Littlejohn v. Trammell, 704 F.3d 817

Lead Opinion

HOLMES, Circuit Judge.

Emmanuel Littlejohn was convicted of two robbery-related charges and a charge of first-degree murder, arising from his role in a 1992 robbery of a Root-N-Scoot convenience store in Oklahoma City. He received extended prison sentences on the robbery charges and a death sentence on the murder charge.

After a long procedural journey through the Oklahoma courts, Mr. Littlejohn filed a Petition for a Writ of Habeas Corpus, seeking relief under 28 U.S.C. § 2254, claiming (relevantly) that his murder conviction and death sentence were obtained in violation of his constitutional rights. The district court denied all relief and Mr. Littlejohn now appeals on multiple grounds. We affirm the district court’s judgment on all grounds except for Mr. Littlejohn’s claims of ineffective assistance of counsel at the penalty phase and cumulative error. As to the ineffective-assistance claim, we reverse the judgment and remand the case to the district court, with directions to conduct an evidentiary hearing and any further appropriate proceedings consistent with this opinion. Additionally, because the resolution of Mr. Littlejohn’s cumulative-error claim may be affected by the district court’s determination of his ineffective-assistance claim, we decline to address the merits of Mr. Little-john’s contentions concerning cumulative error. Instead, we direct the district court to vacate that portion of its judgment upon remand and to consider the cumulative-error claim afresh.

I. Background and Procedural History

The facts are largely undisputed. On June 19, 1992, Mr. Littlejohn and Glenn Bethany robbed a Root-N-Scoot convenience store in Oklahoma City. At the time of the robbery, three individuals were working at the store, one of whom was Kenneth Meers. As the robbery was wrapping up, and Mr. Littlejohn was leaving the store, a shot was fired. The shot struck Mr. Meers in the face, ending his life. The evidence was conflicting concerning the source of the shot, and Mr. Little-john maintained that he did not fire it.

In November 1994, Mr. Littlejohn was charged and convicted by a jury of robbery with a firearm, after conviction of two or more felonies (Count One); murder in the first degree (Count Two);¹ and conspiracy to commit robbery with firearms, after conviction of two or more felonies (Count Three). At sentencing, the jury found three aggravating circumstances under Oklahoma law: (1) that Mr. Littlejohn had been previously convicted of violent felonies; (2) that he knowingly created a great risk of death to more than one person; and (3) that he posed a “continuing threat” to society. See Okla. Stat. tit. 21 § 701.12. His punishment was originally set at 300 and 99 years’ imprisonment on Counts One and Three, respectively, and death on Count Two. Mr. Littlejohn appealed his conviction, and while the appeal was pending, the Supreme Court decided Cooper v. Oklahoma, 517 U.S. 348, 355-56, 368-69, 116 S.Ct. 1373, 134 L.Ed.2d 498 (1996), which found unconstitutional Oklahoma’s rule requiring a defendant bear the burden of proof by clear and convincing evidence that he is incompetent to stand trial. The Oklahoma Court of Criminal Appeals (“OCCA”), in light of Cooper and the fact that Mr. Littlejohn had previously challenged his competency to stand trial, ordered the state trial court, if feasible, to “conduct a retrospective competency hearing utilizing the preponderance of the evidence standard.” R., Vol. 1, pt. II, at 207 (Dist. Ct. Mem. Op., filed May 27, 2010). A jury later found that Mr. Littlejohn had not proven by a preponderance of the evidence that he was incompetent to stand trial in 1994. Mr. Littlejohn then filed another appeal, raising averments of error as to the competency hearings, the guilt phase of trial, and sentencing.

The OCCA denied Mr. Littlejohn relief on his claims of error relating to the retrospective competency hearings and the guilt phase. See Littlejohn v. State (Littlejohn I), 989 P.2d 901, 903-10 (Okla.Crim.App.1998). However, the court found error in the imposition of the death sentence. See id. at 910-12. Specifically, it held that admission of testimony suggesting that Mr. Littlejohn confessed to not only killing Mr. Meers, but also to an unrelated murder, amounted to constitutional error because the confession was uncorroborated by competent evidence, harmfully contributing to the jury’s finding of the continuing-threat aggravator. See id. at 910-11. Moreover, it concluded that there was insufficient evidence that Mr. Littlejohn created a “great risk of death” to more than one person, see Okla. Stat. tit. 21 § 701.12(2), further undermining the second of the three aggravating circumstances that led to his death sentence, see Littlejohn I, 989 P.2d at 911-12. The court remanded the matter for resentenc-ing in light of the fact that “[o]ver ninety percent of the aggravating evidence ... in th[e] case was ... [still] admissible.” Id. at 912.

Mr. Littlejohn’s resentencing trial began on October 30, 2000. He again received a death sentence, based this time on two aggravating circumstances — a previous conviction of a felony involving the use or threat of violence to the person and the continuing-threat aggravator. R., Vol. 1, pt. II, at 208; see also Littlejohn v. State (Littlejohn II), 85 P.3d 287, 290-91 (Okla.Crim.App.2004). He appealed, but the OCCA affirmed his sentence. See Little-john II, 85 P.3d at 290-91, 303. The Supreme Court denied Mr. Littlejohn’s petition for certiorari in October of 2004. See Littlejohn v. Oklahoma, 543 U.S. 947, 125 S.Ct. 358, 160 L.Ed.2d 261 (2004). Mr. Littlejohn filed an application for post-conviction relief in Oklahoma state court, but was unsuccessful.

Mr. Littlejohn then sought federal habe-as relief on February 25, 2005. He raised fourteen claims. See R., Vol. 1, pt. I, at 10-12, 38-150 (Pet. for Writ of Habeas Corpus, filed Sept. 30, 2005). The district court considered his petition, and denied relief on all grounds on May 27, 2010. However, it granted a certificate of appeal-ability (“COA”) to appeal its decision on six claims:

1. The prosecution violated Mr. Little-john’s due process rights by presenting inconsistent theories as to who fired the fatal shot;

2. The prosecution failed to provide adequate notice of certain aggravation evidence;

3. The prosecution improperly presented a transcript of two witnesses who testified at the 1994 trial in absten-tia in violation of Mr. Littlejohn’s confrontation rights;

4. The prosecution engaged in misconduct in violation of Mr. Littlejohn’s constitutional rights;

5. Mr. Littlejohn was prejudiced by counsel’s failure to investigate and present evidence that he has brain damage;² and

6. Cumulative error.

See id., pt. II, at 301 (COA, filed July 6, 2010).

Mr. Littlejohn now appeals, raising eight issues, including the six for which the district court granted the COA. In addition to the six claims, he challenges the district court’s decision to deny habeas relief on his claim that the sentencing judge improperly instructed (and misled) the jury on the meaning of “life without the possibility of parole” as an alternative to the death penalty under Oklahoma law. Further, he contends that the district court wrongly rejected his claim that his rights were violated by virtue of the trial court’s decision to “allow[ ] the jury to hear [his] previously recorded testimony” from the 1994 trial. Aplee. Br. at 3. We granted COAs on these two issues. Consequently, we have jurisdiction to consider the entirety of Mr. Littlejohn’s habeas appeal. See Allen v. Zavaras, 568 F.3d 1197, 1199 (10th Cir.2009).

II. Discussion

The Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”) circumscribes our review of claims adjudicated on the merits in state court proceedings. Under AEDPA, a petitioner is entitled to federal habeas relief on a claim only if he can establish that the state court’s adjudication of the claim on the merits (1) “resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law”; or (2) “resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d)(1), (2). The AEDPA standard is “highly deferential ... [and] demands that state-court decisions be given the benefit of the doubt.” Woodford v. Visciotti, 537 U.S. 19, 24, 123 S.Ct. 357, 154 L.Ed.2d 279 (2002) (per curiam) (quoting Lindh v. Murphy, 521 U.S. 320, 333 n. 7, 117 S.Ct. 2059, 138 L.Ed.2d 481 (1997)) (internal quotation marks omitted).

In applying the legal inquiry under § 2254(d)(1), we ask at the threshold “whether there exists clearly established federal law, an inquiry that focuses exclusively on holdings of the Supreme Court.” Hooks v. Workman (Victor Hooks), 689 F.3d 1148, 1163 (10th Cir.2012). “The absence of clearly established federal law is dispositive under § 2254(d)(1).” House v. Hatch, 527 F.3d 1010,1018 (10th Cir.2008). And, in ascertaining the contours of clearly established federal law, we must look to “the holdings, as opposed to the dicta, of [the Supreme] Court’s decisions as of the time of the relevant state-court decision.” Yarborough v. Alvarado, 541 U.S. 652, 660-61, 124 S.Ct. 2140, 158 L.Ed.2d 938 (2004) (emphasis added) (quoting Williams v. Taylor, 529 U.S. 362, 412, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000)) (internal quotation marks omitted).

If clearly established federal law exists, a state court decision is contrary to it only if the court “contradicts the governing law” or “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 result reached by the Supreme Court.” Bland v. Sirmons, 459 F.3d 999, 1009 (10th Cir.2006) (alteration in original) (quoting Williams, 529 U.S. at 405-06, 120 S.Ct. 1495) (internal quotation marks omitted). A state court decision unreasonably applies federal law if it “identifies the correct governing legal principle from [Supreme Court] decisions but unreasonably applies that principle to the facts of the prisoner’s case.” Id. (alteration in original) (quoting Williams, 529 U.S. at 413, 120 S.Ct. 1495) (internal quotation marks omitted); accord Wiggins v. Smith, 539 U.S. 510, 520-21, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003). “We review the district court’s legal analysis of the state court decision de novo.” Welch v. Workman, 639 F.3d 980, 991 (10th Cir.2011) (quoting Bland, 459 F.3d at 1009) (internal quotation marks omitted).

As a condition for obtaining habeas corpus from a federal court, a state prisoner must show that the state court’s ruling on the claim being presented in federal court was so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fairmind-ed disagreement.

Harrington v. Richter, — U.S.-, 131 S.Ct. 770, 786-87, 178 L.Ed.2d 624 (2011).

Furthermore, in reviewing a state court decision under § 2254(d)(1), we must “limit[]” our inquiry “to the record that was before the state court that adjudicated the claim on the merits.” Cullen v. Pinholster, — U.S. -, 131 S.Ct. 1388, 1398, 179 L.Ed.2d 557 (2011). Factual findings of the state court are presumed correct unless the applicant rebuts that presumption by “clear and convincing evidence.” 28 U.S.C. § 2254(e)(1); accord Welch, 639 F.3d at 991.

“For federal habeas claims not adjudicated on the merits in state-court proceedings, we exercise our ‘independent judgment’ and ‘review the federal district court’s conclusions of law de novo.’ ” Victor Hooks, 689 F.3d at 1163-64 (quoting McCracken v. Gibson, 268 F.3d 970, 975 (10th Cir.2001)); see Sallahdin v. Gibson, 275 F.3d 1211, 1222 (10th Cir.2002). “The district court’s factual determinations are reviewed for clear error.” Victor Hooks, 689 F.3d at 1164. But “[a ]ny state-court findings of fact that bear upon the claim are entitled to a presumption of correctness rebuttable only by ‘clear and convincing evidence.’ ” Id. (emphases added) (quoting 28 U.S.C. § 2254(e)(1)); see Hooks v. Ward, 184 F.3d 1206, 1223 (10th Cir.1999).

Apart from the ineffective-assistance-of-counsel and cumulative-error issues, see infra Parts II.G, II.H, we address Mr. Littlejohn’s challenges on appeal substantially in the order the issues are addressed by the parties in the briefs. At the end, we deny relief on all grounds except for the ineffective-assistance and cumulative-error claims. As to the former (the ineffective-assistance claim), we reverse the judgment and remand the case for further proceedings, most notably an evidentiary hearing. As to the latter (the cumulative-error claim), we decline to address the merits and direct the district court to revisit it upon remand, following its readjudi-cation of Mr. Littlejohn’s ineffective-assistance claim.

A. Instructions to the Jury

Mr. Littlejohn first challenges the constitutionality of the trial court’s instructions to the jury during the second stage. At resentencing, when the jury was considering its sentence, it submitted a note to the trial court asking, “[I]s it possible to change the verdict of life without parole to with parole after our verdict and without another jury verdict [by anyone]?” State R., Vol. VII, Resentencing Tr. at 358. The court seemingly attempted to refer the jury back to the original instructions, which defined the three available sentencing options under Oklahoma law: “death, imprisonment for life without parole, or imprisonment for life.” R., Vol. 1, pt. II, at 227 (quoting Original R., Vol. X, at 1875 (Jury Instructions, given Nov. 7, 2000)) (internal quotation marks omitted). Specifically, the court conveyed to the jury that they “have all the law and evidence necessary to reach a verdict.” Id. (quoting State R., Vol. VII, Resentencing Tr. at 364) (internal quotation marks omitted). It further rejected a request by defense counsel to elaborate on the actual meaning of the three sentencing alternatives.

Mr. Littlejohn raised an averment of error on direct appeal to the OCCA, arguing primarily that the trial court’s response was insufficient and unconstitutional in the context of this case “when considered in conjunction with [its] oral instructions concerning jury questions.” Littlejohn II, 85 P.3d at 291. Specifically, before the jury retired to deliberate, the trial court told the jury that it would answer questions “if it’s appropriate to answer,” and that if, in response, the jury received a “code back that says, you have all the law and evidence necessary to reach a verdict, what that means is the answer to [the] question is in the instructions, it was in the evidence, or you’re asking me something that’s inappropriate for me to answer.” Id. at 291 (emphases added) (quoting relevant portions of the State record) (internal quotation marks omitted). The OCCA rejected Mr. Little-john’s claim, see id. at 291-92, but nonetheless set guidance for “future cases where the jury during deliberations asks ... whether an offender ... is parole eligible”: specifically, “[trial courts] should either refer the jury back to the instructions, ... tell the jury that the punishment options are self explanatory, ... or advise the jury that the punishment options are to be understood in their plain and literal sense.” Id. at 293-94 (citations omitted).

Mr. Littlejohn again raised this argument in his habeas petition, contending in pertinent part that the OCCA’s decision was “contrary to or an unreasonable application of’ the Supreme Court’s decisions in Simmons v. South Carolina, 512 U.S. 154, 114 S.Ct. 2187, 129 L.Ed.2d 133 (1994) (plurality), Shafer v. South Carolina, 532 U.S. 36, 121 S.Ct. 1263, 149 L.Ed.2d 178 (2001), and Kelly v. South Carolina, 534 U.S. 246, 122 S.Ct. 726, 151 L.Ed.2d 670 (2002), because, under the principles embodied in those cases, the “jury [here] did not have conveyed to them and did not understand [in context] the ‘throw away the key' implications of the most attractive alternative to the death penalty.” Aplt. Opening Br. at 32. The district court, relying in part on our decision applying Simmons in Mollett v. Mullin, 348 F.3d 902 (10th Cir.2003), found that the trial court’s response to the jury’s inquiry did not create an inappropriate false choice, and therefore denied Mr. Littlejohn relief.

Mr. Littlejohn raises the same argument on appeal. He identifies numerous Supreme Court cases that stand for the general proposition that adequate juror comprehension of the trial court’s instructions in capital sentencing is vital to a defendant’s constitutional rights. See, e.g., Boyde v. California, 494 U.S. 370, 110 S.Ct. 1190, 108 L.Ed.2d 316 (1990). Most importantly, however, he points out that the OCCA’s decision runs against the holdings in Simmons, Kelly, and Shafer insofar as it ignores the requirement that “jurors ... know if they have a sentencing option that will keep a convicted murderer from ever being paroled to the streets,” and that courts are prevented from “creating a false or unbalanced impression” that the defendant, if not executed, will be eligible for release. Aplt. Opening Br. at 20 (emphasis added).

In Simmons, the Supreme Court held that “due process require[s] a state trial court ‘to instruct the jury in the penalty phase of a capital trial that under state law the defendant [i]s ineligible for parole.’ ” Hamilton v. Mullin, 436 F.3d 1181, 1190-91 (10th Cir.2006) (quoting Simmons, 512 U.S. at 156, 114 S.Ct. 2187). In particular, “whenever future dangerousness is at issue in a capital sentencing proceeding ... due process requires that the jury be informed that a life sentence carries no possibility of parole” if in fact that is the case under state law. Shafer, 532 U.S. at 51, 121 S.Ct. 1263. Put another way, “Simmons rests upon eliminating a jury’s misunderstanding so the jury will not perceive a ‘false choice’ between sentencing to death or a limited period of incarceration when future dangerousness is at issue.” Johnson v. Gibson, 254 F.3d 1155, 1166 (10th Cir.2001).

Oklahoma uses a “three-option sentencing scheme,” which permits the imposition of three different, potential punishments— (1) death, (2) life imprisonment without the possibility of parole, or (3) life imprisonment. See Hamilton, 436 F.3d at 1191. We have held that this scheme is “consistent with the Supreme Court’s rulings since the options do not create a false choice between sentencing petitioner to death and sentencing him to a limited period of incarceration.” Id. (quoting Mayes v. Gibson, 210 F.3d 1284, 1294 (10th Cir.2000)) (internal quotation marks omitted); see Welch, 639 F.3d at 1005 (noting that a basic instruction on Oklahoma’s “three-option” scheme is “constitutionally adequate”). Thus, in circumstances where “the trial court simply directs the jury to review the instructions again” — instructions that clearly set forth the three-option scheme — “the defendant’s due process rights are not violated.” Welch, 639 F.3d at 1005 (citing, inter alia, McCracken, 268 F.3d at 980-81).

However, in further expounding upon Simmons and its offspring, we have held that a due process violation can be created by the trial court, which in some instances may engender juror confusion, thereby creating a “false choice,” even in light of instructions that may correctly state the law. See Mollett, 348 F.3d at 913. In order to be entitled to relief in such a case under the reasoning in Mollet — viz., where the jury demonstrates confusion as to the meaning of possible sentences, including death'—-a petitioner must make four showings:

(1) the prosecution [sought] the death penalty; (2) the prosecution place[d] the defendant’s future dangerousness at issue; (3) the jury ask[ed] for clarification of the meaning of life imprisonment, or a synonymous statutory term; and (4) the judge’s response threaten[ed] to cause a jury’s misunderstanding so the jury [could] perceive a false choice of incarceration when future dangerousness [wa]s at issue.

Hamilton, 436 F.3d at 1191 (quoting Mol-lett, 348 F.3d at 914) (internal quotation marks omitted);³ see Mollett, 348 F.3d at 913-15.

In this case, the district court found, and the parties agree, that factors one through three are satisfied. We are content to proceed assuming as much, without independently examining the merits of each factor. See United States v. McGehee, 672 F.3d 860, 867 (10th Cir.2012). The fourth Mollett factor is at issue. Mr. Littlejohn contends that what was actually conveyed to the jury—both in the trial judge’s answer and in light of his earlier statements regarding what a “code” response would mean—created jury confusion that amounted to the creation of a false choice and a violation of due process. This, according to Mr. Littlejohn, is compounded by the fact, as explained by the OCCA in Littlejohn II, that modern jury pools are perpetually confused about Oklahoma’s three-option sentencing scheme. We reject these arguments.

As Mr. Littlejohn recognizes, we have held that instructions clearly (and correctly) stating Oklahoma’s “three-way choice fulfillf ] the Simmons requirement that a jury be notified [regarding] the defendant's] parole ineligib [ility], ” even where a jury asks the court about the future prospect that the defendant may be released. Mayes, 210 F.3d at 1294. Like in Mayes, the instructions here clearly stated Oklahoma’s three-option scheme. That much is not in dispute.

Of course, the trial judge did not simply refer the jury back to the instructions upon receiving the note requesting clarification on the meaning of the second sentencing option—“life without the possibility of parole.” Rather, upon receiving the note from the jury, the judge referred back to his instructed “code” which told the jury that he would answer questions only “if it’s appropriate to answer,” and that if, in response, the jury received a “code back that says, you have all the law and evidence necessary to reach a verdict, what that means is [1] the answer to [the] question is in the instructions, [2] it was in the evidence, or [3] you’re asking me something that’s inappropriate for me to answer.” Littlejohn II, 85 P.3d at 291 (emphases added) (quoting relevant portions of the State record) (internal quotation marks omitted); see R., Vol. 1, pt. II, at 228.

Primarily at issue is the “third” answer of the “code.” Mr. Littlejohn would have us believe that this answer signaled to the jury that parole was not an appropriate factor for its consideration and, thus, beyond its purview. We believe, however, that the trial judge’s response simply reinforced the plain meaning of the otherwise permissible instructions. The response said nothing—by its precise terms or by its substance-—about the removal from the jury’s sentencing consideration of parole eligibility as a permissible factor. At worst, it suggested that it may be inappropriate for the judge to answer a question, which conceivably, in context, could be a question about parole. The response did not reasonably suggest to the jury that the question of parole eligibility rested elsewhere than in its hands, or that a decision to recommend life without the possibility of parole would cause the sentence to be placed in the hands of another, or otherwise permit Mr. Littlejohn to be released. Cf. Mollett, 348 F.3d at 915-16 (holding that prong four was satisfied where, upon the jury’s note, the trial judge responded that “[m]atters of parole are beyond the purvue [sic] of the jury or the court to consider” because this answer suggested to the jury (incorrectly) that matters of parole were beyond the boundaries of its consideration, indicating that alternative sentences to death could lead to the defendant being released (alterations in original)). This does not bring Mr. Littlejohn’s case within the stable of our decisions finding due process violations where “the trial court informs the jury that it is not to consider the issue of whether the defendant is parole ineligible.” Welch, 639 F.3d at 1005.

Thus, even if we assume the jury applied the third “code” option after the judge’s response—viz., assumed that the question was “inappropriate for [the court] to answer”—there was no “false choice” created. See id. at 1005-06 (finding that the judge’s response to the jury’s question on the meaning of life without the possibility of parole—which noted “that [the court] was not allowed to answer the jury’s questions”—“could not have created a prohibited false choice”). “Failing to clarify the life without parole instruction cannot be ‘taken to mean that parole was available but that the jury, for some unstated reason, should be blind to this fact.’ ” Id. at 1005 (quoting Shafer, 532 U.S. at 53, 121 S.Ct. 1263); see also Littlejohn II, 85 P.3d at 292 (“Neither the trial court’s response to the jury’s question nor its supplemental oral instructions concerning how to decode its answers advised the jury it could not consider parole eligibility in determining the appropriate sentence.”).

Mr. Littlejohn responds that it is clear that juries in Oklahoma have consistently misunderstood the meaning of life imprisonment without the possibility of parole. Thus, he reasons, the trial judge’s answer only perpetuated an inherent confusion that already exists and is deeply entrenched. See Aplt. Opening Br. at 32 (“[T]he problem is not failure to hear, but rather, that the significance of the phrase has not been understood.”). As noted, the OCCA acknowledged this fact and observed “the frequency with which this issue arises ... [required a] reexamin[ation of its] present stance and [a] consideration of] whether future juries who ask about the meaning of life imprisonment without the possibility of parole should be given further explanation.” Littlejohn II, 85 P.3d at 292-93 (emphasis added); see id. at 293 (“[C]ommon sense tells us that some jurors may be confused even when, as here, the jury is presented with one ‘traditional’ life sentencing option, alongside a separate option that specifically states that parole will not be available.”). Therefore, to “provid[e] capital sentencing juries with accurate sentencing information” the OCCA held that in future cases where the jury asks whether a sentence of life imprisonment without the possibility of parole renders an offender parole eligible,

the trial court should either refer the jury back to the instructions, tell the jury that the punishment options are self explanatory, or advise the jury that the punishment options are to be understood in their plain and literal sense and that the defendant will not be eligible for parole if sentenced to life imprisonment without the possibility of parole.^[4]

Id. at 293-94 (citations omitted).

Mr. Littlejohn argues that the OCCA’s decision to provide guidance for future cases, while not applying its ruling to his appeal, was unreasonable. We disagree. To the extent that the trial court’s response referred to the first two options of the “code,” the OCCA’s recommendations encompassed what actually occurred in this case—“where the jury ... asks [about parole eligibility] ... the trial court should[, inter alia,] .. refer the jury back to the instructions.” Id. at 293 (emphasis added); see Welch, 639 F.3d at 1005 (“In applying Simmons, we have concluded that if the trial court simply directs the jury to review the instructions again, the defendant’s due process rights are not violated.”).

Furthermore, “[e]ven assuming the trial court’s statement [otherwise] ... ran afoul of Oklahoma procedural law,”⁵ Welch, 639 F.3d at 1005; see id. at 1004 & n. 13 (noting that Littlejohn II set forth “three options” for a trial court’s response to a jury note “under Oklahoma law ” (emphasis added))—in that it consisted of an option that was not one of the “three” endorsed by the OCCA in Littlejohn II—that would not, absent more, constitute a violation of federal law sufficient to warrant habeas relief, see id. at 1005; Hooks v. Workman (Danny Hooks), 606 F.3d 715, 748 (10th Cir.2010) (“[Fjederal habeas corpus relief does not lie for errors of state law.” (quoting Estelle v. McGuire, 502 U.S. 62, 67, 112 S.Ct. 475, 116 L.Ed.2d 385 (1991)) (internal quotation marks omitted)); see also Johnson, 254 F.3d at 1164; cf. Williams v. Jones, 571 F.3d 1086, 1092 (10th Cir.2009) (per curiam) (“[A]ny correction for a federal constitutional violation must be consistent with federal law.”). And, as to the applicable federal requirements, the OCCA found that the trial judge’s response in this case was consistent with Simmons in that it was not “insufficient, misleading or erroneous” and did not suggest to the jury that it was forbidden to consider parole. Littlejohn II, 85 P.3d at 292; cf. Mollett, 348 F.3d at 915-16; Welch, 639 F.3d at 1005 (“[I]n cases in which the trial court informs the jury that it is not to consider the issue of whether the defendant is parole ineligible, we have found a due process violation.”). This determination was not unreasonable for the reasons given. Cf. Wetzel v. Lambert, — U.S.-, 132 S.Ct. 1195, 1198, 182 L.Ed.2d 35 (2012) (per curiam) (noting that where a state court’s resolution of a claim on the merits on a sufficient substantive ground is not unreasonable, alternative grounds for imputing capricious error are “beside the point”).

While Mr. Littlejohn’s argument may have some appeal insofar as it takes issue with the reality that “a fair number of jurors [still] do not [fully] comprehend the plain meaning of the life imprisonment without the possibility of parole sentencing option [in Oklahoma],” Littlejohn II, 85 P.3d at 293, we are constrained to apply precedent that-relying on Simmons, Shafer, and Kelly on substantially similar facts—compels the denial of relief, see, e.g., Welch, 639 F.3d at 1005 (“The trial court’s response to the jury’s questions did not negate or contradict any of these [three] choices; each were explicitly set forth in the jury instructions and clearly presented in the verdict form.”); see also McCracken, 268 F.3d at 980-81. In sum, for the foregoing reasons, we reject Mr. Little-john’s first argument on appeal.

B. Notice of Testimony

Mr. Littlejohn argues that the district court erred in finding harmless the constitutional error that occurred when the trial court permitted the State to elicit certain damaging testimony from a witness in support of its aggravating-factor presentation at sentencing, although the State had not given notice of this evidence. Bill Meers, the victim’s brother, testified that when he was leaving the courtroom after the first trial, Mr. Littlejohn made certain comments to him, including an admission regarding the killing and—most importantly, for present purposes—a threat to kill Meers. Specifically, Mr. Meers testified that Mr. Littlejohn told him, that “the motherfucker’s [i.e., his brother’s] dead and he ain’t coming back.” State R., Vol. VI, Resentencing Tr. at 21 (Test, of Bill Meers). And he testified further that Mr. Littlejohn said, “I killed the motherfucker, I’ll kill you.” Id. The court reporter captured and memorialized at least a portion of Mr. Littlejohn’s communications to Mr. Meers but, significantly, her transcript did not evince an admission or a threat by Mr. Littlejohn.⁶ See Aplt. Opening Br. at 33 (“The court reporter indeed captured an insensitive comment'by Mr. Littlejohn on exiting the courtroom after the 1994 sentencing, but it did not include either an admission or a threat.”).

Mr. Littlejohn was provided no notice of the alleged admission or threat until the fifth day of the resentencing trial. See Littlejohn II, 85 P.3d at 294. At an in camera hearing, the court notified Mr. Littlejohn of the proposed testimony of Mr. Meers and gave counsel “the weekend and [the following] Monday to prepare for [it].” Id. at 296 n. 10. The OCCA succinctly summarized the unfolding events:

The record shows that after the in camera hearing, the trial court gave the defense some time [i.e., the three days] to try and find the guards who were escorting Littlejohn when he allegedly made the statement to Meers. Although the defense was unable to find those guards, they did locate and call Sgt. Grimsley, who was in charge of security and courtroom guard details during Littlejohn’s first trial. Grimsley explained how defendants were escorted from the courtroom following verdicts at that time. According to Grimsley, it would be highly unlikely that a defendant would be able to stop and make any kind of statement to a victim’s family, and if an incident did occur, a write-up would have been made. The defense also called Littlejohn, who admitted that he made the first remark to Meers [i.e., about the brother being dead and not coming back], but denied telling Meers that he had shot his brother or that he threatened to kill Meers.

Id. at 296 (footnotes omitted). In light of the fact that Mr. Littlejohn had time to meet the allegations with his own investigation, and the fact that the evidence of the continuing-threat aggravating factor was substantial, the OCCA concluded that the lack of sufficient notice of Mr. Meers’s testimony—in particular, regarding Mr. Littlejohn’s alleged threat—did not have the requisite prejudicial effect “on the jury’s verdict” to make the error grounds for reversal. R., Vol. 1, pt. II, at 236.

Mr. Littlejohn argues on appeal that the OCCA’s decision was unreasonable for a few reasons. First, he contends that the error was demonstrably prejudicial in that it limited his trial counsel’s ability to mount a more robust defense. Second, he contends that the prosecutor improperly (and misleadingly) suggested to the jury that she had heard Mr. Littlejohn’s threat, further compounding the prejudice. Finally, he contends that the harm of the error was exacerbated by the prosecution’s rebuttal testimony of Judy Bush, a witness who testified in response to Mr. Little-john’s denial that he made the threat.

It is well-established, as a matter of federal law, that “a defendant must have a meaningful opportunity to deny or explain the State’s evidence used to procure a death sentence.” Walker v. Gibson, 228 F.3d 1217, 1240 (10th Cir.2000) (quoting Duvall v. Reynolds, 139 F.3d 768, 797 (10th Cir.1998)), abrogated on other grounds by Neill v. Gibson, 278 F.3d 1044, 1057 n. 5 (10th Cir.2001); see Gardner, 430 U.S. at 362, 97 S.Ct. 1197 (“We conclude that petitioner was denied due process of law when the death sentence was imposed, at least in part, on the basis of information which he had no opportunity to deny or explain.”); see also Gray v. Netherland, 518 U.S. 152, 163-65, 116 S.Ct. 2074, 135 L.Ed.2d 457 (1996); Wilson v. Sirmons, 536 F.3d 1064, 1102 (10th Cir.2008) (“[W]e construe [the petitioner’s] claim to allege that the late notice deprived him of his ability to provide a proper defense.”).

Furthermore, “while ‘[a] defendant’s right to notice of the charges against which he must defend is well established,’ there is no clearly established constitutional right to non-exculpatory discovery.” Wilson, 536 F.3d at 1103 (alteration in original) (quoting Gray, 518 U.S. at 167-68, 116 S.Ct. 2074). Thus, “[s]o long as Mr. [Littlejohn] had a meaningful ‘opportunity to deny or explain’ ” the adverse evidence, see id. (quoting Gardner, 430 U.S. at 361, 97 S.Ct. 1197), he would not be able to show a “clearly established due process violation,” id. And even if a lack of notice violates an offender’s rights, the resulting error must be prejudicial in order to warrant relief. See id.; Mayes, 210 F.3d at 1292.

We assume without deciding that Mr. Littlejohn has properly established a constitutional violation arising from the State’s failure to provide adequate notice of Mr. Meers’s testimony. We therefore follow the OCCA and the district court and assess only whether that error was prejudicial. We conclude that it was not.

In denying Mr. Littlejohn’s petition, the OCCA relied upon Chapman v. California, 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967), for the proposition that the prejudice from any lack of notice in this case was “harmless beyond a reasonable doubt” because Mr. Littlejohn mounted a defense against the statement and the other evidence presented by the State was overwhelming in support of the continuing-threat aggravator. See Littlejohn II, 85 P.3d at 296. On habeas review, we apply (as did the district court) the standard in Brecht v. Abrahamson, 507 U.S. 619, 113 S.Ct. 1710, 123 L.Ed.2d 353 (1993), in assessing prejudicial effect. See DeRosa v. Workman, 679 F.3d 1196, 1233 (10th Cir.2012) (“[E]ven assuming that the OCCA did err in [applying Chapman ], we are bound to apply ‘the more forgiving standard of review’ outlined in Brecht.” (citation omitted)); Welch, 639 F.3d at 992; Matthews v. Workman, 577 F.3d 1175, 1181 (10th Cir.2009); Herrera v. Lemaster, 301 F.3d 1192, 1200 (10th Cir.2002) (en banc); see also Fry v. Pliler, 551 U.S. 112, 121-22, 127 S.Ct. 2321, 168 L.Ed.2d 16 (2007) (“We hold that in § 2254 proceedings a court must assess the prejudicial impact of constitutional error in a state-court criminal trial under the ‘substantial and injurious effect’ standard set forth in Brecht ... whether or not the state appellate court recognized the error and reviewed it for harmlessness under the ‘harmless beyond a reasonable doubt’ standard set forth in Chapman.... ” (citations omitted)).

“Interests of comity and federalism, as well as the State’s interest in the finality of convictions that have survived direct review within the state court system, mandate a more deferential standard of review in evaluating [the petitioner’s] claim.” Matthews, 577 F.3d at 1181 (quoting Crease v. McKune, 189 F.3d 1188, 1193 (10th Cir.1999)) (internal quotation marks omitted). “Under Brecht, the standard for determining whether habeas relief must be granted is whether the ... error [at issue] had substantial and injurious effect or influence in determining the jury’s verdict.” Selsor v. Workman, 644 F.3d 984, 1014 (10th Cir.2011) (alteration and ellipsis in original) (quoting Brecht, 507 U.S. at 623, 113 S.Ct. 1710) (internal quotation marks omitted).

“[W]hen reviewing errors from a criminal proceeding ... if the harmlessness of the error is in grave doubt, relief must be granted.” O’Neal v. McAninch, 513 U.S. 432, 440, 115 S.Ct. 992, 130 L.Ed.2d 947 (1995) (emphasis added); Bland, 459 F.3d at 1009 (“[A] ‘substantial and injurious effect’ exists when the court finds itself in ‘grave doubt’ about the effect of the error on the jury’s verdict.” (quoting O’Neal, 513 U.S. at 435, 115 S.Ct. 992)); accord United States v. Lawson, 677 F.3d 629, 644 n. 19 (4th Cir.2012). “‘Grave doubt’ exists where the issue of harmlessness is ‘so evenly balanced that [the court] feels [itself] in virtual equipoise as to the harmlessness of the error.’ ” Bland, 459 F.3d at 1009-10 (alterations in original) (quoting O’Neal, 513 U.S. at 435, 115 S.Ct. 992); see DeRosa, 679 F.3d at 1233.

We agree with the district court that the “lack of sufficient notice of Mr. Meers’ testimony [did not have] a substantial and injurious effect on the jury’s verdict.”⁷ R., Vol. 1, pt. II, at 236. In the time that the state trial court gave him, Mr. Little-john was able to secure the testimony of Sgt. Grimsley, which aided his cause. This was significant testimony, in that Sgt. Grimsley was head of courtroom security. Even if he had more time, it is questionable whether Mr. Littlejohn could have obtained more probative evidence on the subject at hand. Indeed, as the district court noted, “Sgt. ... Grimsley testified that he had received an inquiry from defense counsel on Friday afternoon regarding the identities of the guards assigned to Petitioner at his first trial; however, despite an extensive search, issues with record keeping prevented him from discovering this information.” Id. at 234. Furthermore, Mr. Littlejohn was able to offer his own testimony in rebuttal, which directly related to the matter at issue, in that he was the alleged speaker of the threat. While three days to prepare a defense against a statement made many years pri- or is not necessarily a substantial amount of time, Mr. Littlejohn fails to establish that having more time to conduct further investigation would have led to the discovery of any other helpful information about the circumstances surrounding the event.

In addition, before rendering its verdict, the jury was exposed to considerable evidence detailing Mr. Littlejohn’s prior criminal history, which included instances of violence, see Littlejohn II, 85 P.3d at 296— that served as “substantial ... continuing threat evidence,” id. The OCCA summarized the evidence of his criminal history as follows:

[T]he State offered a substantial amount of continuing threat evidence. The State presented evidence that Littlejohn had been incarcerated for all but a few months from the time he was 15-years-old until he committed this crime at the age of 20. The State introduced evidence that showed Littlejohn’s tendency toward violence had begun in elementary school, where he was placed in a class for the emotionally disturbed and he continued to get in trouble for his behavior. The State presented evidence that Littlejohn had been involved in robberies, assaults and a rape. In addition, the State introduced evidence of numerous infractions, some violent, while Little-' john was in the Oklahoma County Jail awaiting his first trial, as well as incidents that occurred while Littlejohn was imprisoned in the Department of Corrections.

Littlejohn II, 85 P.3d at 296. Mr. Littlejohn does not contest the validity of any of this evidence, and it otherwise supports the continuing-threat aggravator found by the jury. See Revilla v. Gibson, 283 F.3d 1203, 1219 (10th Cir.2002) (suggesting that threats of violence suffice to establish that an offender is a continuing threat). Thus, Mr. Meers’s testimony was not the linchpin of the State’s evidentiary presentation concerning the continuing-threat aggravator. And, logically, any prejudicial effect of his testimony would have been lessened by the fact that it was but one item among many that attested to Mr. Littlejohn’s violent disposition.

Mr. Littlejohn’s claims of improper bolstering of Mr. Meers’s testimony by the prosecutor do not alter our conclusion here. As further detailed infra, the record does suggest that the prosecutor, in the context of questioning Mr. Meers, said that she had been present in the courtroom at the first trial and she insinuated— but did not expressly say—that she had heard Mr. Littlejohn make the threat. See State R., Vol. VI, Resentencing Tr. at 26 (Question of Ms. High) (stating, in questioning Mr. Meers, “So in terms of you never having told anyone about those statements, were you aware that I was present and heard?”). As the OCCA observed, however, the trial court sustained defense counsel’s objection to the prosecutor’s statement, and “admonished the jury to disregard it.” Littlejohn II, 85 P.3d at 296. This frequently has the effect of diminishing any discernible prejudice, cf. Wilson, 536 F.3d at 1117 (suggesting that cautionary instructions are a factor to consider under a totality of the circumstances). To be sure, a cautionary instruction is not always sufficient to cure the prejudicial effect of improper statements. Cf. United States v. Sands, 899 F.2d 912, 915 (10th Cir.1990). However, the prosecutor’s brief and oblique statement here is not the kind that would “make a sufficiently strong impression on the jury that it w[ould] be unable to disregard.” Id. (emphasis added). Thus, we cannot say that the OCCA’s determination that the cautionary instruction here diminished the prejudice suffered by Mr. Littlejohn was unreasonable.

Moreover, the prosecution’s unnoticed rebuttal witness, Ms. Judy Bush, essentially confirmed—in very limited testimony— Mr. Meers’s testimony and suggested that Mr. Littlejohn’s version of the story was incorrect. Importantly, Ms. Bush’s testimony was confined to the alleged statements made by Mr. Littlejohn at the 1994 proceeding—nothing more. Thus, it was not outside of the scope of Mr. Meers’s direct examination. It logically follows that any additional prejudicial effect of Ms. Bush’s testimony, arising from the lack of notice of the Meers testimony, was virtually nil. In other words, the substance of her testimony regarding Mr. Littlejohn’s alleged comments was roughly conterminous with that of Mr. Meers; therefore, in terms of preparing a response, Ms. Bush’s testimony did not place an appreciable additional burden on Mr. Littlejohn.

Finally, Mr. Littlejohn contends that the district court and the OCCA failed to consider the reliability concerns associated with this evidence. That is, Mr. Meers’s and Ms. Bush’s testimony “was of very recent vintage and directly in conflict with the official record of what was said in the courtroom.” Aplt. Opening Br. at 47. But these factors would not necessarily militate toward a finding of prejudice on lack-of-notice grounds. Specifically, Mr. Little-john demonstrated that he had the ability to (and did) point out these purported weaknesses in their testimony. In particular, he had access to the fact that the court reporter’s statement was more consistent with Mr. Littlejohn’s view of what occurred. Thus, this factor does not dissuade us from our conclusion that the OCCA acted reasonably in concluding that the notice error was harmless.

At bottom, it must be emphasized that the alleged error here relates to a lack of notice—not the prejudicial content of Mr. Meers’s testimony, which Mr. Littlejohn does not contend would otherwise be inadmissible under Oklahoma law “but for the notice problem.” Littlejohn II, 85 P.3d at 295. The reasonably conceivable prejudicial effect from the lack of notice here does not leave us in “grave doubt” regarding whether it had a “substantial and injurious effect” on the jury’s verdict. Taking into account the totality of the evidence, we are not left to struggle with this conclusion. See Wilson, 536 F.3d at 1103; see also Thomas v. Gibson, 218 F.3d 1213, 1223-24 (10th Cir.2000); Humphries v. Ozmint, 397 F.3d 206, 227 (4th Cir.2005) (en banc). In particular, we look to the fact that Mr. Littlejohn was allowed three days to prepare a response to Mr. Meers’s statement; that, in fact, he did prepare a reasonably cogent affirmative rebuttal case, which featured the testimony of Sgt. Grimsley, who was “in charge of security and courtroom guard details during [his] first trial,” Littlejohn II, 85 P.3d at 296 (emphasis added); that Mr. Littlejohn has not pointed to any additional favorable evidence that he would have acquired if he had been given more notice; and that Mr. Littlejohn tested the credibility of Mr. Meers and Ms. Bush on cross examination.

In sum, the OCCA’s decision thoroughly considered all of the factors at the resen-tencing bearing on any prejudice that Mr. Littlejohn allegedly suffered due to a lack of notice and determined that any error was harmless. We conclude that this decision was unquestionably reasonable. More to the point, we do not have any grave doubts concerning the harmlessness of the (assumed) error involving a lack of notice. Consequently, we reject Mr. Littlejohn’s claim.

C. Prosecutorial Misconduct

Mr. Littlejohn claims that the prosecution made numerous improper comments during his resentencing proceeding, violating his constitutional rights.

1. Improper Vouching

The first allegation of prosecutorial misconduct concerns the allegedly improper statements that the prosecutor made in questioning Mr. Meers (which we have just discussed)—i.e., the prosecutor’s suggestion that she heard Mr. Littlejohn make the disputed statements to Mr. Meers when he left the courtroom after the first trial, including the threat to kill him. The prosecutor’s full statement, embodied in her question to Mr. Meers, was made on re-direct and was as follows: “So in terms of you never having told anyone about those statements, were you aware that I was present and heard?” State R., Vol. VI, Resentencing Tr. at 26. Mr. Meers responded affirmatively that he was “aware [she was] present.” Id. (Test. of Mr. Meers).⁸ Mr. Littlejohn argues that this statement amounted to impermissible prosecutorial vouching.

“Vouching,” or “an assurance by the prosecuting attorney of the credibility of a government witness through personal knowledge or by other information outside of the testimony before the jury,” amounts to improper prosecutorial conduct. Lam v. Kelchner, 304 F.3d 256, 271 (3d Cir.2002); see Matthews, 577 F.3d at 1187; Douglas v. Workman, 560 F.3d 1156, 1177-79 (10th Cir.2009). Generally, there are two ways in which prosecutorial misconduct, like vouching, can result in constitutional error. See DeRosa, 679 F.3d at 1222. “First, [it] can prejudice a specific right ... as to amount to a denial of that right.” Id. (quoting Matthews, 577 F.3d at 1186) (internal quotation marks omitted). Additionally, absent infringement of a specific constitutional right, a prosecutor’s misconduct may in some instances render a habeas petitioner’s trial “so fundamentally unfair as to deny him due process.” Donnelly v. DeChristoforo, 416 U.S. 637, 645, 94 S.Ct. 1868, 40 L.Ed.2d 431 (1974); see Wilson, 536 F.3d at 1117 (“Unless prosecutorial misconduct implicates a specific constitutional right, a prosecutor’s improper remarks require reversal of a state conviction only if the remarks so infected the trial with unfairness as to make the resulting conviction a denial of due process.” (quoting Le v. Mullin, 311 F.3d 1002, 1013 (10th Cir.2002) (per curiam)) (internal quotation marks omitted)); see also Matthews, 132 S.Ct. at 2153-54; Romano v. Oklahoma, 512 U.S. 1, 12-13, 114 S.Ct. 2004, 129 L.Ed.2d 1 (1994).

In determining whether a trial is rendered “fundamentally unfair” in light of the conduct of a prosecutor,

we examine the entire proceeding, “including the strength of the evidence against the petitioner, both as to guilt at that stage of the trial and as to moral culpability at the sentencing phase as well as any cautionary steps—such as instructions to the jury—offered by the court to counteract improper remarks.”

Wilson, 536 F.3d at 1117 (quoting Bland, 459 F.3d at 1024). “[A] court should not lightly infer that a prosecutor intends an ambiguous remark to have its most damaging meaning or that a jury ... will [necessarily] draw that meaning.” Don-nelly, 416 U.S. at 647, 94 S.Ct. 1868; see Banks v. Workman, 692 F.3d 1133, 1148 (10th Cir.2012) (noting that the fundamental-fairness standard for allegedly improper prosecution statements constitutes “a high hurdle”). “[N]ot every improper or unfair remark made by a prosecutor will amount to a federal constitutional deprivation.” Tillman v. Cook, 215 F.3d 1116, 1129 (10th Cir.2000).

Mr. Littlejohn complains that the prosecutor’s statement “implicate[d his] specific constitutional right,” Wilson, 536 F.3d at 1117, to confront a person who in effect was a witness in the case—the prosecutor herself—thus entitling him to heightened scrutiny under Donnelly. He faults the OCCA and the district court for not recognizing this. Moreover, he specifically suggests that the statement was “akin to [ringing a] testimonial-like bell” that could not be “unrung.” See Aplt. Opening Br. at 49, 52 (emphasis added). We disagree.

Despite his oblique references to the Confrontation Clause, Mr. Littlejohn does not actually claim that such a violation took place—only that the prosecutor’s “comments were akin to a violation of the Confrontation Clause.” Id. at 51 (emphasis added). Even if the Supreme Court’s Confrontation Clause jurisprudence were apposite in a capital sentencing context—a matter that we touch on in Part II.D, infra, but need not pursue here—Mr. Litt-lejohn’s argument would not even get out of the gate, because he does not expressly rely upon that jurisprudence.⁹ Accordingly, we would be hard-pressed to conclude that the OCCA violated clearly established federal law under the Confrontation Clause.

As for the fundamental-fairness inquiry, the OCCA found that the prosecutor’s “improper” comment was removed from the purview of the jury’s consideration when the trial court “sustained the defense’s objection to the question and admonished the jury to disregard it.” Littlejohn II, 85 P.3d at 296, 300; see also Greene v. Upton, 644 F.3d 1145, 1159 (11th Cir.2011) (holding that state court’s finding that an admonition eliminated the prejudice of an inappropriate comment was not unreasonable).

The OCCA discussed and assessed all of the pertinent legal factors in arriving at its decision, see Littlejohn II, 85 P.3d at 296, 300, and we cannot say, after a review of the record and the governing law, that its rejection of Mr. Littlejohn’s argument was unreasonable under AEDPA’s deferential standards, see Matthews, 132 S.Ct. at 2153-55 (rejecting due process claim predicated upon the assertion that the prosecutor insinuated that the defendant “had colluded with his lawyer ... to manufacture an extreme emotional disturbance defense” because “even if the comment is understood as directing the jury’s attention to inappropriate considerations, that would not establish that the [state] Supreme Court’s rejection of the ... prose-cutorial misconduct claim” was contrary to clearly established federal law); Wogenstahl v. Mitchell, 668 F.3d 307, 328-29 (6th Cir.2012) (holding that prosecutor’s comments improperly vouching for the State’s witness did not amount to prejudicial error); see also Matthews, 577 F.3d at 1186 (noting that “[w]e assess [the state court’s] decision [on the lack of prejudice of challenged remarks] through AEDPA’s forgiving lens”); cf. Danny Hooks, 606 F.3d at 744-46 (holding that the prosecutor’s conduct prejudiced petitioner where the prosecutor repeatedly “misled [the jury] to believe it was the obligation of a juror holding a minority opinion to abandon that opinion if it was necessary for the jury to reach a unanimous sentence” and other factors were present that contributed to this “coercion”).¹⁰

2. Reference to the Appellate Process

Mr. Littlejohn also contends that the prosecution improperly injected the issue of appellate review into the case. In the prosecutor’s final closing argument at re-sentencing she asserted:

[Mr. Littlejohn’s counsel] has told you that basically Ms. Stensaas [the prosecutor in the Bethany trial] did something that was wrong [i.e., by arguing inconsistent theories], [that] it was improper, there was something wrong with what she did and he full well knows that that has been reviewed by appellate courts.

State R., Vol. VII, Resentencing Tr. at 312 (State’s Closing Argument) (emphasis added).¹¹ The trial court sustained an objection to the extent the prosecution was arguing facts not in evidence. However, because the court found that Mr. Little-john’s counsel “opened the door” to the reference to the prosecution’s conduct, it permitted the prosecution to “address the issue within the record.” Aplt. Opening Br. at 54. The prosecutor subsequently told the jury that, “[i]f there was something wrong with Ms. Stensaas’ closing argument, you would be told that.” State R., Vol. VII, Resentencing Tr. at 314. No objection was made to this latter comment. Both the OCCA, see Littlejohn II, 85 P.3d at 301, and the district court, see R., Vol. 1, pt. II, at 248-49, rejected Mr. Littlejohn’s claim that the prosecutor’s argument was improper.

Mr. Littlejohn claims that the foregoing argument is a clear violation of Caldwell v. Mississippi, 472 U.S. 320, 328-29, 105 S.Ct. 2633, 86 L.Ed.2d 231 (1985), where the Supreme Court held that “it is constitutionally impermissible [under the Eighth Amendment] to rest a death sentence on a determination made by a sentencer who has been led to believe that the responsibility for determining the appropriateness of the defendant’s death rests elsewhere.” This holding has been interpreted by the Supreme Court as simply requiring that “the jury ... not be misled regarding the role it plays in the sentencing decision.” Danny Hooks, 606 F.3d at 743 (quoting Romano, 512 U.S. at 8, 114 S.Ct. 2004) (internal quotation marks omitted) (citing cases); see Darden v. Wainwright, 477 U.S. 168, 183 n. 15, 106 S.Ct. 2464, 91 L.Ed.2d 144 (1986) (“Caldwell is relevant only to certain types of comment—those that mislead the jury as to its role in the sentencing process in a way that allows the jury to feel less responsible than it should for the sentencing decision.”); see also Dugger v. Adams, 489 U.S. 401, 405, 109 S.Ct. 1211, 103 L.Ed.2d 435 (1989); DeRosa, 679 F.3d at 1234.

Here, Mr. Littlejohn has not shown that the district court was incorrect in concluding that the instant comments did not contravene Caldwell—much less that the OCCA’s rejection of the claim was contrary to, or involved an unreasonable application of, Caldwell. The prosecutor’s comments simply do not amount to a Caldwell violation because they referenced “the prosecutor’s actions [in Mr. Bethany’s case] and whether the prosecution had acted inappropriately,” R., Vol. 1, pt. II, at 248 (emphases added); they did not suggest that the jury’s role was “minimize[d in] importance,” Caldwell, 472 U.S. at 333, 105 S.Ct. 2633.

Indeed, considering the closing arguments in context, see Pickens v. Gibson, 206 F.3d 988, 1000 (10th Cir.2000), the comments at issue were merely a response to defense counsel’s suggestion that the prosecution had acted improperly by arguing inconsistent theories at the trials of Mr. Littlejohn and Mr. Bethany. No reasonable jury would have construed the statements as implying that “anyone other than the jury was responsible for fixing the appropriate sentence.” Thornburg v. Mullin, 422 F.3d 1113, 1136 (10th Cir.2005); see id. at 1135-36 (holding that prosecutor’s comments suggesting that the jury was the “State” did not run afoul of Caldwell)-, Moore v. Gibson, 195 F.3d 1152, 1174 (10th Cir.1999) (holding that the prosecutor’s statements, suggesting, inter alia, that the jury was a “little cog in the community” responsible for carrying out the criminal proceedings, did not violate Caldwell (quoting relevant portions of the record) (internal quotation marks omitted)); Sellers v. Ward, 135 F.3d 1333, 1343 (10th Cir.1998) (“Viewing the record with charity, we note the prosecutor’s hyperbole may be regarded as colorful. It is not, however, the stuff from which anyone could perceive an offer to share the burden of the ultimate decisions in this case.”); see also Black v. Workman, 682 F.3d 880, 910-12 (10th Cir.2012) (“Caldwell should not be read too broadly.”)

The prosecutor’s comment here is markedly distinct from statements held to be constitutionally problematic under the principles of Caldwell. See Caldwell, 472 U.S. at 325, 340, 105 S.Ct. 2633 (holding that prosecutor’s comments that the jury’s job was “not the final decision” and “[wa]s reviewable”—which were “focused, unambiguous, and strong”—violated the Eighth Amendment (quoting relevant portions of the record) (internal quotation marks omitted)); Cargle v. Mullin, 317 F.3d 1196, 1223 (10th Cir.2003) (concluding that prosecutor’s argument was inappropriate in that it “suggested] that jurors are part of the team of the prosecution and police, rather than impartial arbiters between the State and the defendant” (internal quotation marks omitted)).

Moreover, the prosecution emphasized multiple times that the decision to impose the death penalty was a tough one, but one that nonetheless rested solely with the jury. See, e.g., State R., Vol. VII, Resentencing Tr. at 316, 329-31, 349-50. Considering everything before the jury, including the entirety of closing arguments which repeatedly emphasized the jury’s correct role, we simply cannot find unreasonable the OCCA’s determination that the jury was not affirmatively misled regarding its part in the sentencing process. See Bland, 459 F.3d at 1019 (considering the prosecution’s comments “in context,” and finding that the OCCA’s rejection of petitioner’s Caldwell argument was not unreasonable, where the prosecutor emphasized that the jury had the final say); cf. Romano, 512 U.S. at 9, 114 S.Ct. 2004 (finding that the introduction of evidence concerning a previously imposed death sentence did not mislead the jury on its sentencing role because, among other reasons, “[t]he jury was instructed that it had the responsibility for determining whether the death penalty should be imposed” (quoting Romano v. State, 847 P.2d 368, 390 (Okla.Crim.App.1993)) (internal quotation marks omitted)). Consequently, this claim of prosecutorial misconduct must fail.

3. Civic Duty and Emotion

Finally, Mr. Littlejohn argues that “[b]oth prosecutors [improperly] expressed personal opinions and urged the jury to return a verdict for the community at large.” Aplt. Opening Br. at 55. For instance, one prosecutor, at the conclusion of the “first closing,” id., noted that “[y]ou are the representatives of the community that have been chosen to be the jurors in this case and it’s your job to decide what is justice in this case,” State R., Vol. VII, Resentencing Tr. at 279. Additionally, the second prosecutor noted that, “you are the twelve representatives of this community in which we all live to decide what is justice. And what is justice in this case is the death penalty.” Id. at 350. Mr. Litt-lejohn did not object to either comment. The OCCA summarily rejected the argument that these comments were improper. See Littlejohn II, 85 P.3d at 301. The district court, moreover, found Mr. Littlejohn’s argument to lack merit because the comments at issue, taken in context, did not “encourag[e] a death sentence based on vengeance and/or outrage.” R., Vol. 1, pt. II, at 249. We agree with the district court.

“It is improper for a prosecutor to suggest that a jury has a civic duty to convict,” Thornburg, 422 F.3d at 1134; see also Viereck v. United States, 318 U.S. 236, 247-48, 63 S.Ct. 561, 87 L.Ed. 734 (1943), by “appealing to] wholly irrelevant” issues, Viereck, 318 U.S. at 247-48, 63 S.Ct. 561. Furthermore, “[i]t is of vital importance to the defendant and to the community that any decision to impose the death sentence be, and appear to be, based on reason rather than caprice or emotion.” Gardner, 430 U.S. at 358, 97 S.Ct. 1197; see Wilson, 536 F.3d at 1121. “Appeals to the jury’s emotion or sense of vengeance call into question the integrity of the criminal justice system by encouraging the jury to convict based on outrage, and not on the evidence,” Wilson, 536 F.3d at 1121 (alteration omitted) (quoting Bland, 459 F.3d at 1028) (internal quotation marks omitted), thereby undermining the “acute need for reliability in capital sentencing proceedings,” Monge v. California, 524 U.S. 721, 732, 118 S.Ct. 2246, 141 L.Ed.2d 615 (1998); see Lankford v. Idaho, 500 U.S. 110, 126, 111 S.Ct. 1723, 114 L.Ed.2d 173 (1991); California v. Brown, 479 U.S. 538, 543, 107 S.Ct. 837, 93 L.Ed.2d 934 (1987); Ohio Adult Parole Auth. v. Woodard, 523 U.S. 272, 294, 118 S.Ct. 1244, 140 L.Ed.2d 387 (1998) (Stevens, J., concurring in part and dissenting in part).

The district court correctly determined that the comments in this case did not play on the jury’s emotions, or otherwise improperly embed into the jury’s sentencing consideration a sense of societal alarm or the need for community justice. In advocating for a result commensurate with “justice,” the prosecution nonetheless emphasized that the jury retained full responsibility for weighing the evidence— which it extensively outlined in closing argument—and that “justice” would be the product of this weighing process. See State R., Vol. VII, Resentencing Tr. at 279. References to “justice” are not necessarily improper, at least where the “prosecutor’s comments [are] firmly rooted in the facts of the case ” and are not otherwise made in a substantially inflammatory manner. Thornburg, 422 F.3d at 1134 (emphasis added); see id. at 1133-34 (concluding that multiple comments from the prosecutor regarding the jury’s role in allowing “justice [to] prevail” were not necessarily inappropriate because there was a nexus between the references and the evidence); see also United States v. Young, 470 U.S. 1, 6-7, 105 S.Ct. 1038, 84 L.Ed.2d 1 (1985); Selsor, 644 F.3d at 1024; Wilson, 536 F.3d at 1120-21. Under clearly established federal law, the OCCA could reasonably conclude that the prosecutor’s comments did not constitute reversible error, because they were sufficiently rooted in the facts of the case, not geared to incite the jury, and not, under the circumstances, sufficient to undermine the fairness of Mr. Littlejohn’s resentencing proceeding. See Spears v. Mullin, 343 F.3d 1215, 1247-48 (10th Cir.2003) (holding that “the OCCA’s decision denying [the petitioner] relief on [his] claim” based, inter alia, on the prosecutor’s argument that “justice cries out for [a conviction]” was not an unreasonable application of Supreme Court precedent and did not result in a “fundamentally unfair trial” in light of the entirety of the record); see also Bland, 459 F.3d at 1026-27 (holding that prosecutor’s comment suggesting that the jury had the ability to prevent the defendant from ever harming anyone else “c[a]me close to the line,” but did not “cross[] it,” because it could be interpreted as implying a “duty to decide whether to sentence [the defendant] to death,” not a duty to so sentence him); accord United States v. Kinsella, 622 F.3d 75, 84-85 (1st Cir.2010).

Consequently, Mr. Littlejohn’s argument must be rejected. The comments at issue were not so plainly “[and] fundamentally unfair as to deny [Mr. Littlejohn] due process.” Donnelly, 416 U.S. at 645, 94 S.Ct. 1868; see Smallwood v. Gibson, 191 F.3d 1257, 1275-76 (10th Cir.1999).

D. Absent Witnesses

Mr. Littlejohn next argues that the OCCA and district court erred in denying his claim based on a Confrontation Clause violation. Specifically, at his 1994 trial, two witnesses—Michelle Ware (Mr. Bethany’s girlfriend at the time of the murder) and Cecilia Harris (Ms. Ware’s sister)—• “offered testimony that Mr. Littlejohn made a statement implicitly admitting he was the shooter.” Aplt. Opening Br. at 56. The prosecution called them as witnesses, but ended up impeaching them in certain instances with prior inconsistent statements. Mr. Littlejohn’s trial counsel further impeached their testimony on the grounds of bias, and on multiple points of inconsistency. See, e.g., State R., Vol. V, 1994 Trial Tr. at 113 (Test, of Cecilia Harris) (showing inconsistencies in Ms. Harris’s story regarding Mr. Littlejohn’s location); id. at 98-95 (Test, of Michelle Ware) (pointing out the obvious bias in Ms. Ware’s testimony—viz., that she had a reason to testify that Mr. Bethany was not the shooter). In Mr. Littlejohn’s resen-tencing proceeding, the prosecution introduced•—over the objection of Mr. Little-john’s counsel—the 1994 testimony of Ms. Ware and Ms. Harris “by reading their testimony from the transcript of Littlejohn’s first trial.” Littlejohn II, 85 P.3d at 296.

On appeal, the OCCA determined that the Confrontation Clause required the proponent (here, the State) to make a showing of unavailability. See id. at 297. It concluded that the prosecution made, if anything, a paltry showing; consequently, the OCCA determined that the Confrontation Clause was violated. See id. (referencing Ohio v. Roberts, 448 U.S. 56, 65-66, 100 S.Ct. 2531, 65 L.Ed.2d 597 (1980)). However, it found that this error was harmless beyond a reasonable doubt, considering the factors identified by the Supreme Court in Delaware v. Van Arsdall, 475 U.S. 673, 106 S.Ct. 1431, 89 L.Ed.2d 674 (1986), stating:

The record shows the entire examination of both witnesses was read to the jury. The testimony had an adequate indicia of reliability as both Ware and Harris were thoroughly cross-examined by defense counsel. In addition, the testimony was taken under oath at Littlejohn’s first trial and was transcribed by a licensed court reporter, thereby providing an adequate record. The cross-examination of both Harris and Ware at Litt-lejohn’s first trial was conducted by the same highly competent defense attorney who represented Littlejohn at his resen-tencing trial. Defense counsel exposed the bias of both Harris and Ware and challenged the accuracy of their perceptions and memories. In addition, their testimony was not the only evidence offered to show that Littlejohn was the triggerman.... On the record before us, we find that the error was harmless beyond a reasonable doubt.

Littlejohn II, 85 P.3d at 298 (citing Chapman, 386 U.S. at 22, 87 S.Ct. 824). The district court, applying Brecht, agreed that the error “did not have a substantial and injurious effect on the jury’s verdict.” R., Vol. 1, pt. II, at 240.

Allowing Mr. Littlejohn maximum latitude in addressing his claim, we assume without deciding that the Confrontation Clause applies in capital sentencing proceedings. Cf. Wilson, 536 F.3d at 1111 (“[W]e have recently stated that it is ‘far from clear’ whether the Confrontation Clause even applies at capital sentencing proceedings.” (quoting United States v. Barrett, 496 F.3d 1079, 1099 (10th Cir.2007))). We further assume without deciding that the OCCA correctly concluded that the State’s introduction of the testimony of the two witnesses violated the

Confrontation Clause.¹²

However, “Confrontation Clause errors[ are] subject to ... harmless-error analysis.” Van Arsdall, 475 U.S. at 684, 106 S.Ct. 1431; see Coy v. Iowa, 487 U.S. 1012, 1021-22, 108 S.Ct. 2798, 101 L.Ed.2d 857 (1988); United States v. Robinson, 583 F.3d 1265, 1275-76 (10th Cir.2009); cf. Bullcoming v. New Mexico, — U.S.-, 131 S.Ct. 2705, 2719 n. 11, 180 L.Ed.2d 610 (2011) (“[W]e express no view on whether the Confrontation Clause error in this case was harmless.”). In the habeas context, as the district court recognized, Brecht must be taken into consideration; it gives the complexion of the Confrontation Clause harmless-error analysis a somewhat different shade:

When a federal court considers a Confrontation Clause violation in a habeas proceeding, the relevant harmless error analysis is whether, assuming that the damaging potential of the cross-examination were fully realized, a reviewing court might nonetheless say that the error had substantial and injurious effect or influence in determining the jury’s verdict. This court’s harmless error review is de novo.

Jones v. Gibson, 206 F.3d 946, 957 (10th Cir.2000) (citations omitted) (quoting Van Arsdall, 475 U.S. at 684, 106 S.Ct. 1431, and Brecht, 507 U.S. at 623, 637-38, 113 S.Ct. 1710) (internal quotation marks omitted).

In determining whether error was harmless in this context, we consider factors such as the “importance of the witness’ testimony in the prosecution’s case, whether the testimony was cumulative, the presence or absence of evidence corroborating or contradicting the testimony of the witness on material points, the extent of cross-examination otherwise permitted, and ... the overall strength of the prosecution’s case.” Van Arsdall, 475 U.S. at 684, 106 S.Ct. 1431; accord Jones, 206 F.3d at 957; see Wiggins v. Boyette, 635 F.3d 116, 121-22 (4th Cir.2011).

Applying these principles, we agree with the district court that the admission of the testimony at issue was harmless. See Welch, 639 F.3d at 993 (“[W]e review [in the Confrontation Clause context] only whether the admission of the testimony is harmless under the Brecht standard.”). First, as the proceeding at issue was a resentencing, much of the prosecution’s proof concerned aggravating evidence supporting the jury’s imposition of the death penalty—quite apart from evidence related to whether Mr. Littlejohn fired the fatal shot. Consequently, the contested testimony was hardly the central evidence in the prosecution’s case.¹³ See Van Arsdall, 475 U.S. at 684, 106 S.Ct. 1431; cf. Merolillo v. Yates, 663 F.3d 444, 455-56 (9th Cir.2011) (applying the first Van Arsdall factor in concluding that the testimony of a state expert witness went “straight to the heart” of the case against the petitioner because it concerned “the issue most argued by both counsel”).

Second, the State offered other competent evidence that Mr. Littlejohn was the triggerman. For instance, “[although he did not see [Mr. Littlejohn] fire the fatal shot, Tony Hulsey, the store clerk ... saw [him] with a gun as Meers approached and believed [Mr. Littlejohn] was the trigger-man.” Littlejohn II, 85 P.3d at 298. Consequently, even if the two witnesses’ “testimony was [not entirely] cumulative ... [other] evidence corroborated it].” Van Arsdall, 475 U.S. at 684, 106 S.Ct. 1431; see Wiggins, 635 F.3d at 125 (“Close examination of the record leaves us with the firm conclusion that the same evidence that makes the improperly admitted statements cumulative corroborates them.... ”); see also Perkins v. Herbert, 596 F.3d 161, 178 (2d Cir.2010); Jones, 206 F.3d at 957; Crespin v. New Mexico, 144 F.3d 641, 650 (10th Cir.1998); cf. Olden v. Kentucky, 488 U.S. 227, 233, 109 S.Ct. 480, 102 L.Ed.2d 513 (1988) (per curiam) (holding that the petitioner’s confrontation rights were violated, and that, under Van Arsdall, the violation was not harmless beyond a reasonable doubt, where it concerned the limitation of cross-examination of the prosecution’s key witness, and the only other corroborating evidence was substantially questionable); Slovik v. Yates, 556 F.3d 747, 755-56 (9th Cir.2009) (holding that a Confrontation Clause violation—• resulting from a limitation imposed on the cross-examination of a key witness— amounted to a “substantial and injurious impact on the jury’s decision” where the only potentially cumulative evidence offered was “shaky” and nonspecific). Mr. Littlejohn has pointed to no compelling reason why the jury could not credit Mr. Hulsey’s testimony, or the other evidence further corroborating it. See, e.g., R., Vol. 1, pt. II, at 239 (noting that “[additional evidence showed that all of the bullet fragments found at the scene were part of a single .380 bullet”—viz., from the type of gun Mr. Hulsey testified to seeing in Mr. Littlejohn’s possession).

Third, Mr. Littlejohn’s attorney thoroughly cross-examined the witnesses at the 1994 trial. See Littlejohn II, 85 P.3d at 298. The entire transcripts of the testimony of Ms. Ware and Ms. Harris were read into evidence, providing context for the jury to consider the deficiencies in the testimony. Furthermore, the prosecution at the 1994 trial repeatedly impeached Ms. Ware and Ms. Harris, in order to elicit answers they had previously given at a preliminary hearing that were inconsistent with the answers they offered in direct examination.

Just to provide one example, Ms. Ware testified that she did not know that money she received from Mr. Bethany came from the robbery, when in fact she had previously testified that she believed the funds were tainted. See State R., Vol. V, 1994 Trial Tr. at 84-85. Similarly, there were discrepancies between their stories. See, e.g., R., Vol. 1, pt. II, at 236 (“Ms. Ware testified that she heard Petitioner say that he did not mean to shoot Mr. Meers who came after him with a broom. Ms. Harris testified similarly, but recalled that Petitioner said that Mr. Meers had a gun in his hand.”). The prosecution’s repeated need to impeach these witnesses, along with the inherent discrepancies between their own stories, undermined the persuasive value of their testimony. These factors collectively diminished any harm that may have attended the introduction of the testimony of the two women without the opportunity for contemporaneous cross-examination of them on the witness stand. Cf. United States v. Burke, 571 F.3d 1048, 1057-58 (10th Cir.2009) (“The extent of the cross-examination [and impeachment] actually conducted ... diminishes the risk of prejudice .... ”); see also Wiggins, 635 F.3d at 121-25.

Mr. Littlejohn stresses that the Confrontation Clause provides him the right to physically face his accusers and to have the jurors assess the demeanor of the witnesses against him. See, e.g., Pennsylvania v. Ritchie, 480 U.S. 39, 51, 107 S.Ct. 989, 94 L.Ed.2d 40 (1987) (“The Confrontation Clause provides two types of protections for a criminal defendant: the right physically to face those who testify against him, and the right to conduct cross-examination.”). He contends that the district court (and the OCCA) improperly overlooked the importance of the foregoing consideration, and “rested [their decisions] largely on the proposition [that] the witnesses’ testimony ... was reliable.” Aplt. Opening Br. at 59.¹⁴

Mr. Littlejohn is correct that he has the right to physically face his accusers. And we acknowledge that the jury’s consideration of a witness’s demeanor is no doubt an important aspect of the Confrontation Clause’s protections. See, e.g., California v. Green, 399 U.S. 149, 157-58, 90 S.Ct. 1930, 26 L.Ed.2d 489 (1970) (referencing Mattox v. United States, 156 U.S. 237, 242-43, 15 S.Ct. 337, 39 L.Ed. 409 (1895)). However, we disagree that this facet of the assumed violation of Mr. Littlejohn’s Confrontation Clause rights definitively tilts the scales in favor of a finding of reversible prejudice. The admission of out-of-court testimony in violation of a defendant’s confrontation rights entails an inquiry somewhat different than that used to assess unconstitutional restrictions on the ability of the defendant to test the witness at trial. See Coy, 487 U.S. at 1020, 108 S.Ct. 2798 (noting that the Confrontation Clause encompasses numerous “implicit” rights, including “the right to cross-examine [and ] the right to exclude out-of-court statements” (citation omitted)). Specifically, the admission of out-of-court testimony in violation of the Confrontation Clause will invariably result in the inability of a jury to assess the demeanor of the declar-ant. However, that does not mean that, in such circumstances, the application of harmless-error review is a futile exercise.

Indeed, in this context, the Supreme Court has recognized that counsel’s prior ability to cross-examine an unavailable declarant in a prior proceeding that was not “significantly limited in any way” counsels in favor of a finding of no underlying violation. Green, 399 U.S. at 166, 90 S.Ct. 1930. Here, we find the same considerations found in Green weigh in favor of a finding of harmless error because Mr. Littlejohn’s counsel was not at all precluded from thoroughly cross-examining Ms. Ware and Ms. Harris at the 1994 trial.

Consequently, Mr. Littlejohn’s arguments under the Confrontation Clause must be rejected. On the facts presented, we cannot say that “[g]rave doubt” exists as to the effect of the (assumed) Confrontation Clause error or that “the matter is so evenly balanced that ... [we feel] in virtual equipoise regarding the error’s harmlessness.” Patton v. Mullin, 425 F.3d 788, 800-01 (10th Cir.2005) (quoting O’Neal, 513 U.S. at 435, 115 S.Ct. 992) (internal quotation marks omitted).

E. Impelling of Testimony

Mr. Littlejohn next claims that the re-sentencing court erred in admitting redacted portions of the testimony that he gave during the 1994 trial because he was “impelled” to testify in order to deny erroneous allegations made by a jailhouse informant. He claims that the prosecution’s use of his 1994 testimony violated Harrison v. United States, which prohibits the introduction into evidence of a defendant’s testimony that is “impelled” by the evidentiary use of the same defendant’s illegally procured confessions. 392 U.S. 219, 222, 88 S.Ct. 2008, 20 L.Ed.2d 1047 (1968); see id. at 223 n. 9, 88 S.Ct. 2008 (“[W]e decide here only a case in which the prosecution illegally introduced the defendant’s confession in evidence against him at trial in its case-in-chief.”). Harrison bars the use of such testimony of a defendant because it is deemed to be “the fruit of the poisonous tree” of his illegal confessions. Id. at 222, 88 S.Ct. 2008.

At the penalty phase of Mr. Littlejohn’s 1994 trial, the State elicited testimony from Lawrence Tingle, a “jailhouse snitch,” who claimed that, while incarcerated, Mr. Littlejohn confessed both to killing Mr. Meers and also to “hiring a hit man to kill his ex-girlfriend and their baby” in Tulsa, Oklahoma. Aplt. Opening Br. at 81. Another individual ultimately was convicted of charges relating to the murder of Mr. Littlejohn’s ex-girlfriend and baby. See id. (citing Young v. State, 992 P.2d 332 (Okla.Crim.App.1998)). On direct appeal, the OCCA found that the State had offered no “true corroborating evidence” for Mr. Tingle’s testimony. See Littlejohn I, 989 P.2d at 911. Accordingly, it found that the admission of the testimony violated Mr. Littlejohn’s due process rights and that the error was not harmless beyond a reasonable doubt in that it contributed to the jury’s finding of the continuing-threat aggravator. See id. Thus, based in part on this finding, it remanded for the resentencing proceeding, which ultimately took place in 2000. See id. at 911-12.

Mr. Littlejohn testified during the 1994 proceedings. See State R., Vol. VIII, 1994 Trial Tr., at 58-198 (Test, of Mr. Littlejohn). In a short portion of his testimony, he denied confessing anything to Mr. Tingle while in his cell. See id. at 107-08. At the 2000 resentencing, the State read much of Mr. Littlejohn’s testimony into the record, but redacted portions dealing with the Tingle incident. See State R., Vol. V, Resentencing Tr. at 799-800 (“But for the record, [the testimony] goes from page 58 to page 198 with the exception of the redacted portions.... ”). On direct appeal from the 2000 resentencing, the OCCA rejected Mr. Littlejohn’s argument that the reading of his testimony was unlawful and should result in another resen-tencing. See Littlejohn II, 85 P.3d at 298-99. Assuming arguendo that his testimony should have been suppressed under Harrison, the OCCA found that any error was harmless beyond a reasonable doubt. See id.

The district court found that Harrison was not clearly established federal law as applied to Mr. Littlejohn’s case because Harrison involved an illegally obtained confession, and otherwise dealt primarily with the fruit-of-the-poisonous-tree doctrine—concepts not at issue here. In any event, applying Brecht, it found that any error under Harrison did not have a substantial or injurious effect on Mr. Littlejohn’s resentencing trial. We agree with the district court that Harrison does not supply clearly established federal law on these facts. Consequently, we need not decide whether the admission of Mr. Littlejohn’s testimony was harmless error.

“ ‘[C]learly established Federal law1 in § 2254(d)(1) ‘refers to the holdings, as opposed to the dicta, of th[e Supreme] Court’s decisions as of the time of the relevant state-court decision.’ ” Musladin, 549 U.S. at 74, 127 S.Ct. 649 (quoting Williams, 529 U.S. at 412, 120 S.Ct. 1495). It “consists of Supreme Court holdings in cases where the facts are at least closely-related or similar to the case sub judice.” House, 527 F.3d at 1016. Whether clearly established federal law exists is a threshold question and is “analytically dispositive in the § 2254(d)(1) analysis.” Id. at 1017.

In Harrison, the prosecution introduced three confessions that the petitioner allegedly made while he was in custody. See 392 U.S. at 220, 88 S.Ct. 2008. After the admission of the confessions, the petitioner took the stand in order to offer his own version of the events in question. The jury ultimately found him guilty, but his conviction was reversed on direct appeal on the grounds that “[his] confessions had been illegally obtained and were therefore inadmissible in evidence against him.” Id. On remand, the prosecution read into evidence relevant portions of the petitioner’s testimony over the defense’s objection—testimony that inferentially incriminated the petitioner. See id. at 221, 88 S.Ct. 2008. The Supreme Court held that the prosecution’s use of the petitioner’s testimony was problematic because “[he] testified only after the Government had illegally introduced into evidence three confessions, all wrongfully obtained.” Id. at 222, 88 S.Ct. 2008. The Court held that “impelled” testimony obtained as a result of a Fifth Amendment violation (i.e., the admission of the illegally obtained confessions) was inadmissible as the “fruit of the poisonous tree.” Id. It concluded that the petitioner’s first-trial testimony was so impelled and, consequently, he was prejudiced by admission of his prior testimony. See id. at 224-26, 88 S.Ct. 2008.

By its terms, Harrison is applicable only where a defendant’s testimony is impelled by the improper use of his own unconstitutionally obtained confessions in violation of the Fifth Amendment. See id. at 222, 88 S.Ct. 2008 (“[T]he same principle that prohibits the use of confessions so procured also prohibits the use of any testimony impelled thereby—the fruit of the poisonous tree.”); see Oregon v. Elstad, 470 U.S. 298, 316-17, 105 S.Ct. 1285, 84 L.Ed.2d 222 (1985) (“If the prosecution has actually violated the defendant’s Fifth Amendment rights by introducing an inadmissible confession at trial, compelling the defendant to testify in rebuttal, the rule announced in Harrison ..., precludes use of that testimony on retrial.” (emphasis added)).

Harrison was concerned with the Fifth Amendment’s prohibition on law enforcement’s unlawful extraction of confessions from defendants. See, e.g., Elstad, 470 U.S. at 316-17, 105 S.Ct. 1285; United States v. Gianakos, 415 F.3d 912, 919 (8th Cir.2005) (noting that Harrison sought to prevent the government from benefitting from the “fruit of an illegally obtained confession” (emphasis added)); Luna v. Massachusetts, 354 F.3d 108, 112 (1st Cir.2004) (“The premise of Harrison was that the original confession (actually several confessions) had been wrongfully obtained under federal law.” (citation omitted)); cf. Cosby v. Sigler, 435 F.3d 702, 707 (7th Cir.2006) (“Since [the petitioner’s] statement was not illegally obtained, and therefore not improperly admitted, the state bears no ... burden [under Harrison]” to “show[ ] that her testimony was not compelled by the admission of the statement.” (emphasis added)).

Here, there is no contention that the State wrongfully obtained a confession from Mr. Littlejohn in violation of the Fifth Amendment—or any other constitutional amendment, for that matter. Rather, these facts only involve Mr. Littlejohn’s alleged admissions-—that were freely made, without coercion—to a private actor, Mr. Tingle. When Mr. Tingle testified about these admissions, Mr. Littlejohn allegedly felt compelled to testify to offer his side of the story. This scenario may be many things, but it is not the stuff of Harrison. See Elstad, 470 U.S. at 316-17, 105 S.Ct. 1285 (noting that “the rule announced” in Harrison applies “[i]f the prosecution has actually violated the defendant’s Fifth Amendment rights by introducing an inadmissible confession at trial, compelling the defendant to testify in rebuttal” but, on the other hand, “the Court has refused to find that a defendant who confesses, after being falsely told that his codefendant has turned State’s evidence, does so involuntarily”).

It is apparent that the rule Mr. Little-john advocates for involves the application of Harrison’s remedial measure (i.e., suppression) where a defendant’s prior testimony is impelled by an alleged due process violation. To adopt such a rule would require us inappropriately to extend Harrison to a novel context. See Premo v. Moore, — U.S.-, 131 S.Ct. 733, 743, 178 L.Ed.2d 649 (2011) (“[Njovelty ... [that] renders [a] relevant rule less than ‘clearly established’ ... provides a reason to reject it under AEDPA.”).¹⁵

Whether Harrison ever may be extended beyond its Fifth Amendment confession context is not the question before us.¹⁶ Rather, giving due deference to state court adjudications as AEDPA commands, our threshold concern must be whether Harrison’s holding furnished the OCCA with clearly established federal law to resolve Mr. Littlejohn’s argument. We answer that question in the negative.¹⁷ For that reason, we reject Mr. Littlejohn’s impelled-testimony argument.

F. Inconsistent Theories

Mr. Littlejohn argues that the prosecution inappropriately took inconsistent positions in his trial and the earlier trial of his co-defendant, Mr. Bethany. Specifically, he contends that, in their separate trials, the prosecution first pointed the finger at Mr. Bethany, and then at Mr. Littlejohn, as the person who fired the shot that killed Kenneth Meers. At the 1994 trial, the State prosecuted Mr. Littlejohn for first degree murder with malice aforethought and, alternatively, for felony murder. However, earlier in March 1993, Mr. Littlejohn claims that the same prosecutor insinuated at Mr. Bethany’s trial that he was the shooter of Mr. Meers. In particular, the prosecutor allegedly “made several arguments that Mr. Bethany ... could have had a gun,” that two witnesses across the street “said that the taller man (Mr. Bethany) was the shooter,” and that “Mr. Bethany came up with his version [of the story, i.e., blaming Mr. Littlejohn] after being warned by police that Mr. Littlejohn was almost certainly going to blame him for the shooting when asked.” Aplt. Opening Br. at 98. Mr. Bethany was ultimately convicted of felony murder and escaped the death penalty.

Mr. Littlejohn sought to prevent the State from taking what he deemed an inconsistent position in his trial. The trial judge rejected Mr. Littlejohn’s attempt and concluded that the prosecution’s theory in Mr. Bethany’s case was not inconsistent with its proposed theory in Mr. Little-john’s case. Mr. Littlejohn claimed on direct appeal that his due process rights were violated by the prosecutor’s conduct. The OCCA rejected this argument, finding that because “the evidence was less than conclusive as to the identity of the shooter,” the issue was appropriately left to the jury to determine as a factual matter. Littlejohn I, 989 P.2d at 909.

On habeas review, the district court agreed. It concluded that, at Mr. Bethany’s trial, “the prosecutor made reference to evidence which might support a finding that [he] was the shooter.” R., Vol. 1, pt. II, at 217 (emphasis added). However, she “did not advocate for [a] malice murder conviction, but repeatedly told the jurors that it was their determination to make.” Id. And, in Mr. Littlejohn’s case, while the prosecutor was “more adamant” that Mr. Littlejohn was the shooter, “the argument was still premised on the presented evidence” and did not violate Mr. Littlejohn’s “general” due process rights. Id. at 217-18. More fundamentally, the district court found that Mr. Littlejohn’s inconsistent-theories argument was not grounded in clearly established federal law—viz., no clearly established federal law would have barred the prosecution from arguing inconsistent theories.

Like the district court, we conclude that Mr. Littlejohn’s inconsistent-theories argument fails at the threshold because it is not based on clearly established federal law. The Supreme Court has had only one occasion to address a State’s use of inconsistent prosecutorial theories. It did so in 2005, after all of Mr. Littlejohn’s state proceedings had concluded. See Bradshaw v. Stumpf, 545 U.S. 175, 125 S.Ct. 2398, 162 L.Ed.2d 143 (2005). In Bradshaw, the Court considered the viability of a prosecutor’s use of inconsistent theories involving two co-defendants who were separately prosecuted. Id. at 187, 125 S.Ct. 2398. Below, the Sixth Circuit had held that the prosecution’s use of inconsistent theories as to the identity of the shooter violated due process and required the invalidation of the defendant’s guilty plea. See id. at 189, 125 S.Ct. 2398 (Souter, J., concurring). The Supreme Court disagreed. It reasoned that the change in the prosecution’s theory was made after the guilty plea and “the precise identity of the triggerman was immaterial to [the petitioner’s] conviction for aggravated murder.” Id. at 187, 125 S.Ct. 2398. The Court did observe, however, that the “use of allegedly inconsistent theories may have a more direct effect on [the petitioner’s] sentence ... for it [wa]s at least arguable that the sentencing panel’s conclusion about [his] principal role in the offense was material to its sentencing determination.” Id. It remanded for a further determination of the claim, “expressing] no opinion on whether the prosecutor’s action amounted to a due process violation, or whether any such violation would have been prejudicial.” Id.

“Before Bradshaw, the Supreme Court had not suggested that inconsistent prose-cutorial theories could constitute a due process violation.” DeCastro v. Branker, 642 F.3d 442, 457-58 (4th Cir.2011). Justice Thomas’s remarks in his Bradshaw concurrence strongly evidence this: The “[Supreme] Court has never hinted, much less held, that the Due Process Clause prevents a State from prosecuting defendants based on inconsistent theories,” 545 U.S. at 190, 125 S.Ct. 2398 (Thomas, J., concurring); accord DeCastro, 642 F.3d at 457-58; Pondexter v. Quarterman, 537 F.3d 511, 526-27 (5th Cir.2008).¹⁸ Even if it were possible to read Bradshaw’s holding as extending due process protections to facts like those before us, that case would not avail Mr. Littlejohn. As noted, Bradshaw was decided after Mr. Littlejohn’s conviction became final, and he does not suggest that Bradshaw has retroactive ef-feet. Thus, it could not constitute clearly established federal law for Mr. Littlejohn’s inconsistent-theories argument. See De-Castro, 642 F.3d at 458 (“Because Bradshaw ... post-dated the finality of Petitioner’s conviction, [it] cannot help him in his quest for habeas relief.”); cf. Thaler v. Haynes, 559 U.S. 43, 130 S.Ct. 1171, 1174 n. 2, 175 L.Ed.2d 1003 (2010) (per curiam) (noting that a certain Supreme Court decision could not be clearly established federal law for petitioner’s claim because it was decided “nearly six years after his conviction became final and more than six years after the relevant state-court decision”). Given this analysis of Bradshaw, we are primed to generally conclude that there is no clearly established federal law to which Mr. Littlejohn can tether his inconsistent-theories claim. In that event, Mr. Little-john’s claim fails at the threshold.

Mr. Littlejohn suggests, however, that Bradshaw does not delimit the universe of possibly relevant clearly established federal law, and that we may look elsewhere. Specifically, Mr. Littlejohn suggests that the Supreme Court’s decisions in, most notably, Berger v. United States, 295 U.S. 78, 55 S.Ct. 629, 79 L.Ed. 1314 (1935), Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), and Turner v. Louisiana, 379 U.S. 466, 85 S.Ct. 546, 13 L.Ed.2d 424 (1965), supply clearly established federal law for these circumstances and that the OCCA’s resolution of his inconsistent-theories claim constituted an unreasonable application of those cases. According to Mr. Littlejohn, collectively, the cases stand for the proposition that where a prosecutor engages in unfair or unjust conduct, the prosecutor may infringe upon a defendant’s due process rights. See, e.g., Aplt. Reply Br. at 32 (noting that these cases invoke “longstanding principles of ... fundamental fairness”).

However, we reject Mr. Littlejohn’s alchemic efforts to transmute the holdings of these cases into clearly established federal law for this particular factual context. Indeed, in House, we expressly noted that the Supreme Court abandoned in Musladin an approach essentially identical to the one that Mr. Littlejohn advances here, which involved “drawing] clearly established federal law from general principles teased from precedent.” House, 527 F.3d at 1015-16. We noted that “in the post-Musladin analysis, clearly established law consists of Supreme Court holdings in cases where the facts are at least closely-related or similar to the case sub judice.” Id. at 1016. We offered further clarification by stating, “Although the legal rule at issue need not have had its genesis in the closely-related or similar factual context, the Supreme Court must have expressly extended the legal rule to that context.” Id.

Post-Musladin, Mr. Littlejohn confronts an insurmountable obstacle. The cases upon which he relies may well be viewed as articulating principles of fundamental fairness, but the Supreme Court articulated those principles in distinct factual contexts that do not resemble the one before us. See Turner, 379 U.S. at 468-74, 85 S.Ct. 546 (holding that the fraternizing between jurors and key prosecution witnesses violated the defendant’s due process rights); Brady, 373 U.S. at 87, 83 S.Ct. 1194 (“We now hold that the suppression by the prosecution of evidence •favorable to an accused upon request violates due process where the evidence is material either to guilt or to punishment, irrespective of the good faith or bad faith of the prosecution.”); Berger, 295 U.S. at 84, 55 S.Ct. 629 (noting a litany of bad faith and inappropriate trial conduct on behalf of the prosecution, including “misstating the facts in ... cross-examination of witnesses,” and “suggesting ... that statements had been made ... out of court” in the absence of proof); see also Darden, 477 U.S. at 180-81, 106 S.Ct. 2464; Donnelly, 416 U.S. at 643-44, 94 S.Ct. 1868. Nor has the Supreme Court expressly extended the principles of those cases to the context of a prosecutor’s use of inconsistent theories. Accordingly, those cases cannot supply clearly established federal law for Mr. Littlejohn’s inconsistent-theories argument.

Furthermore, Mr. Littlejohn’s reliance on various extra-circuit decisions that apply the foregoing Supreme Court cases to claims predicated upon inconsistent prosecution arguments, see, e.g., Smith v. Groose, 205 F.3d 1045 (8th Cir.2000), is unavailing. Putting aside the fact that he is not even able to marshal controlling or persuasive precedent from the Tenth Circuit to support his position, the important point is that the Supreme Court has repeatedly emphasized that the AEDPA analysis should not be focused on “[circuit] precedents, rather than [on] those of th[e Supreme] Court, in assessing the reasonableness of the [state court’s] decision.” Matthews, 132 S.Ct. at 2155; see id. at 2155-56 (“It was plain and repetitive error for the Sixth Circuit to rely on its own precedents in granting [the petitioner] ha-beas relief.”). Consequently, Mr. Little-john’s reliance on scattered circuit precedent is misguided and unpersuasive.

In sum, for the reasons noted above, we conclude that Mr. Littlejohn’s inconsistent-theories argument takes a fatal fall at the threshold of clearly established federal law. No such law was available to guide the OCCA’s adjudication of his argument.

G. Ineffective Assistance at the Penalty Phase

Mr. Littlejohn alleges that his resen-tencing counsel was ineffective for failing to investigate and present evidence that experiences from his birth and childhood caused organic brain dysfunction. We conclude that, in light of the unique procedural circumstances of this case, a remand for further development of the factual record is in order.

1. Standard of Review: De novo

On September 16, 2005, Mr. Littlejohn was evaluated by Dr. Manual Saint Martin, who uncovered evidence of a potential organic brain disorder. Attorneys from the Oklahoma Indigent Defense System (“OIDS”) were appointed to represent Mr. Littlejohn in collateral state proceedings; their representation was focused on the presentation of a post-conviction motion. Twice within a few weeks OIDS filed pleadings before the OCCA, “acknowledging the 60-day [filing] constraint of OCCA Rule 9.7(G)(3) and seeking [for various reasons] additional time to file an application.” R., Vol. 1, pt. II, at 257. See generally Okla. Ct.Crim. Appeals R. 9.7(G)(3) (“No subsequent application for post-conviction relief shall be considered by this Court unless it is filed within sixty (60) days from the date the previously unavailable legal or factual basis serving as the basis for a new issue is announced or discovered.”). The OCCA, however, dismissed the matter for non-compliance with Rule 9.7(G)(3). See Littlejohn v. State (Littlejohn Post-Sentence), No. PCD-2005-1155, at 2-3 (Okla.Crim.App. Dec. 28, 2005).¹⁹ It held that Mr. Littlejohn’s attempt to present the matter to the court was noticed “on the 58th day after [he] contended] the factual basis for the claim became available.” Littlejohn Post-Sentence, at 2-3. However, it deemed this “notice” insufficient to “comply with Rule 9.7” as it could not “be construed as an application [for relief to the court].” Id. Because “[a]ny application [Mr. Littlejohn] could ultimately file in the future in compliance with Rule 9.7[ ] ... would be untimely,” the OCCA dismissed the matter. Id. at 3.

Back in federal court, the State maintained that Mr. Littlejohn failed to exhaust the claim. See R., Yol. 1, pt. II, at 258 (“Respondent contends only that the issue is unexhausted.”). Mr. Littlejohn reasoned that he could no longer exhaust the claim in light of the OCCA’s stance, see Littlejohn Post-Sentence, at 2-3-thus, exhaustion was futile. Moreover, he argued that the procedural bar in Rule 9.7, as applied to him, was inadequate to bar federal habeas review. See Anderson v. Sirmons, 476 F.3d 1131, 1140 (10th Cir.2007) (noting that procedural default bars a ha-beas application only where the rule applied is “independent” and “adequate” (quoting English v. Cody, 146 F.3d 1257, 1259 (10th Cir.1998))). The district court found that any further attempt to exhaust Mr. Littlejohn’s claim would be futile. Moreover, because the State “failed to defend the adequacy of Rule 9.7(G)(3),” the court found the rule inadequate to bar relief. R., Vol. 1, pt. II, at 258. In doing so, the court followed the path that we endorsed in Anderson, where we concluded that, where “[the State] has made no attempt, either before the district court or this court, to defend the adequacy of Rule 9.7(G)(3), ... Rule 9.7(G)(3) is not adequate to bar federal habeas review of [the petitioner’s] claim of ineffective assistance.” 476 F.3d at 1141.

Upon our review of the State’s brief, it appears that the State has (again) failed to challenge Mr. Littlejohn’s assertion that Rule 9.7(G)(3) is inadequate. The State avers in passing that Mr. Littlejohn “never exhausted this claim in the OCCA because he did not file a successor application for post-conviction relief.” Aplee. Br. at 39. This perfunctory statement— not even falling within the framework of a broader argument—falls far short of addressing Mr. Littlejohn’s contentions that exhaustion is futile in the instant circumstances and that a subsequent procedural bar cannot be “adequately” applied to him.²⁰ Nor does it contest the district court’s finding of futility or its decision to apply de novo review. Consequently, like the district court, we review Mr. Little-john’s claim on the merits. See Anderson, 476 F.3d at 1139-40; see also Neill v. Gibson, 263 F.3d 1184, 1197 (10th Cir.2001) (declining to apply a procedural default where, inter alia, the state “d[id] not continue to assert” its applicability), aff'd on reh’g 278 F.3d 1044 (10th Cir.2001); Moore, 195 F.3d at 1167 (addressing an issue on the merits where the State did not adequately argue that the claim was unexhausted or procedurally barred).

Furthermore, because we have at our disposal no merits adjudication of the claim, our standard of review is de novo. See Anderson, 476 F.3d at 1142; see also Davis v. Workman, 695 F.3d 1060, 1073 (10th Cir.2012) (“Because there has been no state-court adjudication on the merits of the claim, AEDPA’s § 2254(d) does not apply.... No prior court having addressed the merits, our review is necessarily de novo.” (citation omitted)); Le, 311 F.3d at 1011 n. 2 (“[Where] there is no adjudication on the particular claim to which deference under § 2254(d) attaches, ... we review that particular claim de novo.”); Romano v. Gibson, 239 F.3d 1156, 1166 (10th Cir.2001); Pickens, 206 F.3d at 1001; accord Breakiron v. Horn, 642 F.3d 126, 137 (3d Cir.2011); Smith v. Mitchell, 567 F.3d 246, 261 (6th Cir.2009).

2. Merits

Mr. Littlejohn contends that his counsel was ineffective for failing to connect the dots between his childhood developmental problems and the fact that he suffers from demonstrated, physiological brain damage, which Dr. Saint Martin discovered. The district court disagreed, finding that Mr. Littlejohn had not shown that he was prejudiced by his counsel’s failure to investigate and present the foregoing evidence. The district court summarily denied Mr. Littlejohn an evidentiary hearing to develop the factual basis of his claims.

We are constrained to disagree with the district court’s resolution of this claim. Under the unique procedural posture of this case, under which our review is de novo, we conclude that Mr. Littlejohn’s ineffective-assistance claim may have merit. More specifically, we determine that Mr. Littlejohn’s trial counsel may have furnished constitutionally deficient representation by failing to investigate the links between the evidence he had at his disposal and potential brain damage, and that Mr. Littlejohn may have been prejudiced as a result. See, e.g., Strickland v. Washington, 466 U.S. 668, 687-88, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984).

At this juncture, however, we would be hard-pressed to affirmatively grant Mr. Littlejohn habeas relief. His ineffective-assistance claim rests primarily on the theory of brain damage that Dr. Saint Martin posits in his declaration and supports with factual averments. As is often true of such claims, however, Mr. Littlejohn’s ineffective-assistance claim is highly fact-bound. In particular, the parties vigorously dispute the scientific persuasiveness of Dr. Saint Martin’s theory and the actual existence of an organic brain disorder. Compare, e.g., Aplt. Opening Br. at 62-63, 73-74, with Aplee. Br. at 43-44. A further exploration of the substance of Dr. Saint Martin’s findings might well reveal significant theoretical or factual holes that would make a finding of deficient performance or prejudice unsound. Consequently, at this time, we cannot conclude that Mr. Little-john is entitled to prevail on his habeas claim. However, that is not the end of the story.

We recognize the heightened care that must be employed in the death-penalty context to ensure that the qualitatively different penalty of death rests on a solid legal footing. See, e.g., Lockett v. Ohio, 438 U.S. 586, 605, 98 S.Ct. 2954, 57 L.Ed.2d 973 (1978) (plurality opinion) (noting that “the imposition of death by public authority is so profoundly different from all other penalties”). Consequently, the existence of genuine factual issues in this case should not cause us—without more— to deny Mr. Littlejohn relief, if there is an appropriate procedural mechanism available to resolve those issues. We conclude that there is such a mechanism available: an evidentiary hearing. See Rule 8 of the Rules Governing Section 2254 Cases in U.S. Dist. Courts advisory committee’s note (1976 adoption) (noting that “the function of an evidentiary hearing is to try issues of fact”); cf. Banks, 692 F.3d at 1144 n. 4 (“[A]n evidentiary hearing is not a fishing expedition. Instead, its function is to resolve disputed facts.”).

We begin with a brief discussion of the legal standards that govern whether Mr. Littlejohn may establish his entitlement to an evidentiary hearing. Then, by reference to the well-settled jurisprudence regarding ineffective-assistance claims in capital cases, we explain why Mr. Little-john has satisfied those legal standards and thus is entitled to a hearing.²¹

a. Evidentiary Hearing Under Pre-AEDPA Standards

A petitioner must satisfy the strict requirements of AEDPA to attain an evidentiary hearing if he “has failed to develop the factual basis of [his] claim in State court proceedings.” 28 U.S.C. § 2254(e)(2); see Fairchild, 579 F.3d at 1145 (noting that “[u]nder AEDPA, a federal habeas court may not grant an eviden-tiary hearing to a defendant who failed to develop his claim in state court, except in a few, narrowly defined circumstances”). Those AEDPA requirements establish a high bar for attaining an evidentiary hearing. A habeas petitioner must demonstrate his claim relies upon “a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court, that was previously unavailable,” 28 U.S.C. § 2254(e)(2)(A)(i), or “a factual predicate that could not have been previously discovered through the exercise of due diligence,” id. § 2254(e)(2)(A)(n). Accord McGee v. Higgins, 568 F.3d 832, 843 (10th Cir.2009). Furthermore, a petitioner must show that “the facts underlying the claim would be sufficient to establish by clear and convincing evidence that but for constitutional error, no reasonable factfin-der would have found the applicant guilty of the underlying offense.” 28 U.S.C. § 2254(e)(2)(B).

However, if a petitioner has not failed to exercise diligence in developing the factual basis for his claim, his request for an evidentiary hearing may be assessed under less-rigorous pre-AEDPA standards. See Coronado v. Ward, 517 F.3d 1212, 1217 n. 2 (10th Cir.2008) (“If we were to credit [petitioner’s diligence] arguments, we would assess his contention of error under pre-AEDPA standards.”); see also Williams, 529 U.S. at 437, 120 S.Ct. 1479 (“If there has been no lack of diligence at the relevant stages in the state proceedings, the prisoner has not ‘failed to develop’ the facts under § 2254(e)(2)’s opening clause, and he will be excused from showing compliance with the balance of the subsection’s requirements.”). Under those pre-AEDPA standards, a “[petitioner is entitled to an evidentiary hearing on the issue of ineffective ... counsel so long as his allegations, if true and not contravened by the existing factual record, would entitle him to habeas relief.” Young v. Sirmons (Kevin Young), 486 F.3d 655, 679 (10th Cir.2007) (alteration and ellipsis in original) (quoting Hammon v. Ward, 466 F.3d 919, 927 (10th Cir.2006)) (internal quotation marks omitted); see also Coronado, 517 F.3d at 1217 n. 2 (“Under that [pre-AEDPA] rubric, a petitioner is entitled to an evidentiary hearing so long as his allegations, if true and not contradicted by the existing record, would entitle him to habeas relief.”).

We conclude that Mr. Littlejohn’s request for an evidentiary hearing may be properly assessed under the pre-AEDPA legal regime. Arguably, Mr. Littlejohn exercised diligence in seeking to develop the factual basis for his claim and, on that basis, could be relieved of the obligation of satisfying AEDPA’s strict standards for an evidentiary hearing.²² In particular, he sought review of his ineffective-assistance claim before the OCCA with evidence from Dr. Saint Martin. The record demonstrates that Mr. Littlejohn promptly sought to file his collateral claim before the OCCA and was prepared to litigate it. Nonetheless, he was precluded from even getting to step one. As noted, the OCCA refused to permit Mr. Littlejohn to obtain the appointment of post-conviction counsel to exhaust this ineffective-assistance claim. OIDS filed numerous pleadings before the OCCA, seeking an opportunity to find Mr. Littlejohn new counsel to pursue the claim on collateral review. The OCCA refused to hear the claim and barred any potential assertion of it. See Littlejohn PosP-Sentence, at 2-3.

However, rather than resolving the merits of the diligence question, we conclude on waiver grounds that AEDPA standards do not apply. We do so, most notably, because the State has failed to specifically challenge Mr. Littlejohn’s diligence in its opening brief, but also relatedly because the State has failed to allege that the procedural bar that the OCCA applied was adequate, or that Mr. Littlejohn failed to timely raise his claims before the OCCA. Where the State fails to argue that the petitioner was less than diligent, we previously have held that a diligence challenge is waived. See Fairchild, 579 F.3d at 1145; see id. at 1145-46 (“[T]he State has not argued that [the petitioner] was less than diligent. The State did challenge [the petitioner’s] diligence in its response to the district court. But it has effectively abandoned the argument by failing to make it in its appellate brief.” (citations omitted)). Thus, guided by the well-settled jurisprudence governing the resolution of ineffective-assistance claims in capital cases, we proceed to determine whether Mr. Littlejohn is entitled to an evidentiary hearing under pre-AEDPA standards. We ultimately conclude that he is.

b. Ineffective Assistance in Capital Sentencing

Under Strickland, a petitioner “must show both that his counsel’s performance ‘fell below an objective standard of reasonableness’ and that ‘the deficient performance prejudiced the defense.’ ” Byrd v. Workman, 645 F.3d 1159, 1167 (10th Cir.2011) (emphasis omitted) (quoting Strickland, 466 U.S. at 687-88, 104 S.Ct. 2052). These two prongs may be addressed in any order, and failure to satisfy either is “dispositive.” Id. at 1168.

“Review of counsel’s performance under Strickland’s first prong is highly deferential.” Danny Hooks, 606 F.3d at 723. “Every effort must be made to evaluate the conduct from counsel’s perspective at the time.” United States v. Challoner, 583 F.3d 745, 749 (10th Cir.2009) (quoting Dever v. Kan. State Penitentiary, 36 F.3d 1531, 1537 (10th Cir.1994)) (internal quotation marks omitted). Furthermore, “counsel is strongly presumed to have rendered adequate assistance and made all significant decisions in the exercise of reasonable professional judgment.” Byrd, 645 F.3d at 1168 (alteration omitted) (quoting Dever, 36 F.3d at 1537) (internal quotation marks omitted); accord Fairchild, 579 F.3d at 1140 (‘We approach these issues with ‘a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance,’ and that ‘the challenged action might be considered sound trial strategy.’ ” (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052)).

However, while we entertain “a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance,” Matthews, 577 F.3d at 1190 (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052) (internal quotation marks omitted), we nevertheless apply “closer scrutiny when reviewing attorney performance during the sentencing phase of a capital case,” Cooks v. Ward, 165 F.3d 1283, 1294 (10th Cir.1998); see also Osborn v. Shillinger, 861 F.2d 612, 626 n. 12 (10th Cir.1988) (“[T]he minimized state interest in finality when resentencing alone is the remedy, combined with the acute interest of a defendant facing death, justify a court’s closer scrutiny of attorney performance at the sentencing phase.”); cf. Wellons v. Hall, 558 U.S. 220, 130 S.Ct. 727, 728, 175 L.Ed.2d 684 (2010) (per curiam) (“From beginning to end, judicial proceedings conducted for the purpose of deciding whether a defendant shall be put to death must be conducted with dignity and respect.”).

Counsel must perform in accordance with “prevailing professional norms.” Young v. Sirmons (Julius Young), 551 F.3d 942, 956-57 (10th Cir.2008) (quoting Wiggins, 539 U.S. at 523, 123 S.Ct. 2527). In capital cases, we refer to the ABA Guidelines for the Appointment and Performance of Counsel in Death Penalty Cases (“ABA Guidelines”) in assessing those “professional norms.” Id. at 957. Generally, “[a]mong the topics defense counsel should investigate and consider presenting include medical history, educational history, employment and training history, family and social history, prior adult and juvenile correctional experiences, and religious and cultural influences.” Id. (emphasis added).

Counsel must conduct a “thorough investigation—in particular, of mental health evidence—in preparation for the sentencing phase of a capital trial.” Wilson, 536 F.3d at 1083. “We recently had occasion to expound on this principle, drawing on a trilogy of Supreme Court cases—Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000), Wiggins v. Smith, 539 U.S. 510, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003), and Rompilla v. Beard, 545 U.S. 374, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005)—involving ineffective assistance at capital-sentencing proceedings.” Victor Hooks, 689 F.3d at 1201 (referring to the discussion in Wilson, 536 F.3d at 1084-85). We set forth “three important principles” that are derived from these cases:

First, the question is not whether counsel did something; counsel must conduct a full investigation and pursue reasonable leads when they become evident. Second, to determine what is reasonable investigation, courts must look first to the ABA guidelines, which serve as reference points for what is acceptable preparation for the mitigation phase of a capital case. Finally, because of the crucial mitigating role that evidence of a poor upbringing or mental health problems can have in the sentencing phase, defense counsel must pursue this avenue of investigation with due diligence.

Wilson, 536 F.3d at 1084-85 (emphasis added) (internal quotation marks omitted) (citations omitted); see id. at 1085 (noting that “[o]ur own Circuit has emphasized this [due-diligence] guiding principle”).

Moreover, in Victor Hooks, we underscored the importance of a specific type of mental-health evidence—that is, evidence relating to organic or physical brain injury:

Evidence of organic brain damage is something that we and other courts, including the Supreme Court, have found to have a powerful mitigating effect.... And for good reason—the involuntary physical alteration of brain structures, with its attendant effects on behavior, tends to diminish moral culpability, altering the causal relationship between impulse and action.

See Victor Hooks, 689 F.3d at 1205 (citations omitted). Because of its central significance, where the defendant’s circumstances put it in play, ordinarily it would be “patently unreasonable for [counsel] to omit this evidence from his case for mitigation.”²³ Smith v. Mullin, 379 F.3d 919, 942 (10th Cir.2004); see id. (noting “evidence of [petitioner’s] mental retardation, brain damage, and troubled background constituted mitigating evidence” and, indeed, is “exactly the sort of evidence that garners the most sympathy from jurors”).

If counsel’s performance at sentencing was deficient, we must then assess whether the petitioner was prejudiced as a result. See Strickland, 466 U.S. at 694, 104 S.Ct. 2052. Prejudice means “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Id. “To assess prejudice arising out of counsel’s errors at a capital-sentencing proceeding, we must ‘reweigh the evidence in aggravation against the totality of available mitigating evidence.’ ” Victor Hooks, 689 F.3d at 1202 (quoting Julius Young, 551 F.3d at 960). “If there is a reasonable probability that at least one juror would have struck a different balance, ... prejudice is shown.” Id. (citations omitted) (quoting Wiggins, 539 U.S. at 537, 123 S.Ct. 2527) (internal quotation marks omitted).

Furthermore, “[bjecause [the instant] ... claim was not decided on the merits by the OCCA, and because it is not procedurally barred, federal habeas review of the claim is de novo.” Anderson, 476 F.3d at 1142 (emphasis omitted) (citing Torres v. Lytle, 461 F.3d 1303, 1311 (10th Cir.2006)). Thus, AEDPA’s deferential provisions do not apply. See id.

i. Performance

Reviewing the matter of counsel’s performance de novo, we conclude that, if the facts validate Mr. Littlejohn’s averments, then we would find that his counsel’s representation was constitutionally deficient. At sentencing, it is true that Mr. Littlejohn’s counsel presented a mitigation case. That case involved the testimony of Wanda Draper, Ph.D. Dr. Draper held “a PhD in development,” “a masters of science in child development,” and had done postdoctoral work in genetic epistomology. State R., Vol. VI, Resentencing Tr. at 78 (Test, of Wanda Draper). She presented a socio-psychological report on Mr. Little-john, concluding essentially that his development “stifled around age ten,” R., Vol. 1, pt. II, at 261, and that, while he understood the difference between right and wrong, he lacked “sensitivity” to it, State R., Vol. VI, Resentencing Tr. at 98. Further, evidence was offered detailing various aspects of Mr. Littlejohn’s “troubled” childhood, including testimony indicating that he “had difficulty making friends and knowing appropriate boundaries,” R., Vol. 1, pt. II, at 261, and that he had a rough home life, id. at 260-61.

Furthermore, the jury heard testimony suggesting that Mr. Littlejohn’s mother used narcotics during the duration of her pregnancy with him. See id. at 260. Indeed, his mother, Ceily Mason, opined that he “had to be ruined from the womb ... [because she] took so much dope till ... [she] delivered, [she] didn’t even know [she] delivered till the next day.” State R., Vol. VI, Resentencing Tr. at 45 (Test. of Ceily Mason). Dr. Draper talked about the effect of a mother’s substance abuse on a fetus. See id. at 89-92 (noting that substance abuse “is considered to be a very significant intrusion on the development of the fetus”).

As noted, however, in connection with post-conviction proceedings, Mr. Littlejohn secured the services of Dr. Saint Martin, a psychiatrist, who conducted a thorough examination of him. Dr. Saint Martin is “licensed to perform psychological and neuropsychological evaluations” in several states. R., Vol. 1, pt. I, at 170 (Deck of Dr. Saint Martin, dated Sept. 26, 2005). In a declaration presented in Mr. Littlejohn’s federal habeas proceeding, Dr. Saint Martin suggested that “Mr. Littlejohn ... suffer[s from] a behavioral disorder manifested by poor impulse control, psychological immaturity and judgment that is caused by neuro-developmental deficits experienced in his peri-natal development,” specifically his mother’s drug abuse. Id. at 171 (emphasis added); see id. (“[T]he evaluation is consistent with neurological deficits caused by insults to Mr. Littlejohn’s developing brain through his mother’s drug abuse pre-nataly and through neglect post-nataly.”). Specifically, he identified “deficits ... existing] in the microscopic structure and neuro-chemical function of the brain” that did not “merely [involve] sociological or psychological considerations.” Id. at 172. As noted, such mental-health evidence is “of vital importance to the jury’s decision at the punishment phase.” Smith, 379 F.3d at 942 (quoting ABA Guidelines §§ 1.1, 4.1, 10.4, 10.7, and 10.11) (internal quotation marks omitted).

While counsel is afforded a great amount of deference in presenting a defense, see Danny Hooks, 606 F.3d at 723, the failure to investigate a compelling mitigation theory can constitute ineffective assistance, see Anderson, 476 F.3d at 1145. “[W]e focus on whether the investigation supporting counsel’s decision not to introduce mitigating evidence ... was itself reasonable.” Wiggins, 539 U.S. at 523, 123 S.Ct. 2527. And, in making this determination, we cannot rely on “hindsight.” Rompilla, 545 U.S. at 381, 125 S.Ct. 2456. Instead, we examine the reasonableness of “counsel’s perspective at the time investigative decisions are made.” Id. (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052) (internal quotation marks omitted).

In this case, while counsel presented some mitigation evidence—relating to Mr. Littlejohn’s socioeconomic and psychological development—there were numerous indicators suggesting that a neurological evaluation could have uncovered evidence of organic brain damage. For instance, in Mr. Littlejohn’s retroactive competency proceedings in state court, Dr. Glenn Barry Robbins testified that he believed that Mr. Littlejohn had “neurological injury [originating] from birth.” State R., Vol. II, Competency Tr. at 13 (Test. of Dr. Glenn B. Robbins). He noted:

We’re talking probably minor cognitive problems that it demonstrated in his academics, and that could have—could be interpreted as poor impulse control or] attention problems----And we see this a lot in head injured patients who are normal individuals. And once you injure the brain cortex, they become aggressive.

Id. (emphasis added). Moreover, Dr. Draper’s testimony inferentially suggested that physical damage could be inflicted on a fetus due to a mother’s drug abuse, see id., Vol. VI, Resentencing Tr. at 89-91, and Mr. Littlejohn’s mother, Ms. Mason, confirmed her drug use during pregnancy, see id. at 45.

“[Reasonably diligent counsel may draw a line when they have good reason to think further investigation would be a waste.” Rompilla, 545 U.S. at 383, 125 S.Ct. 2456. However, counsel is obliged to follow up on leads that are readily identifiable in the evidence. See Wilson, 536 F.3d at 1088 (finding an investigation unreasonable where, despite the fact that “information [relating to the mitigation case] was easily within counsel’s reach,” he never availed himself of that information); see also Anderson, 476 F.3d at 1145 (“The Supreme Court has squarely rejected the notion that, when counsel has ‘some information with respect to petitioner’s background,’ counsel has necessarily fulfilled his constitutional duty to investigate and present a case in mitigation.” (quoting Wiggins, 539 U.S. at 527, 123 S.Ct. 2527)). In this case, there was readily available evidence that should have alerted a reasonable defense counsel to the realistic possibility that Mr. Littlejohn suffered from organic brain damage. Although we do not attempt to make any definitive con-elusion on the matter at this juncture and on this record, there is a very real likelihood that the decision of Mr. Littlejohn’s counsel to focus solely on mitigating evidence relating to the socioeconomic conditions of Mr. Littlejohn’s upbringing and to his psychological makeup—without even investigating whether there was some physical, brain-related condition that would account for Mr. Littlejohn’s behavior—amounted to constitutionally deficient performance.

The State disagrees, suggesting that Dr. Saint Martin’s theory is spurious. Presumably, in the State’s view, it logically follows that reasonable counsel should not have been obliged to investigate and develop evidence related to the theory. Specifically, the State notes that Dr. Saint Martin admittedly relied on evidence that Dr. Stephen Carella offered at Mr. Littlejohn’s 1997 competency proceedings and found that, based on that information, there was no indication that Mr. Littlejohn has “gross damage to his frontal lobes.” R., Vol. 1, pt. I, at 171-72. However, Dr. Saint Martin also opined in his declaration that the physiological problems with Mr. Littlejohn’s brain “are more global and ... distinct from the types of deficits Dr. Carella’s testing was designed to detect.” Id. at 172. Mr. Littlejohn suffers, according to Dr. Saint Martin, from a neurological deficit “caused by insults to [his] developing brain through his mother’s drug abuse pre-nataly and through neglect post-nataly.” Id. at 171. These problems occur at the synapse level between all brain cells, and impair the global, physical development of the brain.

Thus, the State is simply not correct, in suggesting that “gross” damage of the frontal lobe is necessary for a finding of organic damage, according to Dr. Saint Martin’s theory. To the contrary, that theory, if supported and validated by the evidence, could go far in offering a scientifically supported and physical link to Mr. Littlejohn’s crime and “developmental history.” Aplt. Opening Br. at 63; see Porter v. McCollum, 558 U.S. 30, 130 S.Ct. 447, 453, 175 L.Ed.2d 398 (2009) (per curiam) (holding that “[c]ounsel[’s] ... failure] to uncover and present any evidence of [the petitioner’s] mental health or mental impairment,” among other evidence, “did not reflect reasonable professional judgment”); see also Ferrell v. Hall, 640 F.3d 1199, 1233 (11th Cir.2011); Wilson, 536 F.3d at 1092-93; cf. Wackerly v. Workman, 580 F.3d 1171, 1181 (10th Cir.2009) (rejecting the petitioner’s claim that counsel was ineffective for failing to introduce evidence of an organic brain disorder because he proffered “no evidence” that any life event “resulted in any lasting brain damage” and other evidence clearly “suggested] that any injuries [he] suffered had little impact on his mental capacity”).

In sum, there were patent “red flags” in this case, pointing to a possible physiological explanation for Mr. Littlejohn’s violent and anti-social behavior. Dr. Saint Martin’s declaration suggests that in fact Mr. Littlejohn suffered from physical brain damage. As we have noted, it is well-settled that such evidence is of considerable importance to a capital sentencing jury. See, e.g., Smith, 379 F.3d at 942 (noting that this type of evidence is “exactly the sort of evidence that garners the most sympathy from jurors”). Thus, a critical factor becomes whether Dr. Saint Martin’s declaration averments—if developed and probed through the adversary process—would prove worthy of belief. If so, then we would be hard-pressed to conclude that Mr. Littlejohn’s counsel was not constitutionally deficient in failing to follow up on the red flags.

ii. Prejudice

The prejudice inquiry in a capital sentencing case requires a showing that “there is a reasonable probability that, absent [counsel’s] errors, the sentencer ... would have concluded that the balance of aggravating and mitigating circumstances did not warrant death.” Knighton v. Mullin, 293 F.3d 1165, 1178 (10th Cir.2002) (ellipsis in original) (quoting Strickland, 466 U.S. at 695, 104 S.Ct. 2052) (internal quotation marks omitted). In making the latter determination, we must consider “the strength of the State’s case and the number of aggravating factors the jury found to exist, as well as the mitigating evidence the defense did offer and any additional mitigating evidence it could have offered.” Id. “If there is a reasonable probability that at least one juror would have struck a different balance, ... prejudice is shown.” Victor Hooks, 689 F.3d at 1202 (citations omitted) (quoting Wiggins, 539 U.S. at 537, 123 S.Ct. 2527) (internal quotation marks omitted). The Supreme Court has confirmed that prejudice may be found even where “counsel presented what could be described as a superficially reasonable mitigation theory during the penalty phase.” Sears v. Upton, — U.S. -, 130 S.Ct. 3259, 3266, 177 L.Ed.2d 1025 (2010) (per curiam).

On de novo review, we must disagree with the district court’s resolution of the prejudice question. It is undisputed that Mr. Littlejohn has a long history of criminal conduct, including serious violent acts. The district court summarized that evidence:

Petitioner himself relayed his criminal history, [noting that] he learned to hot wire cars at the age of 15 and stole countless cars before being institutionalized in a juvenile facility. Released at age 18, Petitioner committed a robbery just two weeks later. Armed with an Uzi, Petitioner shot at his victim several times before hitting him in the head with the Uzi and taking his money. Petitioner also burglarized an automobile, and because he committed these crimes as an adult, he was sent to the penitentiary. Because of his bad behavior, [he] served almost all of his 24-month sentence .... [After being released in 1992], [he] started selling dope. He and Bethany robbed the Root-N-Scoot ... on June 20,1992.

R., Vol. 1, pt. II, at 263-64 (footnote omitted) (citation omitted). However, Dr. Saint Martin’s declaration suggests at least a partial explanation for it. His history possibly evinces a problem with “impulse control” and a compulsive lack of “judgment” primarily during his youth. See id., pt. I, at 171.

Evidence of organic mental deficits ranks among the most powerful types of mitigation evidence available. See, e.g., Wilson, 536 F.3d at 1083-85. “Counsel in capital cases must explain to the jury why a defendant may have acted as he did— must connect the dots between, on the one hand, a defendant’s mental problems, life circumstances, and personal history and, on the other, his commission of the crime in question.” Victor Hooks, 689 F.3d at 1204 (emphasis added); see id. (collecting cases).

Evidence that an organic brain disorder was a substantial factor in engendering Mr. Littlejohn’s life of deviance probably would have been a significant favorable input for Mr. Littlejohn in the jury’s deci-sionmaking calculus. And, under the particular circumstances of this case, there is a reasonable probability that such evidence would have led at least one juror to support a sentence less than death. Yet, here the jury received virtually no explanation of how Mr. Littlejohn’s alleged mental problems played into the murder. And without this explanation, the prosecution was able to frame the mitigation defense as a mere collection of the social circumstances of Mr. Littlejohn’s upbringing— circumstances that, while unfortunate, do not excuse murder. In this regard, the prosecutor stated, “It is unfortunate that children are raised in [rough] environment[s], but it doesn’t make them killers. Choices make people killers.” State R., Vol. VII, Resentencing Tr., at 346.

The potential prejudice flowing from this omission was heightened in these circumstances by the fact that a considerable portion of the State’s case in aggravation related to the continuing-threat aggravator. Evidence that Mr. Littlejohn possessed physical, neurological deficits of the type suggested by Dr. Saint Martin’s declaration averments would have offered a less blameworthy explanation of Mr. Litt-lejohn’s extensive criminal history. Furthermore, such evidence could have strongly militated against a conclusion that, even given that criminal history, Mr. Littlejohn was actually a continuing threat. Specifically, Dr. Saint Martin opined that, while the “deficits ... are irreparable, ... they are treatable.” R., Vol. 1, pt. I, at 172-73 (emphasis added); see id. at 173 (“[D]rug therapy is available to control the behavior and diminish the impulsivity, which creates most of [Mr. Littlejohn’s] problems in interacting with society.”).

This kind of evidence could have been used to powerful mitigating effect, indicating that Mr. Littlejohn’s criminal past is a product of a physical condition that is treatable, such that his criminal past is not an accurate predictor of his future.²⁴ That is, it could have indicated to a jury that Mr. Littlejohn was not a continuing threat. See Victor Hooks, 689 F.3d at 1205 (“Diagnoses of specific mental illnesses ..., which are associated with abnormalities of the brain and can be treated with appropriate medication, are likely to [be] regarded by a jury as more mitigating than generalized personality disorders.... ” (alterations in original) (quoting Wilson, 536 F.3d at 1094) (internal quotation marks omitted)); cf. Gilson v. Simons, 520 F.3d 1196, 1249-50 (10th Cir.2008) (holding that evidence of organic brain disorder—seemingly with no evidence as to possible treatment—would not have altered the jury’s prejudice analysis because the “presentation of th[e] evidence would likely have weighed against [the petitioner] by erasing any lingering doubts that may have existed as to his role in [the underlying] murder, and by confirming the jury’s conclusion that he represented a continuing threat, even if confined in prison for life”).

The jury that sentenced Mr. Littlejohn to death, however, was not offered anything of the sort through the testimony of Dr. Draper. Indeed, she suggested that Mr. Littlejohn did “[not] have any kind of mental illness,” State R., Vol. VI, Resentencing Tr. at 133, and that his developmental problems could not necessarily be cured by medication, see id. at 103. Dr. Draper did testify regarding the “very significant intrusion,” id. at 89, on fetal development that could have been caused by the drug abuse of Mr. Littlejohn’s mother. This testimony suggested the possibility that Mr. Littlejohn suffered physical brain damage—a possibility that we have noted supra that reasonably competent counsel would have investigated. And her testimony bears slight superficial resemblance to some of Dr. Saint Martin’s declaration averments regarding the implications for Mr. Littlejohn of the substance abuse of his mother. However, it is critical to note that Dr. Draper did not offer any opinion regarding whether Mr. Littlejohn in fact suffered pre-natal brain injuries and, indeed, she would not have been equipped to do so. Dr. Draper was not a psychiatrist—like Dr. Saint Martin—or any other type of physician, for that matter. Therefore, she could not offer the jury any reliable and persuasive evidence on the question of whether Mr. Littlejohn suffered from an organic brain injury that could adversely affect his behavior.²⁵

Absent any evidence indicating that Mr. Littlejohn’s past criminal conduct was due to physical causes that were treatable, the prosecution was free to look askance at the socioeconomic and psychological mitigating evidence that Mr. Littlejohn presented and argue that he was “violent” and “always [has] been.” Id., Vol. VII, Resentencing Tr. at 346; see, e.g., id. at 276 (“Does the fact that he was born to a drug addicted mother in some way reduce his accountability for his actions in 1992[?]”).

It is true, of course, that we have previously found a lack of prejudice in some cases where counsel failed to present additional mental-health evidence in a capital sentencing proceeding. See, e.g., DeRosa, 679 F.3d at 1219-21 (holding that the petitioner had not established that any potential deficiency in counsel’s performance prejudiced him because, inter alia, the evidence presented adequately painted a detailed picture of him); Knighton, 293 F.3d at 1177-80 (finding that defense eoun-sel’s mitigation theory was extensive, and even though no evidence of organic brain damage was presented, there “[was] no[] ... reasonable probability that, had defense counsel presented ... [such evidence] the jury would have imposed a sentence less than death”); Humphreys, 261 F.3d at 1021 (finding no prejudice where counsel put on a reasonable mitigation defense, and additional evidence that “he ... suffered brain damage” would not have altered the verdict of death).

However, DeRosa, Knighton and Humphreys, are distinguishable. First, the latter two cases were issued before the Supreme Court’s decisions in Rompilla and Sears—cases that emphasized the need for courts to consider the prejudicial effect of counsel’s failure to investigate a viable mitigation theory even in the face of an otherwise reasonable mitigation defense. See, e.g., Sears, 130 S.Ct. at 3266. Moreover, in all three cases, the aggravating evidence was more serious than in Mr. Little-john’s case, where only two aggravating factors were found. See, e.g., DeRosa, 679 F.3d at 1219-21 (detailing the extensively brutal nature of the underlying murder which served as a basis for a “heinous, atrocious, or cruel aggravator”); Knighton, 293 F.3d at 1178 (noting that the jury found three aggravating circumstances— (1) prior felony convictions, (2) “continuing threat,” and (3) the creation of a “great risk of death to more than one person”— the factual basis of which included, inter alia, an instance where the defendant broke into an innocent family’s home and held the residents hostage for an extended period of time); Humphreys, 261 F.3d at 1019 (noting that the jury found three aggravating factors). And, as noted already, the continuing-threat aggravator-—which was the focus of much of the State’s case— could have been diminished significantly by evidence of a treatable organic brain disorder. Thus, these cases do not stand in the way of a conclusion that Mr. Little-john has suffered prejudice.

3. Conclusion

In sum, under a de novo standard,²⁶ we conclude that Mr. Littlejohn has alleged a mitigation theory and supporting facts which, if true, would entitle him to relief under Strickland-viz., would justify us in concluding that his counsel was constitutionally deficient in failing to investigate and put on mitigating evidence concerning Mr. Littlejohn’s claimed physical brain injury and that, but for that failure, there is a reasonable probability that the jury would have selected a penalty less than death. Consequently, at this juncture, we are constrained to conclude that the district court erred in finding—as a matter of law—that there was no reasonable probability that at least one juror could have decided against death, in light of counsel’s failure to investigate and present a possible defense based on organic brain damage. We conclude that this case should be remanded to the district court to conduct an evidentiary hearing (and other appropriate proceedings) concerning Mr. Little-john’s ineffective-assistance claim.

H. Cumulative Error

Mr. Littlejohn argues that the cumulative effect of error in this case warrants habeas relief. The OCCA, however, summarily rejected Mr. Littlejohn’s cumulative error claim, finding that the errors “did not deprive [Mr.] Littlejohn of a fair resentencing trial.” Littlejohn II, 85 P.3d at 303; see Short v. Sirmons, 472 F.3d 1177, 1197-98 (10th Cir.2006). The district court essentially agreed. Specifically, it found, at most, five errors—only three of which were relevant to its analysis because they related to Mr. Littlejohn’s resentenc-ing proceeding.²⁷ See R., Vol. 1, pt. II, at 255. The three errors that it considered were: (1) the prosecution’s improper introduction of prejudicial evidence of Mr. Meers’ testimony at the resentencing without providing adequate notice; (2) the prosecution’s presentation of the transcript of the two witnesses—Ms. Ware and Ms. Harris—without a showing of their unavailability, in violation of Mr. Littlejohn’s Sixth Amendment confrontation rights; and (3) the State’s improper introduction of transcripts of Mr. Littlejohn’s own testimony at the 1994 trial. Applying AEDPA deference, the court determined that the OCCA’s decision was neither contrary to, nor an unreasonable application of, any clearly established federal law.

“In the federal habeas context, the only otherwise harmless errors that can be aggregated are federal constitutional errors, and such errors will suffice to permit relief under cumulative error doctrine ‘only when the constitutional errors committed in the state court trial so fatally infected the trial that they violated the trial’s fundamental fairness.’ ” Matthews, 577 F.3d at 1195 n. 10 (quoting Julius Young, 551 F.3d at 972). Mr. Littlejohn argues that the OCCA and the district court erred in finding that the errors in this case were cumulatively harmless. He contends that the magnitude of the error is underscored by the fact that the jury clearly was considering a sentence less than death in this case.

In light of our decision to remand this case, however, we cannot definitively resolve Mr. Littlejohn’s cumulative error claim, or any of the related arguments raised thereunder.²⁸ This is so because, after an evidentiary hearing and other necessary proceedings upon remand, the district court may deem it appropriate to include the allegedly constitutionally deficient performance of Mr. Littlejohn’s counsel in the cumulative-error calculus. See Cargle, 317 F.3d at 1224-25; Short, 472 F.3d at 1197-98; cf. Spears, 343 F.3d at 1251 (considering the cumulative impact of prejudice of ineffective-assistance claims). Thus, we believe that the best course of action is to permit the district court to revisit the cumulative-error claim after conducting an evidentiary hearing on Mr. Littlejohn’s ineffective-assistance claim. See Part II.G.3.

To be sure, the State challenges the threshold ability of this court to even consider this issue. It argues that such a claim is not cognizable under habeas review because there is no clearly established federal law on cumulative error. However, we need not (and do not) definitively resolve this question here.²⁹ In our view, it is the prudent path to permit the district court to consider this claim fully (and arguments related to it) on remand after conducting an evidentiary hearing.

III. Conclusion

In light of the foregoing, we AFFIRM the district court’s judgment on all grounds except for Mr. Littlejohn’s ineffective-assistance claim (Proposition Five) and his cumulative-error claim (Proposition Seven). As to the ineffective-assistance claim, we REVERSE the judgment and REMAND the case to the district court, with directions to conduct an eviden-tiary hearing and any further appropriate proceedings consistent with this opinion. In this regard, as to the cumulative-error claim, we direct the district court to VACATE that portion of its judgment upon remand and reconsider and rule on that claim, in light of its resolution of Mr. Litt-lejohn’s ineffective-assistance claim.

Notes

1 . In the alternative, the State had charged Mr. Littlejohn with felony murder. See Okla. Stat. tit. 21 § 701.7(B).

2 . Later in 2005, Mr. Littlejohn attempted to initiate another state post-conviction motion to exhaust this claim, see R., Vol. 1, pt. II, at 72 (Mot. for Extension of Time to File Appl. for Post-Conviction Relief, filed Dec. 5, 2005), but "[n]o [formal] application was ever filed and the OCCA dismissed the action,” id. at 209.

3 . We acknowledge the Supreme Court’s recent pronouncement in Parker v. Matthews, - U.S. -, 132 S.Ct. 2148, 2155, 183 L.Ed.2d 32 (2012) (per curiam), that the circuit courts may not dispositively apply "[their] own precedents, rather than those of th[e Supreme] Court, in assessing the reasonableness of [a state court's] decision [under AEDPA].” We may, however, consult the precedent of lower courts (including, of course, our own) in order to ascertain the contours of clearly established Supreme Court precedent. See Mollett, 348 F.3d at 913 (suggesting that we may consider lower court case law—including our own application of Supreme Court precedent—in order to determine whether a state court’s decision was contrary to a decision of the Supreme Court); see also id. ("[T]his circuit’s decision in Johnson is relevant because, although not controlling, on-point 'federal case law inferior to Supreme Court precedent, may serve as a guide in determining reasonableness of [a] state court’s application of Supreme Court law.’ ” (second alteration in original) (quoting Bryson v. Ward, 187 F.3d 1193, 1205 (10th Cir.1999))). Here, we apply our "four-pronged” inquiry in Mollett with the assumption that it represents a contextually specific embodiment of Simmons and its progeny. In this AEDPA setting, it is only a non-binding guidepost. Were we to conclude that a state court's resolution of a jury-instruction challenge in a capital case was neither contrary to nor an unreasonable application of Simmons—based upon our best understanding of the import of Simmons, gleaned from its text and our survey of lower-court precedent— then any possible deviation of that state court decision from Mollett's four-prong test would be immaterial. However, because Mr. Little-john cannot make a showing that the OCCA's decision ran afoul of any conceivable, reasonable formulation of Simmons's mandate, we are content to assume that, on these facts, Mollett helps illuminate the correct contours of the applicable law. See Mollett, 348 F.3d at 913.

[4] 4. Mr. Littlejohn questions whether even these instructions would have been sufficient “to comply with the Simmons line of cases.” Aplt. Opening Br. at 28 n.4. However, he submits that they would be a "substantial improvement over what [he] was afforded.” Id.

5 . Mr. Littlejohn claims that he relies on "fundamental principles announced in several other ... Supreme Court decisions ... [suggesting] that the jury must understand the sentencing options before it[ ] and have accurate and complete information to consider when performing a capital sentencing function.” Aplt. Reply Br. at 2 (referencing Walton v. Arizona, 497 U.S. 639, 110 S.Ct. 3047, 111 L.Ed.2d 511 (1990) (plurality opinion), overruled on other grounds by Ring v. Arizona, 536 U.S. 584, 589, 122 S.Ct. 2428, 153 L.Ed.2d 556 (2002); Boyde, 494 U.S. 370, 110 S.Ct. 1190; Mills v. Maryland, 486 U.S. 367, 108 S.Ct. 1860, 100 L.Ed.2d 384 (1988); Gardner v. Florida, 430 U.S. 349, 97 S.Ct. 1197, 51 L.Ed.2d 393 (1977) (plurality opinion); Skipper v. South Carolina, 476 U.S. 1, 106 S.Ct. 1669, 90 L.Ed.2d 1 (1986); Crane v. Kentucky, 476 U.S. 683, 106 S.Ct. 2142, 90 L.Ed.2d 636 (1986)). We would be hard-pressed to conclude that these cases collectively amount to clearly established federal law for this particular factual context. See House, 527 F.3d at 1016. These cases are concerned primarily with instances where the jury is instructed in a manner that inappropriately limits its ability to consider certain evidence, see, e.g., Boyde, 494 U.S. at 380, 110 S.Ct. 1190 ("We think the proper inquiry in such a case is whether there is a reasonable likelihood that the jury has applied the challenged instruction in a way that prevents the consideration of constitutionally relevant evidence.”); see also Rojem v. Gibson, 245 F.3d 1130, 1136-37 (10th Cir.2001), and the defendant's ability to explain evidence to the jury, see Gardner, 430 U.S. at 356, 97 S.Ct. 1197 ("There was ... no ... opportunity for petitioner’s counsel to challenge the accuracy or materiality of ... information [that may have been material].”). Mr. Littlejohn has not demonstrated that the noted authorities involve even remotely comparable factual circumstances. Therefore, they would not have supplied clearly established federal law for the OCCA’s resolution of Mr. Littlejohn’s claim of instructional error. Thus, insofar as Mr. Littlejohn’s argument for relief is predicated on these authorities, it must fail at the threshold. See House, 527 F.3d at 1015 ("Whether the law is clearly established is the threshold question under § 2254(d)(1).”). Indeed, as we explicate at greater length in another setting infra, in House, we recognized that the mode of analysis underlying Mr. Littlejohn's argument—that involves "drawfing] clearly established federal law from general principles teased from precedent,” House, 527 F.3d at 1015-16—was expressly abandoned by the Supreme Court in Carey v. Musladin, 549 U.S. 70, 127 S.Ct. 649, 166 L.Ed.2d 482 (2006).

6 . The court reporter “contemporaneously” recorded Mr. Littlejohn saying, “Fuck you punk. He’s still dead, he ain’t coming back.” Aplt. Opening Br. at 35 (quoting State R., Vol. VIII, 1994 Trial Tr. at 308 (internal quotation marks omitted)). According to the OCCA, "[w]ithout the threat to [Mr.] Meers, the recorded statement [of the court reporter] likely would have been ruled inadmissible.” Little-john II, 85 P.3d at 295. Indeed, there is yet another account of what transpired. In the State’s supplemental notice of aggravating evidence, the words instead were spoken to the victim’s mother, Delores Meers, and were, "Too bad, but your son is still dead!” Id. at 294 (quoting Original R., Vol. IX, at 1682 (internal quotation marks omitted)).

7 . We do not find persuasive Mr. Littlejohn’s suggestion that the OCCA (and the district court, by extension) failed to consider the totality of the circumstances in assessing the prejudicial impact of the lack of notice. In our view, the OCCA and the district court properly considered the entirety of the facts and the record in finding a lack of prejudice, not just, for instance, the "sufficiency of the evidence for the 'continuing threat' aggravator.” Aplt. Opening Br. at 44; see Kennedy v. Lockyer, 379 F.3d 1041, 1054 n. 15 (9th Cir.2004) ("We are obligated to conduct [harmless-error] review on the basis of the 'record as a whole.' " (quoting Brecht, 507 U.S. at 638, 113 S.Ct. 1710)). Under Brecht, "a ha-beas petitioner obtains plenaiy review to determine whether a trial error resulted in actual prejudice." Welch, 639 F.3d at 992 (quoting Brecht, 507 U.S. at 637, 113 S.Ct. 1710) (internal quotation marks omitted).

8 . From a prosecutorial-misconduct perspective, this comment was particularly problematic, reasons Mr. Littlejohn, because, in earlier proceedings, the same prosecutor had informed the court that she heard the version of Mr. Littlejohn’s statement that was recorded by the court reporter—which, significantly, did not contain a threat to Mr. Meers.

9 . Mr. Littlejohn relies primarily on our precedent in Paxton v. Ward, 199 F.3d 1197 (10th Cir.1999). There, we rejected "the state’s contention that the appropriate inquiry is whether the prosecutor's argument denied [the petitioner] his right to a fundamentally fair sentencing proceeding.” Id. at 1217. Instead, on quite different facts, we reasoned: "We agree with the district court that the misconduct which undisputedly occurred here [involving a prosecutor’s misleading and deceptive argument] was an integral part of the deprivation of [the petitioner's] constitutional rights to present mitigating evidence, to rebut evidence and argument used against him, and to confront and cross-examine the state’s witnesses.” Id. at 1218. It goes without saying, however, that Paxton cannot supply clearly established federal law to support Mr. Littlejohn’s claim. See, e.g., House, 527 F.3d at 1016 ("[C]learly established law consists of Supreme Court holdings in cases where the facts are at least closely-related or similar to the case sub judice."). Furthermore, under factual circumstances such as these—where the prosecutor's statements were not admitted into evidence—Donnelly itself indicates that Mr. Littlejohn's argument is a non-starter. See 416 U.S. at 643 n. 15, 94 S.Ct. 1868 ("Respondent does suggest that the prosecutor's statements may have deprived him of the right to confrontation. But this argument is without merit, for the prosecutor here simply stated his own opinions and introduced no statements made by persons unavailable for questioning at trial.” (citation omitted)).

10 . This is particularly true in light of the fact that the prejudicial thrust of the question is subject to some fairminded debate. Specifically, there is a dispute as to whether the prosecutor was attempting to imply that she heard Mr. Littlejohn make the statement. The State contends that the "whole point of the question,” in context, "was patently to explain why [Mr.] Meers did not feel the need to [immediately] report the threat since there were other witnesses present.” Aplee. Br. at 26-27. The State suggests that its interpretation is bolstered by the fact that the question was raised on re-direct examination after Mr. Littlejohn’s counsel questioned why Mr. Meers failed to raise the claim earlier. See State R., Vol. VI, Resentencing Tr. at 24 (Cross-examination of Mr. Meers). Mr. Little-john of course reasons that the question “created a false impression for the jury” that the prosecutor was there and heard the content of the statement. See Aplt. Reply Br. at 9. But, regardless of which view is correct, we do not think that the statement was susceptible only to Mr. Littlejohn’s interpretation. Thus, Mr. Littlejohn has not demonstrated that "the state court’s ruling on the claim [in context] ... was so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement." Richter, 131 S.Ct. at 786-87 (emphases added).

11 . At Mr. Bethany’s trial, “the same prosecutor” from Mr. Littlejohn’s case presented the State’s theory in a manner that permitted the jury to infer that Mr. Bethany fired the fatal shot that killed Mr. Meers. See Part II.F infra. Mr. Littlejohn's counsel introduced portions of the transcript of Mr. Bethany’s trial in an attempt to convince the jury that the State had argued in the respective trials of Mr. Bethany and Mr. Littlejohn that the defendant on trial was the shooter.

12 . In reaching its conclusion that the State violated the Confrontation Clause, the OCCA relied on the Supreme Court’s decision in Roberts, 448 U.S. at 65-66, 100 S.Ct. 2531. See Littlejohn II, 85 P.3d at 297. Following the OCCA’s ruling, the Supreme Court issued its landmark Confrontation Clause decision, Crawford v. Washington, 541 U.S. 36, 124 S.Ct. 1354, 158 L.Ed.2d 177 (2004). The Court recently discussed Crawford's departure from the analytical framework of Roberts:

Before Crawford, this Court took the view that the Confrontation Clause did not bar the admission of an out-of-court statement that fell within a firmly rooted exception to the hearsay rule, see [Roberts, 448 U.S. at 66, 100 S.Ct. 2531], but in Crawford, the Court adopted a fundamentally new interpretation of the confrontation right, holding that "[t]estimonial statements of witnesses absent from trial [can be] admitted only where the declarant is unavailable, and only where the defendant has had a prior opportunity to cross-examine.”

Williams v. Illinois, -U.S. -, 132 S.Ct. 2221, 2232, 183 L.Ed.2d 89 (2012) (plurality opinion) (second and third alteration in original) (quoting Crawford, 541 U.S. at 59, 124 S.Ct. 1354). In fortifying his position here, Mr. Littlejohn places some limited reliance on Crawford. See Aplt. Opening Br. at 57 (noting that the Supreme Court "recently confirmed” in Crawford that "the Confrontation Clause ... required precisely what the defense requested in this case: the live presentation of the witnesses”). The State objects, claiming that Mr. Littlejohn’s reliance on Crawford is impermissible because that decision does not apply retroactively. It is true that Crawford is not "retroactive to cases already final on direct review.” Whorton v. Bockting, 549 U.S. 406, 409, 127 S.Ct. 1173, 167 L.Ed.2d 1 (2007). However, in light of the approach that we take to resolving this claim, we see no need to pause to definitively opine on the implications of Crawford’s non-retroactivity for Mr. Littlejohn’s case; the parties’ dispute amounts to little more than a proverbial tempest in a teapot. Like the district court, we recognize that the OCCA resolved this claim on the merits and also, as noted infra, we proceed to rule in the State’s favor on harmless-error grounds. The propriety of this de-cisional path depends in the first instance not on whether Roberts or Crawford is clearly established federal law, but on whether there is clearly established federal law at all, such that AEDPA’s threshold requirement for ruling on the claim is satisfied. See Victor Hooks, 689 F.3d at 1163 ("Under § 2254(d)(1), the threshold question is whether there exists clearly established federal law....”); cf. infra note 17 (where we focus on the absence of clearly established law, even though the OCCA assumed the applicability of federal precedent and found error). Even assuming arguendo that Roberts supplied the clearly established federal law (and Crawford did not), there still was clearly established federal law for the OCCA to apply and our going forward to adjudicate the claim therefore would be proper. Indeed, Mr. Litt-lejohn may have understood this. See Reply Br. at 15 (noting that the State’s objection is "not an important point” given that Roberts is applicable). Once the threshold requirement is satisfied, under our approach, the focus is on the de novo application of harmless-error standards. At least on the specific facts presented here, the differing substantive standards of Roberts and Crawford for determining whether, as an antecedent matter, the Confrontation Clause has been violated are largely immaterial.

13 . This discussion tracks the last Van Arsdall factor—that is, "the overall strength of the prosecution’s case." Van Arsdall, 475 U.S. at 684, 106 S.Ct. 1431. As discussed, the prosecution presented a substantial amount of aggravating evidence. See Littlejohn II, 85 P.3d at 296. That fact would support the conclusion that the evidence at issue was secondary to the primary focus of its case. In disagreement, relying on the jury's question concerning the death-penalty instructions, see Part II.A, supra, Mr. Littlejohn contends that this was a "close case ... where the jurors were seriously considering a sentence less than death,” Aplt. Opening Br. at 59, and thus, the "mere existence of aggravating factors cannot and does not reasonably establish [harmlessness],” Aplt. Reply Br. at 14. Putting aside our doubts about Mr. Littlejohn’s conjectural inference about the jury's view of the purported closeness of the case, we do not disagree that the strength of the prosecution’s case does not, standing alone, establish harmlessness. However, we may nevertheless consider it as a factor under the totality of the circumstances. See Van Arsdall, 475 U.S. at 684, 106 S.Ct. 1431. In this context, the prosecution’s "case” related to whether the evidence established the propriety of the imposition of the death penalty. And, dispositive or not, Duvall, 139 F.3d at 789-90 ("Under Oklahoma law, a jury is free to decline to impose the death penalty even if it finds that the aggravating circumstances outweigh the mitigating circumstances.”), aggravating evidence is obviously central to the prosecution's case. Furthermore, it is worth noting that such evidence in support of the death penalty did not necessarily need to focus on whether Mr. Littlejohn was the shooter. Cf. Banks, 692 F.3d at 1141 ("[C]ase law has made clear that capital punishment for felony murder charges is both constitutional and not infrequently imposed when the defendant was present during the murder and acted with reckless disregard for human life.”).

14 . Mr. Littlejohn also argues that the district court framed the prejudice inquiry the wrong way because it focused on whether "the prosecution’s failure to make a showing of unavailability prior to the admission of Ms. Ware and Ms. Harris’s testimony” had a prejudicial effect on the verdict. R., Vol. 1, pt. II, at 240 (emphasis added). However, as the State correctly points out, Mr. Littlejohn "makes too much out of the court's use of a shorthand reference to his claim.” Aplee. Br. at 36. The district court clearly addressed whether admission of the testimony without "confrontation” substantially affected the jury’s verdict. See, e.g., R., Vol. 1, pt. II, at 239-40. That is all that Brecht required in this case.

15 .We acknowledge that our decision in Humphreys v. Gibson, 261 F.3d 1016 (10th Cir.2001), may at first blush indicate that Harrison has broader applicability than is suggested here. In Humphreys, the petitioner objected to the admission at resentencing of his own testimony from the guilt stage of his initial trial. See id. at 1023. He argued that he was compelled to testify after the trial court erred in "refusing] to instruct on first degree manslaughter,” his separate theory of the case. Id. However, the OCCA held that the trial court did not err in refusing the instruction. Id. We concluded that, to the extent that the petitioner was "impelled” to testify, he was not "wrongfully impelled.” Id. However, in referencing Harrison, we did not hold that it was clearly established federal law. See id. To the contrary, we noted that "Harrison does not directly address [the petitioner’s] situation,” id. (citing Elstad, 470 U.S. at 316-17, 105 S.Ct. 1285), but nonetheless (in the alternative) we considered his argument because it was without merit, see id. In any event, the petitioner in Humphreys clearly framed his claim to establish that the prosecution’s use of his testimony "violated his Fifth Amendment privilege against self-incrimination.” Id. This is substantially different than Mr. Littlejohn’s argument: that the reading of his testimony violated his due process rights and that the holding of Harrison is sufficiently broad to outlaw it. For these reasons, Humphreys does not advance Mr. Littlejohn’s contention that Harrison's holding is clearly established federal law beyond the Fifth Amendment confession context.

16 . The OCCA in this very case questioned whether Harrison could be limited to the Fifth Amendment context. See Littlejohn II, 85 P.3d at 298 ("We are not convinced Hanison can be so easily limited because in both instances [the Fifth Amendment and due process contexts], the defendant is complaining that he was, more or less, forced to testify because the State used his inadmissible confession against him.”).

17 . Mr. Littlejohn suggests in the alternative that, "since the state court at least assumed Hanison applied, [§ ] 2254(d) is not triggered by its failure to apply Hanison at all, but [only] by its unreasonable harmless error analysis.” Aplt. Opening Br. at 85. In other words, under Mr. Littlejohn’s view, because the OCCA assumed that Harrison was applicable and that there was error under it, AEDPA is not implicated regarding the substance of the constitutional challenge and, consequently, we have no occasion to examine the threshold question of whether Harrison is clearly established federal law, but rather should merely look at whether the OCCA unreasonably applied federal principles of harmless error. We disagree.

We are obliged to examine the state court's adjudication of the claim as a whole in order to ascertain whether it is "on the merits” and, if so, whether there is clearly established federal law for the claim and whether the state court’s "reasoning []or the result ... contradicts” such federal law. Early v. Packer, 537 U.S. 3, 8, 123 S.Ct. 362, 154 L.Ed.2d 263 (2002) (per curiam); cf. Aycox v. Lytle, 196 F.3d 1174, 1177 (10th Cir.1999) (applying AEDPA deference in the face of a summary decision where ”[t]here [wa]s no evidence ... that the state court did not consider and reach the merits” of the petitioner’s claim). Where a state court assumes a constitutional violation in order to address whether the defendant was actually harmed by the violation, as here, the state court takes the claim on the merits; it just disposes of it on alternative merits-based reasoning. Cf. Brown v. Luebbers, 371 F.3d 458, 462-63 (8th Cir.2004) (holding that AEDPA deference applied to a state court's alternative holding where it assumed a trial error and applied harmless error review under Chapman). That is, it renders a decision that is on the merits for purposes of AEDPA, see Richter, 131 S.Ct. at 784; therefore, our inquiry must adhere to the analytical framework of AEDPA, which includes an assay into whether there is clearly established federal law.

18 . Mr. Littlejohn disagrees with this view of the state of Supreme Court precedent. He points to Justice Souter’s concurrence in Bradshaw, which opined that Justice Stevens “observ[ed] 10 years [prior] that 'serious questions are raised when the sovereign itself takes inconsistent positions in two separate criminal proceedings against two of its citizens,’ and that '[t]he heightened need for reliability in capital cases only underscores the gravity of those questions.’ ” 545 U.S. at 189, 125 S.Ct. 2398 (Souter, J., concurring) (third alteration in original) (quoting Jacobs v. Scott, 513 U.S. 1067, 1070, 115 S.Ct. 711, 130 L.Ed.2d 618 (1995) (Stevens, J., dissenting from the denial of certiorari)). However, Justice Stevens’s "on-point” comment was made in a dissent from the denial of certiorari. This hardly contradicts Justice Thomas's observation regarding the actual holdings of the Supreme Court's opinions.

19 . This decision is attached in slip-opinion form to Mr. Littlejohn's opening brief as Attachment F. Citations herein are to the slip opinion.

20 . Here, as in Anderson, we have a “definitive ruling from the state court that it will not review on the merits a successor application from [Mr. Littlejohn] raising a claim that trial counsel was ineffective.” 476 F.3d at 1139. Mr. Littlejohn need not attempt a subsequent application for relief before the OCCA; in light of its order dismissing his action, such an application would amount to "a meaningless and utterly futile act to properly exhaust his state court remedies.” Id.; see Selsor, 644 F.3d at 1026; Fairchild v. Workman, 579 F.3d 1134, 1155 (10th Cir.2009). The State points to no other mechanism by which Mr. Littlejohn could otherwise properly exhaust his ineffective-assistance claim.

21 . Because the OCCA did not adjudicate Mr. Littlejohn's ineffective-assistance claim on the merits, so long as he qualifies for one, a federal evidentiary hearing may be used to develop the factual basis for his claim and the district court (and our court) may consider the evidence produced in the hearing in assessing the merits of his claim, without running afoul of the Supreme Court's recent decision in Cullen, which held that "review under § 2254(d)(1) is limited to the record that was before the state court that adjudicated the claim on the merits." 131 S.Ct. at 1398 (emphasis added); see Black, 682 F.3d at 895 (noting that "even if a federal-court evidentiary hearing is not barred by § 2254(e)(2), the evidence so obtained is inadmissible in reviewing a claim adjudicated on the merits in state court”). Compare Davis, 695 F.3d at 1072 (“In the alternative, Defendant requests an evidentiary hearing to present additional evidence that his statements should have been excluded. But we have already considered all the evidence that he presented to the Oklahoma courts, and under AEDPA our review of the OCCA decision on this issue must be confined to the state-court record.”), with id. at 1076 (“Because the OCCA did not address this coercion issue on the merits and § 2254(d) therefore is inapplicable, evidence from a federal-court evidentiary hearing could be considered in resolving the issue.”).

22 . See Williams v. Taylor, 529 U.S. 420, 443-44, 120 S.Ct. 1479, 146 L.Ed.2d 435 (2000) ("As state postconviction relief was no longer available at the time the facts came to light, it would have been futile for petitioner to return to [state] court[]. In these circumstances, though the state courts did not have an opportunity to consider the new claims, petitioner cannot be said to have failed to develop them in state court by reason of having neglected to pursue remedies available under [state] law.”); see also Wilson, 536 F.3d at 1079 (concluding that the petitioner acted diligently in seeking an evidentiary hearing on an issue supported by non-record evidence by requesting it before the OCCA, even though the OCCA declined to consider the non-record evidence); James v. Ryan, 679 F.3d 780, 804 (9th Cir.2012) (concluding that there was no lack of diligence under § 2254(e)(2) where the state court failed to address the petitioner’s claim on the merits and the petitioner requested an evidentiary hearing in state court); 1 Randy Hertz & James S. Liebman, Federal Habeas Corpus Practice and Procedure § 20.2[b], at 1030 & n.22 (6th ed.2011) (suggesting that a petitioner acts "diligently” under § 2254(e)(2) when a state is "responsible for ... [the failure to] develop[]” the facts in the face of an attempt by the petitioner to seek redress).

23 . Contrary to the Dissent's suggestion, we do not "add another layer of responsibility,” Dissent at 869, on to defense counsel with respect to the investigation and presentation of evidence of organic brain damage. Instead, as our cases like Victor Hooks clearly recognize, where there are credible, reasonably dis-cernable clues that a capital defendant's circumstances will support a mitigation theory based on organic brain damage, it is at the core of a defense counsel’s constitutional responsibilities to conduct a reasonable investigation into the existence of evidence to validate or bolster such a theory and, ordinarily, to present such evidence to the jury, if it is found. Given the powerful mitigative effect of such evidence, reasonably competent counsel would not have settled for some mitigation case based on mental-health evidence such as presented here, regarding largely behavioral abnormalities, if the organic-brain-damage option was available; not only is this behavioral-abnormality evidence different in kind from evidence concerning organic brain damage, but, critically, in most instances it also will be of significantly lesser mitigative value.

24 . The Dissent's assertion that "Dr. Saint Martin's testimony was a classic double-edged sword,” Dissent at 877, is widely off the mark. Instead, such testimony would have offered a physiological explanation for Mr. Littlejohn’s deviant conduct and some assurance that, through medical treatments, his criminal, violent past would not be prologue.

25 . Perhaps the central flaw of the Dissent’s analysis of both Strickland's performance and prejudice prongs lies in its failure to recognize that Dr. Draper simply was not equipped by professional training or experience to offer testimony of the kind reflected in Dr. Saint Martin’s declaration. As for performance, contrary to the Dissent's view, see Dissent at 872, counsel could not reasonably have concluded that Dr. Saint Martin's testimony would be cumulative of Dr. Draper’s. Dr. Draper could not have offered expert testimony regarding whether Mr. Littlejohn's deviant behavior was rooted in physiological deficits of his brain (that is, organic brain damage). In other words, Dr. Draper could not have testified regarding a subject that is well-recognized to have powerful mitigative effect, whereas Dr. Saint Martin's declaration makes patent that he was professionally qualified to speak to this subject. Highlighting its astigmatism, the Dissent bolsters its performance argument by noting that there is no suggestion that "Dr. Draper misdiagnosed [Mr. Littlejohn] or otherwise erred in her evaluation.” Id. at 871. But that is hardly the point: even if Dr. Draper's diagnosis of Mr. Littlejohn had been spol-on in all particulars, she lacked the professional training and experience to offer the kind of diagnosis of physical causes and possible behavioral effects that we have found to be present in Dr. Saint Martin's testimony. The substantive limitations of Dr. Draper’s testimony are further underscored by the Dissent's worrisome suggestion that it would have been constitutionally adequate performance for Mr. Littlejohn's counsel to rely on the jury's "common sense,” id. at 874, regarding the harmful pre-natal effects of maternal substance abuse, because Dr. Draper was available to connect the dots to Mr. Little-john's circumstances. However, as it relates to the particulars of the physiological damage that such substance abuse produced in Mr. Littlejohn and its likely negative behavioral effects, Dr. Draper actually was not competent to speak. Indeed, the Dissent concedes that Dr. Draper was unable to explain "in scientific terms” "the physical mechanism that may have inflicted prenatal injury” on Mr. Littlejohn. Id. at 875 (emphasis added). Dr. Saint Martin, on other hand, certainly was qualified to undertake this task. For similar reasons, on the question of prejudice, we cannot accept the Dissent's suggestion that it is "unlikely” that the kind of testimony that Dr. Saint Martin is seemingly prepared to offer “would have made any difference,” id. at 877. That suggestion is predicated on the Dissent's oft-expressed view that “Dr. Saint Martin's testimony would have created, at best, a marginal increase in the amount of information available,” id. at 875, and that "Dr. Saint Martin's evaluation resulted in a similar, if not virtually identical, diagnosis as Dr. Draper's,” id. at 872. Put bluntly, the Dissent is simply wrong and the record does not support its view.

26 . We pause to underscore the unique procedural posture of this case. We are not obliged here to defer to a state-court adjudication of the ineffective-assistance claim. It is certainly possible—and perhaps likely—that we would have reached a different conclusion, if reviewing under the deferential standards of AEDPA a decision of the OCCA rejecting on the merits Mr. Littlejohn’s claim of ineffective assistance. See, e.g., Humphreys, 261 F.3d at 1018-21 (holding that the state court's decision that counsel was not ineffective for failing to put forth additional evidence in mitigation—including evidence of “brain damage”—"was not contrary to, nor an unreasonable application of, clearly established Supreme Court precedent”). However, those standards of review are not in play here because we have no merits decision from the OCCA. Furthermore, the absence of such a merits decision also opens the door in this case for a possible federal evidentiary hearing to develop the factual basis of Mr. Littlejohn's claim—a door that the Supreme Court firmly shut in Cullen where the state court has ruled on the merits of the federal claim. See Cullen, 131 S.Ct. at 1398; see also supra note 21.

27 . Because Mr. Littlejohn's “death sentence was determined in a resentencing proceeding before a different jury,” the district court found "no first stage carryover concern.” R., Vol. 1, pt. II, at 255.

28 . For that reason, we need not address Mr. Littlejohn’s argument that the OCCA applied the wrong standard when it noted that the errors were “harmless in the aggregate” and “did not deprive Littlejohn of a fair resentencing.” Littlejohn II, 85 P.3d at 303.

29 . We have noted a divergence between circuits on this issue. See Victor Hooks, 689 F.3d at 1194 n. 24 (comparing, inter alia, Williams v. Anderson, 460 F.3d 789, 816 (6th Cir.2006), with Parle v. Runnels, 505 F.3d 922, 928 (9th Cir.2007)). But “[a]lthough we have never expressly held ... that cumulative-error analysis is clearly established federal law, we have long conducted cumulative-error analyses in our review of federal habeas claims.” Id.; see, e.g., Banks, 692 F.3d at 1150-51; Davis, 695 F.3d at 1082; Thackerv. Workman, 678 F.3d 820, 849 (10th Cir.2012); Matthews, 577 F.3d at 1195 n. 10.

---

Concurrence Opinion

TYMKOVICH, Circuit Judge,

concurring and dissenting in part.

I agree with the majority on all issues except its conclusion Littlejohn’s counsel may have failed to meet Strickland requirements for ineffective assistance of counsel. In my view, counsel’s performance was neither deficient nor prejudicial. But more fundamentally, I disagree with the majority’s conclusion that counsel’s failure to develop additional neurological evidence'—even when a constitutionally adequate mental health mitigation defense was presented at trial—requires habeas relief. I therefore respectfully dissent.

To show ineffective assistance of counsel, Littlejohn “has the twofold burden of establishing that (1) defense counsel’s performance was deficient, i.e., counsel’s ‘representation fell below an objective standard of reasonableness’ as measured by ‘prevailing professional norms,’ and (2) defendant was prejudiced thereby, i.e., ‘there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.’ ” United States v. Rushin, 642 F.3d 1299, 1302 (10th Cir.2011) (quoting Strickland v. Washington, 466 U.S. 668, 688, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984)).

Counsel’s performance in this case was neither deficient nor prejudicial.

As part of the mitigation trial, Little-john’s counsel undertook the following tasks: (1) conducted an adequate investigation of mitigating factors, (2) presented Littlejohn’s mental defects to the jury through expert and lay witnesses, and (3) explained what may have caused those mental defects and how they contributed to the commission of murder. Littlejohn and the majority would add another layer of responsibility—counsel has a constitutional obligation to articulate from a neurological/psychiatric perspective additional potential physical brain damage.

Strickland imposes no such requirement. While applicable Supreme Court and Tenth Circuit precedent requires capital defense counsel to present evidence relating to a defendant’s mental health where relevant, none go so far as to enshrine the neuro-psychiatric approach now advocated by the majority opinion as the only acceptable lens through which such evidence may be presented. In my view, Littlejohn’s counsel conducted an adequate mental health investigation and elected to present the results through the expert testimony of a developmental psychologist. This choice was well within the bounds of tactical discretion the law gives defense counsel in determining whether and how to present mitigation evidence.

But even assuming for the sake of argument that counsel’s performance was deficient, Littlejohn was not prejudiced by the error. The affidavit submitted by the post-conviction psychiatrist, Dr. Saint Martin, who evaluated Littlejohn for purposes of his habeas petition, offers only the mundane observation that Littlejohn exhibits symptoms “consistent with” a behavioral disorder “caused by neuro-developmental deficits experienced in his perinatal development.” R., Vol. I, pt. 1, at 170, 172. Dr. Saint Martin underscores the ambiguity of his observations with a disclaimer that “the specifics of these ‘wiring’ problems are not yet well understood.” Id. at 171. While suggestive of another approach to mental health evidence, his observations do not fundamentally undercut the approach actually taken. Moreover, such testimony, if believed, would constitute a double-edged sword—while it might reduce Littlejohn’s moral culpability in the eyes of a jury, it could strengthen a jury’s concerns about future dangerousness, since it would suggest Littlejohn could not control his behavior.

I.

Counsel’s performance was not deficient

The majority identifies two flaws in Litt-lejohn’s counsel’s performance. First, it finds counsel did not conduct an adequate investigation of Littlejohn’s mental health pathologies. Second, and more alarmingly, it concludes the failure to develop additional mental health evidence linked to “organic brain damage” was inherently both deficient and prejudicial.

I disagree on both points.

Defense counsel in a death penalty sentencing trial is required to conduct a reasonable investigation of mitigating evidence. Porter v. McCollum, 558 U.S. 30, 130 S.Ct. 447, 453, 175 L.Ed.2d 398 (2009). What constitutes a reasonable investigation is judged by “prevailing professional norms.” Rompilla v. Beard, 545 U.S. 374, 380, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005). “In judging the defense’s investigation, as in applying Strickland generally, hindsight is discounted by pegging adequacy to ‘counsel’s perspective at the time’ investigative decisions are made, and by giving a ‘heavy measure of deference to counsel’s judgments.’” Id. at 381, 125 S.Ct. 2456 (citations omitted) (quoting Strickland, 466 U.S. at 689, 691, 104 S.Ct. 2052). Basic expectations for such an investigation include reviewing prison, educational, and medical records, as well as attempting to contact and interview family members. See id.; Bobby v. Van Hook, 558 U.S. 4, 130 S.Ct. 13, 18, 175 L.Ed.2d 255 (2009). Where there are signs that a defendant may suffer from a mental defect, a mental health evaluation is generally required. See Anderson v. Sirmons, 476 F.3d 1131, 1143 (10th Cir.2007). But it is not our job to second guess a reasonable mitigation strategy, even where the passage of time or the retention of new experts suggest better ways to package a mental health defense.¹

Applying the teachings of the Supreme Court, I find counsel’s investigation was adequate. First of all, counsel hired Dr. Wanda Draper, a former professor at the University of Oklahoma School of Medicine. Dr. Draper, a psychologist with a doctorate in child development, interviewed Littlejohn and reviewed his court records, prison records, medical records, and school records in preparation for her testimony. She met personally with Litt-lejohn for a mental-health evaluation and determined that he was of average intelligence and was not mentally ill. She did determine, however, that Littlejohn suffered from a behavioral disorder that made it difficult for him to fully control his behavior in some circumstances. She also interviewed Littlejohn’s mother, grandmother, and sister to gain a better understanding of the circumstances of Little-john’s upbringing. There is no evidence that this investigation was rushed or that Dr. Draper was precluded from developing any relevant evidence.

Nevertheless, Littlejohn and the majority argue the investigation was inadequate because counsel did not specifically order a psychiatric or neurological evaluation. The cases they rely on for this argument, however, do not support the notion that a psychological evaluation, such as that conducted by Dr. Draper, is constitutionally insufficient. In those cases, defense counsel procured no mental health evaluation whatsoever. See, e.g., Sears v. Upton, — U.S. -, 130 S.Ct. 3259, 3264, 177 L.Ed.2d 1025 (2010) (“[T]he cursory nature of counsel’s investigation into mitigation evidence—limited to one day or less, talking to witnesses selected by [Sears’s] mother—was on its face ... constitutionally inadequate.” (alteration in original) (quotation marks omitted)); Porter, 130 S.Ct. at 453 (2009) (“Counsel thus failed to uncover and present any evidence of Porter’s mental health or mental impairment, his family background, or his military service.”); Anderson, 476 F.3d at 1143 (“Anderson was not evaluated by any mental health or other expert qualified to ascertain whether Anderson suffered from neurological or other deficits that would mitigate his moral culpability.” (emphasis added)). Here, Dr. Draper’s investigation was far more comprehensive and thorough.

While there may be some mental conditions for which a psychological evaluation—as opposed to a psychiatric evaluation-—is insufficient, this is not one of those cases. Littlejohn does not argue that Dr. Draper misdiagnosed him or otherwise erred in her evaluation. To the contrary, Dr. Saint Martin arrived at a nearly identical diagnosis: that Littlejohn suffered from “a behavioral disorder.” Thus, this is not a case where a mental health diagnosis, which should have been discovered, was overlooked due to an unqualified evaluator or time constraints. See Dunlap v. Clements, 476 Fed.Appx. 162, 166-67 (10th Cir.2012) (unpublished) (“On occasion, the Supreme Court has condemned an attorney’s failure to thoroughly investigate sentencing-phase mitigation evidence as constitutionally deficient assistance. But.... [t]hose cases involve inexcusable neglect.”); cf. Wilson v. Sirmons, 536 F.3d 1064, 1085 (10th Cir.2008) (finding constitutional error where an adequate investigation could have allowed an evaluator to confirm a diagnosis of schizophrenia).

Of course, it is possible that Littlejohn’s counsel could have procured an evaluation from Dr. Draper and a psychiatrist like Dr. Saint Martin. The law recognizes, however, that at some point additional investigation results in diminishing returns. See Strickland, 466 U.S. at 699, 104 S.Ct. 2052 (“Counsel’s strategy choice was well within the range of professionally reasonable judgments, and the decision not to seek more character or psychological evidence than was already in hand was likewise reasonable.”). In any given case, “there [are] any number of hypothetical experts ... whose insight might possibly have been useful.... Counsel [is] entitled to formulate a strategy that [is] reasonable at the time and to balance limited resources in accord with effective trial tactics and strategies.” Harrington v. Richter, — U.S. -, 131 S.Ct. 770, 789, 178 L.Ed.2d 624 (2011); see also Dunlap, 476 Fed.Appx. at 167 (“Lawyers often disagree on trial strategy, a fact disregarded in the vast number of ineffectiveness claims.”). Here, Littlejohn’s counsel reasonably could have concluded that additional mental health evaluations could “be expected to be only cumulative, and ... distractive from more important duties.” Bobby, 130 S.Ct. at 19. And, in fact, this conclusion is borne out by the evidence before us: Dr. Saint Martin’s evaluation resulted in a similar, if not virtually identical, diagnosis as Dr. Draper’s.

In the end, I cannot' agree that the failure to develop additional evidence of organic brain injury was required in this case. “There are ... countless ways to provide effective assistance in any given case. Even the best criminal defense attorneys would not defend a particular client in the same way. Rare are the situations in which the ‘wide latitude counsel must have in making tactical decisions’ will be limited to any one technique or approach.” Harrington, 131 S.Ct. at 788-89 (citation omitted) (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052). Therefore, courts apply “a ‘strong presumption’ that counsel’s attention to certain issues to the exclusion of others reflects trial tactics rather than ‘sheer neglect.’ ” Id. at 790 (quoting Yarborough v. Gentry, 540 U.S. 1, 8, 124 S.Ct. 1, 157 L.Ed.2d 1 (2003) (per curiam)). In particular, “the decision of which witnesses to call is quintessentially a matter of strategy for the trial attorney.” Boyle v. McKune, 544 F.3d 1132, 1139 (10th Cir.2008). This inquiry into the reasonableness of counsel’s decisions is an objective one; even if additional evidence could have supported a defense argument, courts are not supposed to find ineffective assistance on that basis if they “conclude that a competent attorney might elect not to use it.” Harrington, 131 S.Ct. at 789 (emphasis added).²

In the course of her evaluation, Dr. Draper found that Littlejohn did not suffer from a severe mental disorder. She also found that he was of average intelligence. But, she did diagnose him with a behavioral disorder, which she characterized as an “emotional disturbance” that created a reduced capacity to control his actions and empathize with others. Dr. Draper surmised that this emotional disturbance was a result of Littlejohn’s early life experiences, including his mother’s prenatal substance abuse, as well as neglect by his mother, father, and grandmother. Counsel decided to humanize Littlejohn with this evidence in the hope of generating enough sympathy from the jury to negate the government’s case for death.

At trial, Dr. Draper testified that Little-john’s behavioral development was negatively impacted by several factors. First, she testified that his mother’s substance abuse during pregnancy negatively impacted his development, though she did not explain at length the physical mechanism by which this occurred. She then walked through various stages of Littlejohn’s upbringing, and explained in detail how the severe neglect he experienced resulted in an inability to trust others and control his impulses. On cross-examination, she admitted Littlejohn was not mentally ill, but diagnosed his condition as “emotional disturbance.” 2000 Tr., Vol. VI, at 133. She also admitted he was able to “distinguish between reality and fantasy.” Id. While Dr. Draper’s testimony focused more on Littlejohn’s upbringing than on his exposure to prenatal substance abuse, she did clearly bring that abuse to the jury’s attention. Those facts were also separately communicated to the jury by Littlejohn’s mother during her testimony.

Dr. Saint Martin’s testimony would have added little to the evidence already before the jury. His affidavit, in summary, establishes three major points: (1) Littlejohn has a behavioral disorder; (2) Littlejohn’s disorder is “consistent with neurodevelop-mental deficits experienced in his perinatal development”; and (3) “[sjtudies demonstrate” that such deficits are “corre-lat[ed]” with observable effects “at the level of the synapse,” with the caveat that “the specifics of these ... problems are not yet well understood.” I understand this to mean that Littlejohn’s prenatal substance abuse and upbringing may have contributed to his anti-social behavior later in life, and that such abuse also may have had some physically detectable, albeit ill-defined, impact on Littlejohn’s brain cells.

Dr. Draper’s testimony had already established the first two propositions advanced by Dr. Saint Martin; Littlejohn suffered from a behavioral disorder caused by his exposure to prenatal substance abuse and the neglect he experienced throughout his childhood. Thus, the only significant addition in Dr. Saint Martin’s affidavit is the third point: the articulation of a certain pattern of brain activity correlated with prenatal substance abuse. But as everyone who has ever taken a statistics class learns, “correlation and causation are two different things.” Arredondo v. Locklear, 462 F.3d 1292, 1301 (10th Cir.2006); see also Norris v. Baxter Healthcare Corp., 397 F.3d 878, 885 (10th Cir.2005) (“A correlation does not equal causation.”). Dr. Saint Martin’s affidavit, on its face, does not purport to provide, with anything approaching scientific confidence, a missing causal link between Littlejohn’s prenatal exposure to alcohol and his mental defects.

But even if Dr. Saint Martin’s affidavit did provide a definitive causal link, there is no support in law for the proposition that additional testimony on prenatal neurological development is necessary for a jury to grasp the causal link between prenatal drug exposure and brain damage. It is common knowledge that pregnant women are not supposed to drink alcohol, smoke, or use drugs because doing so may harm the fetus. It is likewise common knowledge—or common sense—that this harm is accomplished by a physical mechanism, that is, the fetus is physically exposed to substances in the mother that harm it. And, in fact, Dr. Draper explained this, testifying that “because of the passage of nutrients through the placenta, there’s a great deal of concern about the use of drugs and alcohol, particularly, in the mother.” 2000 Tr., Vol. VI, at 91 (noting that “alcohol and substance abuse does impact the developing fetus,” id. at 89). She also testified that Littlejohn was hospitalized at birth in part because of his mother’s prenatal exposures. Further scientific explication of the specific biological pathways may have been interesting, and even may have lent an air of authoritativeness, but was by no means necessary to explain the idea: mother ingests substance, substance reaches fetus, substance harms fetus. The harm to Littlejohn and its resulting influence on his future behavior was the very point of Dr. Draper’s testimony.

To be sure, it is important for the expert to articulate the type of harm the substance causes. But here, Dr. Draper did exactly that—she identified the harm as a failure to develop normally, which, combined with childhood neglect, resulted in a behavioral disorder. That is essentially the same harm identified by Dr. Saint Martin.

The majority points to Rompilla, 545 U.S. 374, 125 S.Ct. 2456, and Sears, 130 S.Ct. 3259, arguing they require us to remand when counsel fails to explain the biological mechanisms underlying a defendant’s mental disorder. But while they address the same general issues as the present case—mental health investigation and presentation of mitigation evidence— the factual particulars are almost completely opposite.

First, in Rompilla, counsel failed to present any mental health evidence, even when it was readily and easily developable. And in Sears, counsel’s investigation was blatantly inadequate—counsel failed to conduct any mental health testing, and then failed to inquire into Sears’s abusive upbringing. See id. at 3264-65. The unreasonableness of the investigation itself casts serious doubt on the reasonableness of counsel’s trial strategy. See id. at 3265 (“[A]ny finding with respect to the reasonableness of the mitigation theory counsel utilized ... is in tension with the trial court’s unambiguous finding that counsel’s investigation was itself so unreasonable as to be facially unconstitutional.”).

Here, in contrast, counsel’s investigation was more than adequate. Counsel investigated Littlejohn’s troubled past and procured a mental health evaluation that uncovered evidence of a behavioral disorder. Littlejohn’s subsequent evaluation by Dr. Saint Martin does not indicate that he was misdiagnosed by Dr. Draper. Thus, unlike in Sears, the investigation supports rather than undermines the reasonableness of counsel’s trial strategy.

And, moreover, Sears suffered from severe mental dysfunction. He was “described as severely learning disabled and as severely behaviorally handicapped.” Id. at 3262. In addition to these “gross[ ]” impairments, Sears exhibited something less than a full grasp on reality—a “grandiose self-conception and evidence of [] magical thinking.” Id. at 3264. This evidence paints a graphic picture of an individual significantly more impaired than Littlejohn, who has average intelligence and does not have a severe mental disability, even according to his own experts.

Perhaps most significantly, Sears’s jurors were kept completely in the dark regarding his mental impairments; as far as they knew, Sears was an average individual, with an upbringing that was “stable, loving, and essentially without incident.” Id. at 3261. Littlejohn’s jurors, in contrast, were made well-aware of Little-john’s behavioral disorder, the effects of which were described in detail by Dr. Draper. Whereas the new information about Sears likely would have caused a significant paradigm shift in the minds of most jurors, Dr. Saint Martin’s testimony would have created, at best, a marginal increase in the amount of information available.

Like Sears, the other cases Littlejohn relies on involved defendants with severe mental defects that were totally hidden from the jury. See, e.g., Jefferson v. Upton, - U.S. -, 130 S.Ct. 2217, 2219, 176 L.Ed.2d 1032 (2010) (“[A]s far as the jury knew, Jefferson did not suffer from brain damage or neurological impairment; he had no organic disorders; and his emotional stability, impulse control, and judgment were perfectly normal.” (quotation marks omitted)); Williams v. Taylor, 529 U.S. 362, 396, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000) (“Counsel failed to introduce available evidence that Williams was ‘borderline mentally retarded’ and did not advance beyond sixth grade in school.”); Anderson, 476 F.3d at 1144 (“[Rjather than offering the jury a potential explanation for Anderson’s actions relating to the murders he participated in, trial counsel’s case in mitigation was limited to a simple plea for mercy.”); Smith v. Mullin, 379 F.3d 919, 942 (10th Cir.2004) (“[Evidence of Mr. Smith’s mental retardation, brain damage, and troubled background constituted mitigating evidence-It was patently unreasonable for Mr. Watson to omit this evidence from his case for mitigation.”); see also Porter, 130 S.Ct. at 453 (“Counsel ... failed to uncover and present any evidence of Porter’s mental health or mental impairment... .”).³

The majority points to Victor Hooks v. Workman, 689 F.3d 1148 (10th Cir.2012), but that case is also unpersuasive. First, Hooks’s mental defects were more severe than Littlejohn’s. Hooks was mildly retarded, whereas Littlejohn was of average intelligence. On top of that, Hooks’s brain injury was linked to chronic psychotic disorder. As I discussed above, Littlejohn’s post-conviction examination by Dr. Saint Martin did not result in a clear diagnosis of significant brain damage, but only tentative conclusions of possible brain injury. Second, Hooks’s counsel’s preparation was inadequate—the family history and mental health work-up was only perfunctory. Third, unlike in Hooks, Littlejohn’s expert, Dr. Draper, did “connect the dots” from Littlejohn’s mental impairment to his later crimes. She explained the physical mechanism that may have inflicted prenatal injury (albeit not in scientific terms); she explained Littlejohn’s upbringing in detail; and she explained how all these experiences manifested themselves in a behavioral disorder involving poor impulse control—the exact conclusion the majority and Dr. Saint Martin want us to reach and think would be powerful mitigation evidence: “[Littlejohn suffers from a] behavioral disorder manifested by poor impulse control.” R. Vol. I, pt. 1, at 171. An opinion that Littlejohn’s “poor impulse control” was “caused” by one type of mental disorder rather than another is a thin reed on which to hang deficient performance. It certainly does not point to a failure to discover “significant mitigation evidence” as required by the Supreme Court. Sears, 130 S.Ct. at 3266.

The ready differences between the Supreme Court and Tenth Circuit cases and this one strengthen my view “that a competent attorney might elect not to use” evidence of the type proffered by Dr. Saint Martin. Harrington, 131 S.Ct. at 789. Littlejohn’s counsel’s strategy evidently was to evoke the jury’s sympathies by humanizing Littlejohn. Counsel reasonably could have concluded that the less technical developmental-psychology approach offered by Dr. Draper best accomplished this goal. In light of Littlejohn’s rather moderate mental defects, counsel may have concluded that focusing on the biological aspects of Littlejohn’s mental condition would distract or confuse the jury. Counsel may also have believed, particularly in light of Littlejohn’s average intelligence, that the jury would not be overly convinced that neurological injuries'—even if they existed—were sufficiently serious to significantly impair his judgment. Finally, counsel may have been reluctant to present scientific evidence the specifics of which, by Dr. Saint Martin’s own admission, “are not yet well understood.” R., Vol. I, pt. 1, at 171.

While other counsel may have chosen a different approach, I cannot say that Littlejohn’s counsel’s performance was outside the “wide latitude counsel must have in making tactical decisions.” Harrington, 131 S.Ct. at 788-89 (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052). And the majority’s elevation to a hallowed status certain types of mental health evidence is not required, let alone compelled, by Supreme Court precedent.

II.

Even assuming counsel’s performance was deficient, Littlejohn was not prejudiced

To establish prejudice, Littlejohn “must show a reasonable probability that the jury would have rejected a capital sentence after it weighed the entire body of mitigating evidence ... against the entire body of aggravating evidence.” Wong v. Belmontes, 558 U.S. 15, 130 S.Ct. 383, 386, 175 L.Ed.2d 328 (2009). As the Supreme Court recently explained:

[T]he question is not whether a court can be certain counsel’s performance had no effect on the outcome or whether it is possible a reasonable doubt might have been established if counsel acted differently. Instead, Strickland asks whether it is “reasonably likely” the result would have been different. This does not require a showing that counsel’s actions “more likely than not altered the outcome,” but the difference between Strickland’s prejudice standard and a more-probable-than-not standard is slight and matters “only in the rarest case.” The likelihood of a different result must be substantial, not just conceivable.

Harrington, 131 S.Ct. at 791-92 (citations omitted) (quoting Strickland, 466 U.S. at 693, 696-97, 104 S.Ct. 2052). Littlejohn cannot meet this burden.

As explained above, Littlejohn’s mental defects were not hidden from the jury. Dr. Draper described Littlejohn’s condition, its causes, and its effects on his behavior. It is unlikely that the additional information provided by Dr. Saint Martin would have made any difference, particularly given the tentative nature of his analysis. Indeed, no prejudice has been found in cases where the new information that would have been provided by additional testimony was significantly greater. See, e.g., Knighton v. Mullin, 293 F.3d 1165, 1179 (10th Cir.2002) (“Defense counsel ... did present a great deal of psychiatric evidence at sentencing, although the defense did not do so under the rubric of organic brain damage.”); Humphreys v. Gibson, 261 F.3d 1016, 1021 (10th Cir.2001) (finding new psychiatric testimony of organic brain damage and addiction “essentially cumulative” with prior testimony of depression, severe alcohol abuse and a personality disorder). It is Littlejohn’s burden to show prejudice, and the evidence he marshals does not surmount that hurdle.

The majority’s argument, in essence, is that a more technical presentation of organic brain injury evidence is inherently more persuasive than a more holistic presentation of the kind offered by Dr. Draper. I find no support for this argument in orn-ease law. Nor do I agree that the kind of testimony offered by Dr. Draper is inherently less convincing than the variety offered by Dr. Saint Martin. Some jurors may be swayed by testimony that sounds more technical and scientific; others may be confused or skeptical. It can be a mixed bag. See Frederick Schauer, Can Bad Science Be Good Evidence? Neuroscience, Lie Detection, and Beyond, 95 Cornell L.Rev. 1191, 1210 (2010) (finding the empirical evidence regarding the extent to which juries rely on scientific testimony “decidedly mixed”); Daniel A. Krauss & Bruce D. Sales, The Effects of Clinical and Scientific Expert Testimony on Juror Decision Making in Capital Sentencing, 7 Psychol. Pub. Pol’y & L. 267, 305 (2001) (finding “less scientific” presentation of evidence may be more convincing to jurors); see also Bradley D. McAuliff et al., Can Jurors Recognize Missing Control Groups, Confounds, and Experimenter Bias in Psychological Science?, 33 Law & Hum. Behav. 247, 255 (2009) (“[JJurors may be unable to evaluate statistical and methodological issues in a sophisticated manner.”). Here, the jury was faced with a defendant afflicted with a moderate, but by no means severe, mental disorder. In truth, Dr. Saint Martin’s approach may have been less effective than Dr. Draper’s more humanizing approach in evoking the jury’s sympathies.⁴

Littlejohn’s prejudice argument fails for an additional reason: Dr. Saint Martin’s testimony was a classic double-edged sword that “[t]he jury could have perceived ... as aggravating rather than mitigating.” Wackerly v. Workman, 580 F.3d 1171, 1178 (10th Cir.2009) (quoting Duvall v. Reynolds, 139 F.3d 768 (10th Cir.1998)); accord Dunlap, 476 Fed.Appx. at 166 (“Evidence of mental illness can arouse sympathy and diminish culpability—or it can raise the specter of an irrational, incorrigible predator.”). Evidence of brain defects might have reduced Littlejohn’s culpability somewhat, but also could have strengthened the jury’s finding of future dangerousness, insofar as it suggested Littlejohn’s behavior was beyond his control, even if partially treatable.⁵ See Atkins v. Virginia, 536 U.S. 304, 321, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002) (“[R]eliance on mental retardation as a mitigating factor can be a two-edged sword that may enhance the likelihood that the aggravating factor of future dangerousness will be found by the jury.”); Penry v. Lynaugh, 492 U.S. 302, 324, 109 S.Ct. 2934, 106 L.Ed.2d 256 (1989) (“[Defendant’s] mental retardation and history of abuse is thus a two-edged sword: it may diminish his blameworthiness for his crime even as it indicates that there is a probability that he will be dangerous in the future.”), abrogated on other grounds by Atkins, 536 U.S. at 321, 122 S.Ct. 2242.

And there is no evidence presented here that violence linked to organic brain damage is any more treatable than violence linked to a behavioral disorder. Thus, Littlejohn cannot rely on a claim that a more physiological account of his behavior would have diminished the aggravating effect of the mental health evidence presented.⁶

As a final word, the central flaw in the majority’s analysis is its erection of a categorical invocation of “organic brain injury” evidence as different in kind for purposes of a Strickland analysis. Nothing in the Supreme Court’s cases suggests it is per se error to fail to develop one type of mental health evidence over another. While in some cases, neurological evidence may be more persuasive, in others, such as this one, it would not be. But to give a talismanic quality to one type of mental health evidence without any showing that it is inherently more persuasive to juries than other evidence stretches the cases past recognition and goes far beyond what the Supreme Court requires when reviewing Strickland claims.

Littlejohn’s counsel’s performance was not deficient, and even if it was, Littlejohn was not prejudiced by his counsel’s error. Therefore, I would affirm the district court’s denial of Littlejohn’s ineffective-assistance claim.

1 . As the Supreme Court said in Bobby v. Van Hook, as is equally true here, "This is not a case in which the defendant's attorneys failed to act while potentially powerful mitigating evidence stared them in the face, cf. Wiggins, 539 U.S., at 525, 123 S.Ct. 2527, or would have been apparent from documents any reasonable attorney would have obtained, cf. Rompilla v. Beard, 545 U.S. 374, 389-393, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005). It is instead a case, like Strickland itself, in which defense counsel's 'decision not to seek more’ mitigating evidence from the defendant's background 'than was already in hand' fell 'well within the range of professionally reasonable judgments.' " 130 S.Ct. at 19 (some citations omitted).

2 . Littlejohn points us to an affidavit from his trial counsel stating that his choice not to engage a psychiatric expert was not motivated by strategic considerations. The Supreme Court, however, has instructed us not to look to defense counsel's subjective motivations when determining reasonableness, particularly when counsel only articulates his motivations long after the trial: Although courts may not indulge “post hoc rationalization’’ for counsel’s decisionmak-ing that contradicts the available evidence of counsel's actions, neither may they insist counsel confirm every aspect of the strategic basis for his or her actions.... After an adverse verdict at trial even the most experienced counsel may find it difficult to resist asking whether a different strategy might have been better, and, in the course of that reflection, to magnify their own responsibility for an unfavorable outcome. Strickland, however, calls for an inquiry into the objective reasonableness of counsel’s performance, not counsel’s subjective state of mind. Harrington, 131 S.Ct. at 788-89; see also Allen v. Mullin, 368 F.3d 1220, 1240 (10th Cir.2004) (noting the motive trial counsel may have at the habeas stage to " ‘fall on the sword’ in order to derail a death sentence”).

3 . Even the investigation in Wilson, 536 F.3d 1064, was less substantial than the one here.

4 . Even to the extent some of our cases accentuate the difference between mental impairments generally and physical brain damage in particular, see Victor Hooks, 689 F.3d at 1207, Wilson, 536 F.3d at 1094, nothing was presented in this case that would show an Oklahoma jury would view one type of evidence more credible and persuasive than the other. Without more, I cannot presume trial counsel should have done more, or that if he did it would have generated convincing testimony.

5 . Unlike Smith, 379 F.3d at 943 n. 11, this case is not one in which the defendant can argue the "aggravating 'edge' " of the double-edged sword was already before the jury. The aggravating edge here is Littlejohn's inability to control himself. Littlejohn's primary argument is that Dr. Draper’s testimony failed to convince the jury of his lack of self-control. He cannot simultaneously claim that Dr. Draper’s testimony succeeded in convincing the jury of this fact.

6 . The majority seems to assume in general that organic brain injuries are treatable, but that behavioral disorders are not (or at least less so). I have no basis to make that assumption, and my guess is that the vast majority of mental health disorders are at least treatable in part. But I do not think the cases support a categorical presumption that juries are likely to be more sympathetic to a "treatable” physical brain injury than a "treatable” behavioral disorder.

Case Details

Case Name: Littlejohn v. Trammell

Court Name: Court of Appeals for the Tenth Circuit

Date Published: Jan 7, 2013

Citations: 704 F.3d 817; 2013 WL 64372; 2013 U.S. App. LEXIS 415; 10-6148

Docket Number: 10-6148

Court Abbreviation: 10th Cir.