After shooting and killing his former girlfriend, Charles Christopher Williams was convicted by a South Carolina jury of kidnapping, murder, and possession of a firearm during a violent crime. He was sentenced to death for the murder. After exhausting state remedies, Williams petitioned the United States District Court for the District of South Carolina for a writ of habeas corpus pursuant to 28 U.S.C. § 2254 . The district court denied or stayed all of Williams' claims, except Ground Six, which asserted a claim of ineffective assistance of counsel resulting from trial counsel's failure to investigate potentially mitigating evidence of Fetal Alcohol Syndrome ("FAS"). On this ground, the district court granted Williams' petition and the State now appeals. For the reasons that follow, we affirm the judgment of the district court.

I.

On the morning of September 3, 2003, Williams entered a Greenville, South Carolina grocery store where his former girlfriend, Maranda Williams, worked. He confronted her, then forced her into a store office, where he held her at gunpoint for approximately 90 minutes. During this period she called 911 and hostage negotiators tried to convince Williams to release her. She eventually attempted to escape, but Williams pursued her, shooting her four times and killing her. Upon hearing the shots, law enforcement officers entered the store and apprehended Williams. Following his arrest, Williams gave a statement in which he confessed to the crimes for which he was later charged. In February 2005, a Greenville County, South Carolina, jury convicted Williams of kidnapping, murder, and possession of a firearm during a violent crime.

At trial, Williams was represented by attorneys William Nettles and John Mauldin, both of whom were experienced in capital cases. Nettles had handled approximately five death penalty cases through trial and sentencing, as well as a handful of post-conviction relief cases. Mauldin had overseen "close to a dozen [capital cases] to verdict" and worked on nearly three times as many cases after a death notice had been filed. J.A. 493-94.

In preparation for the penalty phase, Nettles and Mauldin assembled a defense team that included, among others, social worker Jan Vogelsang, clinical neuropsychologist Dr. James Evans, clinical psychiatrist Dr. Robert Richards, neurologist Dr. David Griesemer, and forensic psychiatrist Dr. Seymour Halleck. As part of the investigation, Vogelsang gathered information about Williams' upbringing. She interviewed Williams' father, who told her that he had observed Williams' mother, Daisy Huckaby, drinking while pregnant, though he was unable to provide any additional details. Vogelsang also interviewed Williams' sister, who recalled that Huckaby drank while pregnant with Williams, but could not say how much. (The record indicates that Vogelsang either failed to ask Huckaby about her drinking or that Huckaby denied drinking while pregnant.)

The defense team experts assessed Williams for neurological and psychological issues. Following an evaluation, Dr. Evans concluded that Williams suffered neurological impairments as the result of frontal lobe damage and, consequently, had learning difficulties. Dr. Richards examined Williams and diagnosed him with bipolar and obsessive-compulsive disorder. Finally, following an MRI and neurological exam the week prior to the trial, Dr. Griesemer reported that, though there were some cognitive issues, Williams' MRI showed a normal brain.

During the penalty phase, 1 defense counsel presented mitigating evidence of Williams' troubled childhood-including his mother's alcoholism-as well as his mental illness and difficulties in school. To this end, counsel presented testimony from Williams' father and sister; Williams' first grade teacher; a co-worker of Daisy Huckaby; and their experts, including Dr. Richards, who testified about his diagnoses, and Dr. Halleck, who opined that Williams suffered from major depressive disorder and obsessive-compulsive disorder but was able to, with difficulty, conform his behavior to the requirements of the law. Moreover, through his cross-examination of the state psychiatrist, Nettles elicited additional mitigation testimony, including information about Williams' trouble with his parents' divorce, Huckaby's alcoholism, Williams' difficulty in school, and his untreated attention deficit disorder. In turn, the State alleged a single aggravating factor: "Murder was committed while in the commission of kidnapping." J.A. 809.

On the second day of penalty phase deliberations, the jury sent a note to the trial court stating it was deadlocked nine to three in favor of death. Williams moved for a mistrial but the trial court denied the motion and instead gave an Allen 2 charge. The jury resumed its deliberations and, after three hours and 45 minutes, returned a sentence of death. The Supreme Court of South Carolina affirmed Williams' convictions and death sentence, State v. Williams , 386 S.C. 503 , 690 S.E.2d 62 (2010), and the United States Supreme Court denied his petition for a writ of certiorari, Williams v. South Carolina , 562 U.S. 899 , 131 S.Ct. 230 , 178 L.Ed.2d 153 (2010).

In November 2010, Williams filed a petition for post-conviction relief in the Greenville County, South Carolina Circuit Court ("PCR court"), asserting errors that included trial counsel's failure to investigate signs that Williams suffered from FAS-namely, evidence of Huckaby's drinking during her pregnancy and Williams' corresponding brain damage. In January 2013, the PCR court held an evidentiary hearing at which three FAS experts testified on Williams' behalf. Dr. Richard Adler, a forensic psychiatrist, diagnosed Williams with Partial Fetal Alcohol Syndrome, a form of FAS. Neuropsychologist Dr. Paul Connor testified that his assessment of Williams indicated severe functional impairments and damage to the corpus callosum, all consistent with or symptomatic of FAS. Finally, Dr. Natalie Novick Brown, a forensic psychologist, concluded that Williams' executive functions-including "self-regulation" and "behavior control"-were impaired due to FAS, leading to behavioral difficulties, including impulse control problems and coping skills equivalent to those of a nine year old. J.A. 588. All three experts acknowledged that at the time of the trial in 2005, a widely recognized protocol to forensically assess FAS in the criminal justice context had not yet been fully developed, but that individual practitioners had been addressing FAS and had developed a framework for diagnosing the condition and treating its symptoms. 3

Trial counsel also testified, but neither could recall a mitigation investigation into FAS, or why such an investigation was not conducted. Mauldin testified that although FAS awareness had become much more prevalent in the years since Williams' trial, the issue "certainly existed well before"

the 2003 American Bar Association Guidelines for the Appointment and Performance of Defense Counsel in Death Penalty Cases. 4 J.A. 500. The commentary in these Guidelines designated FAS as a potentially mitigating factor to be investigated by counsel in capital cases. He further acknowledged that, in hindsight, several issues should have indicated a potential FAS diagnosis for Williams when he was preparing for trial. First, Mauldin testified that he had reports in his files that indicated Huckaby drank during her pregnancy. Mauldin acknowledged that at the time of the trial, such drinking should have signaled a potential FAS issue for him. With this information about Huckaby's drinking, he should have, as a first step, directed a neurologist to conduct whatever testing would have been necessary to determine whether Williams was affected by FAS. Nonetheless, Mauldin testified, "I honestly cannot say why [Huckaby's drinking] wasn't a red flag for me eight years ago." J.A. 500. "[A]s extraordinary as that seems," he continued, "I can't explain why there was no discussion or follow-up on that." J.A. 512. Second, Mauldin testified that the developmental delays and learning problems exhibited by Williams were issues he should have associated with FAS. Finally, Mauldin also explained that some of the follow-up information the defense team experts were seeking was of the type he should have associated with FAS. Specifically, Mauldin testified that at the time of the trial he was aware of the correlation between a significantly smaller head circumference at birth and FAS and knew that Dr. Richards, as of August 2004, had become interested in potential brain damage and had requested records containing the circumference of Williams' head at birth and recommended an MRI of Williams' brain. Nonetheless, Mauldin was unable to explain why the records were not produced to Dr. Richards, or why an MRI was not conducted until the week prior to the beginning of the trial in February 2005, rather than in August 2004.

Nettles testified that he was aware of the ABA Guidelines mandating investigation of mitigating evidence, including personal, family, and medical history, as part of penalty phase preparations. 5 But he could not remember at what point he developed an "understanding" of FAS. J.A. 465. He did recall that the subject of Huckaby's drinking came up, but testified that he was focused on it as evidence of Williams' difficult childhood, not of FAS. He also recalled some evidence of neurological damage. Nonetheless, he did not recall any discussion about FAS or FAS being considered as a potentially mitigating factor.

In denying Williams' petition, the PCR court concluded:

[T]his Court finds that trial counsel had evidence that [Williams'] mother drank during pregnancy, and that trial counsel was aware of the resulting complications, including brain damage. Trial counsel also had evidence that [Williams] possibly suffered brain damage, based on Dr. Evans' reports. Trial counsel presented this information, along with other mitigation evidence, to the defense experts. Considering all of the information it had available and in consultation with its experts, trial counsel developed a cogent strategy to present mitigation evidence-including evidence of the mother's alcohol addiction-but also made a strategic decision not to present to the jury evidence of brain damage or a diagnosis of Fetal Alcohol Syndrome ( though trial counsel was unable to articulate the reasons for that strategic decision ). Instead, trial counsel's strategy was to present mitigation evidence regarding [Williams'] troubled childhood and his [other disorders], as diagnosed by defense experts.

J.A. 665 (emphases added). Finally, the PCR court also found that, even if Williams had presented evidence of FAS to the jury, it was unlikely that the jury would have returned a different sentence. The PCR court based its conclusion in part on a survey of eight jury verdicts from other jurisdictions demonstrating that defendants are sentenced to death in spite of mitigating evidence of FAS or organic brain damage. The South Carolina Supreme Court dismissed Williams' petition for writ of certiorari, Williams v. South Carolina , No. 2016-MO-012, 2016 WL 1458174 (S.C. Apr. 13, 2016), as did the United States Supreme Court, Williams v. South Carolina , --- U.S. ----, 137 S.Ct. 1812 , 197 L.Ed.2d 762 (2017).

After initiating habeas proceedings in the district court in November 2016, Williams filed an amended § 2254 petition in February 2017, asserting 15 grounds for relief, of which only the first six are at issue on appeal. Ground One asserted that the trial court's Allen charge was improperly coercive. Ground Two asserted that the State elicited prejudicial testimony from its forensic psychiatrist by asking her if she became involved only in cases in which "the death penalty may be considered." Compare J.A. 23-25, with J.A. 276. Grounds Three and Four asserted that trial counsel failed to properly object to a series of allegedly prejudicial comments made during the State's closing argument. Ground Five asserted that trial counsel was ineffective for failing to assert Williams' right to seek assistance from the German government under the Vienna Convention on Consular Relations based on his German citizenship. And of particular importance to this appeal, Ground Six asserted Williams was denied effective assistance of counsel after trial counsel failed to investigate evidence of FAS.

The case was referred to a magistrate judge, who recommended the petition be granted as to Ground Six, and that Williams' death sentence be vacated as a result. The magistrate judge concluded that the PCR court's finding that trial counsel "made a strategic decision" was unreasonable given that this finding was directly contradicted by trial counsel's PCR testimony. Compare J.A. 885, with J.A. 665. The magistrate judge also concluded that Williams established prejudice: because the State put forward only one aggravating factor and "the jury was deprived of powerful [mitigating] evidence," a reasonable probability existed that the jury would have returned a life sentence had this additional mitigating evidence been presented and credited by the jury. J.A. 888. Finally, the magistrate judge also recommended granting summary judgment to the State as to Grounds One through Five and Seven through Ten, and dismissing without prejudice Grounds Eleven through Fifteen. 6

The district court adopted the magistrate judge's recommendations as to Grounds One through Ten, including granting the petition as to Ground Six. The district court also granted Williams a stay as to Grounds Eleven through Fifteen, pending exhaustion of those claims in state court. Consequently, the district court vacated the death sentence and "suggest[ed]" a resentencing trial. J.A. 959.

The State filed a timely appeal as to Ground Six. 7 This Court has jurisdiction pursuant to 28 U.S.C. §§ 1291 and 2253.

II.

28 U.S.C. § 2254 (a) provides that a federal district court "shall entertain an application for a writ of habeas corpus" filed by a state prisoner "only on the ground that he is in custody in violation of the Constitution or laws or treaties of the United States." Generally speaking, before filing a § 2254 petition, a petitioner must exhaust all state court remedies. Id. § 2254(b) ; see also Jones v. Sussex I State Prison , 591 F.3d 707 , 712-13 (4th Cir. 2010) (explaining that § 2254 's exhaustion requirement means that a state prisoner must have first presented his claim before every available state court).

Once a state prisoner has exhausted his claims in state court and filed a federal habeas petition, "[i]f a state court has already resolved the merits of a claim for post-conviction relief, a federal court may not grant a writ of habeas corpus [under § 2254 ] unless the state court's decision" meets the requirements of § 2254(d). Byram v. Ozmint , 339 F.3d 203 , 206 (4th Cir. 2003). Specifically, § 2254(d) provides:

An application for a writ of habeas corpus on behalf of a person in custody pursuant to the judgment of a State court shall not be granted with respect to any claim that was adjudicated on the merits in State court proceedings unless the adjudication of the claim-

(1) resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States; 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.

Section 2254(d) further "require[s] us to limit our analysis" of the state PCR court's decision "to the law as it was 'clearly established' by [the Supreme Court] at the time of the [PCR] court's decision." Wiggins v. Smith , 539 U.S. 510 , 520, 123 S.Ct. 2527 , 156 L.Ed.2d 471 (2003).

Under § 2254(d)(1), such a decision is "contrary to" Supreme Court precedent "if the state court applie[d] a rule that contradicts the governing law set forth in" Supreme Court cases, or "confront[ed] a set of facts that are materially indistinguishable from a [Supreme Court decision] and nevertheless arrive[d] at a result different from [that] precedent." Williams v. Taylor , 529 U.S. 362 , 405-06, 120 S.Ct. 1495 , 146 L.Ed.2d 389 (2000). A decision is an "unreasonable application" of clearly established Supreme Court precedent if the PCR court "correctly identifie[d] the governing legal rule but applie[d] it unreasonably to the facts of a particular prisoner's case." Id. at 407-08 , 120 S.Ct. 1495 . "In order for a federal court to find a state court's application of [Supreme Court] precedent unreasonable, the state court's decision must have been more than incorrect or erroneous. The state court's application must have been objectively unreasonable." Wiggins , 539 U.S. at 520-21 , 123 S.Ct. 2527 (internal citation and quotation marks omitted); see also Harrington v. Richter , 562 U.S. 86 , 103, 131 S.Ct. 770 , 178 L.Ed.2d 624 (2011) ("[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 fairminded disagreement.").

Alternatively, a state prisoner may be granted relief pursuant to § 2254(d)(2) if the PCR court decision's was based on a factual determination "sufficiently against the weight of the evidence that it is objectively unreasonable." Winston v. Kelly , 592 F.3d 535 , 554 (4th Cir. 2010). As with legal conclusions, "[f]or a state court's factual determination to be unreasonable under § 2254(d)(2), it must be more than merely incorrect or erroneous." Id. (internal citation omitted).

This Court's review is de novo when a federal district court's habeas decision is based on the state court record. Gray v. Branker , 529 F.3d 220 , 228 (4th Cir. 2008). State court factual determinations are presumed correct and may be rebutted only by clear and convincing evidence. 28 U.S.C. § 2254 (e)(1). Upon review, we conclude that the PCR court's determination of Williams' Ground Six ineffective assistance claim involved both an unreasonable application of federal law clearly established by Supreme Court precedent at the time of the PCR hearing, and an unreasonable determination of the facts in light of the record before it. Consequently, as described in greater detail below, we affirm the district court's grant of habeas corpus relief under § 2254.

III.

In Ground Six of his petition, Williams contends that his attorneys' performance during the penalty phase violated his Sixth Amendment right to effective assistance of counsel. A prisoner petitioning for habeas relief based on ineffective assistance of counsel must meet two components: "[a] petitioner must show that counsel's performance was deficient, and that the deficiency prejudiced the defense." Wiggins , 539 U.S. at 521 , 123 S.Ct. 2527 (citing Strickland v. Washington , 466 U.S. 668 , 687, 104 S.Ct. 2052 , 80 L.Ed.2d 674 (1984) ). We address each in turn, mindful that on appeal our inquiry is limited to whether the PCR court's ineffective assistance determination was contrary to or an unreasonable application of Supreme Court precedent or an objectively unreasonable factual determination.

A.

1.

The district court determined that defense counsel were deficient at the sentencing phase because of their failure to investigate evidence indicating that Williams had FAS. To establish deficient performance, a petitioner must demonstrate that counsel's representation "fell below an objective standard of reasonableness."

Strickland , 466 U.S. at 688 , 104 S.Ct. 2052 . The performance inquiry here focuses on the standard of reasonableness related to counsel's duty to investigate mitigating evidence for sentencing in a capital case. Strickland does not require investigation of every conceivable line of mitigating evidence but does impose "a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary." Id. at 691 , 104 S.Ct. 2052 . "[S]trategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable; and strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation." Id. at 690-91 , 104 S.Ct. 2052 .

Review of trial counsel's investigation is considered "from counsel's perspective at the time," id. , and the professional norms then prevailing. Padilla v. Kentucky , 559 U.S. 356 , 366-67, 130 S.Ct. 1473 , 176 L.Ed.2d 284 (2010). A "well-defined norm" at the time of Williams' trial provided "that investigations into mitigating evidence 'should comprise efforts to discover all reasonably available mitigating evidence.' " Wiggins , 539 U.S. at 524 , 123 S.Ct. 2527 (internal citation omitted) (applying norm to a trial that occurred in 1989). "[A reviewing] court must consider not only the quantum of evidence already known to counsel, but also whether the known evidence would lead a reasonable attorney to investigate further." Id. at 527 , 123 S.Ct. 2527 .

In turn, "[p]revailing norms of practice as reflected in the American Bar Association standards and the like are guides to determining what is reasonable[.]" Strickland , 466 U.S. at 688 , 104 S.Ct. 2052 . With respect to investigating mitigating evidence in preparation for the penalty phase of capital proceedings, the ABA Guidelines at the time of trial noted that a defendant's psychological history and mental status could "explain or lessen the client's culpability for the underlying offense[ ]," and therefore should be considered as part of the mitigation investigation. ABA Guidelines § 10.11(F)(2), reprinted in 31 Hofstra L. Rev. at 1056. Commentary to § 10.11 explained that expert testimony concerning "the permanent neurological damage caused by fetal alcohol syndrome" could "lessen the defendant's moral culpability for the offense or otherwise support[ ] a sentence less than death." Id. at 1060-61 ; see also id. at 956-57 (noting, with respect to § 4.1, that because "the defendant's psychological and social history and his emotional and mental health are often of vital importance to the jury's decision at the punishment phase," the defense team should include at least one person qualified to screen for mental or psychological defects so as to "detect the array of conditions (e.g., post-traumatic stress disorder, fetal alcohol syndrome , pesticide poisoning, lead poisoning, schizophrenia, mental retardation ) that could be of critical importance" (emphasis added) ). Furthermore, the 2003 Fourth Circuit case Byram -which was decided about 18 months before Williams' sentencing-also recognized FAS could be a mitigating factor in a capital case. 339 F.3d at 209-10 (describing that trial counsel's investigation of potential mitigating evidence indicated that the defendant's mother had not abused alcohol during her pregnancy, which, coupled with "the absence of any evidence of organic brain dysfunction, [led] trial counsel [to] conclude[ ] that they did not have a sufficient factual basis to present FAS as evidence in mitigation").

2.

We note at the outset that most of trial counsels' decisions and actions on issues unrelated to FAS did bear the hallmarks of effective assistance: trial counsel had experience in capital cases; counsel consulted with numerous experts in developing a mitigation case; and counsel spent a significant amount of time developing mitigation arguments. See id. at 205-11 (listing similar factors to bolster conclusion that counsel's performance was not deficient). But as Wiggins makes abundantly clear, an inadequate investigation into potentially mitigating evidence can be, by itself, sufficient to establish deficient performance. 539 U.S. at 534 , 123 S.Ct. 2527 .

Here, counsel's investigation into potentially mitigating evidence of FAS failed to meet an objective standard of reasonableness. By counsel's own PCR-court admission, their failure to further investigate signs of FAS fell below the then-current standard for mitigation investigations: both attorneys acknowledged they were aware of the mitigating value of neurological defects at the time of Williams' sentencing-with Mauldin specifically testifying that he was aware at the time of the importance of FAS as a potential mitigating factor-yet they failed to investigate this issue. Mauldin noted that he should have been aware of the issue because the evidence of Williams' brain damage and Huckaby's alcohol consumption during her pregnancy, as well as Dr. Richards' request for medical records concerning the circumference of Williams' head at birth and an MRI, should have alerted counsel to this issue at the time of sentencing. But, as Mauldin testified, he was unable to explain why this information did not raise a red flag: "I wish I could say I connected that, but I did not. ... I really don't have an explanation for why I was missing those kinds of indicators." J.A. 519. As he further stated, "I am dumbfounded about why a certain course of action did not occur [as the result of being aware of Huckaby's drinking during her pregnancy]. ... [I]t is unexplainable to me." J.A. 525. Mauldin was similarly unable to explain why, despite awareness of Huckaby's drinking, counsel did not even consider whether Williams had FAS. Nettles' testimony confirmed Mauldin's: despite numerous indicators of FAS, they did not consider whether to pursue that evidence.

Consequently, because there was no recognition of a potential FAS diagnosis by trial counsel, there was no further exploration of FAS as a potential mitigating factor. And because there was no further exploration, there was necessarily no opportunity for counsel to make a strategic decision about whether or not to further develop the FAS evidence or present it in mitigation. Rather, the investigation here was deficient for the same reasons that Wiggins found counsel's investigation to be deficient: the lack of an informed decision regarding mitigating evidence. In Wiggins , there was evidence of a Maryland death row inmate's alcoholic mother and his problems in foster care. Despite this evidence, counsel failed to follow up on these leads for potentially mitigating evidence. 539 U.S. at 525 , 123 S.Ct. 2527 . The Supreme Court concluded that "any reasonably competent attorney would have realized that pursuing these leads was necessary to making an informed choice among possible defenses, particularly given the apparent absence of any aggravating factors in petitioner's background." Id.

A comparison to our analysis in Byram also highlights the deficiencies in trial counsel's investigation here. In Byram , this Court affirmed a district court's denial of a § 2254 petition, concluding that trial counsel's failure to present mitigating FAS evidence did not amount to deficient performance. 339 F.3d at 211 . The Court held that counsel's failure to develop FAS evidence was reasonable in light of two factors: first, there was no indication that the birth mother drank during her pregnancy (specifically, the birth mother denied abusing alcohol during her pregnancy and her records contained no contrary evidence); and second, there was no evidence of organic brain damage or FAS, particularly in the test results evaluated by the defense team. Id. at 210 ("Based upon [the] investigation and the absence of any evidence of organic brain dysfunction, trial counsel concluded that they did not have a sufficient factual basis to present FAS as evidence in mitigation."). Consequently, trial counsel did not fall short of "well-defined norms requiring the discovery of all reasonably available mitigating evidence," nor did they "abandon their investigation at an unreasonable juncture." Id. (internal quotation marks omitted).

In contrast to Byram , both of these red flags were present here. First, although evidence of Huckaby's drinking during pregnancy was mixed, there was sufficient evidence of alcohol abuse that Vogelsang flagged it for general concern. Second, there was evidence of Williams' brain damage, including impairment of the front lobe. Furthermore, even though evidence of brain damage led Dr. Richards to suggest ordering an MRI of Williams' brain and to request medical records of Williams' head circumference at birth-information often correlated with a FAS diagnosis-the team failed to provide the medical records or to obtain the MRI until the week prior to trial. Consequently, evidence of FAS was reasonably available, but counsel failed to connect the indicators suggesting further investigation. And given that FAS evidence was widely acknowledged to be a significant mitigating factor that reasonable counsel should have at least explored-as outlined in the ABA Guidelines and caselaw at the time, and by counsel during their PCR testimony-counsel's actions were deficient. To the point, because counsel failed to conduct any investigation despite the red flags, their conduct fell well short of the conduct Byram concluded would have actually been deficient: abandoning an investigation into FAS "at an unreasonable juncture." 339 F.3d at 210 ; see also Wiggins , 539 U.S. at 527-28 , 123 S.Ct. 2527 ("[C]ounsel chose to abandon their investigation at an unreasonable juncture, making a fully informed decision with respect to sentencing strategy impossible."); Strickland , 466 U.S. at 690-91 , 104 S.Ct. 2052 ("[S]trategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation.").

An investigation into FAS evidence would also have been substantively different from the defense team's investigation into other mental illnesses and behavioral issues because FAS could have established both cause and effect for Williams' criminal acts whereas the other mitigation evidence went more to effects on behavior. That is, FAS evidence could have provided to the jury evidence of an overarching neurological defect that caused Williams' criminal behavior. See ABA Guidelines, 31 Hofstra L. Rev. at 1060-61 ("If counsel cannot establish a direct cause and effect relationship between any one mitigating factor and the commission of a capital offense, counsel may wish to show the combination of factors that led the client to commit the crime."); id. at 1061 ("[I]t is critically important to construct a persuasive narrative in support of the case for life, rather than to simply present a catalog of seemingly mitigating factors."). Without the information on FAS, the jury could have assumed that Williams was an individual who-despite a challenging childhood, learning disabilities, and other mental health issues-was generally responsible for his actions, and therefore would have assigned greater moral culpability to him for his criminal behavior.

Of course, counsel would not have been required to present evidence of FAS. Indeed, counsel may have concluded, after investigating and considering FAS as a mitigating factor, that it was an unsound strategy to present this information to the jury because, for example, it could indicate future dangerousness. But that analysis can justify a decision only after a reasonable investigation into FAS. Here, counsel did not collect any FAS evidence or consider its resulting import as part of the mitigation strategy.

3.

But, as noted earlier, it is not enough for us to determine that trial counsel failed to meet the Strickland standard for performance. In the § 2254 context, we must also determine whether the district court erred in concluding the PCR court's determination was "contrary to, or involved an unreasonable application of," Supreme Court caselaw or was based on "an unreasonable determination of the facts." 28 U.S.C. § 2254 (d). The State contends that the district court failed to afford the appropriate deference to the PCR court's determination-based on "competing evidence of what the defense team knew, and what the defense team did"-and that counsel made a reasonable strategic decision in preparing a mitigation case that excluded presentation of a FAS diagnosis. Opening Br. 31.

We disagree. Applying the correct standard here, we conclude that the PCR court's determination that the investigation was not deficient involved both an unreasonable application of the law and an unreasonable determination of the facts.

As an initial matter, the PCR court's application of Strickland and its progeny to the present case was objectively unreasonable. 28 U.S.C. § 2254 (d)(1). In reaching its conclusion, the PCR court confused a strategic decision not to further develop FAS evidence after some investigation into its potential mitigating value-which could have complied with Strickland -with a complete failure to investigate the FAS evidence for any potential mitigating value, a failure that plainly falls below an objective standard of reasonableness. As Wiggins concluded, even if trial counsel "would not have altered their chosen strategy" of presenting other mitigating factors, "counsel were not in a position to make a reasonable strategic choice ... because the investigation supporting their choice was unreasonable." 539 U.S. at 536 , 123 S.Ct. 2527 . "[T]his case is therefore distinguishable from our precedents in which we have found limited investigations into mitigating evidence to be reasonable." Id. at 525 , 123 S.Ct. 2527 . Consequently, even under the highly deferential standard afforded to the PCR court, that court's conclusion was unreasonable: "In deferring to counsel's decision not to pursue a mitigation case despite their unreasonable investigation, the [state court] unreasonably applied Strickland ." Id. at 534 , 123 S.Ct. 2527 . In short, the PCR court could not reasonably find trial counsel made a strategic decision in accord with Strickland where counsel was unaware of the decision.

The PCR court's determination of the facts was also objectively unreasonable. 28 U.S.C. § 2254 (d)(2) ; see also Harrington , 562 U.S. at 100, 131 S.Ct. 770 . Specifically, the PCR court relied on the factual assumption that trial counsel made a strategic choice not to present the FAS evidence. But, as recounted above, it was impossible for trial counsel to have made a strategic choice because there was no investigation into FAS. Both Nettles and Mauldin testified repeatedly that FAS was never considered, while Vogelsang also testified that nobody ever ruled out FAS. Therefore, counsel could not, as the PCR court found, have made a choice between mitigation strategies. Rather, "[t]he record of the actual sentencing proceedings underscores the unreasonableness of counsel's conduct by suggesting that their failure to investigate thoroughly resulted from inattention, not reasoned strategic judgment." Wiggins , 539 U.S. at 526 , 123 S.Ct. 2527 .

Additionally, the PCR court erroneously assumed that a lack of an established protocol assessment of FAS in the forensic context meant that FAS was not a widely understood condition at the time of trial; in fact, the ABA Guidelines at the time flagged FAS as a potentially mitigating factor, and trial counsel testified they were sufficiently aware of FAS such that certain issues that arose during their investigation should have triggered an investigation into a possible FAS diagnosis. See also Moore v. Texas , 518 U.S. ----, ----, 137 S.Ct. 1039 , 1049, 197 L.Ed.2d 416 (2017) ("[B]eing informed by the medical community does not demand adherence to everything stated in the latest medical guide. But neither does our precedent license disregard of current medical standards."). The PCR court's reliance on this factual determination to reach the conclusion that trial counsel made a strategic decision to exclude FAS evidence underscores the unreasonableness of the PCR court's decision.

* * * *

For these reasons, we agree with the district court that the PCR court erred in concluding that Williams had failed to establish deficient performance of counsel.

B.

This does not end our inquiry either, however, because Williams must also establish that the PCR court's prejudice determination was contrary to or an unreasonable application of Supreme Court precedent, or an objectively unreasonable factual determination. 28 U.S.C. § 2254 (d).

1.

To establish Strickland prejudice, Williams was required to demonstrate "a reasonable probability that at least one juror would have struck a different balance." Wiggins , 539 U.S. at 537 , 123 S.Ct. 2527 . "A reasonable probability is a probability sufficient to undermine confidence in the outcome." Strickland , 466 U.S. at 694 , 104 S.Ct. 2052 .

Furthermore, "[i]n assessing prejudice, [the Court] reweigh[s] the evidence in aggravation against the totality of available mitigating evidence." Wiggins , 539 U.S. at 534 , 123 S.Ct. 2527 . Specifically, the Court evaluates both the evidence adduced at trial and in the state PCR proceedings.

2.

The PCR court concluded that no prejudice occurred because the addition of FAS evidence "would have 'merely resulted in a 'fancier' mitigation case, [with] no effect on the outcome of the trial.' " J.A. 667. This conclusion by the PCR court was based in part on a survey of jury verdicts in other jurisdictions demonstrating "that defendants are often sentenced to death in spite of evidence offered in mitigation that the defendant had fetal alcohol syndrome or organic brain damage." J.A. 668. Along the same lines, the State argues that Williams cannot establish prejudice because a main indicator of FAS (Huckaby's drinking), as well as two FAS-related or FAS-like symptoms (Williams' mental illnesses and learning disabilities) were already before the jury. Consequently, the State argues, the addition of a diagnosis of FAS would not have changed the outcome. The State also contends that even if FAS evidence had been presented, any mitigation value would have been undercut by it simultaneously suggesting future dangerousness to the jury. See Brown v. Thaler , 684 F.3d 482 , 499 (5th Cir. 2012) (concluding FAS evidence is "double-edged" because "although it might permit an inference that [a defendant] is not as morally culpable for his behavior, it also might suggest that he, as a product of his environment, is likely to continue to be dangerous in the future" (internal quotation marks and alteration omitted) ).

"When a defendant challenges a death sentence such as the one at issue in this case, the question is whether there is a reasonable probability that, absent the errors, the sentencer ... would have concluded that the balance of aggravating and mitigating circumstances did not warrant death." Strickland , 466 U.S. at 695 , 104 S.Ct. 2052 . We conclude that Williams has established prejudice: had the FAS evidence been presented, there was a reasonable probability that, given the balance of aggravating and mitigating factors, the jury would have returned a different sentence. First, as discussed previously, the FAS evidence was different from the other evidence of mental illness and behavioral issues because it could have established cause and effect for the jury-specifically, a FAS diagnosis could have provided to the jury evidence of a neurological defect that caused Williams' criminal behavior. Without this information, the jury could have assumed that Williams was an individual who-despite challenges in his home life, education, and mental health-was generally responsible for his actions, and therefore would have assigned greater moral culpability to him for his criminal behavior. 8

At the PCR hearing, experts testified that FAS impaired Williams' judgment, as well as his ability to control his impulses and consider the consequences of his actions. This could have been persuasive mitigating evidence for a jury-particularly a deadlocked one-considering the death penalty, and could have been outcome-determinative because of how it framed a defendant's culpability, particularly in comparison to the other mitigating factors submitted for the jury's consideration. See Rompilla v. Beard , 545 U.S. 374 , 391-93, 125 S.Ct. 2456 , 162 L.Ed.2d 360 (2005) (linking brain damage caused by FAS and petitioner's capacity to appreciate the criminality of his conduct).

Further, the State only presented one aggravating factor: that the murder occurred in the commission of a kidnapping. Consequently, had this solitary aggravating evidence been weighed against the totality of the mitigating evidence presented during both the penalty phase and the PCR proceedings, there is a reasonable probability the jury would have determined the balance of factors did not warrant a death sentence.

The district court thus correctly determined that Williams had established Strickland prejudice.

3.

We also agree with the district court's conclusion that the PCR court's prejudice determination involved an unreasonable application of clearly-established law. 28 U.S.C. § 2254 (d)(1). As an initial matter, we note that, by relying on the survey of jury verdicts, the PCR court failed to examine the facts of this case in view of the Strickland requirements and instead made a generalized assessment unrelated to the case before it. Relatedly, the PCR court's failure to reweigh the totality of the available mitigation evidence against the aggravating evidence in this specific case is evidenced by two additional points. See Williams , 529 U.S. at 397-98 , 120 S.Ct. 1495 . First, although the mitigation evidence may have been mixed, it was error for the state court to fail to "entertain [the] possibility" that the mitigating FAS evidence could have "alter[ed] the jury's selection of penalty" because it "might well have influenced the jury's appraisal of [the defendant's] moral culpability." Id. at 398 , 120 S.Ct. 1495 . As discussed above, the mitigating FAS evidence here could have been significant for the jury because it could have established cause and effect, thereby diminishing Williams' culpability. The evidence's significance is further heightened here given that the jury was initially deadlocked on whether to impose the death penalty. Second, as outlined previously, the aggravating evidence was minimal. When compared to the totality of the mitigating evidence, it is clear that the PCR court assigned unreasonable weight to the sole aggravating factor.

Given the aggravating and mitigating evidence in the context of this particular case, it is evident that the presentation of the FAS evidence would have resulted in, at a minimum, a reasonable probability of a different sentence, even if it did not guarantee one. This is all the law requires. As a result, the district court properly found that the PCR court's prejudice determination was unreasonable.

IV.

Finally, Williams argues the district court erred in granting summary judgment to the State on Grounds One through Five. Because we affirm the district court's grant of summary judgment to Williams as to Ground Six, we do not address these issues further.

V.

For the foregoing reasons, we affirm the district court's judgment.

AFFIRMED