approximately” 70, these statements are best read simply to mean that it is important to account for measurement error. {See Tr. at 1906 (Dr. Denney: “[M]y judgment is that [‘approximately’] means we’re talking about standard error.”).) See also AAIDD 2010 Manual at 40 (“AAIDD (just as the [APA]) does not intend for a fixed cutoff score to be established for making a diagnosis of ID. Both systems (AAIDD and APA) require clinical judgment regarding how to interpret possible measurement error.” (emphasis added)); id. at 27 (defining subaverage intellectual functioning as “an IQ score that is approximately two standard deviations below the mean, considering the standard error of measurement for the specific assessment instruments used and the instruments’ strengths and weaknesses” (emphasis added)). But if the bottom of the confidence interval for a well-administered IQ test is above a 70 — that is, after adjusting for measurement error — this fact would be strongly indicative that the test-taker is not mentally retarded. For example, if a defendant scores a 71 on the latest Wechsler test, his 66% confidence interval would be from 68.4 to 73.6; such a score may or may not (depending on clinical judgment) be deemed indicative of mental retardation. But if a defendant’s 66% confidence interval ranges from, say, 71.4 to 76.6 (the result of a score of 74 on that test), then we would be at least 66% confident that his true score is higher than two standard deviations below the mean, strongly suggesting that he is not retarded. This principle is consistent with the approaches of the APA, see DSM-IV-TR at 41 (“[TJhere is a measurement error of approximately 5 points in assessing IQ.... Thus, it is possible to diagnose Mental Retardation in individuals with IQs between 70 and 75 .... ” (emphasis added)), and those federal courts .that have interpreted IQ scores using an SEM of 5 and an upper bound of 75, see, e.g., Thomas, 614 F.Supp.2d at 1281 (“A court must [ ] consider ... the standard error of measurement in determining whether a petitioner’s IQ score falls within a range containing scores that are less than 70.”); Smith, 790 F.Supp.2d at 490; Bourgeois, 2011 WL 1930684, at *25; Hardy, 762 F.Supp.2d at 857; Blue, 2010 WL 8742423, at *9; Davis, 611 F.Supp.2d at 475; see also Atkins, 536 U.S. at 309 n. 5, 122 S.Ct. 2242. ■ In sum, when interpreting Wilson’s IQ scores in light of their inherent potential to be affected by measurement error, the court will: (1) apply the particular SEMs for the different IQ tests Wilson has taken; and (2) use a 66% confidence interval — the range from one SEM below to one SEM above the observed score. An IQ score for which the 66% confidence interval encompasses scores of 70 or below may be indicative of mental retardation depending on clinical judgment, but an IQ score with a 66% confidence interval beginning above 70 will be considered strong evidence against a finding of mental retardation. 2. The Flynn Effect A second important issue is whether the court should take the so-called “Flynn Effect” into account when considering Wilson’s IQ scores. The court concludes that it should. The Flynn Effect is a phenomenon discussed in a series of widely-cited papers by James R. Flynn. See, e.g., James R. Flynn, Tethering the Elephant: Capital Cases, IQ, and the Flynn Effect, 12 Pysch., Pub. Policy, & L. 170 (2006). Flynn’s studies show that in almost all nations in the developed world, there is an upward trend in performance on a given IQ test after the test is normed. See Davis, 611 F.Supp.2d at 485. In other words, populations as a whole will do better on an IQ test as time passes after the norming of the test. Id. “The cause of this increase is largely unknown, although some speculate that improved socioeconomics, education and even better nutrition have increased the scores, that the test[s] themselves have become more sophisticated, or that perhaps people are simply getting ‘smarter.’ ” Hardy, 762 F.Supp.2d at 858; see also AAIDD 2010 Manual at 37. The amount of increase varies depending on the test, but it is approximately 3 IQ points per decade, or 0.33 points per year. Davis, 611 F.Supp.2d at 485-86. What this means is that over time, the norms for an IQ test become outdated. Id. at 486. Because the population has improved (or has gotten “smarter”) since the time the test was normed, the average score for the population would be above 100 if the test were given to a representative sample today. Hardy, 762 F.Supp.2d at 858. This also means that a score two standard deviations below the average— the approximate score for a finding of mental retardation — will be higher than 70. Davis, 611 F.Supp.2d at 486. In other words, although mentally retarded people are getting “smarter” along with the rest of the population, they remain mentally retarded because of their level of intelligence relative to the average member of the population (two standard deviations away). For example, someone who receives a score of 72 on a test that was normed a decade ago would be expected, on average, to score a 69 on a newly-normed test. Hardy, 762 F.Supp.2d at 858. His intellectual functioning is still two standard deviations below the mean in spite of his above-70 score on the outdated test. “Professionals in the field do not disagree that the [Flynn Effect] exists, but rather, there is professional disagreement regarding whether to adjust an individual’s score” to correct for the Flynn Effect. Wiley, 668 F.Supp.2d at 894; see also Thomas v. Allen, 607 F.3d 749, 753 (11th Cir.2010) (“The parties in this case agree that the Flynn effect is an empirically proven statistical fact; however, they disagree on the extent to which an individual test subject’s IQ score should be adjusted to take into consideration this phenomenon.”). Generally, adjustments for the Flynn Effect are made by lowering the individual’s IQ score by 0.33 points for each year after the test was normed, before accounting for the SEM. See Lewis, 2010 WL 5418901, at *8; AAIDD 2010 Manual at 37. In Flynn’s view, adjusting IQ scores on the basis of the Flynn Effect is necessary because, “[o]therwise, one person will meet the criterion of mental retardation, and another person will be judged not to have done so, purely because one took a test with current norms and the other took a test with obsolete norms.” Flynn, supra, at 176. The AAIDD recommends adjusting for the Flynn Effect for similar reasons. See AAIDD 2010 Manual at 37 (stating that “best practices require recognition of a potential Flynn Effect when older editions of an intelligence test (with corresponding older norms) are used in the assessment or interpretation of an IQ score,” and that “[i]n cases where a test with aging norms is used, a correction for the age of norms is warranted”); AAIDD 2012 User’s Guide at 23. Experts in this case from both sides have recognized the existence of the phenomenon. (See, e.g., Tr. at 1959, 1998-99, 2001 (Dr. Denney’s testimony); Shapiro Rep. at 7; James Rep. at 11; Woods Rep. at 5.) And among courts, “[w]hile support for the use of the Flynn effect to adjust IQ scores ... may not be universal, it is widespread.” Davis, 611 F.Supp.2d at 486; see, e.g., Walker v. True, 399 F.3d 315, 322 (4th Cir.2005) (criticizing a district court for refusing to consider the Flynn Effect and directing the court to consider it on remand); Smith, 790 F.Supp.2d at 491 (“The Court [ ] finds the Flynn Effect should be applied to the [IQ] scores.”); Lewis, 2010 WL 5418901, at *11 (“[T]he court recognizes the Flynn Effect as a best practice for an intellectual disability determination.”); Hardy, 762 F.Supp.2d at 862-63 (“In light of the substantial evidence supporting the existence of the Flynn Effect, the Court concludes that [defendant’s] score of 73 should be corrected to take it into account.”); Wiley, 668 F.Supp.2d at 894 (accounting for the Flynn Effect); Davis, 611 F.Supp.2d at 488 (“[T]he Court finds the defendant’s Flynn effect evidence both relevant and persuasive, and will, as it should, consider the Flynn-adjusted scores in its evaluation of the defendant’s intellectual functioning.”); Thomas, 614 F.Supp.2d at 1278 (“It [] is undisputed that Professor Flynn’s recommendation ... is a generally accepted adjustment.”); Green v. Johnson, No. 05-CV-3540, 2006 WL 3746138, at *45 (E.D.Va. Dec. 15, 2006) (“[T]he Flynn Effect is recognized throughout the profession.”). Despite this substantial authority, and without actually contesting the validity of the Flynn Effect, the Government argues that the court should not adjust Wilson’s scores based on the Flynn Effect for two reasons. First, the Government argues that the Flynn Effect was not widely recognized until well after Atkins, and thus its acceptance by the AAIDD and other clinical publications should not be deemed “part of the national consensus against executing the mentally retarded.” (Gov’t Mem. at 28-29.) The court has already rejected the argument that it is precluded from considering post-Atkins developments in the clinical understanding of mental retardation, and the use of the Flynn Effect is one such development. (See Part II.A.) Second, the Government argues that the court should not adjust Wilson’s scores based on the Flynn Effect because such adjustments are not typically made in standard clinical practice (as opposed to the forensic context). (See Gov’t Mem. at 26.) See also Candelario-Santana, 916 F.Supp.2d at 208, 2013 WL 101615, at *15 (“[T]he government’s experts could not point to a single instance in their professional experience where they applied, or could recall a colleague’s application of, the Flynn Effect.”); Ledford v. Head, No. 02-CV-1615 (JEC), 2008 WL 754486, at *7 (N.D.Ga. Mar. 19, 2008) (“There was testimony at the hearing that the Flynn effect is a generally recognized phenomenon, but experts for both petitioner and respondent agreed that it is not used in clinical practice to reduce IQ scores.”). “That may be so, but it does not justify ignoring the phenomenon in the face of its unchallenged existence.” Thomas, 614 F.Supp.2d at 1277. The court’s duty is to determine as best it .can whether Wilson has significantly subaverage intellectual functioning — an IQ approximately two standard deviations below the mean. Attempting to do so without adjusting for indisputably outdated norms seems senseless. 3. The Practice Effect In addition to the Flynn Effect, the AAIDD recommends considering a phenomenon called the “practice effect.” AAIDD 2010 Manual at 35, 38, 102. The court will do so, but not in the manner that Wilson requests. The AAIDD describes the practice effect as follows: The practice effect refers to gains in IQ scores on tests of intelligence that result from a person being retested on the same instrument.... For example, the [Wechsler Adult Intelligence Scale-Third Edition] Manual presents data showing the artificial increase in IQ scores when the same instrument is re-administered within a short time interval. The [ ] Manual also reports average increases between administrations with intervals of 2 to 12 weeks. For this reason, established clinical practice is to avoid administering the same intelligence test within the same year to the same individual because it will often lead to an overestimate of the examinee’s true intelligence. Id. at 38 (citation and italics omitted); see also AAIDD User’s Guide at 23. The theory behind the practice effect “is that because IQ assessments rely upon novel tasks and instructions to assess ability and performance, an instruction given on a test will be more familiar to the examinee and more quickly implemented on subsequent presentations.” Wiley, 668 F.Supp.2d at 896. The effects are normally greater on performance items than on verbal items. (See James Rep. at 12 (“Performance scales are more susceptible to practice effects because the tasks rely, in part, on the novelty of items and familiarity with those items takes away the novelty, improving scores.”).) Wilson argues that “the Court should apply at least a 5 to 8 point adjustment to the more recent full scale IQ scores similar to the Flynn Effect.” (Def. Mem. at 32.) He apparently bases this argument entirely on the views — relied upon in Dr. James’s report (see James Rep. at 12) — of Alan. S. Kaufman, a psychology professor who did not testify at the Atkins hearing (see Def. Mem. at 32). According to Dr. Kaufman: Clinicians should understand the average practice effect gains in intelligence scores for children, adolescents, and adults. The expected increase of about 5 to 8 points in global IQ renders any score obtained on a retest as a likely overestimate of the person’s true level of functioning — especially if the retest is given within about six months of the original test, or if the person has been administered a Wechsler scale (any Wechsler scale) several times in the course of a few years. Alan. S. Kaufman, Practice Effects, in Robert J. Sternberg, 2 Encyclopedia of Human Intelligence 828 (1994). As an initial matter, the court is reluctant to place significant reliance upon Dr. Kaufman’s views. Because he did not testify at the Atkins hearing, he has not been qualified as an expert in this case, the Government has not had the opportunity to cross-examine him, and the court has not had the opportunity to judge his credibility. In any event, neither Dr. Kaufman’s article nor any other psychological literature provided by Wilson supports the application of a blanket 5 to 8 point adjustment for all IQ retests. Unlike with the Flynn Effect, there does not appear to be an accepted method in the psychological community for adjusting IQ scores to account for the practice effect. (See Tr. at 626 (Dr. Olley’s testimony that “the Practice Effect is known to inflate scores, but there’s no formula that says if you take this test, this often, this much time in between tests, that it will inflate the score by this number of points”).) See also Green, 2006 WL 3746138, at *44 (“There was no consensus among the experts on the degree to which the practice effect would effect a test score, especially for someone with low intellectual functioning.”). Contrary to Wilson’s suggestion, Dr. Kaufman does not recommend adjusting an individual’s IQ scores by 5 to 8 points for all retests; he recommends only that “[c]linicians should understand the average practice effect gains.” Id. (emphasis added). The AAIDD also does not recommend any particular adjustment of scores to account for the practice effect (as it- does with the Flynn Effect); it recommends only that clinicians “avoid administering the same intelligence test within the same year.” AAIDD 2010 Manual at 38 (emphasis added). Indeed, Wilson’s own expert, Dr. Olley, suggested that although IQ scores should be “interpreted” based on the practice effect, they should not be adjusted — or “reduced” on this basis. (Tr. at 625-27.). See also Green, 2006 WL 3746138, at *44 (refusing to adjust a petitioner’s score to account for the practice effect because “a conclusion that the practice effect increased [petitioner’s] test scores by a certain number of points would be purely speculative.”). It is unsurprising that Wilson’s proposed adjustment lacks support because the practice effect is heavily dependent upon the length of time between the original test and the retest. Dr. Denney, for example, testified that there is an important difference between a retest within “a short period” and a retest after a period of more than a year, and that practice effects tend to “fall away” after seven years. (Tr. at 1920.) Cf. Kaufman, supra, at 828 (practice effect overestimates a person’s intellectual functioning “especially if the retest is given within about six months of the original test, or ... several times in the course of a few years”). A number of courts have also recognized this principle. See, e.g., Blue, 2010 WL 8742423, at *13 (“[T]he practice effect only applies when there is a short interval between tests. The nine-month period here should have dispelled any lingering effect from the first test.”); Garcia Briseno, 2007 WL 998743, at *8 (“[I]n a two-to-twelve week period of retesting, full-scale IQ can be as much as five points higher. For performance IQ scale practice effects will be minimized after between a year to two years.”); Green, 2006 WL 3746138, at *44 (“The practice effect refers to an increase in a person’s score on an IQ test when it is administered within a short time after taking the same or [a] similar teste ... [T]he effect is more pronounced the closer in time the tests are given.” (emphasis added)); Bowling v. Commonwealth, 377 S.W.3d 529, 539 (Ky.2012) (the practice effect “refers, to only relatively short term periods between tests”). To be sure, experts in this case have suggested that the practice effect could occur after even a very long interval between- administrations. (See, e.g., Tr. at 1214-1215 (Dr. James: “[Tjhere’s research that shows practice effects can happen after a few weeks, they can happen after a few years, they can happen as much as 13 years on from the original test.... They can occur at any interval.”); id. at 1920-21 (Dr. Denney’s testimony that practice effects could “[potentially” affect scores seven years after the first test and, in “rare instances,” twelve to thirteen years later).) But no expert has suggested that, after a very long interval, the practice effect would still be expected to increase a person’s test scores by an average of 5 to 8 points, or for that matter by any particular number of points. Wilson has provided the court with no persuasive guidance as to the proper adjustment (if any) for IQ tests taken a substantial amount of time after the original test, and the evidence suggests that any such adjustment should be minimal.. To summarize, the court will — as the AAIDD recommends — take into account the practice effect in interpreting Wilson’s IQ scores. See AAIDD 2010 Manual at 35, 38, 102. But the court will not apply any particular point adjustment to his IQ scores on the basis of the practice effect, as it finds no basis for doing so in the psychological literature or case law. Cf Green, 2006 WL 3746138, at *44. In other words, th