ANDERSON, Circuit Judge:

This death penalty case is before us on an appeal from a grant of a writ of habeas corpus limited to resentencing. Dugger appeals the district court's determination that Doyle's trial counsel was ineffective at the sentencing phase.

I. FACTS

Before trial, the state court judge ordered several psychiatrists to examine Doyle. Dr. Zager was appointed to investigate a number of issues for the defense, including whether Doyle's capacity to appreciate the criminality of his conduct or to conform his conduct to the requirements of the law was substantially impaired at the time of the offense; and whether the offense was committed while Doyle was under the influence of extreme mental or emotional disturbance. Ex. 6, Order Appointing Defense Expert at 1457-58. These two issues related to mitigating circumstances in preparation for the possibility of a sentencing phase. Dr. Zager's report is not included in the record on appeal.

Although the court orders appointing Drs. Krieger and McClure to examine the defendant are not included in the record on appeal, it is apparent that the orders did not direct these two doctors to consider mitigating circumstances in their analyses of the defendant. Ex. 7, Evidentiary Hearing at 66-67. Neither of the doctors' written reports mentions either of the two relevant mitigating circumstances. Ex. 8, Appeal from Denial to Vacate Judgment and Sentence, App. 1, 2. In addition, Dr. McClure states in his report that he was appointed to examine Doyle for competency and the possibility of involuntary hospitalization. Id. at App. 2, p. 1. Dr. Krieger testified that he examined Doyle on the issues of competency to stand trial and insanity. Ex. 5, Penalty Phase at 1334.

Dr. Eichert also examined Doyle to determine if he was competent to stand trial and competent at the time of the offense. Ex. 5, Penalty Phase at 1325-26. At trial, Dr. McClure testified for Doyle. Doyle was convicted of first degree murder and sexual battery.

None of the doctors examined Doyle between the guilt and sentencing phases, and no further written reports were filed. At sentencing, Dr. Eichert testified for the State. Dr. Eichert testified that Doyle was competent to stand trial and competent at the time of the offense. He testified Doyle had above average intelligence but brain damage prevented his use of that intelligence. Dr. Eichert mentioned that Doyle's brother was shot in a hunting accident in 1981, and that Doyle's girlfriend started going out with other men. Doyle told him one or both of these incidents resulted in Doyle's drug abuse. Dr. Eichert also admitted Doyle had an emotional problem. Ex. 5, Penalty Phase at 1325-30.

Dr. Krieger testified for Doyle at sentencing. He stated that Doyle was depressed and that Doyle began abusing intoxicants after his brother's death. He also said that Doyle had flashbacks of his brother's accident. Dr. Krieger testified that Doyle was just above the intelligence level for mental retardation. Ex. 5, Penalty Phase at 1334-42.

Dr. McClure also testified at sentencing. He stated that Doyle was suffering from borderline mental retardation and a severe form of personality disorder. Dr. McClure also testified that Doyle lacked the capacity to control his behavior, that this was a chronic problem, and that the death of a close relative could produce psychotic symptoms in someone suffering from a personality disorder. Dr. McClure stated that he believed the crime was committed in an impulsive manner rather than in a cold, calculating way, and he responded affirmatively to trial counsel's question as to whether Doyle suffered from an extreme emotional disturbance. Ex. 5, Penalty Phase at 1345-52.

Doyle's mother also testified at sentencing. She stated that Doyle changed after his brother died and that Doyle's live-in girlfriend had moved out shortly before the offense. Ex. 5, Penalty Phase at 1355-62.

The jury was instructed as to a number of mitigating circumstances: (1) whether Doyle has no significant history of past criminal activity; (2) whether the offense was committed while Doyle was under the influence of extreme mental or emotional disturbance; (3) whether Doyle had the capacity to appreciate the criminality of his conduct or to conform his conduct to the requirements of the law was substantially impaired; (4) the age of Doyle at the time of the offense; and (5) any other aspect of Doyle's character or record and any other circumstance of the offense. The jury returned a recommendation for the death penalty by a vote of 8-4. On May 13, 1982, the state court judge sentenced Doyle to death. He found the following aggravating circumstances: (1) the capital felony was committed during the commission of rape; (2) the capital felony was committed to avoid or prevent a lawful arrest; and (3) the capital felony was especially heinous, atrocious, or cruel. He found no statutory mitigating circumstances, and he also found that there were no mitigating circumstances which could outweigh the aggravating factors. With respect to the mitigating circumstance of capacity to conform his conduct to the law, the state court found that Doyle had held jobs and attended school, that the doctors determined Doyle was sane and able to aid his attorney, and that the court had found Doyle competent to stand trial. Ex. 6, State Court's Findings in Support of Sentence at 1578-79.

Doyle then appealed to the Florida Supreme Court. See Doyle v. State, 460 So.2d 353 (Fla. 1984). With respect to the mitigating circumstance of capacity to conform his conduct to the law, the court found that the trial judge did not improperly rest his determination on competence to stand trial, but rather on Doyle's ability to function in society. With regard to the aggravating factor of commission of the capital felony to avoid lawful arrest, the court reversed the trial court's determination. However, the court determined that the death penalty was appropriate since the trial judge correctly found two aggravating factors and no mitigating circumstances. Justices Overton and McDonald dissented from the majority's decision on the applicability of mitigating circumstances. The dissenters argued that the record indicates that the trial judge applied the wrong standard in its determinations of extreme mental or emotional disturbance and capacity to conform conduct to the law. They stated that those two mitigating circumstances are present and that the case should be remanded for the trial judge to consider those circumstances.

Doyle then filed a post-conviction motion to vacate judgment and sentence. Although Doyle presented a number of issues, an evidentiary hearing held in state court was restricted to ineffective assistance of counsel at sentencing. The same judge that presided over Doyle's trial conducted the hearing. Dr. Bauer testified that Doyle was under extreme emotional disturbance at the time of the crime and that he was unable to appreciate the criminality of his acts and to conform his conduct to the law. Ex. 7, Evidentiary Hearing at 10-12. Tenbrook, Doyle's trial counsel, testified that the report of Dr. Zager, the first psychiatrist to examine Doyle, did not support his arguments as much as he had hoped, and, therefore, Tenbrook requested the appointment of other psychiatrists. Ex. 7, Evidentiary Hearing at 39, 47-48. Tenbrook stated that he hoped a psychiatrist could give testimony that would be helpful at the penalty phase. Id. at 39-40. He also testified that he mentioned the penalty phase to the doctors prior to their initial examinations of Doyle, id. at 48-49, 60, 69, and that he discussed impaired capacity in preparing for the penalty phase, id. at 57. Tenbrook also stated he was not sure whether he had concentrated on the death of Doyle's brother as a possible trigger of extreme emotional disturbance. Id. at 56. He explained that he did not emphasize testimony about the combination of Doyle's retardation and his brother's death because he was trying to establish a longstanding problem, which he thought the jury would be more likely to accept than a sudden change in behavior due to the brother's death, id. at 58-59, and because he was concerned about a potential challenge to Doyle's account of his brother's accidental death, id. at 70. Tenbrook admitted that the mitigating circumstances were not addressed in any of the doctors' written reports. Id. at 65-66. He also admitted that the judge's orders appointing Drs. McClure and Krieger did not mention mitigating circumstances. Id. at 66-67. However, his testimony was clear that he did orally discuss mitigating circumstances with both doctors. He stated that, prior to the penalty phase, he asked both doctors if they needed to interview Doyle further, and the doctors responded in the negative. Id. at 67. Shortly after the evidentiary hearing, Doyle filed an affidavit sworn by Dr. Krieger, which stated that the doctor examined and evaluated Doyle only on his competency to stand trial, his sanity at the time of the offense, and whether he met the criteria for involuntary hospitalization and that he never reached any conclusions directed toward mitigating circumstances. The state trial judge denied relief, stating:

It is clear that defense [counsel] sought mental health expertise to assist him before trial, and that certain mental health experts were presented on the Defendant's behalf. . . . Additionally, the Court finds that the opinions of Doctors Eichert, McClure, Krieger, and Zager were all considered appropriately by defense counsel prior to the trial and sentencing in this cause, and were further used appropriately.

Ex. 7, Order Denying Motion to Vacate Judgment and Sentence at 172-73 (emphasis in original).

Doyle then appealed to the Florida Supreme Court. The court affirmed the trial court's decision, finding that "[d]efense counsel sought mental health expertise to assist him before trial and presented testimony of these experts at both the guilt and penalty phases of the trial." Doyle v. State, 526 So.2d 909, 911 (Fla. 1988).

Doyle then filed for a writ of habeas corpus in the Southern District of Florida. After counsels' oral arguments, the district court ordered resentencing. The district court found:

DOYLE's counsel never consulted with the experts regarding [mitigating] factors and he made no requests for further examination once DOYLE was convicted. The experts have since admitted that their examinations of DOYLE were limited to competence and sanity.

R1-12-13. The court also found that, with regard to trial counsel's closing argument, he did not delineate the mitigating circumstances so that the jury would clearly understand them, and he did not specifically address the aggravating factors at all. R1-12-14-15. As an alternative, the court also granted relief based on Doyle's contention that comments made by the prosecutor and the state trial judge diminished the jury's sense of responsibility in violation of Caldwell v. Mississippi, 472 U.S. 320, 105 S.Ct. 2633, 86 L.Ed.2d 231 (1985). On a motion for relief from judgment, the district court reversed its determination on the Caldwell claim on the ground that Dugger v. Adams, 489 U.S. 401, 109 S.Ct. 1211, 103 L.Ed.2d 435 (1989), requires that a defendant raise a Caldwell claim at the state level in order to avoid a procedural bar. Respondent Dugger then appealed to this court, arguing that Doyle was not given ineffective assistance of counsel at sentencing. Doyle did not cross-appeal.

II. ANALYSIS

A claim of ineffective assistance of counsel must meet the familiar two prong test identified in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). First, the defendant must show that counsel's performance fell below an objective standard of reasonableness. 466 U.S. at 687, 104 S.Ct. at 2064. The reason for this requirement is that "[c]ounsel has a duty to bring to bear such skill and knowledge as will render the trial a reliable adversarial testing process." 466 U.S. at 688, 104 S.Ct. at 2065 (citation omitted). However, we employ a presumption that counsel's performance is within the wide range of acceptable trial conduct. 466 U.S. at 689, 104 S.Ct. at 2065. Second, the defendant must show that the errors deprived the defendant of a reliable result — i.e., "there is a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." 466 U.S. at 687, 694, 104 S.Ct. at 2064, 2068. This prong is often called the prejudice prong. This two prong standard for ineffectiveness claims clearly applies to the penalty phase in a capital case. 466 U.S. at 686-87, 104 S.Ct. at 2064.

Before we apply the Strickland test, we first determine the facts on which we base our decision. As noted above, in granting the writ of habeas corpus, the district court placed significant reliance on a finding of fact that Doyle's trial counsel never consulted with the experts regarding mitigating circumstances. However, federal courts must defer to state courts' findings of fact as provided in 28 U.S.C. § 2254(d). Strickland, 466 U.S. at 698, 104 S.Ct. at 2070. In contrast to findings of fact, conclusions of law — i.e., whether or not counsel's performance was effective and whether or not the performance prejudiced the defendant — are subject to plenary review in federal court. Id. Therefore, our first step is to determine what the factual findings of the state court are. As stated above, the state court determined:

It is clear that defense [counsel] sought mental health expertise to assist him before trial, and that certain mental health experts were presented on the Defendant's behalf. . . . Additionally, the Court finds that the opinions of Doctors Eichert, McClure, Krieger, and Zager were all considered appropriately by defense counsel prior to the trial and sentencing in this cause, and were further used appropriately.

Ex. 7, Order Denying Motion to Vacate Judgment and Sentence at 172-73 (emphasis in original). The first sentence is a factual finding and is also undisputed: trial counsel did seek mental health expertise and did present certain experts on Doyle's behalf. The question of whether the second sentence is a finding of fact or a conclusion of law is more difficult. Doyle contends that the second sentence is merely a legal conclusion that counsel's performance with respect to the doctors' opinions was "appropriate" or effective. Respondent contends that it is also an implicit finding of fact that counsel did consult with the doctors about mitigating circumstances.

During the proceedings before the state court judge, Doyle's collateral counsel cross-examined Tenbrook by attempting to suggest that Tenbrook had not consulted with Drs. Krieger and McClure about mitigating circumstances. Counsel elicited the fact that the orders appointing both doctors mentioned only that the doctors examine Doyle to determine whether he was competent to stand trial and whether he was sane at the time of the crime; counsel then suggested that the doctors had indicated that they had examined Doyle solely for that purpose. Tenbrook acknowledged that the orders appointing the doctors mentioned only those two purposes, but responded that he had talked to both doctors and requested that they also explore mitigating circumstances. Following the evidentiary portion of the hearing before the state judge, collateral counsel argued to the court that Tenbrook had not consulted with the doctors about mitigating circumstances. The state judge expressly noted his disagreement with that conclusion. Ex. 7, Evidentiary Hearing at 101.

In light of the arguments before the state judge, the colloquy between collateral counsel and the judge, and the written finding quoted above, we conclude that the state judge made a finding of fact that Tenbrook did consult with Drs. Krieger and McClure about mitigating circumstances.

Such state court findings of fact are binding in federal habeas corpus proceedings unless one of the exceptions provided for in 28 U.S.C. § 2254(d) apply. We conclude that no exceptions apply. The state court finding is "fairly supported by the record." 28 U.S.C. § 2254(d)(8). Tenbrook clearly testified that he did consult the doctors with respect to mitigating circumstances. Tenbrook's testimony is corroborated by the fact that Dr. McClure's testimony at sentencing did address mitigating circumstances. We conclude that the district court erred in failing to defer to this finding of fact.

Having determined that Tenbrook did consult with Drs. Krieger and McClure about mitigating circumstances does not end our inquiry. Doyle also asserts that Tenbrook's presentation of the mental health evidence and his argument to the jury were so deficient that his performance fell below the wide range of acceptable trial conduct and thus was ineffective under Strickland, 466 U.S. at 689, 104 S.Ct. at 2065. Doyle argues that the medical testimony at trial and at the penalty phase did not focus on the relevant mitigating circumstances — i.e., extreme emotional disturbance and substantial impairment of his capacity to appreciate the criminality of his conduct or to conform his conduct to the requirements of the law. The ultimate conclusion whether counsel's performance fell below the requirements of Strickland is a question of law which we must determine de novo. 466 U.S. at 698, 104 S.Ct. at 2070.

After a careful review of the record, we conclude that counsel's performance was not below the wide range of acceptable advocacy. Contrary to Doyle's contentions, Tenbrook did present substantial evidence that Doyle's condition met both mitigating circumstances. Dr. McClure testified at the penalty phase that Doyle suffered from borderline mental retardation and a personality disorder. Ex. 5, Penalty Phase at 1345. He defined the latter as falling somewhere between neurosis (anxiety disorder) and psychosis (e.g., schizophrenia where one is obviously out of control). He responded affirmatively to a question by Tenbrook as to whether Doyle's problems would constitute an extreme mental or emotional disturbance. Id. at 1345. Dr. McClure also testified that Doyle lacked the capacity to control his mood, impulses, and behavior, and lacked the intellectual capacity to understand how to control his behavior or the standard to which he should conform. Id. at 1346. He testified that Doyle's problems were longstanding, were evidenced at an early age in Doyle's school records, and were exaggerated by Doyle's recent traumatic experience when his brother was killed in a hunting accident. Id. at 1346. Finally, Dr. McClure testified that, in his opinion, Doyle's action in this case was impulsive, and not cold, calculating, and premeditated. Id. at 1352.

Dr. Krieger also testified at the penalty phase. Although Dr. Krieger's testimony was not as specifically focused on the precise language of the statutory mitigating circumstances, his testimony provided substantial support for a finding of mitigating circumstances. He testified that Doyle was borderline mentally retarded and discussed Doyle's impaired capacity to control himself. Ex. 5, Penalty Phase at 1338.

On cross-examination by Tenbrook, the expert called by the state, Dr. Eichert, acknowledged that Doyle was brain-damaged and had emotional problems. Ex. 5, Penalty Phase at 1329-30.

We conclude that Tenbrook adduced testimony before the sentencing jury that communicated to the jury substantially the same evidence that collateral counsel asserts should have been adduced.

Finally, the record does not support Doyle's contention that Tenbrook failed to argue these two mitigating circumstances to the jury. In fact, Tenbrook argued: "Mitigating circumstances; is there any doubt that at the time of the offense he was under the influence of any extreme or emotional disturbance?" Ex. 6, Penalty Phase at 1384. He also argued that Doyle did not have the ability to control his conduct. He argued that Doyle did not kill for the so-called "normal" reasons (e.g., money, love, hate, revenge), but rather because he was sick — i.e., Doyle suffered from a well-documented and longstanding problem Doyle had not asked for. Id. at 1383-84. Finally, Tenbrook argued that the purposes behind the death penalty would not be served by imposing the death penalty upon this "pathetic, sick boy." Id. at 1385.

While Tenbrook's argument was very brief and did not discuss the details of the doctors' testimony, Tenbrook may well have concluded that such details were unnecessary because the jury had heard that testimony immediately before the arguments. The record reflects a clear communication to the jury of the defense position, i.e., that the death penalty should not be imposed because Doyle was under the influence of extreme emotional disturbance and because his capacity to control his behavior was impaired.

For the foregoing reasons, we conclude that Doyle has failed to establish that his trial counsel's performance at the sentencing phase constituted ineffective assistance of counsel under Strickland v. Washington. Therefore, Doyle's contention does not meet the first prong of the Strickland test.

For the foregoing reasons, we reverse the district court's grant of the writ.

REVERSED.