Virgil Euristi MARTINEZ, Petitioner-Appellee, v. Nathaniel QUARTERMAN, Director, Texas Department of Criminal Justice, Correctional Institutions Division, Respondent-Appellant, 481 F.3d 249

PRADO, Circuit Judge: Respondent-Appellant Nathaniel Quar-terman (“Quarterman”) appeals the opinion and order of the district court granting Petitioner-Appellee Virgil Euristi Martinez’s (“Martinez”) petition for a writ of habeas corpus based on a claim of ineffective assistance of counsel. Martinez alleged, and the district court agreed, that his trial attorneys, Jerri Yenne and Stan McGee, provided ineffective assistance of counsel by inadequately investigating temporal lobe epilepsy (“TLE”) as mitigating evidence at the punishment phase of his trial.

However, we conclude that reasonable professional judgments supported counsel’s limited investigation into TLE and that Martinez failed to establish prejudice as a result of counsel’s limited investigation. The Texas Court of Criminal Appeals’s denial of Martinez’s application for habeas relief was not “objectively unreasonable.” We therefore REVERSE. I. FACTUAL AND PROCEDURAL HISTORY This is the second time that this court has considered Martinez’s petition for ha-beas corpus.

Martinez alleges that his counsel provided ineffective assistance at the punishment phase of his trial because they failed to fully investigate TLE as mitigating evidence. Martinez contends that the TLE evidence “would have rebutted the State’s case of future dangerousness, provided the jury with a vehicle to spare his life, both in terms of future dangerousness and mitigation, and provided an explanation for his behavior and violent crime.” Ill Fed.Appx. at 225. In support of his claim, Martinez submitted affidavits from Drs.

Theodore Pearlman and Anand Mehendale, in which the doctors opined that TLE played a role in Martinez committing the murders. Although the state habeas record contained affidavits from Martinez’s trial counsel, those affidavits did not clearly demonstrate the extent of counsel’s investigation into and knowledge of TLE. Given the indeterminacy of the record, we vacated the district court’s denial of habeas and remanded for further development of the record.

We instructed the district court to conduct an evidentiary hearing to determine “whether counsel’s investigation of Martinez’s temporal lobe epilepsy was unreasonably deficient and, if so, whether counsel’s failure to investigate this condition and produce evidence relating to it amounted to ineffective assistance of counsel.” Martinez 111 Fed.Appx. at 230. Specifically, we asked the district court to clarify: (1) how much of the information in Dr. Pearlman’s February 27, 1997, report did Yenne learn in her investigation, and whether the report should have triggered further investigation; (2) whether Dr.

Mehendale told Yenne that Martinez suffered from TLE or about that condition’s effect on aggressive behavior; and (3) whether Yenne read Martinez’s school records and considered how TLE might relate to the behavioral problems noted therein. Id. at 227-28. A magistrate judge held a two-day hearing on June 6-7, 2005, and the parties submitted additional deposition evidence to the court. The magistrate judge determined that Yenne read Dr. Pearlman’s report and knew of his diagnosis of TLE.

The report and recommendation concluded that Yenne did not understand the relationship between TLE and post-seizure aggression or Martinez’s future dangerousness because she failed to ask Dr. Me-hendale his medical opinion on these subjects. The magistrate judge also found that, though Yenne read the school records, she never asked either Dr. Pearlman or Dr. Mehendale about how TLE might explain Martinez’s behavioral problems in school. According to the magistrate judge, counsel’s failure to further investigate TLE constituted ineffective assistance of counsel.

The report and recommendation concluded that counsel’s failure to fully investigate TLE prejudiced Martinez because, with further investigation, counsel could have rebutted much of the State’s aggravating evidence and could have given the jury an explanation for Martinez’s crime. Accordingly, on November 9, 2005, the magistrate judge issued a report and recommendation advising that the district court grant habeas relief. On February 7, 2006, the district court issued an opinion and order accepting the magistrate judge’s report and recommendation and granted Martinez’s petition for habeas relief.

Quarterman now appeals the district court’s opinion and order. II. STANDARD OF REVIEW In a habeas appeal, this court reviews the district court’s findings of fact for clear error and its conclusions of law de novo, applying the same standards to the state court’s decision as did the district court. Busby v. Dretke, 359 F.3d 708, 713 (5th Cir.2004). Martinez filed his habeas petition after the effective date of the Antiterrorism and Effective Death Penalty Act (“AEDPA”), 28 U.S.C. § 2254; therefore, AEDPA governs this appeal. Lindh v. Murphy, 521 U.S. 320, 336, 117 S.Ct. 2059, 138 L.Ed.2d 481 (1997).

Under AEDPA, this court may not grant habeas relief on a claim that a state court has adjudicated on the merits “unless the adjudication of the claim ... resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States.” Riddle v. Cockrell, 288 F.3d 713, 716 (5th Cir.2002) (quoting 28 U.S.C. § 2254(d)(1)) (internal quotations omitted).

A state court’s decision is “contrary to” clearly established federal law if “it relies on legal rules that directly conflict with prior holdings of the Supreme Court or if it reaches a different conclusion than the Supreme Court on materially indistinguishable facts.” Busby, 359 F.3d at 713 (citing Williams v. Taylor, 529 U.S. 362, 405-06, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000)). A decision constitutes an “unreasonable application” of clearly established federal law if it is “objectively unreasonable.” Pondexter v. Dretke, 346 F.3d 142, 146 (5th Cir.2003).

The decision of the state court might be incorrect, but still fall below the “objectively unreasonable” threshold. See Neal v. Puckett, 286 F.3d 230, 236 (5th Cir.2002). This court must presume that a state court’s findings of fact are correct, and the petitioner has the burden of rebutting the presumption of correctness by clear and convincing evidence. See 28 U.S.C. § 2254(e)(1). Ten years after AEDPA’s enactment, its standards are by now familiar, but Quar-terman vigorously objects that neither the magistrate judge nor the district court applied AEDPA deference.

We need not decide this particular issue because we are persuaded that, irrespective of AEDPA deference, the district court erred in finding ineffective assistance under Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). III. DISCUSSION In Strickland, the Supreme Court articulated the standard for establishing an ineffective assistance of counsel claim. Martinez must demonstrate both that: (1) his counsel’s performance was deficient; and (2) counsel’s deficient performance prejudiced his defense. Strickland, 466 U.S. at 687, 104 S.Ct. 2052.

Counsel’s performance is deficient if it “fell below an objective standard of reasonableness.” Id. at 688, 104 S.Ct. 2052. The Supreme Court has instructed that judicial scrutiny of counsel’s performance must be “highly deferential.” Id. at 689, 104 S.Ct. 2052. A reviewing court should make every effort “to eliminate the distorting effects of hindsight” and to “evaluate the conduct from counsel’s perspective at the time.” Id.

Further, “strategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable; and strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation.” Id. at 690-91, 104 S.Ct. 2052 (emphasis added). In addition to deficient performance, Martinez must demonstrate prejudice.

Deficient performance results in prejudice when “there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Strickland, 466 U.S. at 694, 104 S.Ct. 2052. More precisely, in a capital case such as this one, the standard 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.” Id. at 695, 104 S.Ct. 2052. 1.

Deficient Performance and Inadequate Investigation The gravamen of Martinez’s ineffective assistance of counsel claim is that his counsel prejudiced his defense by failing to conduct a reasonably adequate investigation of TLE as mitigating evidence. The magistrate judge found that counsel read Dr. Pearlman’s report and knew of Dr. Pearlman’s opinion that Martinez committed murder during the course of a TLE-induced seizure.

Nevertheless, both Martinez and the district court fault counsel for not further investigating TLE and discovering the links between TLE and post-seizure aggression, TLE and Martinez’s poor disciplinary record at school, and how TLE could have explained Martinez’s bizarre and, at times, violent behavior at the Kerrville State Hospital. However, Strickland does not require counsel to fully investigate all mitigating evidence. The Strickland Court recognized there would be times when “reasonable professional judgments supported] limitations on investigation.” 466 U.S. at 690-91, 104 S.Ct. 2052; see also Wiggins v.

Smith, 539 U.S. 510, 527, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003) (“In assessing the reasonableness of an attorney’s investigation, however, a court must consider not only the quantum of evidence already known to counsel, but also whether the known evidence would lead a reasonable attorney to investigate further.”). Certainly counsel did not fully explore TLE, but we must decide whether counsel’s decision to forego further investigation was based on reasonable professional judgments. Without a doubt, Dr.

Pearlman’s report brimmed with information which could have been useful to Martinez’s mitigation case, but it also teemed with damaging information which convinced counsel not to pursue TLE any further. Pearlman’s report contained information which counsel rightly did not want the jury to hear. First and foremost, counsel thought that it would be more harmful than beneficial for the jury to learn that Martinez had a mental disorder which, in Dr. Pearlman’s words, caused “savage and uncontrolled” aggressiveness. Yenne Dep. Vol. 8 at 160-61.

Counsel believed that this might cause the jury to believe that Martinez was a “complete danger to society” and that he was “incapable of controlling any of his behavior.” Id. at 161. Counsel thought that evidence of Martinez’s aggressiveness, even if it were caused by a physical condition, would not sit well with a Brazoria County jury. Id. at 161-62. The evidence for Martinez’s TLE embodies the type of “double-edged” evidence which this circuit has repeatedly stated that counsel may elect not to present to the jury. Martinez v. Dretke, 404 F.3d 878, 889 (5th Cir.2005) (Martinez II), Johnson, 306 F.3d at 253.

The TLE evidence failed to impress counsel not only because it suggested that Martinez was prone to aggressiveness, but also because counsel feared that the jury simply would not believe it. Counsel suspected that the jury would not accept that epilepsy caused the murders because epilepsy is a fairly common disorder and, in most people’s experience, does not result in such catastrophic violence. Yenne Dep. Vol. 6. at 33. Further, Yenne believed that Martinez’s lack of violent incidents in jail was inconsistent with someone who could not control his behavior. Id. Vol. 6 at 38. Counsel were also skeptical of Dr.

Pearlman’s opinion that Martinez committed the murders while having a seizure. McGee questioned how a seizure could last long enough to encompass four murders in which the victims were shot multiple times, requiring Martinez to reload. McGee Dep. at 64-65. The law permits counsel to question Dr. Pearlman’s conclusions based on their review of the evidence. See Riley v.

Dretke, 362 F.3d 302, 305-06 (5th Cir.2004) (allowing counsel not to put on evidence of mental retardation where counsel’s subjective belief that his client was not retarded based on counsel’s observations of the client, information from the family, and school records). In addition to suggesting that Martinez was prone to aggressiveness, Dr. Pearl-man’s report was based, in part, on school records which showed that Martinez engaged in antisocial behavior from a young age. Counsel feared that letting Dr. Pearlman testify about TLE would open the door to Martinez’s troubling school records.

A fairly representative sample of those records noted that Martinez (1) exhibited “explosive behavior,” (2) “thinks about and plans what he can do to get back at those who have bothered him,” and (3) once brought live .22 caliber cartridges to class. Res. Ex. Vol. 1 Tab E at 666, 676, & 625. Dr. Pearlman’s report gave no indication that TLE caused or contributed to these behavioral problems. Even if Dr. Pearlman’s report had suggested that TLE was the cause, such evidence would have been double-edged. Finally, Dr.

Pearlman’s report contained information which counsel believed undermined their overall trial strategy. The report referenced a quotation from Martinez’s mother in which she stated that Martinez was jealous in his love for Veronica Fuentes. Counsel concluded, not unreasonably, that this provided the State with a motive for the murders where before it did not have one. Yenne Dep. Vol. 8 at 158. Counsel believed that evidence of jealousy or stalking would have cinched the death penalty for Martinez. Id. at 177. The report also mentioned Martinez’s confession to Pearlman that he had killed John Gomez.

Counsel believed that conceding Martinez had killed Gomez would have conflicted with their strategy of arguing mistaken identity at the guilt/innocence phase of the trial. While there is certainly no formal rule against switching theories between the punishment and guilt/innocence phases of the trial, in this case, counsel believed that switching theories would make them lose credibility with the jury and appear hypocritical. Yenne Dep. Vol. 8 at 151. Indeed, Stan McGee testified that, in his experience, juries did not react well to a switch in theories between the different phases of the trial.

Counsel chose to argue residual doubt rather than presenting inconsistent theories to the jury. See e.g., Moore v. Johnson, 194 F.3d 586, 618 (5th Cir.1999) (noting that this circuit has held that arguing residual doubt may be a reasonable, even highly beneficial, strategy in a capital case). After reading Dr. Pearlman’s report, counsel knew that Dr. Pearlman believed that Martinez suffered from a mental disorder which made him prone to aggressive behavior and that Martinez committed the murders during a seizure. Therefore, this case is unlike Lockett v.

Anderson, in which we found that counsel provided ineffective assistance where counsel failed to discover evidence of brain abnormalities because counsel did not follow up on evidence which suggested psychological problems. 230 F.3d 695 (5th Cir.2000). Given all of the damaging information contained in Dr. Pearlman’s report, counsel made a reasonable professional judgment to limit their investigation into TLE as mitigating evidence. Despite counsel’s reasonable reservations about presenting TLE to the jury, counsel did make some attempts to follow up on the TLE evidence discovered in Dr. Pearlman’s report.

Jerri Yenne met with Dr. Mehendale and asked the doctor what he thought of Dr. Pearlman’s opinion that Martinez committed the murders during an epileptic seizure. Dr. Mehendale responded that he believed it was unlikely that Martinez committed the murders while Martinez was having a seizure. Faced with conflicting expert testimony about the role that TLE played in the commission of the crimes, it was reasonable for counsel to conclude that TLE was not worth pursuing. Counsel believed that it made no sense to put on experts with different opinions. Yenne Dep. Vol. 8 at 160.

Nevertheless, the district court and Martinez both fault counsel for not asking Dr. Mehendale about post-seizure aggression or how TLE might explain Martinez’s poor behavioral record at school and at the Kerrville state hospital. To fault counsel for not asking these particular questions is to engage in the kind of hindsight second-guessing that Strickland, warned against. 466 U.S. at 689, 104 S.Ct. 2052. Perhaps different counsel might have asked those questions, but this does not mean that Martinez’s counsel’s actions “fell below an objective standard of reasonableness” because they failed to do so.

Id. at 688, 104 S.Ct. 2052. Yenne pointedly asked Dr. Mehendale to evaluate Dr. Pearlman’s opinion that Martinez committed the murders during a seizure and Dr. Mehendale rejected that position. In a forty-five minute conversation with Yenne, Dr. Mehen-dale never suggested that, while it was unlikely Martinez committed the murders during a seizure, post-seizure aggression could have accounted for the murders. Mehendale Dep. at 59. Yenne is a lawyer, not a medical doctor. Yenne and McGee’s personal experiences with and knowledge of epilepsy did not put them on notice of post-seizure aggression.

McGee Dep. at 64-65; Yenne Dep. Vol. 6 at 52; see also Martinez II, 404 F.3d at 886 (taking into account counsel’s personal and professional experience in evaluating whether counsel should have been put on notice to investigate further). Further, and perhaps more importantly, there was simply nothing in Dr. Pearlman’s report which would have alerted counsel to the possibility of post-seizure aggression or to a link between Martinez’s behavioral problems in school and TLE. Instead of pursuing TLE, counsel made the strategic choice to argue residual doubt at the punishment phase.

Counsel believed that they had a strong chance of prevailing on a direct appeal with respect to some exclusion of evidence issues. Yen-ne Dep. Vol. 8 at 168. Counsel also supposed that they might prevail on direct appeal because of insufficient evidence to conclude that Martinez murdered the children. Id. Vol. 6 at 41. Counsel concluded that having Dr.

Pearlman testify would be counter-productive to this potential appeal because he would have to admit that Martinez confessed to killing Gomez and, more importantly, the TLE testimony would have suggested that TLE-induced aggression also prompted Martinez to kill the children. The district court attempted to discount this strategic choice when it notes that Martinez “was convicted with the support of significant eyewitness testimony concerning the identity of the murderer.” R. Excerpts Tab E at 6.

While this may have been true with respect to Veronica Fuentes and John Gomez, this was not true for the children, who were killed in a trailer away from the eyes of witnesses. Residual doubt, especially as it concerns the children, was therefore a strategic choice entitled to deference. See Moore, 194 F.3d at 618. After the evidentiary hearing, there is no doubt that counsel both knew of Dr. Pearlman’s opinion that TLE played a role in the commission of the murders and failed to fully investigate TLE. However, Supreme Court precedent does not require a full investigation into all mitigating evidence. Burger v.

Kemp, 483 U.S. 776, 794, 107 S.Ct. 3114, 97 L.Ed.2d 638 (1987). After reading Dr. Pearlman’s report and discussing it with each other, counsel had sufficient information to determine that TLE was not worth pursuing. Counsel’s decision is entitled to deference and was not deficient. 2. Prejudice Martinez cannot prove that his counsel’s decision not to fully investigate TLE resulted in prejudice.

In determining prejudice, we must decide “whether there is a reasonable probability that, absent the errors, the sentencer ... would have concluded that the balance of aggravating and mitigating circumstances did not warrant death.” Strickland, 466 U.S. at 695, 104 S.Ct. 2052. Our duty requires us to “compare the evidence actually presented at sentencing with all the mitigating evidence contained in the postconviction [sic] record.” Neal, 286 F.3d at 241; see also Williams, 529 U.S. at 397-98, 120 S.Ct. 1495 (reviewing court must re-weigh the totality of the mitigating evidence against the aggravating evidence).

The district court described the mitigating evidence in this case as “potentially of significant help” to Martinez. R. Excerpts Tab E at 6. We disagree with this conclusion. The magistrate judge’s report and the district court’s opinion focus too narrowly on the beneficial aspects of the TLE evidence, while overlooking its many drawbacks. Looking at all the mitigating evidence contained in the post-conviction record, the TLE evidence would not have been a significant help to Martinez.

As a whole, the evidence was not so compelling that there was a reasonable probability that the sen-tencer would have determined that death was an inappropriate sentence. First, as has been previously discussed, rather than cutting solely in favor of Martinez, the TLE evidence was double-edged. As the Supreme Court has noted, “[mitigation, after all, may be in the eye of the beholder.” Burger, 483 U.S. at 794, 107 S.Ct. 3114 (citations omitted).

The jury could have felt that TLE made Martinez a future danger because it inclined him toward uncontrolled aggression, or the jury could have accepted TLE as evidence that Martinez acted with diminished capacity. The TLE evidence also suffered from the fact that Martinez’s experts disagreed over significant aspects of the TLE evidence. Drs. Pearlman and Mehendale broadly agree that TLE-fueled aggression played a role in Martinez’s commission of the murders, but the devil is in the details. First, Dr. Pearlman believes that Martinez committed the murders while having a seizure, but Dr.

Mehendale believes that scenario is unlikely. Instead, Dr. Mehendale opines that Martinez murdered while experiencing post-seizure aggression. Second, Dr. Pearlman does not believe that TLE contributed to Martinez’s antisocial behavior in school, whereas Dr. Mehendale contends that TLE was obtusely related to those behaviors. Finally, Dr. Pearlman’s report states that with treatment, “there is no likelihood that [Martinez] will commit future acts of dangerousness to society.” Dr. Mehendale’s affidavit is less definitive, stating that Martinez’s future dangerousness could be “somewhat diminished” with treatment.

Further, at his deposition, Dr. Mehendale conceded that there was a possibility that Martinez’s TLE disorder could not be controlled — there were no guarantees. Mehendale Dep. at 56. Faced with significant disagreement between Martinez’s experts, a jury might well have been unimpressed with TLE as mitigating evidence. Of course, counsel could have elected to present only one expert to the jury, but a jury would have had sufficient reason to find each expert’s testimony less than compelling. We have already noted the potential pitfalls of putting Dr. Pearlman on the stand in the

section addressing whether counsel’s performance was deficient. Putting Dr. Mehendale on the stand would have been even less beneficial to Martinez. As we have already seen, though Dr. Me-hendale opines that Martinez’s capacity for future dangerousness could be reduced with treatment, he conceded that there was a possibility that Martinez’s TLE could not be controlled with treatment. Unfortunately for Martinez, that was not the last of Dr. Mehendale’s damaging admissions. Although there is a link between TLE and violence, Dr.

Mehendale, on a couple of occasions, stated that there is no data quantifying the degree of violence associated with TLE. Mehendale Dep. at 20-21 & 53. In fact, Dr. Mehen-dale reports that in a study of 5400 epileptics “none of them committed murder.” Id. at 77. He stated that although epileptics have “bad brains” their actions very rarely result “in a horrid tragedy like this.” Id. Dr. Mehendale concluded by saying that the reason most epileptics stop short of committing murder is because “epileptic brains have [a] conscience, and [Martinez] didn’t.” Id. at 78. Dr.

Mehen-dale, Martinez’s own expert, would have undermined any argument by Martinez that TLE reduced his moral culpability for the murders. Surely, this is not compelling mitigation testimony which undermines the outcome of the state trial. After considering all of the mitigating evidence, we hold that the additional mitigating evidence was not so compelling, especially in light of the horrific facts of the crime, that the sentencer would have found a death sentence unwarranted.

At the very least, the Texas Court of Criminal Appeals’s decision finding no ineffective assistance of counsel was not “objectively unreasonable.” IV. CONCLUSION For the reasons stated above, we REVERSE the decision of the district court. REVERSED. . A jury convicted Martinez of murdering his ex-girlfriend Veronica Fuentes; Veronica’s two children, five-year-old Joshua and three-year-old Cassandra; and a bystander John Gomez. The jury subsequently sentenced Martinez to death. Martinez properly pursued and exhausted his stale remedies.

The Texas Court of Criminal Appeals ultimately denied Martinez's application for habeas relief. Our previous opinion contains a fuller account of the factual and procedural history. See Martinez v. Dretke, 111 Fed.Appx. 224 (5th Cir.2004) (Martinez I). . In the words of Stan McGee, “my sense of our investigation about mitigation and future dangerousness was everything that we came up with or everything that Ms. Yenne came up with seemed to me to be — it hurt more than it helped.” McGee Dep. at 47. .Admittedly, Dr.

Pearlman’s report states that with treatment "there is no likelihood that [Martinez] will commit future acts of dangerousness to society,” but it is counsel’s decision to decide whether, on balance, the TLE evidence was more helpful than harmful. See Johnson v. Cockrell, 306 F.3d 249, 253 (5th Cir.2002) (noting decision not to present double-edged testimony even less susceptible to judicial second-guessing). Furthermore, Dr. Pearlman’s opinion as to future dangerousness was based, in part, on his belief that Martinez lacked either a criminal history or a prior history of catastrophic violence.

Yenne, however, knew that Dr. Pearlman was unaware of some of Martinez’s prior bad acts, such as his history of stalking women, and she wanted to avoid exposing Dr. Pearlman to this potential line of cross-examination. Yen-ne Dep. Vol. 8. at 162, 142 (mentioning history of stalking women). . Martinez and the district court accuse counsel of not understanding the distinction between TLE and other types of epilepsy or the relationship between TLE and violence. These objections are overstated because, as will be discussed in the

section on prejudice, scientists currently do not have a complete understanding of how TLE relates to violence, especially the catastrophic violence of this case. . Counsel’s skepticism about Dr. Pearlman’s explanation for the murders was reasonable, for Dr. Mehendale also disagrees with it. . Counsel also worried that it would come out that Dr. Pearlman did not believe Martinez’s account of the murders. Yenne Dep. Vol. 8 at 153. . The district court found that counsel "simply did not know the link between TLE and violence.” R. Excerpts Tab E at 5.

This finding of fact is clearly erroneous because it is not supported by the record. While it is true that counsel did not know the relationship between TLE and post-seizure aggression, after reading Dr. Pearlman’s report, counsel knew that Dr. Pearlman believed Martinez committed murder while having a seizure and that TLE caused aggressive behavior. . The relationship between TLE and Martinez's behavioral problems in school is far from self-evident. Dr. Mehendale believes that TLE contributed to Martinez's antisocial behavior in an "obtuse way,” but Dr.

Pearl-man does not agree that Martinez's childhood behavior was caused by or a sign of TLE. Mehendale Dep. at 29; Pearlman Dep. at 100-01.