DETRICH v. RYAN

w-ref384 F.3d 567, 622 (9th Cir. 2004). Although a defendant's informed wishes can justify failing to present mitigating evidence, it cannot justify failing to investigate such evidence because counsel retains a duty to inform his client about the risks and potential benefits of presenting a mitigation case. See id. b

Detrich's counsel's sentencing-phase strategy of emphasizing residual doubt about whether Detrich had actually committed the murder likewise did not make a thorough mitigation investigation unnecessary. First, the record reveals that Higgins made no strategic choice to emphasize residual doubt to the exclusion of a classic mitigation presentation here. Higgins avowed in a declaration that his failure to conduct a thorough investigation of Detrich's background, or to present a fuller mitigation case at the penalty hearing, was not based on a sentencing strategy. Indeed, he avowed that, had he known the mitigating information about Detrich's past and mental health problems, he would have presented it. The fact that Higgins did present a limited mitigation case to the sentencing court corroborates this claim. See Wiggins, 539 U.S. at 526, 123 S.Ct. 2527 (reasoning that counsel's presentation of a "halfhearted mitigation case" demonstrated that the failure to put on a stronger mitigation case was not a strategic choice).

Second, even if Higgins had made such a strategic choice, that choice would not have been reasonable. The Supreme Court has made clear that the investigation supporting a decision not to introduce mitigating evidence must be reasonable because a thorough investigation is necessary to make "a fully informed decision with respect to sentencing strategy." Wiggins, 539 U.S. at 523, 527, 123 S.Ct. 2527. Where a strategic choice is made "`after less than complete investigation,'" a court must defer to that choice "only to the extent that `reasonable professional judgments support the limitations on investigation.'" Id. at 533, 123 S.Ct. 2527 (quoting Strickland, 466 U.S. at 690-91, 104 S.Ct. 2052). As explained above, Higgins did not reasonably limit the investigation here because the information that Higgins knew about Detrich's past would have led "any reasonably competent attorney" to pursue further investigation. See id. at 525, 123 S.Ct. 2527. Because it would not have been supported by a reasonable investigation, a strategy focusing on residual doubt to the exclusion of a classic mitigation defense would not have been reasonable.

Moreover, we doubt that such a strategy would have been reasonable here, even if supported by an adequate investigation. Where a judge is likely to find an aggravating factor that would make the death penalty mandatory in the absence of sufficient counterbalancing mitigating evidence, counsel's failure to make a strong mitigation case falls short of professional standards. Summerlin v. Schriro, 427 F.3d 623, 640 (9th Cir. 2005). Here, it was likely that the sentencing judge would find such an aggravating factor: the previous sentencing judge had found that the crime was especially heinous, cruel, or depraved. And under Arizona law at the time, the presence of this factor required imposition of the death penalty unless the sentencing court found mitigating circumstances substantial enough to outweigh it. Ariz.Rev. Stat. § 13-703. Under Summerlin, professional standards therefore required Higgins to present a strong mitigation case. He did not. Indeed, he did little to supplement the mitigation case from Detrich's first sentencing that had led the first sentencing judge to find no mitigating factors, much less sufficiently weighty ones. The only new evidence Higgins offered in support of leniency were the three letters from Detrich's sister.

For these reasons, no special consideration justified Higgins's failure to pursue a thorough mitigation investigation here. Neither Detrich's purported wish not to involve his family nor his counsel's purported strategic choice to pursue a residual doubt strategy at sentencing made a thorough mitigation investigation unnecessary. Higgins's performance was therefore deficient under Strickland, and the state court's decision to the contrary was objectively unreasonable.

B. Prejudice

Having concluded that the state court unreasonably held that Detrich's counsel did not perform deficiently, we must next determine whether the deficient performance prejudiced Detrich's defense. See Strickland, 466 U.S. at 687, 104 S.Ct. 2052. Under Strickland, counsel's performance is prejudicial if "there is a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Id. at 694, 104 S.Ct. 2052. A "reasonable probability" of prejudice exists "even if the errors of counsel cannot be shown by a preponderance of the evidence to have determined the outcome"; indeed, a "reasonable probability" need only be "a probability sufficient to undermine confidence in the outcome." Id. In other words, Detrich suffered prejudice if "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." Id. at 695, 104 S.Ct. 2052. In assessing that probability, we "consider `the totality of the available mitigation evidence . . .' and `reweig[h] it against the evidence in aggravation.'" Porter, 130 S.Ct. at 453-54 (quoting Terry Williams, 529 U.S. at 397-98, 120 S.Ct. 1495). In so doing, we do not consider the "idiosyncracies of the particular' decisionmaker, such as unusual propensities toward harshness or leniency." Strickland, 466 U.S. at 695, 104 S.Ct. 2052.

Detrich presented volumes of new evidence of his extremely abusive childhood and new expert assessments of his brain damage and neuropsychological deficits for the first time in federal district court. Even without considering this compelling new evidence, however, we conclude that the state court unreasonably determined that Detrich was not prejudiced by his counsel's deficient performance. In particular, we conclude that there is a reasonable probability that the sentencing judge would have imposed a sentence less than death had Detrich's counsel obtained and presented an expert evaluation of Detrich's neuropsychological functioning.

1. State Court Decision

The only evidence of Detrich's mental health before the sentencing judge was a 1991 psychological evaluation by a court clinic psychologist conducted at the request of Detrich's first counsel and two 1985 reports, one by a state psychologist and the other by a state psychiatrist, from when Detrich was incarcerated for writing worthless checks. These reports provided only a snapshot of Detrich's psychological profile. The 1985 reports mentioned in passing that Detrich was impulsive, noting that he was "an impulsive individual" and that he had "occasional impulsive responses," while the 1991 report indicated to the contrary that a test that could reflect impulsivity was "not elevated." The reports concluded that he had above-average intelligence, with cognitive functioning largely intact. The 1985 reports contradictorily noted that he had poor judgment and that his judgment was "grossly intact." They noted that he externalized problems, was sensitive to rejection, had an impaired ability to relate to others, and had a low tolerance for frustration. The 1991 report concluded that he exhibited antisocial attitudes, beliefs, and behaviors, and that he had probable antisocial personality disorder.

Had Detrich's trial counsel enlisted the assistance of a neuropsychological expert, he would have been able to offer an evaluation with much greater mitigating weight. An evaluation by neuropsychologist Dr. Briggs that Detrich presented to the state post-conviction court indicated that Detrich suffered neuropsychological deficits and opined that his crime was driven by instinct that grew out of his abusive childhood, not by consequence-driven thought or reason.

The state post-conviction court concluded that there was no "reasonable probability" that this neuropsychological evaluation "would have compelled this Court to impose a sentence less than death." In support of this conclusion, the state court reasoned, "Dr. Briggs' report was not significantly different from the report considered by this Court. Indeed, Dr. Briggs found that Petitioner's general neuropsychological functioning was normal and showed an absence of cognitive dysfunction."

This conclusion 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)(2). To be sure, Dr. Briggs's report did state that Detrich earned scores on neuropsychological tests that were "in the normal range of neuropsychological functioning." But the state court failed to mention a critical part of Dr. Briggs's assessment: his report later makes clear that this reflected a "recovered picture," as "improvement in function occurs as time (and sobriety) from the incidents [of head injuries and drug use] increase." Significantly, the report indicated that Detrich scored 25 on a scale for which 0-26 represented "normal" neuropsychological functioning. In other words, over ten years after the crime, at the time of these tests, Detrich's neuropsychological functioning was just two points shy of abnormal. Thus, considering Dr. Briggs's finding of "normal" functioning in context, along with his statement that this represented a "recovered picture," the most reasonable inference is that Detrich's neuropsychological functioning was not normal at the time of the crime.

The state court's failure to account for Dr. Briggs's statement that Detrich's functioning had improved since the time of the crime resulted in an unreasonable determination of the facts. We have previously made clear that a state court unreasonably determines the facts when it "overlook[s] or ignore[s] evidence [that is] highly probative and central to petitioner's claim." Taylor, 366 F.3d at 1001. Evidence is sufficiently "probative" and "central" if it is "sufficient to support petitioner's claim when considered in the context of the full record bearing on the issue presented in the habeas petition." Id. The portion of Dr. Briggs's report suggesting that Detrich's neuropsychological functioning was almost certainly abnormally impaired at the time of the crime is central to Detrich's claim. As we explain at length below, if an expert would have testified about such abnormality at the sentencing, there is a reasonable probability that Detrich would not have received the death penalty. The "state court's failure to consider, or even acknowledge" this highly probative evidence therefore "casts serious doubt on the state-court fact-finding process and compels the conclusion that the state-court decision[][was] based on an unreasonable determination of the facts." Id. at 1005.

2. De Novo Prejudice Assessment

We proceed to consider whether Dr. Briggs's report, read as a whole, shows that there is a reasonable probability that Detrich's sentencing counsel's failure to seek and introduce expert neuropsychological evidence prejudiced Detrich's defense. As an initial matter, we must first determine whether we must continue to apply AEDPA deference after having concluded that the state court's decision was "based on an unreasonable determination of the facts" under § 2254(d)(2). We conclude that the AEDPA imposes no such requirement.

The Supreme Court has made clear that "[w]hen a state court's adjudication of a claim is dependent on an antecedent unreasonable application of federal law, the requirement set forth in § 2254(d)(1) is satisfied[, and a] federal court must then resolve the claim without the deference AEDPA otherwise requires." Panetti, 551 U.S. at 953, 127 S.Ct. 2842. In Panetti, the Supreme Court considered a post-AEDPA habeas petition by a prisoner alleging that he was incompetent to be executed. Id. at 934-35, 127 S.Ct. 2842. In reviewing this claim, the Court first concluded that the state court had unreasonably applied clearly established law in implicitly concluding that it had offered constitutionally adequate procedures to resolve the prisoner's claim of incompetence. Id. at 952-53, 127 S.Ct. 2842. Because the state court's determination of the merits of the prisoner's competency claim depended on those clearly inadequate procedures, the Supreme Court did not defer to that determination, but rather considered the merits de novo. See id. at 954, 127 S.Ct. 2842.

We see no reason why our approach should differ where a state court's adjudication of a claim is dependent on an antecedent unreasonable determination of fact. Under § 2254(d), the unreasonable determination of fact alone authorizes a federal court to grant habeas relief-provided, of course, that the petitioner's constitutional rights were in fact violated. Section 2254(d) is phrased in the disjunctive: we may grant relief only if the state court decision was contrary to or involved an unreasonable application of clearly established Supreme Court law or if the state court decision was based on an unreasonable determination of fact. See 28 U.S.C. § 2254(d). Section 2254(d)'s plain language therefore does not require that a state court decision involve both an unreasonable determination of fact and an unreasonable application of law before we may grant relief.

We therefore hold that, when a state court adjudication is based on an antecedent unreasonable determination of fact, the requirement set forth in § 2254(d) is satisfied, and we may proceed to consider the petitioner's claim de novo. The Eleventh Circuit, in a recent en banc decision with only one judge dissenting on an unrelated issue, agrees, and we have found no case from any other circuit that holds otherwise. See Jones v. Walker, 540 F.3d 1277, 1288 n. 5 (11th Cir. 2008) (en banc).

This understanding is consistent with the principle of deference. Because we do not know what the state court would have decided had it applied the law to the correct facts, there is no actual decision to which we can defer. Continuing to apply AEDPA deference even after concluding that the state court had unreasonably determined the facts to which it applied the law would therefore require us to assess the reasonableness of a decision that the state court never actually reached. Yet it is beyond question that the AEDPA does not require us to defer to such hypothetical decisions. For instance, on several occasions, the Supreme Court has considered ineffective assistance of counsel claims under the AEDPA in cases where the relevant state courts had not reached one of the two Strickland prongs. See Porter, 130 S.Ct. at 452; Rompilla, 545 U.S. at 390, 125 S.Ct. 2456. In those cases, the Supreme Court did not ask whether it would have been objectively unreasonable for the state court to decide the unreached Strickland prong adversely to the habeas petitioner, but rather reviewed that prong de novo, as there was no actual state court decision to which it could defer. Porter, 130 S.Ct. at 452; Rompilla, 545 U.S. at 390, 125 S.Ct. 2456. Like the state courts that never reached one of the Strickland prongs in those cases, the state court here similarly never considered whether Detrich's defense was prejudiced by his counsel's failure to present expert evidence of Detrich's abnormal neuropsychological functioning. We therefore likewise examine this question de novo.

Evaluating Detrich's claim of prejudice de novo, we conclude that, had his trial counsel presented expert evidence like that provided by Dr. Briggs, there is a "reasonable probability" that Detrich would have received a sentence less than death, particularly in light of the sentencing scheme in effect at the time. Under Arizona law at the time of Detrich's sentencing, the sentencing judge had to "weigh the mitigating circumstances against the aggravating circumstances to determine if leniency is required." State v. Gretzler, 135 Ariz. 42, 659 P.2d 1, 13 (1983) (internal quotations and citation omitted). The law enumerated ten aggravating factors whose presence would require imposition of the death penalty unless the judge also found "mitigating circumstances sufficiently substantial to call for leniency." Ariz.Rev. Stat. § 13-703(E), (F). In weighing the aggravating and mitigating factors, "the number of aggravating and mitigating circumstances is not dispositive, but rather their gravity." Gretzler, 659 P.2d at 13. In Detrich's case, expert evidence demonstrating that he suffered from neuropsychological dysfunctions would have affected the "gravity" of the circumstances on both sides of the balance: it would have not only strengthened the mitigating circumstances, but also weakened the aggravating factor — that the crime was especially cruel, heinous, and depraved — that authorized Detrich's death sentence. We are therefore convinced that there is a reasonable probability that Detrich would have received a more lenient sentence had his trial counsel presented such evidence.

The statute enumerated five mitigating circumstances, but also authorized sentencing judges to consider other, non-statutory mitigating factors. Ariz.Rev. Stat. § 13-703(G); accord State v. Castaneda, 150 Ariz. 382, 724 P.2d 1, 12 (1986) ("[The statute] lists some mitigating circumstances though the court is not limited to those listed but must set forth any mitigating factor which might call for leniency.").

a

Dr. Briggs's report provides qualitatively new information that the sentencing judge did not consider: it indicated for the first time that the crime may have been driven by an instinctive, learned behavior that stemmed from Detrich's abusive past and by Detrich's neuropsychological deficits. This expert evidence would have added significant mitigating weight.

First, Dr. Briggs explained how Detrich's upbringing shaped his psychological profile. He explained that Detrich's abusive background — in which he was "taught to fear and hate" — created "an atmosphere where the problem solving process is more `take advantage of the person before the person does it to me.'" According to Dr. Briggs, "this stance makes perfect sense in the world where adults have and teach no boundaries and no respect, much less sympathy and feelings other than inappropriate touch and abuse." He further opined that the abuse Detrich suffered caused him to develop a mindset "in which instinct took over and reason could not be accessed." Dr. Briggs's analysis thus explains both the mindset that led Detrich to commit this crime and the cause of that mindset.

By suggesting a "causal nexus" between Detrich's abusive childhood and the crime, this expert analysis could "impact `the quality and strength of the mitigation evidence.'" State v. Tucker, 215 Ariz. 298, 160 P.3d 177, 201 (2007) (quoting State v. Newell, 212 Ariz. 389, 132 P.3d 833, 849 (2006)). We have repeatedly noted that expert testimony like this can offer a powerful explanation of a defendant's crimes, and that the failure to introduce such evidence can therefore prejudice a defendant. In Douglas v. Woodford, we held that counsel's failure to present expert testimony explaining the possible causal link between the defendant's childhood and his crime was prejudicial. Douglas v. Woodford, 316 F.3d 1079, 1090 (9th Cir. 2003). Although counsel had argued that the defendant's disadvantaged childhood had created a "demon" within him that had contributed to the crimes, that argument "lacked force without some expert testimony to back it up." Id. Similarly, in Caro v. Calderon, we held that counsel may have prejudiced the defendant where he failed to present expert testimony "to explain the ramifications" of the defendant's life experiences on his behavior. Caro v. Calderon, 165 F.3d 1223, 1227 (9th Cir. 1999). There is a reasonable probability that Detrich's counsel's failure to provide available expert testimony about how Detrich's abusive childhood influenced his psychological development and contributed to his crime likewise prejudiced Detrich's defense.

Second, Dr. Briggs's report not only offers an expert explanation of the causal link between Detrich's horrific childhood and his crime, but also raises the possibility that his neuropsychological deficits may have contributed to the crime. In addition to indicating that Detrich's neuropsychological functioning was almost certainly deficient at the time of the crime, Dr. Briggs's report also explains that Detrich's neuropsychological deficits interacted with his emotional status to cause "a greater overall impairment in function" and that being under the influence of drugs further inhibits Detrich's abilities. Neuropsychological deficits would likely have had greater mitigating weight than the diagnoses reflected in the psychological reports that the sentencing judge considered, which concluded only that Detrich had probable antisocial personality disorder. See State v. Walton, 159 Ariz. 571, 769 P.2d 1017, 1034 (1989) (explaining that "some types of[mental] impairments clearly bear more weight than others," and that "personality disorders have not sufficed to tilt the balance in favor of leniency").

Although Dr. Briggs's report, standing alone, does not explain how Detrich's neuropsychological deficits would have affected his behavior, the state post-conviction court denied Detrich the opportunity to develop this evidence at an evidentiary hearing at which Dr. Briggs could have provided more detail about Detrich's deficits and their effects. In any event, the report does at least suggest some ways in which Detrich's deficits might have contributed to his crime. The report remarks on Detrich's impulsivity, calling him "very impulsive," noting impulsivity in several areas of testing, and characterizing his "basic problem" as being "impulsive and insist[ing] on having his own way regardless of the law or the feelings of other people." The report concludes that, given Detrich's history, his "decision-making, especially when compromised by alcohol, was not based on any consequence-driven thought process, but rather a leaned [sic] behavior that bypassed right or wrong." Dr. Briggs's report thus at least raises the possibility that Detrich's neuropsychological deficits caused the impulsivity that contributed to his crime. There is therefore a reasonable probability that the failure to introduce such expert evidence prejudiced Detrich's defense.

Moreover, expert analyses presented to the federal district court confirm that Detrich's neuropsychological deficits may have contributed to Detrich's crime in this way. Although Dr. Briggs's report alone convinces us that Detrich suffered prejudice from his counsel's failure to obtain and present an expert assessment of Detrich's neuropsychological functioning and of the effects of his horrific childhood on his mental state, the additional evidence presented in federal court reinforces this conclusion.

As an initial matter, we can properly consider this new evidence that Detrich presented for the first time in federal court, and that the state courts therefore had no opportunity to consider. We may consider such evidence so long as two conditions are met: First, the petitioner must have been "diligent in his efforts" to develop the evidence in state court. Michael Williams v. Taylor, 529 U.S. 420, 435, 120 S.Ct. 1479, 146 L.Ed.2d 435 (2000); accord Holland v. Jackson, 542 U.S. 649, 652-53, 124 S.Ct. 2736, 159 L.Ed.2d 683 (2004) (per curiam) (holding that new evidence can be considered so long as the petitioner "was not at fault in failing to develop that evidence in state court"). Second, the new evidence must not so "fundamentally alter the legal claim already considered by the state courts" as to render it unexhausted. Vasquez v. Hillery, 474 U.S. 254, 260, 106 S.Ct. 617, 88 L.Ed.2d 598 (1986).

Here, Detrich exercised diligence by presenting the state court with evidence of neuropsychological impairment and requesting a hearing to develop that evidence further. See Michael Williams, 529 U.S. at 437, 120 S.Ct. 1479 ("Diligence will require in the usual case that the prisoner, at a minimum, seek an evidentiary hearing in state court in the manner prescribed by state law."). He is not to blame for the state court's denial of that request.

Nor does the new evidence render Detrich's ineffective assistance of counsel claim unexhausted. "[N]ew factual allegations do not render a claim unexhausted unless they `fundamentally alter the legal claim already considered by the state courts.'" Weaver v. Thompson, 197 F.3d 359, 364 (9th Cir. 1999) (internal quotations and citation omitted). The new evidence that Detrich presents does not. Detrich provided the state court with a neuropsychological expert's initial analysis indicating he had some neuropsychological deficits and explaining he exhibited "impulsive" behaviors. Moreover, his post-conviction counsel specifically informed the state court that "complete" testing could show Detrich's impaired ability to reason or function in stressful situations and a biological cause of his impairments, and requested a hearing at which evidence of Detrich's brain damage could be developed. The new evidence presented in federal court simply offers details about Detrich's brain damage, the biological causes of his impairments, and his impaired ability to function. The new evidence thus does not "fundamentally alter" Detrich's claims, and the state court had "a meaningful opportunity to consider" them. Vasquez, 474 U.S. at 257, 106 S.Ct. 617. Particularly because the failure to present the new evidence to the state court "stemmed from the state courts' refusal to grant [the petitioner] an evidentiary hearing on the matter, rather than from any failure of diligence on his part," the new evidence does not render Detrich's claim unexhausted. Weaver, 197 F.3d at 364.

The expert testimony that Detrich presented to the federal district court provides significant new details about Detrich's neuropsychological deficits, their behavioral effects, and their causes. Particularly considering this new evidence, we are left with no doubt that, had Detrich's trial counsel enlisted the assistance of a mental health expert, there is a reasonable probability that Detrich would have received a sentence less than death.

In the district court, Detrich presented the expert opinion of neuropsychiatrist Dr. Amezcua-Patiño, who diagnosed Detrich with cognitive disorder secondary to congenital deficits and concluded that Detrich has "severe Neuro Psychiatric conditions that he acquired at birth, most likely as a result of Congenital malformations, worsened by significant abuse and neglect, and eventually further deteriorated by the use of drugs and alcohol." Dr. Amezcua-Patiño opined that Detrich's abnormal cognitive functioning "directly affected[his] ability to problem solve and control his actions . . ., including during the commission of [his] crimes," and "affected his ability to appreciate the wrongfulness of his actions." Another doctor, Dr. Froming, similarly noted damage in the frontal-subcortical areas of functioning, which resulted in impulsivity, impulsive errors, and problems in self-monitoring. Dr. Froming also described Detrich as "very fast responding with an inability to stop once he had started on something." Dr. Cuniff, a medical geneticist, similarly noted that when Detrich attempted a difficult task, he often would not abandon an ineffective strategy. According to Dr. Froming, when "[u]nder the influence of alcohol and other drugs, Mr. Detrich's ability to inhibit, plan ahead, make decisions, and perform in an error-free fashion would be significantly reduced, especially given his neurological damage." Dr. Froming even remarked that Detrich's behavior during the testing "was so extraordinary that I have not witnessed it in any of my previous evaluations since 1979." Finally, she opined that Detrich's functioning would have improved over time as he remained sober, indicating that Detrich's functioning would have been even more impaired at the time of the crime.

All of these expert analyses indicate how Detrich's deficits could have affected his behavior in a way that could explain his crime: cognitive deficits caused him to act impulsively and impaired his ability to control his actions. Such an explanation could have significant mitigating weight. As the Supreme Court has recognized, evidence that a defendant's "violent behavior was a compulsive reaction rather than the product of cold-blooded premeditation" could alter the selection of a penalty. Terry Williams, 529 U.S. at 398, 120 S.Ct. 1495. Evidence that such a compulsive reaction stemmed from neuropsychological deficits would have even greater mitigating force. Cf. Caro, 280 F.3d at 1258 (holding that the failure to introduce evidence of physiological defects that can cause "impulse discontrol" can prejudice a defendant because such evidence could reduce a defendant's "moral culpability").

Dr. Amezcua-Patiño also offered more detail on how Detrich's abusive childhood could have led to his neuropsychological deficits. The development of the prefrontal cortex — which controls reasoning, problem solving, motivation, and response flexibility — is "vitally dependent on reciprocal interactions with an emotionally attuned caregiver." Abuse impedes the normal development of the brain and likely leads to an underdeveloped cortex and a hyperactive response to stress. Extreme stress interferes with the functioning of the thinking part of the brain that is "particularly important in inhibiting the stress response." By establishing another causal link between Detrich's abusive childhood and his crime, this evidence would have increased the mitigating weight of Detrich's horrific childhood. See Tucker, 160 P.3d at 201.

In sum, had Detrich's sentencing counsel consulted a mental health expert to assist in the penalty phase of Detrich's second trial, he would have been able to present expert analyses showing how both neuropsychological deficits and Detrich's abusive childhood contributed to his crime. Had the sentencing judge heard such testimony, there is a reasonable probability that he would have afforded more weight to the mitigating circumstances.

The fact that the state has presented experts who disagree with Detrich's experts' analyses does not alter our conclusion. As the Supreme Court recently recognized, even where state experts identify problems with the tests and conclusions of defense experts, "it [is] not reasonable to discount entirely the effect that [the expert's] testimony might have had on the jury or the sentencing judge." Porter, 130 S.Ct. at 455.

Further, this is not a case where expert testimony is unnecessary because laypeople can easily understand the mitigating evidence without assistance. Recently, in Wong v. Belmontes, the Supreme Court held that a defendant did not suffer prejudice from his attorney's failure to introduce expert testimony about how the defendant's childhood could have contributed to his crimes because "the body of mitigating evidence . . . was neither complex nor technical. It required only that the jury make logical connections of the kind a layperson is well equipped to make." Wong v. Belmontes, ___ U.S. ___, 130 S.Ct. 383, 388, 175 L.Ed.2d 328 (2009). In Belmontes, however, the mitigating evidence was not about brain development and cognitive impairments, but rather about how traumas the defendant experienced as a child "caused him to `los[e] ground in comparison with his peers [both] academically [and] socially' and `intensified [his] sense of himself as defective, something from which he never recovered.'" See Belmontes v. Ayers, 529 F.3d 834, 853 (9th Cir. 2008), rev'd,

___ U.S. ___, 130 S.Ct. 383, 175 L.Ed.2d 328 (2009). Whereas a jury can use "its common sense or own sense of mercy," Belmontes, 130 S.Ct. at 388, to understand the psychological impact of an abusive childhood, a sentencer would likely need expert testimony to understand how a traumatic childhood could shape brain development in a way that would lead to impulsive behavior. Because expert testimony would provide additional explanations beyond a layperson's understanding of Detrich's inability to curb his impulses, there is a reasonable probability that this testimony would have increased the weight of the mitigating evidence.

b

In addition to strengthening the mitigating evidence, the expert evidence of Detrich's neuropsychological dysfunctions would have weakened the statutory aggravating factor that the sentencing court found to authorize imposition of the death penalty, i.e., that Detrich's crime was especially cruel, heinous, and depraved. Under Arizona law, "cruelty involves the pain and distress visited upon the victims, and . . . heinous and depraved go to the mental state and attitude of the perpetrator as reflected in his words and actions." Gretzler, 659 P.2d at 10. Based on the evidence before it, the sentencing court found that the number and severity of wounds, some of which were defensive, showed that the crime was especially cruel, and that Detrich's words and actions at the time of the crime evidenced a heinous and depraved state of mind. For the reasons we explain below, evidence of Detrich's neuropsychological deficits could have provided an alternative explanation of Detrich's words and actions and accordingly could have decreased the weight of the "especially cruel, heinous, and depraved" aggravating factor.

To be sure, evidence of Detrich's neuropsychological dysfunctions would not bear on the cruelty of the crime, which focuses on the victim's suffering. See Gretzler, 659 P.2d at 10. Because there is therefore no reasonable probability that this expert evidence would have led the sentencing court to find no aggravating circumstances whatsoever, the death penalty would have remained an option. See Ariz.Rev. Stat. § 13-703(E), (F). Nonetheless, under Arizona law, sentencing judges look not only to whether an aggravating circumstance is present, but also to the weight of that aggravating circumstance. See State v. Canez, 202 Ariz. 133, 42 P.3d 564, 596 (2002) (assessing the weight of the (F)(6) aggravating circumstance after reversing the finding of heinousness and depravity but upholding the finding of cruelty); State v. Gulbrandson, 184 Ariz. 46, 906 P.2d 579, 604 (1995) (noting that "[t]he (F)(6) aggravating circumstance would have even more weight if defendant had relished the murder"). Thus, even assuming that Detrich could not successfully rebut the state's argument that the murder was especially cruel, the neuropsychological evidence could diminish the weight of the aggravating factor by showing that the crime was not especially heinous and depraved.

Contrary to the district court's understanding, the heinous/depraved inquiry turns on the defendant's subjective state of mind, not on his mere words and acts. The Arizona Supreme Court has made clear that, to determine whether a crime was committed in an especially heinous or depraved manner, a court must consider "the killer's state of mind at the time of the offense," which "may be shown by his behavior." State v. Lujan, 124 Ariz. 365, 604 P.2d 629, 636 (1979); accord State v. Fulminante, 161 Ariz. 237, 778 P.2d 602, 620 (1988) ("The terms `heinous' and `depraved' focus upon a defendant's state of mind at the time of the offense, as reflected by his words and acts."); State v. Martinez-Villareal, 145 Ariz. 441, 702 P.2d 670, 680 (1985) ("In order to determine depravity the court must focus on defendant's state of mind. Defendant's state of mind may be inferred from behavior at or near the time of the offense." (citation omitted)). Because this factor focuses on a defendant's subjective state of mind, expert psychiatric evidence can demonstrate that a defendant's crime was not especially heinous or depraved. See Summerlin, 427 F.3d at 641-42 (holding that counsel's failure to present psychiatric evidence of a defendant's "lack of impulse and emotional control and organic brain dysfunction" prejudiced the defendant because that evidence "could directly counter . . . that the crime had been committed in a `heinous, cruel, or depraved manner'").

The Arizona Supreme Court has identified five factors that can lead to a finding of heinousness or depravity: the defendant's relishing of the murder, infliction of gratuitous violence, mutilation of the victim, senselessness of the crime, and helplessness of the victim. Gretzler, 659 P.2d at 11. In Detrich's case, the sentencing judge found that four of the five factors — all but mutilation — were present. According to the sentencing judge, Detrich's asking Charlton if he wanted "a shot" at the "dead" but "warm" body indicated that Detrich relished the crime. The forty wounds evidenced gratuitous violence. And the crime was senseless in that the victim posed no threat to Detrich, and the victim was helpless. These factors led the sentencing judge to conclude that the crime was especially heinous and depraved.

Expert testimony could have rebutted the finding that the gratuitous violence and Detrich's apparent relishing of the crime evidenced a heinous and depraved state of mind. First, expert testimony that Detrich exhibited an "inability to stop once he had started on something," and that his "ability to inhibit, plan ahead, [and] make decisions . . . would be significantly reduced, especially given his neurological damage," could have provided the sentencing judge with a different understanding of the victim's forty wounds. Rather than evidencing a "shockingly evil state of mind," Gretzler, 659 P.2d at 11, the sheer number of wounds reflected Detrich's impulsive and persistent behavior, which was caused by his neurological deficits. The state argues that Detrich's brain dysfunctions do not explain the gratuitous violence because there is no evidence that he was out of control, as Detrich apparently directed Charlton to drive out of town, told him to pull over, and later lied that he had been in a fight when his friend asked why he was covered in blood. That Detrich may have regained control after committing the murder, however, does not indicate that he also had control over his impulses while inflicting the wounds.

Second, given the neuropsychological evidence, there is a reasonable probability that the sentencing judge would have concluded that Detrich's comment offering Charlton "a shot" at the dead body evidenced impulsivity and immaturity, not relishing of the killing. Indeed, the Arizona Supreme Court has acknowledged that "post-murder statements suggesting indifference, callousness, or a lack of remorse constitute `relishing,' only when they indicate, beyond a reasonable doubt, that the killer savored or enjoyed the murder at or near the time of the murder." State v. Greene, 192 Ariz. 431, 967 P.2d 106, 115-16 (1998). The Arizona Supreme Court has, for example, found that a defendant's statement that he had "clubbed" a "faggot," while callous, did not establish relishing. Id. at 115. Similarly, that court rejected a finding of relishing where a defendant bragged that his victim had "squealed like a rabbit." State v. Graham, 135 Ariz. 209, 660 P.2d 460, 463 (1983). In that case, the court attributed the defendant's statement not to his savoring of the crime, but to his immaturity and need to impress his peers, which a psychological report had noted. Id. Similarly, here, expert analyses would have reported that Detrich was "very immature" and impulsive. These neuropsychological characteristics, explained in part by Detrich's brain development and abusive childhood, could have provided an alternative explanation of Detrich's callous statement to Charlton: his statement reflected immaturity and impulsivity, not his enjoyment of the act of killing Souter.

Thus, expert testimony about Detrich's neuropsychological dysfunctions could have shed a different light on Detrich's words and actions at the time of the crime. Had the sentencing judge heard expert testimony about Detrich's impulsivity, inability to change a course of action once started, and immaturity, there is a reasonable probability that he would have concluded that the callous statement to Charlton and the number of wounds reflected these neuropsychological characteristics, not relishing or knowing infliction of gratuitous violence that reflected a heinous and depraved state of mind. And there is more than a reasonable probability that this expert evidence would have at least precluded the sentencing judge from finding beyond a reasonable doubt that Detrich's actions reflected a heinous and depraved state of mind. See State v. Jordan, 126 Ariz. 283, 614 P.2d 825, 828 (1980) (holding that aggravating circumstances must be proven beyond a reasonable doubt). Moreover, had the sentencing judge not found gratuitous violence or relishing, he almost certainly would not have found the crime to have been especially heinous or depraved, because the other two factors urged by the state — senselessness of the crime and helplessness of the victim — rarely support a finding of heinousness and depravity on their own. See State v. Hyde, 186 Ariz. 252, 921 P.2d 655, 684 (1996).

There is therefore a reasonable probability that expert evidence of Detrich's neuropsychological deficits would have changed the aggravation side of the sentencing balance. Because there is a reasonable probability that the sentencing judge would not have found Detrich's crime to be especially heinous and depraved, there is a reasonable probability that he would have afforded less weight to the aggravating factor that authorized imposition of the death penalty here. If the aggravator had less weight, it would of course take less mitigating evidence to outweigh it.

c

Had Detrich's trial counsel presented expert evidence of Detrich's Neuropsychological dysfunction, there is a reasonable probability that the sentencing judge would have ascribed more weight to the mitigating circumstances and less weight to the aggravating circumstance that the crime was especially heinous, cruel, or depraved. Critically, there is a reasonable probability that these changes in weights on both sides of the sentencing balance would have resulted in a sentence less than death.

This is not a case where the aggravating factors are so overwhelming that a death sentence was all but assured. First, even for very gruesome crimes, the death penalty is not necessarily unavoidable. Douglas, 316 F.3d at 1091. Second, we have found that aggravating circumstances are so significant as to preclude a finding of prejudice only where the aggravating circumstances were worse, and the omitted mitigating evidence weaker, than those here. For example, in Bible v. Ryan, we held that the severity of the aggravating factors precluded a finding of prejudice where the defendant had kidnaped, stripped, molested, tied up, and murdered a nine-year-old girl, and where the omitted mitigating evidence showed only that the defendant had high fevers as a child that theoretically could have resulted in organic brain damage. Bible v. Ryan, 571 F.3d 860, 863-64, 870 (9th Cir. 2009). Here, not only are the aggravating circumstances of Detrich's crime less egregious, but he has offered evidence that he actually suffers organic brain damage that could help explain his crime. Cf. Bible, 571 F.3d at 871 (emphasizing that the defendant did not contend that he actually had organic brain damage in concluding that the failure to introduce this evidence did not prejudice him). The aggravating factors similarly far outweighed the omitted mitigating evidence in Woodford v. Visciotti where the Supreme Court held that the state court was not objectively unreasonable in finding no prejudice where a defendant, who had previously knifed a man and stabbed a pregnant woman, had committed a pre-planned armed robbery that involved an execution-style killing and another attempted killing. Woodford v. Visciotti 537 U.S. 19, 26-27, 123 S.Ct. 357, 154 L.Ed.2d 279 (2002). There, the mitigating evidence that counsel failed to present was far weaker than that here: the defendant had been berated as a child, had no self-esteem, moved twenty times growing up, was born with club feet, and suffered from depression, feelings of inadequacy and incompetence, and a possible seizure disorder. Id. at 26, 123 S.Ct. 357. Similarly, in Cox v. Ayers, we recently found no prejudice where the "weight of the aggravating factors was staggering," as the petitioner had "set out to kill everyone in [a] house, including children sleeping in their beds, . . . for money." Cox v. Ayers, 613 F.3d 883, 900 (9th Cir. 2010). Balanced against these "staggering" aggravating factors was "mostly cumulative" evidence of the petitioner's abusive childhood, but no evidence, new or otherwise, of any neuropsychological deficiencies that could explain the crime. Id. at 896-97, 900.

Because the mitigating evidence that Detrich's counsel failed to present is powerful, and because the aggravating circumstances surrounding Detrich's crime are not so severe as practically to preclude a finding of prejudice, there is a reasonable probability that the new evidence of Detrich's neuropsychological dysfunctions would have led the sentencing judge to conclude that the mitigating circumstances outweighed the aggravating circumstances and accordingly to impose a sentence less than death.

Because we conclude that the failure to introduce such expert mental health evidence alone prejudiced Detrich's defense, we need not decide whether Detrich's counsel's additional failure to introduce evidence of Detrich's abusive childhood, his successful adaptation to prison, or the impact that his execution would have on his family also prejudiced him.

In sum, Detrich's trial counsel's failure to conduct an adequate penalty phase investigation, and the resulting failure to present powerful available mitigating evidence, deprived Detrich of his Sixth Amendment right to the effective assistance of counsel. The state court unreasonably applied Strickland in concluding that counsel's performance was not deficient. Moreover, the state court's conclusion that Detrich's counsel's performance did not prejudice Detrich's penalty-phase defense was based on an unreasonable determination that Dr. Briggs's report indicated that Detrich's neuropsychological functioning was normal. Because available expert neuropsychological evidence that Higgins failed to obtain and present would have provided a powerful explanation of Detrich's crime, we conclude that there is a reasonable probability that Detrich would have received a sentence less than death if Higgins had provided adequate representation. We therefore hold that Detrich is entitled to habeas relief on his penalty phase ineffective assistance of counsel claim.

IV. CONCLUSION

For the reasons explained above, we reverse the district court's denial of habeas relief on Detrich's penalty phase ineffective assistance of counsel claim. The case is remanded for the district court to issue a writ of habeas corpus vacating Detrich's death sentence unless the state re-sentences Detrich within a reasonable time set by the district court. If the state chooses not to re-sentence, Detrich's sentence will automatically be converted to life in prison in accordance with Arizona law.

REVERSED and REMANDED.