United States v. Barrett

Full title: UNITED STATES OF AMERICA, Plaintiff - Appellee, v. KENNETH EUGENE BARRETT…

Court: UNITED STATES COURT OF APPEALS FOR THE TENTH CIRCUIT

Date published: Jan 19, 2021

985 F.3d 1203 (10th Cir. 2021)

Opinion

No. 19-7049

01-19-2021

UNITED STATES of America, Plaintiff - Appellee, v. Kenneth Eugene BARRETT, Defendant - Appellant.

Table of Contents

I. Introduction...1207

II. Background...1207

A. Factual Background...1208

1. Police Obtained a Warrant to Search Mr. Barrett's House...1208

2. Police Surveilled Mr. Barrett's House...1208

3. Police Attempted to Execute the Warrant...1208

4. Mr. Barrett Killed Trooper David Eales...1208

B. Procedural Background...1209

1. Trials and § 2255 Petition...1209

a. State trial proceedings and sentencing...1209 b. Federal trial...1209

c. Federal sentencing...1209

d. Direct Appeal – Tenth Circuit (Barrett I)...1212

e. 28 U.S.C § 2255 motion – District Court...1212

f. 28 U.S.C § 2255 appeal – Tenth Circuit (Barrett II)...1212

2. Evidentiary Hearing...1212

a. Mr. Barrett's evidence...1213

i. Dr. George Woods...1213

1) Direct examination by Mr. Barrett...1213

2) Cross-examination by the Government...1215

ii. Dr. Deborah Miora...1215

1) Direct examination by Mr. Barrett...1215

2) Cross-examination by the Government...1216

3) Redirect examination by Mr. Barrett...1216

iii. Dr. Jeanne Russell...1216

1) Direct examination by Mr. Barrett...1216

2) Cross-examination by the Government...1216

iv. Dr. Bill Sharp's report...1217

v. Family witnesses...1217

b. Government's evidence...1217

i. Dr. Steven Pitt...1217

1) Direct examination by the Government...1217

2) Cross-examination by Mr. Barrett...1218

ii. Dr. J. Randall Price...1218

1) Direct examination by the Government...1218

2) Cross-examination by Mr. Barrett...1219

3. Magistrate Judge's and District Judge's Decisions...1219

a. Magistrate Judge's report and recommendation...1219

i. Deficient performance...1219

ii. Prejudice...1219

b. Objections to the report and recommendation...1220

c. District court's decision...1220

i. Deficient performance...1220

ii. Prejudice...1220

III. Discussion...1221

A. Standard of Review...1221

B. Legal Background...1221

1. Ineffective Assistance of Counsel...1221

2. Mental Health Evidence...1222

a. Nature of the impairments...1222

b. Connection to the offense...1222

3. Abusive Background Evidence...1223

C. Analysis...1223

1. Aggravating Evidence...1224

2. Postconviction Mitigating Evidence...1224

a. Evidence of mental impairments...1224

i. Evidence of brain damage...1224

1) Mr. Barrett's evidence...1224

2) Government's response...1225

ii. Evidence of mental illnesses...1225

1) Bipolar disorder...1226

a) Mr. Barrett's evidence...1226

b) Government's response...1226

2) PTSD...1227

iii. Connection to the offense...1227

b. Evidence of an abusive upbringing...1229

3. Balancing the Aggravating and Mitigating Evidence...1229

a. Strengthened mitigation case...1230 b. Weakened aggravation case...1231

c. Minimal post-sentencing aggravation evidence...1231

IV. Conclusion...–1233

I. INTRODUCTION

Kenneth Eugene Barrett moved under 28 U.S.C. § 2255 to vacate his death sentence based on ineffective assistance of counsel in violation of the Sixth Amendment to the Constitution.¹ The district court denied relief. On appeal, this court remanded for an evidentiary hearing to determine whether defense counsel's performance at sentencing was deficient and prejudicial. Following the hearing, the district court found deficient performance but no prejudice, and again denied relief. Mr. Barrett has appealed. Exercising jurisdiction under 28 U.S.C. § 1291, we reverse.

1 Section 2255 provides that a federal prisoner "claiming the right to be released upon the ground that the sentence was imposed in violation of the Constitution or laws of the United States, ... may move the court which imposed the sentence to vacate, set aside or correct the sentence." 28 U.S.C. § 2255(a).

Mr. Barrett's death sentence was based on his intentionally killing a state trooper when drug task force officers attempted to execute a warrant at his home. The evidence at trial and at the sentencing hearing depicted him as having planned a lethal attack on police officers. In mitigation, defense counsel introduced testimony that he was a loved family member and good person who was sorry for his crime. But his counsel did not introduce evidence that he experienced abuse as a child; suffered from brain-damage, bipolar disorder, and post-traumatic stress disorder ("PTSD"); and struggled to exercise judgment in pressured situations. The jury recommended, and the court imposed, the death penalty.

After filing his § 2255 motion in district court, Mr. Barrett presented declarations and assessments from experts describing his abusive upbringing and mental impairments, information his trial counsel failed to present at sentencing. The district court denied relief. On appeal, we remanded for the district court to conduct an evidentiary hearing to determine (1) whether defense counsel was deficient in failing to develop and present mitigating evidence and, if so, (2) whether there was a reasonable probability the jury would not have recommended a death sentence if counsel had introduced the mitigating evidence.

The district court referred the evidentiary hearing to a magistrate judge, who conducted the hearing over seven days. Fourteen witnesses testified, including experts on both sides. After the hearing, the magistrate judge recommended the district court find that Mr. Barrett's counsel performed deficiently under the Sixth Amendment and that such performance prejudiced Mr. Barrett at sentencing. The Government conceded deficient performance but objected to the recommended finding of prejudice.

Without rehearing any testimony, the district court rejected the recommendation, held Mr. Barrett could not establish prejudice, and denied relief. Although it found counsel's performance deficient, the court observed that "the jury would not have been impressed by [Mr. Barrett's] experts' opinions." Barrett v. United States , No. CIV-09-105-RAW, 2019 WL 1406957, at *6 (E.D. Okla. Mar. 28, 2019).

The district court issued a certificate of appealability ("COA"), and Mr. Barrett appealed to this court. After reviewing the evidence presented at trial, sentencing, and the postconviction evidentiary hearing, we reverse.

II. BACKGROUND

We provide background on (A) the facts underlying Mr. Barrett's conviction, and (B) the procedural history of this case, including the evidence adduced at the evidentiary hearing on Mr. Barrett's family history and mental health.

A. Factual Background

We take the following facts from our previous opinions in this case. See United States v. Barrett , 496 F.3d 1079 (10th Cir. 2007) ( Barrett I ) (direct appeal); United States v. Barrett , 797 F.3d 1207 (10th Cir. 2015) ( Barrett II ) (first § 2255 appeal).

1. Police Obtained a Warrant to Search Mr. Barrett's House

In 1999, the Oklahoma State Drug Task Force ("Task Force") received a tip from a confidential informant that Mr. Barrett was manufacturing and distributing methamphetamine at his house. The confidential informant told the Task Force that Mr. Barrett had promised to kill any officer who came to arrest him and that he was operating his drug business at night because he believed law enforcement could not execute a search warrant at night.

The Task Force obtained a no-knock, day-or-night search warrant for the house. They planned to seize drugs and drug paraphernalia. The warrant allowed entry "without the normally required knocking and announcing due to the violent and unstable nature of [Mr. Barrett] and the danger posed to law enforcement personnel." Barrett I , 496 F.3d at 1082-83 (quotations and alterations omitted). Given the perceived risk, the Task Force obtained assistance from the Oklahoma Highway Patrol Tactical Team ("Tact Team").

2. Police Surveilled Mr. Barrett's House

Three state troopers surveilled Mr. Barrett's house in a white, unmarked Ford Bronco. Mr. Barrett's cousin was with him at the time. According to the cousin, Mr. Barrett saw the Bronco pass, recognized it as belonging to law enforcement, and said he did not care and "was going out in a blaze of glory." Barrett II , 797 F.3d at 1211 (quotations omitted).

The Tact Team devised a plan to serve and execute the warrant. Under the plan, two white Broncos and a marked patrol car with its emergency lights activated, each containing two troopers, would approach Mr. Barrett's house, single-file from the east, while three troopers in another patrol car would approach the fence south of the house. The six troopers approaching from the east would enter and secure the house. Of the three troopers approaching from the south, one would stay at the fence to provide sniper cover and the other two would apprehend anyone fleeing west.

3. Police Attempted to Execute the Warrant

Shortly after midnight, the Tact Team met with members of the Task Force, who planned to follow two minutes behind the Tact Team as it approached Mr. Barrett's property. The Tact Team then executed its plan. The lead Bronco approached from the east with its emergency lights off. The second Bronco and a patrol car followed. At least one vehicle activated its emergency lights. As the vehicles neared the house, bullets began hitting the lead Bronco's windshield.

Meanwhile, the troopers coming from the south arrived at the fence. Two of them scaled it and approached the house. They saw Mr. Barrett's son outside and ordered him to get on the ground. He eventually complied, and the officers handcuffed him as he yelled for Mr. Barrett.

4. Mr. Barrett Killed Trooper David Eales

The lead Bronco continued to receive gunfire as it neared the house. The driver, Oklahoma State Trooper John Hamilton, ducked between the seats. The passenger, Trooper David Eales, exited. Mr. Barrett shot him three times.

Trooper Hamilton threw a "flash-bang" stun grenade out the window, which temporarily stopped the gunfire. He exited the Bronco, and Mr. Barrett shot him in the shoulder as he moved toward Trooper Eales, who was lying face-down. Trooper Hamilton saw Mr. Barrett standing in his doorway holding a rifle. He shot at Mr. Barrett but missed as Mr. Barrett went inside. Another officer shot at Mr. Barrett through a window and hit his legs.

Officers dragged Mr. Barrett out to the front yard. He reached for a pistol in his waistband, but the officers subdued him and took the gun.

Trooper Eales died. An autopsy revealed that one of the three bullets entered his upper back, broke four ribs, and perforated his left lung and aorta.

B. Procedural Background

Unless otherwise noted, we take the following procedural history from our previous opinions in this case.

1. Trials and § 2255 Petition

a. State trial proceedings and sentencing

Mr. Barrett was charged in state court with one count of first-degree murder, one count of shooting with intent to kill, and two counts of discharging a firearm with intent to kill. The case proceeded to trial but ended in a hung jury.

Mr. Barrett was retried almost two years later. The jury acquitted him on the first-degree murder charge and found him guilty of the lesser-included crime of first-degree manslaughter. The jury also acquitted on the charge of shooting with intent to kill and instead found him guilty of the lesser-included offense of assault and battery with a dangerous weapon. It acquitted him on the two charges of discharge of a firearm with intent to kill.

The court sentenced Mr. Barrett to consecutive terms of 20 years in prison on the manslaughter conviction and 10 years on the assault and battery conviction. He did not appeal his convictions or sentences.

b. Federal trial

After the state trials, a federal grand jury indicted Mr. Barrett on three counts:

• Count 1 – using and carrying a firearm during and in relation to drug trafficking crimes and possessing a firearm in furtherance of such drug trafficking offenses, resulting in death, in violation of 18 U.S.C. § 924(c)(1)(A) and (j).

• Count 2 – using and carrying a firearm in relation to a crime of violence—the killing of a state law enforcement officer engaged in or on account of the performance of such officer's official duties—and possessing a firearm in furtherance of such crime of violence, in violation of 18 U.S.C. § 924(c)(1)(A) and (j).

• Count 3 – intentionally killing, during the commission of a drug trafficking crime, a state law enforcement officer engaged in the performance of his official duties, in violation of 21 U.S.C. § 848(e)(1)(B).

See Barrett I , 496 F.3d at 1086. The case went to trial, and the jury found Mr. Barrett guilty of all three counts.

c. Federal sentencing

In 1994, Congress enacted the Federal Death Penalty Act, 18 U.S.C. §§ 3591 - 3598. Under the Act, when a defendant is convicted of federal capital offenses, the government may petition the court to hold a "separate sentencing hearing to determine the punishment to be imposed." 18 U.S.C. § 3593(b). At this hearing, the factfinder "shall consider whether [the aggravating factors] sufficiently outweigh [the mitigating factors] to justify a sentence of death." Id. § 3593(e).²

2 Statutory aggravating and mitigating factors are listed in § 3592, but the jury also may consider "any other aggravating factor for which notice has been given," and any other mitigating factor. 18 U.S.C. § 3592(a), (d).
"The burden of establishing the existence of any aggravating factor is on the government, and is not satisfied unless the existence of such a factor is established beyond a reasonable doubt." Id. § 3593(c). "A finding with respect to any aggravating factor must be unanimous." Id. § 3593(d).
"The burden of establishing the existence of any mitigating factor is on the defendant, and is not satisfied unless the existence of such a factor is established by a preponderance of the information." Id. § 3593(c). "A finding with respect to a mitigating factor may be made by 1 or more members of the jury." Id. § 3593(d).

When the court held the § 3593 sentencing hearing in this case, the Government presented aggravating evidence, and Mr. Barrett presented mitigating evidence. The Government also relied on the guilt phase evidence to support its aggravation case.³

3 The court instructed the jury that it could consider "all the evidence received at the guilt phase of this trial to the extent that it is relevant to ... the existence of any ... aggravating or mitigating factors." Suppl. ROA at 5267-68.

The Government pointed to evidence that Mr. Barrett created a grave risk to other people in addition to Trooper Eales and planned and premeditated the shooting. In Barrett I , we recounted the trial evidence:

[Mr.] Barrett had been aware for some time of the outstanding warrant for his arrest, and anticipated that law enforcement officials would come to his house to arrest him at some point. Despite that awareness, or perhaps because of it, [Mr.] Barrett exhibited a defiant attitude towards law enforcement officials. On the front gate leading to his residence, [Mr.] Barrett had installed a sign reading: "Keep Out. I don't give a shit who you are, if you cross my gate or come on my property, I'll shoot." Further, in the months and weeks leading up to the date of the shooting, [Mr.] Barrett regularly told friends and family that if law enforcement officers came to his house, "there was going to be a shootout," "he would shoot the first police that came through his door," and "he was going to take out as many law enforcement officers as he could before they got him." Indeed, on the evening of [the shooting], [Mr.] Barrett observed three of the Tact Team members drive by his residence in an unmarked vehicle, and subsequently stated to his cousin, Travis Crawford, that he knew the vehicle belonged to law enforcement officials, that he didn't "give a fuck" if they came back to serve the arrest warrant, and that "he was going out in a blaze of glory" if they did so.

[Mr.] Barrett's conduct in the months, weeks, and days leading up to the shooting incident suggests his threats were far from idle. [Mr.] Barrett possessed multiple firearms at his residence, including five rifles, three shotguns, and two pistols. During the day, [Mr.] Barrett typically kept a rifle nearby. He also carried a nine millimeter pistol in his pants at all times.

As for the night of the shooting incident, the evidence presented at trial was more than sufficient to have allowed the jury to reasonably find that [Mr.] Barrett knew it was law enforcement officials

who were approaching his residence en masse.

496 F.3d at 1113 (citations and brackets omitted).

The Government introduced evidence at the sentencing hearing concerning the impact of Trooper Eales's death on his family, including a statement from his daughter and testimony from his sister, mother, and widow. Id. at 1099-1100.

In Barrett II , we summarized Mr. Barrett's mitigating evidence as follows:

[Mr. Barrett's] trial attorneys called his mother, father, stepmother, brother, ex-wife, uncle, and cousin to testify that [Mr. Barrett] was a loved family member and good person who was sorry for killing [Trooper] Eales. The defense also called a state-court clerk to testify about [Mr. Barrett's] state-court proceedings and his lack of prior felonies, several state-prison employees to testify that [he] was a well-behaved prisoner, two neighbors to testify that [he] was a good mechanic and a nonviolent person, and a bondsman to testify that he had no concerns about [Mr. Barrett's] being a danger while his bench warrant was outstanding. None of the witnesses discussed [Mr. Barrett's] mental health or troubled background in any significant detail, and [Mr. Barrett's] closing argument included only one mention of his commitment to a hospital almost 20 years earlier "for drug abuse and other treatment," and that was in the context of describing his "rocky" marriage.

Barrett II , 797 F.3d at 1224 (citation omitted).

After hearing the aggravation and mitigation cases, the jury returned findings of aggravating and mitigating factors. It found three aggravating factors on Count 3:

• "[Mr.] Barrett committed the offense after substantial planning and premeditation," Barrett I , 496 F.3d at 1087 ;

• "in the commission of the offense or in escaping apprehension for the offense, [Mr. Barrett] knowingly created a grave risk of death to one or more persons in addition to [Trooper] Eales," id. ; and

• Mr. Barrett caused "injury, harm, and loss to the victim's family," id.

As mitigating factors on all three counts, some or all the jurors found that Mr. Barrett:

• had accepted responsibility for the death of Trooper Eales (found by five jurors with respect to each count);

• had been convicted and punished for the death of Trooper Eales in state court (found by five jurors with respect to each count);

• at the time of the shooting incident, had no prior felony convictions (found by all twelve jurors with respect to each count);

• was a father (found by all twelve jurors with respect to each count);

• was a loved son and stepson (found by all twelve jurors with respect to each count);

• was a good neighbor and friend (found by seven jurors with respect to each count);

• would not present a future danger to society by being imprisoned for life without possibility of release (found by two jurors with respect to each count);

• never left his residence during 1999 (found by one juror with respect to each count); and

• had been mistreated by a police officer when Mr. Barrett was seventeen years old (by breaking Mr. Barrett's jaw during a fight) (found by six

jurors with respect to each count). Also found were:

• other factors in Mr. Barrett's childhood, background, or character mitigated against imposition of the death sentence (found by one juror with respect to Counts 1 and 2, and by two jurors with respect to Count 3); and

• death would affect his child, family, and friends (found by all 12 jurors with respect to each count).

Id. at 1087-88. The jury found no mitigating factors related to Mr. Barrett's mental health.

The jury recommended sentences of life imprisonment without the possibility of release on Counts 1 and 2, and death on Count 3. The district court imposed the recommended sentences.

d. Direct Appeal – Tenth Circuit (Barrett I)

Mr. Barrett appealed his convictions, arguing various procedural and substantive errors, as well as cumulative error. See generally Barrett I , 496 F.3d 1079. We affirmed. Id. at 1121.

e. 28 U.S.C § 2255 motion – District Court

Mr. Barrett petitioned for relief under 28 U.S.C. § 2255, raising 19 challenges, including ineffective assistance of counsel. See Barrett v. United States , No. 09-CIV-105-JHP, 2012 WL 3542609, at *1, *5 (E. D. Okla. Aug. 16, 2012). The district court denied the motion and declined to issue a COA. Id. at *94.

f. 28 U.S.C § 2255 appeal – Tenth Circuit (Barrett II)

Mr. Barrett then sought a COA from this court. See Barrett II , 797 F.3d at 1213. We granted a COA on seven issues related to ineffective assistance of counsel. Id.

We affirmed the district court's denial of all but one of Mr. Barrett's ineffective assistance claims. Id. at 1232. We reversed and remanded "for the district court to hold an evidentiary hearing on whether the performance of trial counsel was deficient in not investigating [Mr. Barrett's] background and mental health," and to determine "whether [Mr. Barrett] suffered prejudice from any deficiency during the penalty phase of his trial." Id.

As to deficient performance, we said Mr. Barrett "ha[d] presented a case that his defense team did little to investigate his background and mental condition," despite the fact that "the record suggest[ed] [his] trial attorneys had indications that [his] mental health and background merited further investigation." Id. at 1225-26.

As to prejudice, we found that, "[t]aking Mr. Barrett's proffered mitigation evidence as true, it is sufficient to require an evidentiary hearing." Id. at 1229.

2. Evidentiary Hearing

On remand, the district court assigned the evidentiary hearing to the magistrate judge who had been involved with the case for more than a decade. The magistrate judge assessed (1) whether Mr. Barrett's counsel performed deficiently by failing to investigate and present evidence of his alleged abusive upbringing and mental impairments, and, if so, (2) whether that deficient performance prejudiced Mr. Barrett. Because the district court found deficient performance, we discuss only the evidence related to whether there was prejudice. As we describe in more detail below, "the [prejudice] question is whether 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." Wilson v. Trammell , 706 F.3d 1286, 1305 (10th Cir. 2013) (quotations and alterations omitted).

We summarize the key testimony and exhibits from the evidentiary hearing. See id. at 1307 ("In evaluating these claims of prejudice, we look to the testimony at the evidentiary hearing, together with the exhibits offered in evidence at that hearing, to see what likely would have been presented at trial if Defendant's counsel had done what he contends they should have."). The prejudice issue calls for a "probing and fact-specific analysis, which will necessarily require [the] court to speculate as to the effect of the new evidence" in light of the evidence presented at trial and sentencing. See Barrett II , 797 F.3d at 1229 (quotations omitted). We recount (a) Mr. Barrett's evidence, and (b) the Government's rebuttal evidence.

a. Mr. Barrett's evidence

i. Dr. George Woods

Mr. Barrett's primary mental health expert was Dr. George Woods, a physician specializing in psychiatry and neuropsychiatry. His neuropsychiatric practice focused on the relationship between "brain and behavior, brain functioning and behavior, [and] brain anatomy and behavior." ROA, Vol. 4 at 529. He was formerly the chief of staff at three hospitals. In the mid-1990s, he was the clinical director of the New Beginnings Clinical Dependence Program, which "was one of the first programs in California to really look at both the [chemical dependency] component as well as the psychiatric component [of cooccurring disorders] in order to try to parse those out." Id. at 531. He also taught medical school courses in forensic psychiatry and geriatric psychiatry, and a law school course on mental health and the law.

1) Direct examination by Mr. Barrett

Dr. Woods testified that he performed comprehensive psychiatric examinations of Mr. Barrett in 2009 and 2017. He interviewed Mr. Barrett and reviewed his medical history, previous neuropsychological evaluations, family history, "records of the ... offense itself," "neuropsychological testing of Dr. Myla Young," and academic records. Id. at 537-38. He noted that, before the examination in 2009, "a comprehensive understanding of [Mr. Barrett's] social history had not been developed," which he called "the most scientific tool one has in understanding both medical and psychiatric disorders." Id. at 537.

According to Dr. Woods, "psychiatric disorders ... are familial," and "there can be a genetic component." Id. at 538. He said a subject's family history thus provides "a real understanding of what may be occurring," and that it can "occasionally [allow for] diagnosis." Id.

Dr. Woods described Mr. Barrett's family history—"go[ing] back three or four generations"—of suicidal behavior, bipolar disorder, mood disorders, depression, and "[r]eactive violence." Id. at 539-40. He explained that these are "heterogenous disorders," relevant in "the same way that one would look for a family history of diabetes or a family history of hypertension." Id. at 540-41. He also observed that Mr. Barrett's mother drank while pregnant, see id. at 545, and that Mr. Barrett's father hit him and his mother, see id. at 546.

Dr. Woods described Mr. Barrett's "lifetime history of being suicidal" and his three psychiatric hospitalizations in 1986, 1994, and 1995. Id. at 546, 547. He explained that, in 1995, Mr. Barrett was treated as a mental illness patient when he was hospitalized, and that he was given Haldol, an antipsychotic medication. See id. at 547-48. Although Dr. Woods noted that Mr. Barrett was using drugs when he was admitted, he said that the doctors "didn't just see [his behavior] as a function of Mr. Barrett being a drug user." Id. at 548.

Dr. Woods said he had diagnosed Mr. Barrett with bipolar disorder. He based his diagnosis on Mr. Barrett's (1) "family history [of] bipolar disorder" and mood disorders; (2) "[m]ood lability," which "describes this swinging of mood that one often sees" in bipolar disorder ; (3) "pressured speech," which is a symptom of bipolar disorder ; and (4) history of "significant drug use ... that started at 10 or 11" and continued throughout his life. Id. at 548-49.

Dr. Woods explained the connection between chemical dependency and bipolar disorder, noting that "[a]bout 65 percent of people that ... carry a diagnosis of chemical dependency of one substance or another also meet[ ] the diagnosis of bipolar disorder." Id. at 542. But he noted that several of Mr. Barrett's symptoms "aren't necessarily consistent with drugs." Id. at 551. He noted that pressured speech, for example, is not a symptom of methamphetamine abuse but is a symptom of bipolar disorder.

Dr. Woods also testified that Mr. Barrett displayed symptoms "consistent with [PTSD]," likely resulting from his father's violence and exacerbated by his mother's alcoholism and neglect, which left Mr. Barrett without any "coping resources" for the violence he witnessed and experienced. Id. at 557, 570. He explained that Mr. Barrett's drug use "couldn't have helped" with his disorders, but that drug use was not the basis for the diagnoses. Id. at 574-75.

Dr. Woods agreed with Dr. Myla Young, who gave Mr. Barrett neuropsychological tests and found "significant impairments" in his orbital, frontal lobe, and temporal lobe. Id. at 560. He explained these impairments can compromise both executive functioning—such as "being able to understand the context of situations"—and academic functioning. Id. Dr. Woods said that Mr. Barrett had been hit in the head with a steel ball when he was in elementary school, see id. at 685, and reported head injuries stemming from a broken jaw when he was 16 or 17, see id. at 686-87.

At the end of his direct examination, Dr. Woods had the following exchange with defense counsel:

Q. All right. Now, are you familiar, Doctor, generally with what the facts of this case are as to how the offense was committed, under what circumstance it was committed, and things of that nature?

A. Yes, sir.

...

Q. What sort of effect or impact do Mr. Barrett's neurological deficits and his psychiatric illnesses have in his ability to perceive and react to that kind of situation?

A. I -- I think the first thing we have to take into consideration, Mr. Autry, is that these symptoms of psychiatric and neurological disorder are comorbid. And what that means is that they interact with each other, and they undermine each other so that his brain impairments really do a job on his psychiatric disorder. And vice versa. And they create circumstances that ... is even worse than would be from day-to-day. The idea of a hyperreactive response is completely consistent with both his history, his mood disorder, as well as what we would see in someone that [has] these kinds of brain impairments, this misperception of -- of the circumstances. Once again, we talk about executive functioning, being able to weigh and deliberate -- effectively weigh and deliberate. And this clearly was an example of misperceiving and not being able to effectively weigh and deliberate.

Id. at 568-70. Dr. Woods also noted that "[t]here appears to be a deterioration in [Mr. Barrett's] mental state in the ... weeks and perhaps the days before this event in 1999." Id. at 571.

2) Cross-examination by the Government

During cross-examination, Dr. Woods admitted that no other doctor had diagnosed Mr. Barrett with bipolar disorder or PTSD. He also agreed that, "to make a diagnosis of bipolar, a person has to be separate from the use of substances for a period of time." Id. at 657-58. Although he acknowledged that Mr. Barrett said "there was never a time [he] wasn't using methamphetamine," id. at 661, he noted that "if you look at the medical records from his hospitalizations, you will find that, even though ... he reported [using methamphetamine], his toxicology screens were negative for methamphetamines," id. at 659.

Finally, Dr. Woods suggested that the reason Mr. Barrett was not diagnosed with bipolar disorder in 1995 was simply because "[t]hey didn't have all the data that we have." Id. at 680. He concluded that "[t]he behaviors that are described in ... 1995 are not the behaviors that are consistent with someone that's high on methamphetamines." Id. The "symptoms of methamphetamines ... don't include mood lability, they don't include mood swings, they don't include rapid thinking, exactly the kinds of symptoms that they described in [Mr. Barrett's] hospitalizations." Id.

ii. Dr. Deborah Miora

Mr. Barrett called Dr. Miora to explain the findings of Dr. Myla Young, a neuropsychologist who evaluated Mr. Barrett in 2009 but died before the hearing. Given Mr. Barrett's history of fetal alcohol exposure,⁴ head injuries, difficulties in school, suicide attempts, and substance abuse, Dr. Young had administered a "battery" of neuropsychological tests to assess his brain function. See id. at 1659, 1676-80. Mr. Barrett was incarcerated when Dr. Young administered the tests.

4 Despite this fetal alcohol exposure, no doctor diagnosed Mr. Barrett with fetal alcohol syndrome, and Dr. Miora expressly declined to do so. See ROA, Vol. 4 at 1818.

Dr. Miora was a clinical neuropsychologist and had worked in neuropsychology for more than 30 years. After earning her Ph.D., she obtained a post-doctoral certificate in neuropsychology, "which required two years of training in neuro anatomy [and] administration of tests." Id. at 1649. She became a full-time faculty member at a forensic graduate program that "emphasized neuropsychological assessment, as well as other ways of evaluating within the forensic world." Id. at 1650.

1) Direct examination by Mr. Barrett

According to Dr. Miora, Mr. Barrett's test results reflected weaknesses in his temporal, frontal, and parietal lobes, as well as his dorsal, lateral, and orbitofrontal areas of the brain. See id. at 1709, 1742.⁵ He showed impaired visual attention, which, Dr. Miora explained, causes "difficulty in pressured situations where he has to visually attend to what is going on and generate a response." Id. at 1726. He also demonstrated a possible learning disability. Finally, Dr. Miora observed that Mr. Barrett's impairments could compromise daily functioning and could inhibit judgment in unpredictable situations. See id. at 1757-59. 2) Cross-examination by the Government

5 Using Dr. Young's data, Dr. Miora estimated Mr. Barrett's I.Q. to be 82, which is roughly the 12th percentile. See ROA, Vol. 4 at 1712.

On cross-examination, Dr. Miora agreed that, although Mr. Barrett performed poorly on some tests, his performance was average on most.⁶ She also acknowledged that, in 1986, Mr. Barrett was diagnosed with mixed substance abuse. She admitted she did not know the cause of Mr. Barrett's impairments, and she agreed it "very well may have" been his substance abuse. Id. at 1818.

6 The Government pointed out, and Dr. Miora admitted, that on one of the tests she claimed showed impairments in Mr. Barrett's processing speed, she overlooked additional data suggesting the impairment was not as pronounced as she had suggested. See ROA, Vol. 4 at 1797-1803. The additional data showed Mr. Barrett performed worse than 95 percent of test-takers, rather than 99 percent, which, Dr. Miora conceded, "is not in the severely impaired range as [she] testified" on direct examination. Id. at 1803.

The Government called attention to a note Dr. Young made on Mr. Barrett's tests that "[c]urrent research demonstrates a significant relationship between" poor performance on one of the tests and inability to perform "daily functioning such as driving an automobile." Id. at 1830. The Government pressed Dr. Miora to explain how Mr. Barrett is able to drive given his low score. She responded that she did not think Dr. Young's note suggested Mr. Barrett would be unable to drive.

Finally, Dr. Miora admitted that she was unfamiliar with the details of Mr. Barrett's offense and could not "develop any opinion about what was going on with him at the time of the crime." Id. at 1834.

3) Redirect examination by Mr. Barrett

On redirect examination, Dr. Miora first noted that Mr. Barrett's average performance on most tests did not diminish the probative value of his poor performance on others.

Next, she explained that although she could not discern the injury that caused Mr. Barrett's impairments, she concluded he had brain damage based on his test results, not on records of a triggering event.

iii. Dr. Jeanne Russell

Dr. Jeanne Russell conducted a risk assessment of Mr. Barrett in 2003 for the defense in his state case. See id. at 1198-99. "[A] risk assessment involves determining in what situations or in what context a person will be dangerous." Id. at 1199. When she testified, Dr. Russell was a licensed psychologist in private practice. She formerly worked for the Oklahoma Forensic Center, previously known as the Eastern State Hospital, where she evaluated "20 to 25" inmates in death-penalty cases. Id. at 1196. She did not perform a comprehensive mental health examination on Mr. Barrett.

1) Direct examination by Mr. Barrett

Dr. Russell declined Mr. Barrett's trial counsel's request to be the mitigation investigator in Mr. Barrett's federal case. She testified at the evidentiary hearing that, although she could have updated her risk assessment of Mr. Barrett, there was insufficient time to be a mitigation investigator. She agreed that "a competent mitigation investigation [would not] consist solely of doing a risk assessment or updating a risk assessment." Id. at 1220. Dr. Russell testified that because she had not conducted a comprehensive mental health evaluation, she had not ruled out whether Mr. Barrett had a major mental illness such as bipolar disorder.

2) Cross-examination by the Government

During cross-examination, Dr. Russell acknowledged that, in 2004, she told Mr. Barrett's state court counsel that his "[r]ecords consistently indicated a substance abuse problem" and that "no major mental illness was noted." Id. at 1226; Gov't Exh. 63 at 250. She also acknowledged that, when she met with Mr. Barrett, he exhibited no symptoms of a major mental illness. She wrote in her report that "[n]either Dr. [Bill] Sharp[,] [the psychologist who evaluated Mr. Barrett for the defense,] nor the previous risk assessment found symptoms of major mental illness, such as schizophrenia or schizoaffective disorder or bipolar or major depression, at the time of his assessment." ROA, Vol. 4 at 1229-30 (quoting Gov't Exh. 35 at 8240).

iv. Dr. Bill Sharp's report

Dr. Sharp, a clinical psychologist, evaluated Mr. Barrett for the defense in 2002. He did not testify at the evidentiary hearing, but his report from the 2002 evaluation was entered into evidence. In the report, Dr. Sharp found that Mr. Barrett had a history of abusing alcohol, cannabis, methamphetamine, cocaine, and PCP, among other substances. He diagnosed Mr. Barrett with (1) amphetamine dependence, (2) cannabis dependence, (3) alcohol dependence, (4) nicotine dependence, (5) an unspecified learning disorder, (6) attention deficit/hyperactivity disorder, (7) avoidant personality disorder, (8) paranoid personality disorder, and (9) an "[o]rganic [i]mpairment [r]elative to [t]remor and [l]ong [t]erm [m]emory [d]ifficulty." Gov't Exh. 16 at 7. He recommended that Mr. Barrett be assessed regarding the possibility of organic neurological damage.

v. Family witnesses

Several of Mr. Barrett's family members testified at the evidentiary hearing. They reported that Mr. Barrett was a "hyper" child, ROA, Vol. 4 at 396, and that members of his family suffered from mental health issues, see id. at 405-06.

Stephen Barrett, Mr. Barrett's brother, provided the most pertinent family information. He testified that he believed Mr. Barrett's mother had "mental or emotional problems" and was an alcoholic. Id. at 433, 437. He noted that Mr. Barrett "failed at school," likely due to "the breakup of the family." Id. at 441-42. He further believed Mr. Barrett "had psychological issues growing up," and that "drugs and alcohol ... compounded those." Id. at 443. He also said that, as a child, he saw his mother and Mr. Barrett "in a physical altercation" when an argument turned into "a wrestling match on the floor." Id. at 446. Although Stephen Barrett testified at Mr. Barrett's federal trial, he did not testify about this family history.

b. Government's evidence

i. Dr. Steven Pitt

Dr. Steven Pitt is a forensic psychiatrist who evaluated Mr. Barrett in 2017. When Dr. Pitt testified, he was a clinical associate professor in the Department of Psychiatry at the University of Arizona Health Sciences Center in Phoenix. Previously, he was the director of forensic psychiatry at Arizona State Hospital and had a private practice in general psychiatry.

1) Direct examination by the Government

According to Dr. Pitt, Mr. Barrett started abusing drugs at about age 10. See id. at 1457. He repeatedly commented that Mr. Barrett has been a drug addict most of his life.

Dr. Pitt observed that, in several medical documents, Mr. Barrett denied a history of head injuries. See id. at 1482-83. He testified that the Social Security Administration denied Mr. Barrett benefits and noted that "[m]edical reports show that [Mr. Barrett is] moderately depressed, but there are no signs of a severe mental illness," such as bipolar disorder or PTSD. Id. at 1465 (quoting Gov't Exh. 10 (dated Feb. 20, 1987)).

As Dr. Pitt reported, Mr. Barrett was not diagnosed with bipolar disorder when he was discharged from his hospitalization in 1995. And Mr. Barrett's treatment then with Haldol, the antipsychotic drug, could have been "for a number of different reasons," including "agitation." Id. at 1468.

Dr. Pitt opined that, for two reasons, Mr. Barrett suffered from neither bipolar disorder nor PTSD. First, he pointed to Mr. Barrett's "history of drug use." Id. at 1471. He noted that Mr. Barrett might have had a "substance-induced mood disorder" but that he did not have bipolar disorder. Id. at 1473. He explained that "you don't make a diagnosis of something like bipolar disorder if the symptoms can be better explained by the physiological effects of a medical condition or a substance." Id. at 1477. Contrary to Dr. Woods's testimony, Dr. Pitt testified that methamphetamine use can cause pressured speech and mood lability.

Second, he noted that at least four psychologists had interviewed Mr. Barrett, and none diagnosed him with bipolar disorder. See id. at 1479. He explained that a person suffering from bipolar disorder cannot hide the symptoms for long.

Finally, Dr. Pitt explained that Mr. Barrett did not have PTSD because "there's no threshold event [that] qualifies for meeting the initial criteria for having [PTSD]," and "nowhere in any records ... was [Mr. Barrett] diagnosed with PTSD." Id. at 1493. He noted that, if "the symptoms ... [are] better explained by, say, substance use or some other organic process, then you need to consider that." Id. at 1383. He concluded that, "as of September 24th, 1999"—the date of the crime—he would diagnose Mr. Barrett with "[a]mphetamine dependence with physiologic dependence, features of an antisocial personality disorder, cannabis dependence, and [would] rule out a learning disorder not otherwise specified." Id. at 1498.

2) Cross-examination by Mr. Barrett

During cross-examination, Dr. Pitt agreed that (1) Mr. Barrett's family history of mental health problems heightened the risk he would suffer from similar issues; (2) Mr. Barrett suffered at least some head injuries, despite inconsistencies in the record regarding whether he ever lost consciousness from them; (3) substance abuse can be comorbid with bipolar disorder and PTSD; and (4) a diagnosis of depressive disorder would be reasonable given Mr. Barrett's suicide attempt with a shotgun when he was 24.

ii. Dr. J. Randall Price

Dr. Price is a clinical and forensic psychologist who evaluated Mr. Barrett in 2005. He has been licensed as a psychologist since 1983.

1) Direct examination by the Government

Dr. Price testified about Mr. Barrett's education and occupations. He explained that Mr. Barrett "wasn't interested [in school] and was truant frequently," and was ultimately suspended or expelled in eighth grade when he refused to cut his hair. Id. at 1273-74. Mr. Barrett then began working and eventually became a car mechanic.

Dr. Price next explained the tests he performed on Mr. Barrett. On the Reynolds Intellectual Assessment System ("RIAS"), Mr. Barrett "tested in the range of average intellectual functioning." Id. at 1296. Dr. Price noted that someone who had sustained a significant head injury and lost consciousness would be unlikely to perform as well as Mr. Barrett. See id. at 1297.

On the Repeatable Battery for the Assessment of Neuropsychological Status ("RBANS"), Mr. Barrett demonstrated better "visual-spatial constructional ability" than 95 percent of people, worse "language functioning" than 58 percent, and worse attention than 95 percent. Id. at 1301-02.

On the Personality Assessment Inventory ("PAI"), Mr. Barrett self-reported that his own problems stemmed from substance abuse. He also displayed symptoms of depression and paranoia. See id. at 1304-06.

On the Minnesota Multiphasic Personality Inventory-2 ("MMPI-2"), Mr. Barrett showed antisocial attitude and paranoia. See id. at 1306-09. Dr. Price explained that Mr. Barrett's symptoms were consistent with paranoid personality disorder (e.g., heightened sensitivity to criticism), as opposed to paranoid delusional disorder (e.g., delusional beliefs that "people are doing things to [him]"). Id. at 1309.

Dr. Price diagnosed Mr. Barrett with (1) a learning disorder, (2) polysubstance dependence, (3) a depressive disorder, and (4) a personality disorder. See id. at 1318-19. He did not believe Mr. Barrett suffered from PTSD because he did not show symptoms. Id. at 1358-60.

2) Cross-examination by Mr. Barrett

During cross-examination, Dr. Price acknowledged that a person's brain can function well in some areas and poorly in others. He admitted that the difference between Mr. Barrett's high scores in some areas and low scores in others is "not typical." Id. at 1334. But he attributed the disparity to "a problem with attention" rather than a specific mental illness. Id.

3. Magistrate Judge's and District Judge's Decisions

a. Magistrate Judge's report and recommendation

The magistrate judge's report and recommendation ("R&R") recommended that the district court find Mr. Barrett's counsel performed deficiently and that the deficiency prejudiced Mr. Barrett.

i. Deficient performance

First, the magistrate judge found that Mr. Barrett's "federal trial counsel neither hired a mitigation or mental-health professional nor attempted to investigate themselves in any depth [his] mental health or family background." Barrett v. United States , No. CIV-09-105-JHP, 2018 WL 7889966, at *4 (E.D. Okla. Aug. 10, 2018).

Second, the magistrate judge found that Mr. Barrett's counsel provided inadequate reasons for failing to investigate his client's mental health. He explained: (1) counsel did not sufficiently question Mr. Barrett and his family about his mental health, despite evidence warranting such questions; (2) Mr. Barrett did not oppose a mitigation strategy concerning his mental health; and (3) counsel's choice of a different mitigation strategy did not offset his failure to investigate Mr. Barrett's mental health.

ii. Prejudice

The magistrate judge recommended that the district court find counsel's deficient performance caused prejudice. He found that Mr. Barrett had a family history of mental health problems, violence, and substance abuse. He noted that Mr. Barrett (1) had been physically and emotionally abused as a child, (2) was developmentally delayed, (3) consumed alcohol and illegal drugs at a young age, and (4) was hospitalized once for a suicide attempt in 1986 and again in 1995 when he believed he was "losing his mind." Id. at *7.

Next, the magistrate judge addressed Mr. Barrett's psychological evaluations. He reviewed Dr. Sharp's report and diagnoses, noting that, "[a]mong Dr. Sharp's recommendations was that [Mr. Barrett] should pursue the possibility of organic neurological damage." Id. at *8. The magistrate judge then reviewed Dr. Woods's testimony and conclusions regarding Mr. Barrett's family history and mental disorder and Dr. Miora's testimony about Mr. Barrett's test results and organic brain damage. He concluded this evidence "could have had a powerful mitigating effect on the jury in this case." Id. at *10 (quotations omitted).

The magistrate judge turned to the Government's evidence. He summarized Dr. Price's testimony and the results of his tests on Mr. Barrett, and Dr. Pitt's testimony contradicting Dr. Woods's diagnoses.

The magistrate judge concluded that "[t]he evidence potentially available to [Mr. Barrett] regarding his mental health history and his family background would in all likelihood have aided substantially in mitigation, and the rebuttal evidence provided by the Government would not appear to have added substantially to the calculus of aggravation." Id. at *13. He thus recommended the district court find prejudice and vacate Mr. Barrett's death sentence on Count 3. Id.

b. Objections to the report and recommendation

Both Mr. Barrett and the Government objected to the R&R. Mr. Barrett objected to the magistrate judge's recommendation that Mr. Barrett be resentenced only on Count 3, rather than on all three counts.

The Government did not object to the deficient performance finding. It objected to the magistrate judge's recommended finding that defense counsel's deficient performance prejudiced Mr. Barrett.

c. District court's decision

After the magistrate judge issued his recommendation, the district court trial judge was recused and the case was reassigned to a different district judge. The judge rejected the magistrate judge's recommendation that the court vacate Mr. Barrett's death sentence.

i. Deficient performance

Because "the government [did not] object[ ] to the magistrate judge's finding that counsel's performance was deficient," the district court adopted the magistrate judge's conclusion that "trial counsel rendered constitutionally deficient performance in developing a mitigation strategy." Barrett , 2019 WL 1406957, at *3.

ii. Prejudice

The district court rejected the magistrate judge's finding of prejudice. Id. at *4.

First, it questioned whether "the jury would have given much weight" to Mr. Barrett's experts' testimony in light of "the evidence it heard over the course of the entire trial." Id. It quoted the factual background in Barrett I and concluded that Mr. Barrett "had clearly resolved to murder Trooper Eales or any other law enforcement officer long before this incident played out." Id. at *6.

Second, it said that "several aspects of the testimony presented at the evidentiary hearing convince this court that the jury would have rejected the mental health diagnosis provided by [Mr. Barrett's] expert witnesses." Id. It noted Mr. Barrett's history of drug abuse, inconsistent recollection of his head injuries, and his ability to drive, work as a mechanic, and effectively carry out his threats. Id. at *6-7. The court concluded it was "much more likely that the jury would have found that [Mr. Barrett] did, in fact, have a significant drug problem as opposed to a major mental health disorder." Id. at *7.

* * * *

The district court thus adopted the magistrate judge's finding of deficient performance and rejected the finding of prejudice. It granted a COA on the prejudice issue. Id. at *9.

III. DISCUSSION

On appeal, Mr. Barrett argues the district court should have adopted the magistrate judge's recommended finding that counsel's deficient performance caused prejudice. We agree. We therefore reverse and remand to the district court to grant habeas relief, vacate the death sentence on Count 3, and resentence Mr. Barrett on that count.

A. Standard of Review

In reviewing an ineffective assistance of counsel claim, we "accept the district court's underlying factual findings unless clearly erroneous," and "we review de novo whether ... any deficiencies prejudiced the defendant." United States v. Rodriguez-Rivera , 518 F.3d 1208, 1216 (10th Cir. 2008) (quotations omitted). The movant bears the burden of showing prejudice. See Wong v. Belmontes , 558 U.S. 15, 20, 130 S.Ct. 383, 175 L.Ed.2d 328 (2009) (per curiam). In assessing whether defense counsel's deficient performance prejudiced Mr. Barrett, the district court addressed whether there is a reasonable probability the jury would have reached a different result absent counsel's ineffectiveness. This is a "mixed question of law and fact with a significant legal component, which we ordinarily review de novo." Littlejohn v. Royal , 875 F.3d 548, 558 n.3 (10th Cir. 2017).

B. Legal Background

1. Ineffective Assistance of Counsel

The Sixth Amendment provides: "In all criminal prosecutions, the accused shall enjoy the right ... to have the Assistance of Counsel for his defence." U.S. Const. amend. VI. "[T]he right to counsel is the right to the effective assistance of counsel." Strickland v. Washington , 466 U.S. 668, 686, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984) (quotations omitted).

"We review claims of ineffective assistance of counsel according to the two-pronged test in [ Strickland ]." Barrett II , 797 F.3d at 1213. "Under this test, the defendant must show that his counsel's performance ‘fell below an objective standard of reasonableness,’ and that the deficient performance resulted in prejudice." Rodriguez-Rivera , 518 F.3d at 1216 (quoting Strickland , 466 U.S. at 688, 104 S.Ct. 2052 ).

"In [an ineffective assistance] challenge to a capital sentence, the [prejudice] 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." Wilson , 706 F.3d at 1305 (quotations and alterations omitted). "A reasonable probability is less than a preponderance of the evidence, but sufficient to undermine confidence in the outcome." Smith v. Mullin , 379 F.3d 919, 942 (10th Cir. 2004) (quotations omitted). "In a system ... where only a unanimous jury may impose the death penalty, the question is whether it's ‘reasonably probable that at least one juror would have struck a different balance.’ " Grant v. Trammell , 727 F.3d 1006, 1018-19 (10th Cir. 2013) (quoting Wiggins v. Smith , 539 U.S. 510, 537, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003) ) (brackets omitted); see also 18 U.S.C. § 3593(e) ("[T]he jury by unanimous vote, ... shall recommend whether the defendant should be sentenced to death.").

"To resolve whether there was prejudice, we do not consider omitted mitigation evidence in a vacuum." Wilson , 706 F.3d at 1305. "[A] reviewing court must consider all the relevant evidence that the jury would have had before it if the defendant had pursued the different path—not just the mitigation evidence the defendant could have presented." Id. at 1306 (quotations and brackets omitted). That is, "we must consider not just the mitigation evidence that Defendant claims was wrongfully omitted, but also what the prosecution's response to that evidence would have been." Id.

2. Mental Health Evidence

a. Nature of the impairments

When this panel discussed the prejudice issue in Barrett II , we noted that "evidence of mental impairments ‘is exactly the sort of evidence that garners the most sympathy from jurors,’ and that this is especially true of evidence of organic brain damage." Barrett II , 797 F.3d at 1231 (quoting Smith , 379 F.3d at 942 ); see also Grant v. Royal , 886 F.3d 874, 920 (10th Cir. 2018) ("Evidence of organic brain damage is something that we and other courts, including the Supreme Court, have found to have a powerful mitigating effect.") (quotations omitted); Littlejohn v. Trammell , 704 F.3d 817, 864 (10th Cir. 2013) ("Evidence of organic mental deficits ranks among the most powerful types of mitigation evidence available.").⁷ We also have noted the mitigating value of "[d]iagnoses of specific mental illnesses." Hooks v. Workman , 689 F.3d 1148, 1205 (10th Cir. 2012) (quotations omitted); see also Wilson , 706 F.3d at 1310 (noting the potential mitigating value of a "plausible diagnosis of paranoid schizophrenia" if the "Defendant's behavior during the crime could [be] tied to that disorder"). "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." Hooks , 689 F.3d at 1205.

7 Organic brain damage often stems from "severe head trauma." See Alverson v. Workman , 595 F.3d 1142, 1162 (10th Cir. 2010) ; see also Hooks v. Workman , 689 F.3d 1148, 1205 (10th Cir. 2012).

By contrast, "evidence suggesting a defendant suffers from generalized personality [dis]orders may be less powerful." Grant , 727 F.3d at 1021 (quotations omitted). Such disorders "are diagnosed on the basis of reported behavior, are generally inseparable from personal identity, and are often untreatable through medical or neurological means." Wilson , 536 F.3d at 1094.

Further, although "[e]vidence of organic mental deficits ranks among the most powerful types of mitigation evidence available," this proposition is "qualified in two salient respects." Littlejohn , 875 F.3d at 559. First, not all evidence of brain damage "has the same potency in the Strickland prejudice analysis," and we must "examine the precise nature of the alleged organic brain damage." Id. Second, "in some instances," brain damage evidence may be "just as likely—if not more likely—to have ... an aggravating effect rather than a mitigating effect on a sentencing jury." Id. at 560. For example, "evidence about mental health problems ... often possesses a ‘double-edged nature,’ " as jurors may conclude that the defendant is "simply beyond rehabilitation." Grant , 727 F.3d at 1021 (quotations omitted); see also Gilson v. Sirmons , 520 F.3d 1196, 1250 (10th Cir. 2008) (finding counsel's failure to introduce evidence of the defendant's organic brain damage did not prejudice him because it highlighted his violent nature and portrayed him as a "continuing threat").

b. Connection to the offense

Binding precedents recognize that evidence of mental impairments has substantially more mitigating value when it helps explain the defendant's criminal behavior. See, e.g. , Williams v. Taylor , 529 U.S. 362, 398, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000) (finding prejudice when the evidence indicated the defendant's "violent behavior was a compulsive reaction rather than the product of cold-blooded premeditation"); Smith , 379 F.3d at 943 (finding prejudice when the jury had not heard "an explanation of how [the defendant's] organic brain damage caused" him to commit the crime).

For example, in Anderson v. Sirmons , 476 F.3d 1131 (10th Cir. 2007), we found prejudice stemming from counsel's failure to investigate and present evidence that the defendant had an abusive mother, suffered damage to his frontal lobe ("the area of the brain that affects reasoning, problem solving, and judgment"), and "suffered chronic drug addiction" that "exacerbate[d] [his] mental deficits and impairments." Id. at 1147. Despite the "horrific" nature of his crime, we reasoned this evidence could have "both explained to the jury the reasons [the defendant] was predisposed to act in concert with [accomplices] on the night of the murders and demonstrated [he] was less morally culpable than the average defendant for committing the murders." Id. at 1147-48. The "mitigation evidence [at sentencing had] left the jury with no explanation for the murders other than the prosecution's assertion [the defendant] was ‘evil.’ " Id. at 1148 ; see also Smith , 379 F.3d at 943 (finding prejudice although "[t]he jury already had evidence of [the defendant's] impulsiveness and lack of emotional control [at sentencing]," because it "lacked ... an explanation of how [his] organic brain damage caused [his] outbursts of violence").

By contrast, in Wilson v. Trammell , we found no prejudice despite the defendant's postconviction diagnosis of paranoid schizophrenic disorder. 706 F.3d at 1310. The diagnosis was not "plausible" given the doctor's "sloppy analysis and bias," and "there was no credible evidence that Defendant acted as a result of delusions or hallucinations." Id. at 1308-10.⁸ We noted that "this would be a different case" "if (1) the record had supported a plausible diagnosis of paranoid schizophrenia and (2) Defendant's behavior during the crime could have been tied to that disorder." Id. at 1310.⁹

8 We also noted that the defendant's postconviction diagnostic tests might have allowed the prosecution to portray him "as a dangerous criminal who could not safely be allowed to live." See Wilson , 706 F.3d at 1308.

9 See also Littlejohn , 875 F.3d at 564 (finding no prejudice where the defendant's experts failed to show "that [his] alleged organic brain damage played a substantial role in engendering his life of criminal deviance"); Young v. Sirmons , 551 F.3d 942, 965, 968 (10th Cir. 2008) (declining to find prejudice in part because the defendant's "Compulsive Personality Disorder," "mild impulsiveness," and "possible low frustration tolerance" did not "substantially reduce [his] ‘moral culpability’ for the two murders." (quoting Williams , 529 U.S. at 398, 120 S.Ct. 1495 )).

3. Abusive Background Evidence

As we noted in Barrett II , "evidence of childhood abuse, neglect, and instability can play a significant role in mitigation." 797 F.3d at 1229-30. The Supreme Court and this court have emphasized the mitigating value of an abusive upbringing. See, e.g. , Anderson , 476 F.3d at 1147 ("The Supreme Court has similarly noted that evidence of borderline mental retardation and childhood poverty and abuse are highly relevant to the question of moral culpability.").

C. Analysis

Proceeding to our analysis of prejudice, we discuss (1) the aggravating evidence the prosecution introduced at trial and sentencing, and (2) the mitigating evidence Mr. Barrett introduced at the evidentiary hearing and the Government's attempts to discredit that evidence. We then (3) balance the aggravating and mitigating evidence. The Government mounted credible challenges to the mitigation evidence at the evidentiary hearing. But if defense counsel had presented the evidence of Mr. Barrett's mental impairments and childhood abuse at sentencing, "there is a reasonable probability" that at least one juror "would have concluded that the balance of aggravating and mitigating circumstances did not warrant death." See Wilson , 706 F.3d at 1305 (quotations and alterations omitted). We therefore reverse.

1. Aggravating Evidence

At sentencing, the jury found that Mr. Barrett (1) "committed the offense after substantial planning and premeditation," (2) "knowingly created a grave risk of death to one or more persons in addition to [Trooper] Eales," and (3) caused "injury, harm, and loss to the victim's family." Barrett I , 496 F.3d at 1087. In rejecting his claims on direct appeal, we summarized the evidence pertaining to each factor.

First, the evidence of premeditation included Mr. Barrett's statements "in the months and weeks leading up to the date of the shooting ... that if law enforcement officers came to his house, ‘there was going to be a shootout,’ ‘he would shoot the first police that came through his door,’ and ‘he was going to take out as many law enforcement officers as he could before they got him.’ " Id. at 1112-14 (citations, brackets, and footnote omitted).¹⁰

10 Mr. Barrett argues "this evidence lacked credibility because it came from [seven] late-coming informant witnesses, ... [a]t least two of [whom] have recanted their accounts." Aplt. Br. at 47. Even if this were so, the jury heard evidence of the events described in Barrett I . Mr. Barrett must show a reasonable probability his mental impairments would have caused the jury to impose a lesser sentence given the evidence it "would have had before it" at the time. Wong , 558 U.S. at 20, 130 S.Ct. 383.

Second, the evidence of grave risk to others included Mr. Barrett's shooting Trooper Hamilton and firing bullets at the approaching cars. See id. at 1086-87 & n.3.

Third, the evidence of loss to the victim's family included statements from Trooper Eales's sister, mother, widow, and daughter. See id. at 1099.

2. Postconviction Mitigating Evidence

At the evidentiary hearing, Mr. Barrett presented evidence of (a) mental impairments and (b) an abusive upbringing. The Government challenged the validity and mitigating weight of that evidence. We assess the competing arguments with respect to each category.

a. Evidence of mental impairments

During the evidentiary hearing, Dr. Miora reported, and Dr. Woods agreed, that Mr. Barrett suffers from organic brain damage. Dr. Woods testified that Mr. Barrett suffers from bipolar disorder and PTSD. According to Dr. Woods, Mr. Barrett's shooting Trooper Eales was "clearly was an example of misperceiving and not being able to effectively weigh and deliberate." ROA, Vol. 4 at 570.

We discuss below the evidence showing that Mr. Barrett suffered from (i) brain damage, (ii) mental illnesses, and (iii) difficulty exercising judgment as a result of those impairments.

i. Evidence of brain damage

1) Mr. Barrett's evidence

Mr. Barrett presented evidence showing he suffered brain damage from several head injuries. As a child, Mr. Barrett was hit in the head with a steel ball. As an adult, he was involved in a motorcycle accident that damaged his periorbital area. He may have suffered "neuropsychological consequences" because his mother drank while pregnant. Id. at 1676.

Dr. Young's tests, which Dr. Miora analyzed, reflected weaknesses in Mr. Barrett's temporal lobe, frontal lobe, and parietal lobe. He showed impaired "visual attention," which, Dr. Miora explained, causes "difficulty in pressured situations where he has to visually attend to what is going on and generate a response." Id. at 1726. Dr. Miora observed that Mr. Barrett's impairments can compromise daily functioning and can inhibit judgment in unpredictable situations. See id. at 1756-59. We have recognized this type of evidence can be powerfully mitigating. See Grant , 886 F.3d at 920 ("Evidence of organic brain damage is something that we and other courts, including the Supreme Court, have found to have a powerful mitigating effect." (quotations omitted)).

2) Government's response

The Government challenged Dr. Miora's testimony.

First, during Dr. Miora's cross-examination, it produced Dr. Young's note that "[c]urrent research demonstrates a significant relationship between" results such as Mr. Barrett's and inability to perform "daily functioning such as driving an automobile." ROA, Vol. 4 at 1830. Because Mr. Barrett could drive and work as a mechanic, the Government argues Dr. Young's and Dr. Miora's findings were flawed.

Although a jury may find these points persuasive, it also would hear Dr. Miora's clarification that she did not "believe [Dr. Young] was suggesting" that Mr. Barrett could not drive or work. Id. at 1830-31. She conceded she did not know "why [Dr. Young] put [the note] in [the report]," but said it did not affect her diagnoses. Id. at 1831. Further, Dr. Young's note suggested only that "[c]urrent research demonstrates a significant relationship" between Mr. Barrett's test results and the inability to drive. Id. at 1830. Dr. Young did not say every person with Mr. Barrett's test results cannot drive.

Second, as Dr. Price noted, Mr. Barrett did not report head injuries in many of his medical documents, and in at least one record, he denied having suffered head injuries. See id. at 1463, 1482-83; see also Gov't Exh. 5 at 1. Dr. Price testified that someone who had sustained a significant head injury and lost consciousness would be unlikely to perform as well as Mr. Barrett had on the tests he administered. A jury could credit Dr. Price's testimony, but Dr. Pitt acknowledged there were records of at least two head injuries that Mr. Barrett suffered.

Given Mr. Barrett's recollections of childhood and adult head injuries, his performance on Dr. Young's tests, and the records of at least some head injuries, the jury could reasonably have concluded Mr. Barrett suffered from brain damage.¹¹

11 The Government also argues that the record belies Dr. Miora's conclusions because "[Mr.] Barrett effectively attended to the complex visual tasks of target selection, weapon aiming, and tactical concealment despite darkness, flashing emergency lights, and the presence of his son in the front yard." Aplee. Br. at 36. But Mr. Barrett argues his judgment was compromised, not his ability to use a firearm.

ii. Evidence of mental illnesses

Dr. Woods diagnosed Mr. Barrett with bipolar disorder and PTSD. The Government disputes Dr. Woods's diagnoses. It points to Dr. Pitt's testimony that Mr. Barrett exhibited only "[a]mphetamine dependence with physiologic dependence, features of an antisocial personality disorder, [and] cannabis dependence." Id. at 1498. In assessing a disputed diagnosis of schizophrenia in Wilson , we declined to find prejudice in part because the defendant failed to show the diagnosis was "plausible." 706 F.3d at 1310. We thus consider the plausibility of Dr. Woods's diagnoses.

1) Bipolar disorder

a) Mr. Barrett's evidence

Dr. Woods's diagnosis of bipolar disorder is plausible. He based it on Mr. Barrett's (1) "family history [of] bipolar disorder" and mood disorders, (2) "mood lability," (3) "pressured speech," and (4) history of "significant drug use." ROA, Vol. 4 at 548-49. He observed these circumstances and symptoms while conducting two comprehensive psychological assessments of Mr. Barrett.

b) Government's response

The Government argues Dr. Woods's bipolar diagnosis is flawed for two reasons.

First, it points out that Dr. Woods is the only doctor to diagnose Mr. Barrett with bipolar disorder despite his previous psychological evaluations and hospitalizations. For example, the doctor who discharged Mr. Barrett after his 1995 hospitalization diagnosed him with only marital problems, mixed substance abuse, alcohol abuse, and mixed personality disorder.¹²

12 Mr. Barrett notes that he was provisionally diagnosed with bipolar disorder on his intake form in his 1995 hospitalization. See Aplt. Br. at 43 (citing ROA, Vol. 4, at 650, 710, 1563-64; Gov't Exh. 9). But the doctor who discharged Mr. Barrett after that hospitalization did not diagnose him with bipolar disorder. See Gov't Exh. 6.

Although this may have persuaded some jurors to question Dr. Woods's diagnosis, he testified that Mr. Barrett was not diagnosed with bipolar disorder previously because "[t]hey didn't have all the data that we have." Id. at 680, 104 S.Ct. 2052. He also noted that, in 1995, doctors "treat[ed] [Mr. Barrett] with medications that are used in a psychiatric setting" reflecting "the psychiatric disorder that he had." Id. at 679, 104 S.Ct. 2052.¹³ Most telling, Dr. Woods is the only doctor to perform a comprehensive psychiatric examination of Mr. Barrett. And as the magistrate judge observed, "Dr. Pitt is not a neuropsychologist and thus did not conduct a neuropsychological examination." Barrett , 2018 WL 7889966, at *12. These points address the Government's objection, and likely would have convinced at least one juror to accept Dr. Woods's diagnosis.

13 The Government argues it is "inappropriate [to] attempt to substantiate the bipolar disorder by drawing the Court's attention to drugs [Mr. Barrett] had received." Aplee. Br. at 39. But even if the drugs Mr. Barrett received in 1995 do not alone allow for a diagnosis of bipolar disorder, that evidence supports the notion that doctors may have noticed symptoms of bipolar disorder in 1995 even if they lacked the data to make a formal diagnosis.

Second, the Government contends that Mr. Barrett's history of drug abuse casts doubt on Dr. Woods's diagnosis. Dr. Pitt opined that drug abuse would "better explain[ ]" Mr. Barrett's symptoms. ROA, Vol. 4 at 1490. Dr. Woods acknowledged that "[the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition] says a primary diagnosis of bipolar disorder must be established based on symptoms that remain once substances are no longer being used." Id. at 657.

Mr. Barrett does not contest that he abused drugs throughout his life, including when he shot Trooper Eales. Instead, he argues that he also suffered from underlying mental illnesses. Dr. Woods testified that mood lability and pressured speech are not symptoms of drug abuse but are symptoms of bipolar disorder. See id. at 551-52. He maintained, for example, that "[t]he behaviors that are described in ... 1995 are not the behaviors that are consistent with someone that's high on methamphetamines." Id. at 680, 104 S.Ct. 2052. Dr. Woods also reported that Mr. Barrett's toxicology screens showed he was not under the influence of methamphetamine at least some of the time he exhibited bipolar symptoms. See id. at 659-63.

Further, Dr. Pitt conceded that substance abuse can exacerbate the effects of bipolar disorder and PTSD in certain cases. Courts have explained that drug addiction that contributes to or compounds the effects of brain impairments does not diminish the mitigating weight of those impairments. In Anderson , for example, we explained that evidence of the defendant's drug addiction supported finding prejudice because it "exacerbate[d] [his] mental deficits and impairments." 476 F.3d at 1147 ; accord Rompilla , 545 U.S. at 390-91, 125 S.Ct. 2456 (noting that the defendant's history of alcoholism was mitigating evidence).

Mr. Barrett's drug abuse does not necessarily undermine Dr. Woods's diagnosis. Unlike in Wilson , "the record [of the evidentiary hearing here] support[s] a plausible diagnosis" of bipolar disorder. Wilson , 706 F.3d at 1310.

2) PTSD

Dr. Woods testified that Mr. Barrett displayed symptoms "consistent [with] [PTSD]," likely resulting from his father's violence and exacerbated by his mother's alcoholism and neglect. ROA, Vol. 4 at 557. On cross-examination, the Government did not undermine Dr. Woods's conclusion that the violence of Mr. Barrett's father could have been the precipitating event for Mr. Barrett's PTSD. Dr. Pitt, by contrast, declined to diagnose Mr. Barrett with PTSD, explaining that "there's no threshold event ... that ... qualifies for meeting the initial criteria for having [PTSD]," and "nowhere in any records ... was [Mr. Barrett] diagnosed with PTSD." Id. at 1493.

Dr. Woods's assessment that Mr. Barrett has PTSD is plausible. He did not rely on "irrelevan[t]" facts or interpret Mr. Barrett's past actions and statements out of context. Wilson , 706 F.3d at 1308-09. Even if some reasonable jurors could side with Dr. Pitt, there is a reasonable probability at least one would credit Dr. Woods's testimony.

iii. Connection to the offense

Dr. Woods concluded that Mr. Barrett's "symptoms of psychiatric and neurological disorder [s] ... interact with each other, and they undermine each other so that his brain impairments really do a job on his psychiatric disorder. And vice versa." ROA, Vol. 4 at 569. He explained that "[t]he idea of a hyperreactive response is completely consistent with both [Mr. Barrett's] history, his mood disorder, as well as what we would see in someone that ha[s] these kinds of brain impairments, this misperception of ... the circumstances." Id. at 570.

Dr. Woods also testified that Mr. Barrett's alleged impairments undermined his "executive functioning," and his ability to "effectively weigh and deliberate." Id. And he noted that "[t]here appears to be a deterioration in [Mr. Barrett's] mental state ... in the weeks and perhaps the days before this event in 1999." Id. at 571. He concluded that Mr. Barrett's shooting Trooper Eales "clearly was an example of misperceiving and not being able to effectively weigh and deliberate." Id. at 570.

The Government attempts to discredit Dr. Woods, asserting he "did not understand what had occurred" because he mistakenly believed Mr. Barrett "killed Trooper Eales during an exchange of gunfire, rather than during an ambush." Aplee. Br. at 45. The Government points to Dr. Woods's declaration in 2009 that "Mr. Barrett's description of the events leading to his arrest indicates that he did not leave his shack before he was shot by the police." ROA, Vol. 1 at 1318. It also notes the following exchange between Mr. Barrett's counsel and Dr. Woods at the evidentiary hearing:

Q. All right. And shots were fired?

A. Yes, sir.

Q. Trooper Eales was killed?

A. Yes, sir.

Q. There was an exchange of gunfire. Mr. Barrett himself was shot?

A. Yes, sir.

ROA, Vol. 4 at 569. "Given Dr. Woods' ignorance of these salient facts," it argues, "he did not shed any mitigating light on [Mr.] Barrett's actions." Aplee. Br. at 46.

The Government's attempt to discredit Dr. Woods does not account for the fact that Dr. Woods's testimony, fairly read, does not show he believed the police shot at Mr. Barrett before Mr. Barrett shot Trooper Eales. Rather, he acknowledged that "Trooper Eales was killed" and then agreed that "[t]here was an exchange of gunfire" and "Mr. Barrett himself was shot." ROA, Vol. 4 at 569. This is consistent with the facts of the case. See Barrett II , 797 F.3d at 1212 (noting that "[Trooper] Eales, exited [the car] and was shot three times," and then "[Trooper] Hamilton fired two shots at [Mr. Barrett] that missed, but [Trooper] Manion shot him through a window and hit his legs"). Further, Dr. Woods testified he was familiar with "records of the ... offense itself from police records, etc." ROA, Vol. 4 at 537.¹⁴

14 The Government also points out that Dr. Miora admitted she was unfamiliar with the details of Mr. Barrett's crime and could not "develop any opinion about what was going on with him at the time of the crime." ROA, Vol. 4 at 1834. But, as noted, Dr. Woods "reviewed the [same] neuropsychological testing" as Dr. Miora and agreed with her findings of organic brain damage. Id. at 537, 558-60, 123 S.Ct. 2527. And he explained how Mr. Barrett's "psychiatric and neurological disorder [s]" influenced his criminal behavior. Id. at 569-70.

The Government also relies on Mr. Barrett's statements before the shooting indicating planning and premeditation. It contends that, "as reflected in the finding of substantial planning and premeditation and other evidence offered at trial, [Mr.] Barrett attacked the police because he recognized them," not as a misperception or compulsive reaction. Aplee. Br. at 21.

Mr. Barrett's statements cut against Dr. Woods's theory that the crime "clearly was an example of misperceiving and not being able to effectively weigh and deliberate." ROA, Vol. 4 at 570. And the planning and premeditation evidence should be part of the prejudice calculus. Williams , 529 U.S. at 398, 120 S.Ct. 1495 (noting the importance of evidence showing that the defendant's "violent behavior was a compulsive reaction rather than the product of cold-blooded premeditation"). Given the evidence of Mr. Barrett's statements, some jurors could attribute less weight to his mental health evidence.

Still, "[a] reasonable probability" that the jury would have recommended a life sentence "is less than a preponderance of the evidence" and need only "undermine confidence in the outcome." Smith , 379 F.3d at 942 (quotations omitted). We doubt Mr. Barrett's statements would have caused every juror to reject Dr. Woods's explanation for his criminal behavior or discount his evidence of mental impairments.

As Dr. Woods explained, "[t]here appears to be a deterioration in [Mr. Barrett's] mental state ... in the weeks and perhaps the days before this event in 1999." ROA, Vol. 4 at 571. A reasonable juror could have attributed Mr. Barrett's behavior during that time to his "deteriorati[ng]" mental state and thus considered him less morally culpable for his statements and subsequent actions. Id.

Further, Mr. Barrett's mental health evidence may have convinced at least one juror that he carried out his threats at least in part due to the influence of brain damage, bipolar disorder, or PTSD. Talking about killing police officers is not the same as deciding and acting to do it. Faced with Mr. Barrett's mental health evidence, a reasonable juror could have attributed his statements to bravado and his actions to brain damage, and also could have concluded he reacted to multiple police cars approaching his property in the middle of the night at least in part due to his mental impairments.

* * * *

Although the prejudice issue is close, the jury heard none of Mr. Barrett's mental health explanation for his criminal behavior at sentencing. At the evidentiary hearing, he introduced the type of mental impairment evidence that, as this court has recognized, can carry substantial mitigating weight. Dr. Woods connected that evidence to his offense. The Government cast some doubt on that evidence at the hearing, but not enough to erase a reasonable probability that a single reasonable juror would find it sufficient to vote against death.

b. Evidence of an abusive upbringing

Dr. Woods also reported that Mr. Barrett's mother drank while pregnant and that Mr. Barrett's father beat Mr. Barrett and his mother. He said Mr. Barrett had "a family history of pretty significant violence by [his] father that lasted up until about the age of 13." ROA, Vol. 4 at 557. Mr. Barrett also experienced "neglect by [his] mother," who "was a heavy drinker" and who provided "a lack of coping resources." Id. Stephen Barrett, Mr. Barrett's brother, noted that, as a child, he saw his mother and Mr. Barrett "in a physical altercation" when an argument turned into "a wrestling match on the floor." Id. at 446.

In light of this evidence, the magistrate judge observed that Mr. Barrett's "childhood was characterized by early exposure to alcohol and alcohol abuse, as well as physical and emotional abuse and neglect." Barrett , 2018 WL 7889966, at *7. As we noted in Barrett II , "[s]uch evidence of childhood abuse, neglect, and instability can play a significant role in mitigation." 797 F.3d at 1229-30. A jury could reasonably have credited as mitigating the testimony that Mr. Barrett experienced an abusive upbringing.

3. Balancing the Aggravating and Mitigating Evidence

The district court found that defense counsel's performance at sentencing was deficient under the Sixth Amendment. The Government does not contest that finding. The only question on appeal is "whether there is a reasonable probability that, absent the errors," Wilson , 706 F.3d at 1305 (quotations omitted), "at least one juror would have struck a different balance" between the aggravating and mitigating circumstances, Grant , 727 F.3d at 1019 (quotations omitted). See also Smith , 379 F.3d at 942 ("A reasonable probability is less than a preponderance of the evidence, but sufficient to undermine confidence in the outcome." (quotations omitted)).

Balancing the aggravating and mitigating evidence, viewing the evidence as a whole, and drawing on the relevant case law, we conclude there is a reasonable probability at least one juror would not have recommended a death sentence had defense counsel introduced evidence of Mr. Barrett's mental impairments and abusive upbringing at the sentencing hearing.

a. Strengthened mitigation case

As explained above, Mr. Barrett provided plausible evidence that significant mental impairments compromised his judgment on the night of the shooting. Dr. Woods said his behavior "clearly was an example of misperceiving and not being able to effectively weigh and deliberate." ROA, Vol. 4 at 570. And his history of drug addiction, including methamphetamine abuse, "if anything, exacerbated or made worse the symptoms of his mental illness and neurocognitive deficits." Id. at 574. Mr. Barrett also introduced evidence of an abusive upbringing at the hands of parents who were violent toward him and each other, and a mother who neglected him and drank excessively, including while pregnant.

The jury heard almost none of this evidence at sentencing. See Barrett II , 797 F.3d at 1224-25. Defense counsel relied instead on evidence that Mr. Barrett "was a loved family member and good person who was sorry for killing [Trooper] Eales"; "lack[ed] ... prior felonies [and] was a well-behaved prisoner"; "was a good mechanic and a nonviolent person"; and would not be "a danger while his bench warrant was outstanding." Id. at 1224. "None of the witnesses discussed [Mr. Barrett's] mental health or troubled background in any significant detail." Id.

Even if Mr. Barrett's mitigation case at sentencing amounted to more than a "few naked pleas for mercy," see Rompilla , 545 U.S. at 393, 125 S.Ct. 2456, his postconviction evidence "bore no relation" to what he introduced at the sentencing hearing, id. , which "left the jury with no explanation for the murders," see Anderson , 476 F.3d at 1148. This case thus differs from those where "the additional mitigating evidence [would not] have added anything significant to the jury's understanding of [the defendant's] mental deficiencies." DeRosa v. Workman , 679 F.3d 1196, 1217, 1221 (10th Cir. 2012) (declining to find prejudice despite postconviction mental health and abuse evidence because this evidence "[did] not differ in a substantial way from the evidence actually presented at sentencing" (quotations and alterations omitted)).

Mr. Barrett's case resembles Anderson v. Sirmons , where we found counsel's failure to introduce mental health and abusive upbringing evidence prejudicial despite the defendant's "callous and brutal" murders and the jury's finding of three aggravating factors. 476 F.3d at 1146. Like Mr. Barrett, Mr. Anderson introduced evidence of significant damage to his frontal lobe, of a chronic addiction to methamphetamine that "exacerbate[d] [his] mental deficits and impairments," and that he had an "abusive mother." Id. at 1147. As in this case, "the absence of this readily available mitigation evidence [at sentencing] left the jury with no explanation for the murders other than the prosecution's [characterization of the defendant]." Id. at 1148. We said in Anderson that "this court cannot overstate the importance of the type of evidence that was available in this case but was never presented to the jury." Id. at 1147-48 ; see also Hooks , 689 F.3d at 1207 (finding prejudice where, like here, "[a] proper presentation of the family-history and mental-health evidence, not to mention the circumstances surrounding [the] conviction, would have been powerful mitigation." (quotations omitted)).¹⁵ b. Weakened aggravation case

15 This case differs from those involving implausible diagnoses or impairments unconnected to the petitioner's criminal behavior. See, e.g. , Young v. Sirmons , 551 F.3d 942, 965 (10th Cir. 2008) (declining to find prejudice when the petitioner was diagnosed with a "Compulsive Personality Disorder" and "some mild impulsiveness and possible low frustration tolerance"); Littlejohn , 875 F.3d at 561, 564 (declining to find prejudice where the petitioner established only an attention deficit disorder and a "garden-variety" impulse control disorder and failed to substantiate his expert's pre-hearing claim that "organic brain damage played a substantial role in engendering his life of criminal deviance"); Wilson , 706 F.3d at 1310 (declining to find prejudice but noting that "this would be a different case" "if (1) the record had supported a plausible diagnosis of paranoid schizophrenia and (2) Defendant's behavior during the crime could have been tied to that disorder").

Although the Government challenged the mitigation evidence at the evidentiary hearing, it introduced no additional aggravating evidence that it would have offered in response to Mr. Barrett's new mitigating evidence. It introduced little if any evidence of Mr. Barrett's drug addiction or misbehavior that the jury had not already heard. As we explained in Barrett II , at the sentencing hearing

the jury heard extensive testimony of [Mr. Barrett's] abuse of his ex-wife, ... significant evidence of [his] drug use, ... [and] evidence of [his] outstanding warrant for failure to appear in state court on drug charges and the drug paraphernalia found on [his] person and property after the shooting.

797 F.3d at 1232 (citation omitted). As a result, "the evidence of spousal abuse and drug involvement would not have surprised the jury and presented little downside risk to [Mr. Barrett's] offering evidence of his mental condition." Id.

Nor did the Government at the evidentiary hearing attempt to show Mr. Barrett's mental health evidence would have had a "double-edged" effect. See, e.g. , Gilson 520 F.3d at 1249-50 (finding no prejudice in part because the defendant's "inability to regulate behavior or inhibit impulses" supported the jury's finding of the aggravating factor "that he represented a continuing threat, even if confined in prison for life" (quotations omitted)); McCracken v. Gibson , 268 F.3d 970, 980 (10th Cir. 2001) (same). Despite our suggestion in Barrett II that the Government could rebut Mr. Barrett's mental health evidence if it could show he "was a psychopath and would be at a high risk of committing violent offenses if freed," 797 F.3d at 1232, the Government introduced no such evidence at the evidentiary hearing. And it did not argue Mr. Barrett's evidence would have had this double-edged effect.¹⁶

16 Although the Government claimed at oral argument that Mr. Barrett's evidence would have portrayed him as a continuing threat, Oral Arg. at 47:12-52:20, we decline to consider this argument given the Government's lack of briefing, see United States v. DeRusse , 859 F.3d 1232, 1240 n.3 (10th Cir. 2017) ("Arguments that are raised for the first time at oral argument come too late to merit our attention." (quotations and brackets omitted)).

In opposing Mr. Barrett's § 2255 motion on appeal, the Government relies on the aggravating evidence that it already presented to the jury at sentencing. Mr. Barrett's evidence of mental impairments and childhood abuse not only would have offset the weight of that evidence, it also "would have softened [the] edge" of the planning and premeditation factor. Hooks , 689 F.3d at 1207. And evidence of a "deterioration in [Mr. Barrett's] mental state ... in the weeks and perhaps the days before this event in 1999," ROA, Vol. 4 at 571, may well have caused the jury to attribute less weight to his threatening statements in that period and for his actions.

c. Minimal post-sentencing aggravation evidence

The district court characterized as aggravating Dr. Woods's testimony that Mr. Barrett refused to "accept[ ] responsibility for [Trooper] Eales's death" while maintaining "that he and his son were the victims in this case." Barrett , 2019 WL 1406957, at *8. Even if this could be aggravating evidence, the jury had already heard most of it. For example, "[Mr.] Barrett's defense during the first-stage proceedings was that he was unaware that the persons entering his property were law enforcement officials, and that he was simply reacting in defense of himself and his son." Barrett I , 496 F.3d at 1114-15.

The district court also pointed to evidence that Mr. Barrett admitted to Dr. Pitt during his psychological evaluation that he had "obtained and us[ed] drugs while in jail" and "talked a jailer into letting him out of his cell for the purpose of engaging in sexual activity with another inmate." Barrett , 2019 WL 1406957, at *8 n.26. The court opined this would have shown Mr. Barrett "was not afraid to violate the rules [in prison]" and was thus "a continuing danger to others." Id. at *8. But the court did not explain how these alleged nonviolent infractions would have convinced the jury that Mr. Barrett was dangerous to others, let alone offset the additional mitigating evidence.

Finally, the district court characterized as aggravating the testimony at the evidentiary hearing from Mr. Barrett's family that he engaged "in physical altercations [with his mother]" and "violently attacked his younger brother on at least two separate occasions." Id. But the court again failed to explain how this information would make a jury more likely to recommend the death penalty. This case differs from Littlejohn , for example, where we emphasized that the defendant's mental health evidence "would have opened the door for the prosecution to introduce (1) harmful evidence that [the defendant] suffered from an antisocial personality disorder,¹⁷ (2) testimony concerning the limited treatment options available for [his] disorders, along with (3) damaging evidence regarding [his] post-offense misconduct." 875 F.3d at 562.

17 Although Dr. Woods noted that people with bipolar disorder can engage in "antisocial behavior," ROA, Vol. 4 at 584-85, and Dr. Pitt said Mr. Barrett suffered from "features of an antisocial personality disorder," id. at 1498, the Government does not argue this evidence would have had any aggravating effect.

The district court gave two primary reasons for declining to find prejudice. First, it was skeptical that "the jury would have given much weight" to Mr. Barrett's experts in light of "the evidence it heard over the course of the entire trial." Barrett , 2019 WL 1406957, at *4. Second, it concluded that Mr. Barrett "had clearly resolved to murder Trooper Eales or any other law enforcement officer long before this incident played out." Id. at *6. Both reasons are tenable. But we agree with the magistrate judge, who heard the testimony at the evidentiary hearing firsthand, that "[t]he evidence potentially available to [Mr. Barrett] regarding his mental health history and his family background," which the jury did not hear, "would in all likelihood have aided substantially in mitigation." Barrett , 2018 WL 7889966, at *13. And Mr. Barrett's additional mitigating evidence would likely have softened the edge of the aggravating evidence the jury heard at trial and sentencing.¹⁸ * * * *

18 Both Mr. Barrett's state and federal proceedings suggest only limited jury support for the death penalty, even without consideration of the mental health evidence. The state court jury did not even find him guilty of a capital offense, acquitting him of first-degree murder and instead convicting him of the lesser-included offense of first-degree manslaughter. It also acquitted him of shooting with intent to kill and instead convicted him of the lesser-included offense of assault and battery with a dangerous weapon. See Barrett I , 496 F.3d at 1086. The federal jury recommended the death penalty on only one of the three death penalty counts.

In summary, at the evidentiary hearing, Mr. Barrett introduced substantial mitigating evidence that bore no resemblance to the constitutionally deficient mitigation case his counsel presented at sentencing. Although the Government cast some doubt on the testimony of Mr. Barrett's experts, the evidence plausibly showed that Mr. Barrett suffers from organic brain damage, bipolar disorder, and PTSD. Dr. Woods linked the combined effect of Mr. Barrett's mental impairments to his criminal behavior. Mr. Barrett also introduced new evidence of an abusive upbringing that some jurors likely would have found sympathetic.

In contrast to the mitigating evidence Mr. Barrett presented, the Government offered no additional aggravating evidence, relying exclusively on what the jury had already heard. It pointed to Mr. Barrett's statements showing substantial planning and premeditation and his ability to carry out the crime. It also noted that no medical professional before Dr. Woods had diagnosed Mr. Barrett with bipolar disorder or PTSD, and argued Mr. Barrett's drug abuse better explained his behavior.

In reversing, we conclude that, viewing the totality of the evidence at trial, sentencing, and the postconviction hearing, there is a reasonable probability at least one juror would have recommended a life sentence. As noted, "[a] reasonable probability is less than a preponderance of the evidence" and need only "undermine confidence in the outcome." Smith , 379 F.3d at 942 (quotations omitted). Because Mr. Barrett's postconviction evidence shifts the balance in favor of mitigation, we are not "confiden[t]" all twelve jurors would have recommended a death sentence had defense counsel introduced this evidence. See id . (quotations omitted).¹⁹

19 Mr. Barrett raises several additional arguments he casts as procedural challenges. We need not address these claims because we reverse on the merits.

IV. CONCLUSION

We reverse the district court's decision finding no prejudice, vacate Mr. Barrett's sentence on Count 3, and remand for resentencing on Count 3.²⁰

20 Mr. Barrett also argues he is entitled to resentencing on all three counts, including those for which he received life sentences. Aplt. Br. at 52. We disagree. In 2015, we reversed and remanded Mr. Barrett's "death sentence for the district court to hold an evidentiary hearing on whether the performance of trial counsel was deficient ... and whether [Mr. Barrett] suffered prejudice." Barrett II , 797 F.3d at 1232 (emphasis added). The district court issued a COA for "review of a death sentence ." Barrett , 2019 WL 1406957, at *9 (quotations omitted and emphasis added); 28 U.S.C. § 2253(c)(3) ("The [COA] ... shall indicate which specific issue or issues satisfy the showing required ...."). Further, Mr. Barrett argued throughout this proceeding that he "should not have been sentenced to death." ROA, Vol. 1 at 300. On appeal, he relies on cases finding a reasonable likelihood the jury would have imposed a sentence less than death. For these reasons, we reject his request that we also vacate his life sentences on Counts 1 and 2.

HARTZ, J., Circuit Judge, concurring

I concur in the judgment and all the panel opinion except section III(C)(2)(a).

Though I agree with the conclusion of that section—that the mental-impairment evidence created a reasonable probability that a juror would vote against death—I think the section's treatment of the government's contrary evidence understates its impact by employing a divide-and-conquer approach that addresses each of the government's points separately. Viewed cumulatively, those points would likely have undermined most jurors' confidence in Mr. Barrett's experts. Cf. United States v. Arvizu , 534 U.S. 266, 274–75, 122 S.Ct. 744, 151 L.Ed.2d 740 (2002) (criticizing "divide-and-conquer analysis" in assessing reasonable suspicion). I can certainly understand the district court's belief that the jury would not be impressed by those experts. The government mounted a strong counterattack. I will not recap each of the points scored by the government; the panel opinion thoroughly discusses the evidence. But among other things, the government pointed out (1) that Mr. Barrett's mental condition had been examined repeatedly over an extended period of time yet no prior mental-health professional had diagnosed him with the conditions offered by the defense experts, and (2) his symptoms could be readily explained by his lifetime of serious drug abuse. Perhaps more importantly, the expert opinion that Mr. Barrett's shooting of Trooper Eales "clearly was an example of misperceiving and not being able to effectively weigh and deliberate," ROA Vol. 4 at 570, appears inconsistent with the findings that the jury must have made regarding his behavior—in particular, his competence in reacting to the police raid and his lengthy planning for such an event.

Nevertheless, I lack confidence that the outcome would have been the same had this mitigation evidence been presented to the jury. The experts were well-credentialed, there was supporting evidence, and their examination of Mr. Barrett was apparently more extensive than those conducted in the past. The damage to Mr. Barrett's reasoning capacity may have been responsible for his suicidal plan to go down in flames. And there is a difference between blustering about having a shootout with the police and actually deciding to go through with it, so impairment of his judgment during the raid may have been ultimately responsible for his deadly actions. A reasonable juror might well have decided that the mitigating evidence precluded a death sentence. In that regard, I note that the state jury did not find Mr. Barrett guilty of first-degree murder and the federal jury imposed the death penalty on only one of the three death-eligible offenses.