OPINION
KELLER, P.J.,
delivered the opinion of the Court
in which MEYERS, PRICE, WOMACK, KEASLER, HERVEY and COCHRAN, JJ., joined.
In accordance with instructions from the United States Supreme Court, we reconsider this case in light of Atkins v. Virginia.1 We shall affirm.
I. BACKGROUND
A. Procedural history
In January of 2000, appellant’s trial for capital murder began. At trial, evidence was introduced by both parties regarding whether appellant was mentally retarded. This evidence was introduced primarily in the punishment phase in connection with a determination of the mitigation special issue. At no point did appellant request that the trial judge or the jury make a specific fact-finding as to whether appellant was in fact mentally retarded.2
Appellant was subsequently convicted of capital murder and sentenced to death. In his eighth and ninth points of error on direct appeal, he alleged that inflicting the death penalty on the mentally retarded violates due process and constitutes cruel and unusual punishment under the United States Constitution. On January 16, 2002, this Court affirmed his conviction and sentence.3 In a published opinion, we held that there was no per se bar to executing mentally retarded persons.4 We also pointed out that the State presented some evidence .that appellant was not mentally retarded and that the jury had the opportunity to observe appellant’s behavior firsthand in a videotaped interview.5
After our decision was handed down, appellant petitioned the United States Supreme Court for a writ of certiorari. He also filed a state application for writ of habeas corpus, pursuant to Article 11.071.6 In both of these proceedings, he pursued his mental retardation claim. On June 20, 2002, the Supreme Court decided Atkins.7 On August 5, 2002, in the habeas action, the trial court designated the issue of whether appellant was mentally retarded [27]*27as a previously unresolved fact issue and ordered a hearing by way of affidavits. On October 7, 2002, the Supreme Court vacated our judgment on direct appeal and remanded the case to us for reconsideration in light of Atkins.8 On December 3, 2002, after reviewing the trial record and the affidavits submitted by the parties, and relying upon personal recollection of the events occurring at trial,9 the habeas trial court adopted the State’s proposed findings of facts and conclusions of law, concluding that appellant is not, in fact, mentally retarded. On February 26, 2003, we denied relief on the habeas application in an order adopting the trial court’s findings.10
B. Factual History
Appellant and his friend, Robert Neville, decided to kill someone. In pursuance of their plan, they purchased various weapons, including rifles, pellet guns, and a crossbow, along with ammunition. They decided to kill Amy Robinson, a former coworker of appellant’s at Kroger, because she was an easy target. Robinson suffered from a genetic disorder called Turner’s syndrome. As a result, she stood only four feet five inches tall at the age of nineteen and had the mental capacity of a third or fourth grader. She was physically slow and trusted everyone. For this reason, appellant and Neville believed she would not put up a fight.
Appellant and Neville went to the Kroger store where Robinson worked, checked her schedule, and then waited for her to ride her bicycle down the street on her way to work. They persuaded her to go with them for a drive and promised they would take her back to work. They drove Robinson twelve miles away from the store to a remote field, where Neville shot at her with the crossbow but failed to hit her. Appellant then shot her in the leg with a pellet gun, and Neville shot her in the chest with a .22 caliber rifle. Appellant also shot her in the chest with the pellet gun several times. Neville then shot Robinson in the head, killing her instantly, because appellant worried someone would hear the loud noises that Robinson was making. Appellant and Neville returned to the scene of the crime a few days later. Appellant took keys and money from the victim’s pocket, and Neville fired some more shots into her dead body. A few weeks later, the police arrested the two men as they tried to flee to Mexico.
C. Mental retardation evidence
1. Trial
During the guilt phase of trial, one State’s witness, Tamara Campbell, testified briefly about appellant’s mental ability. Campbell had been appellant’s supervisor at Kroger. When asked whether appellant appeared to be mentally challenged, Campbell replied, “No. He was lazy, but he wasn’t mentally challenged, in my opinion.” However, according to Campbell, Amy Robinson was mentally challenged.
During the punishment phase, the parties introduced a substantial amount of evidence regarding appellant’s level of intelligence. Karen Hall, appellant’s mother, testified that appellant had always been slower than other children. At age five, he could not stack blocks. He was in special education classes from the first [28]*28through the eighth grade, was placed in regular classes in ninth grade but could not handle them, and did not advance beyond the tenth grade. Karen further testified that appellant plays like an eight-year-old and associates with children who are eight, nine, and ten years old. She catalogued a number of areas in which appellant was deficient, when compared to others his age: he could not count change, tell time from a traditional clock with hands, read a menu, use public transportation, use a vacuum cleaner without tearing it up, make his bed, and wash dishes. He also would not use a table knife; where a piece of meat, such as a pork chop, needed to be cut, appellant would tear it up with his hands. He also became easily lost within a few blocks from home, and he often chewed with his mouth open. She conceded that appellant could read and write at a fourth grade level, use the phone, operate a microwave, load and unload a dishwasher, use a pencil and pen, make a sandwich, brush his teeth, shave, and dress himself. She also testified that he read children’s books and the Bible, and she admitted that appellant was able to read well enough to pass a written driver’s license examination (after failing the computerized version). She also admitted that appellant has had at least one girlfriend his own age, and possibly two.
Damon Lee Hall, appellant’s older brother, testified that, at age fourteen to fifteen, appellant associated with children ages eight to nine. Appellant had trouble explaining things-taking a long time to do so. He could understand directions only when they were given slowly. Appellant had difficulty understanding how to play pool and had trouble reading clocks with hands. Appellant also had trouble counting money and did not know when he was shortchanged. Damon testified that he once took advantage of appellant’s deficiency to cheat him out of some of the proceeds of the sale of a video game console system.
Ken Traynor was a wood shop teacher when appellant went to school. At first, Traynor believed appellant was mentally disturbed, but he came to believe that appellant was simply lacking intelligence. Appellant could not do the simplest of tasks. He could not understand concepts ' even after repetition, and he would not remember how to do something from one day to the next. Mathematics was largely beyond appellant’s ability. Even the simplest addition was sometimes a challenge, while multiplication and division were simply out of the question. Traynor recalled one assignment that took appellant eight weeks to complete, when it should have taken only two. However, Traynor also recalled that appellant had difficulty getting motivated to work-sometimes he would just stay in one place and sit. Appellant did successfully complete a video game console-something he was actually highly motivated to do. He sometimes stayed late to work on it. When asked about appellant’s verbal abilities, Traynor replied that appellant did not have any trouble verbalizing his responses.
Cheryl Conner was a school psychologist at the time she testified, but she had taught appellant English, reading, and math during his freshman and sophomore years in high school. She estimated that appellant had the reading comprehension of a first-grader and the math ability of a third-grader. Appellant could add and subtract on paper but not in his head. He could not write complete sentences or paragraphs without significant prompting. He could follow only one instruction at a time-and sometimes he would fail at that. Conner had to set a five-minute task timer to keep appellant on task or he would fall asleep or let his attention drift to the point where he would simply sit and stare.
[29]*29Some teachers commented that appellant drooled in class. He would often sleep in class and not do assignments. Some of the other teachers thought appellant was lazy, but Conner believed that appellant was very depressed. She believed appellant’s inability to understand things caused him to give up and tune out. She did testify that appellant was apparently good at video games.
According to a Test of Non-verbal Intelligence (TONI), appellant had an IQ score of 84, while a Wechsler Intelligence Scale for Children — Revised exam (WISC-R) indicated an IQ of 71. Appellant’s school classified him as “learning disabled,” not as “mentally retarded.” Appellant was classified as learning disabled because, in the opinion of the assessors, he failed to live up to his potential. Linda Akin assessed appellant’s “adaptive behavior” using “informal measures.” Akin concluded: “Results showed that the student’s level of intellectual functioning is consistent with his or her adaptive behavior, with no significant deficits in either area.” Conner disagreed with that assessment. Conner admitted, however, that depression can lower an IQ score.
Chris Bybee was appellant’s math teacher during the 1995-96 school year. He characterized appellant as slow and difficult to motivate. Appellant could not perform multiplication or division. Based upon appellant’s work, Bybee guessed that appellant’s IQ was in the upper 60s. By-bee admitted that appellant was unmotivated most of the time, was sometimes lazy, and sometimes slept in class.
Dr. Mark Cunningham was a clinical and forensic psychologist hired by the defense.
Dr. Cunningham administered the Wech-sler Adult Intelligence Scale, third edition, exam (WAIS-III)to appellant and obtained a full-scale IQ score of 67 with a three percent margin of error. Dr. Cunningham noted that a 1991 WISC-R exam had placed appellant’s IQ at 71 with a five percent margin of error. At the time of the 1991 evaluation, appellant was classified as learning disabled, but at an earlier evaluation, conducted in 1988, appellant had been classified as mentally retarded.
Based upon interviews with Bybee, Conner, and Tina Dodson (appellant’s stepmother), Dr. Cunningham concluded that appellant possessed adaptive behavior deficits in seven different areas: (1) independent functioning (eating, dressing, transportation), (2) economic activity (handling money), (3) language development, (4) self-direction (excessive passivity), (5) socialization (ability to interact with others), (6) social engagement, and (7) functional academics. In support of this conclusion, Dr. Cunningham cited the following observed behavior: appellant could not play cards, give directions to his home, identify nearby streets, or travel to his workplace on his own. Appellant would not brush his teeth or use a table knife.11 Appellant had no peer-age friends or girlfriends beyond relatively brief relationships. He could not name artists in the music groups to which he listened and was amazed that others could do so. Appellant was easily cheated in trades and in selling things. He was generally gullible and easy to manipulate. Appellant required very specific, concrete instructions and was slow to comprehend or explain things. Appellant took little [30]*30interest in the world around him and displayed low initiative.
Dr. Cunningham also administered the Wide Range Achievement Test, third edition, exam (WRAT III), designed to show a person’s ability in the area of functional academics. Appellant’s reading score placed him at an IQ of 59, his spelling score at an IQ of 51, and his arithmetic score at an IQ of 55. Dr. Cunningham concluded that appellant was indeed failing to live up to his potential, but his potential was mental retardation and appellant was scoring even worse than that. Dr. Cunningham commented on the TONI-II exam, which had yielded an IQ of 84, by saying that the potential for inaccuracy in that test was high. The TONI test was designed primarily for students who were difficult to test, sometimes because they did not speak English, or were paralyzed, or had some other disability not necessarily related to intelligence that impeded the effectiveness of the more comprehensive, individually-administered intelligence exams. Dr. Cunningham observed that the TONI-II exam was only fifteen minutes long, while an individually-administered intelligence exam took one and a half hours. In addition, the TONI-II was administered in a group setting, which made cheating (looking on someone else’s paper) possible. Dr. Cunningham concluded that appellant would be classified as “mildly mentally retarded” under the DSM-IV12 classification scheme. Dr. Cunningham also noted that appellant was exempted from taking the TAAS13 test.
During cross-examination, Dr. Cunningham conceded that appellant had told some lies during their initial interview. Appellant had stated that he never fired a gun, that he never returned to the victim’s body, and that he was under the influence of methamphetamine and marijuana. Appellant later admitted that he had stolen money from the victim’s dead body. Appellant also remarked that his rights had not been read to him when he was arrested and that the police had no right to question him., Dr. Cunningham also testified that appellant said he was reading a Stephen King novel, a Dean Koontz novel, and the Bible.
The State called several rebuttal witnesses. The first was Dr. J. Randall Price, a clinical and forensic psychologist and a neuropsychologist. Dr. Price accepted the results obtained by Dr. Cunningham’s IQ testing14 but differed somewhat as to their significance. In Dr. Price’s estimation, appellant might be borderline rather than mentally retarded:
Q. Were you able to determine whether or not Michael Hall was mentally retarded?
A. Well, I’m not as convinced that he is as Dr. Cunningham is. He is at that level where it’s either borderline, right at the level of mild mental retardation, or he’s mildly mentally retarded. It’s— it’s sort of a judgment call.
Dr. Price did engage in his own adaptive behavior testing, using the Street Survival Skills Questionnaire. The results of that testing showed appellant’s adaptive skills to be “pretty average.” Appellant was [31]*31borderline in three areas: time, using money, and measurements. He was average on tools, domestics, health and safety, personal services, and the use of functional signs.15
Dr. Price further testified that, during his interview with appellant, appellant acted and talked like an adolescent or young adult. Appellant spoke in a fluent, articulate manner about his crime and related events in a coherent, logical way. His thought processes were on topic, logical, and goal-oriented. Appellant’s thought processes appeared to flow “pretty normally” and the interview was like a normal conversation.
During cross-examination, Dr. Price testified that, taking into account the margin of error, Dr. Cunningham’s IQ exams of appellant yielded an IQ range of 64 to 71 within a 90 percent confidence level. IQ testing at age twelve yielded an IQ range of 68 to 74 within a 90 percent confidence level. In response to defense questioning, Dr. Price also remarked that appellant used the word “concordance” on his own. In comparing IQ tests, Dr. Price estimated appellant’s IQ to be right around 70. Dr. Price conceded that appellant could not think as fast or as well as 97 to 98 percent of the human population. On redirect, Dr. Price testified that, during interviews with himself and Dr. Cunningham, appellant had given a variety of statements minimizing his role in the offense and a variety of excuses for participating in the offense (alcohol, drugs, being “out of it,” etc.).
Alan Boles, an eighteen-year-old college freshman, also testified. At age 16, Boles had worked at Kroger along with appellant. Boles testified that appellant taught him how to sack groceries. Boles never noticed anything “slow” about appellant.
Although he was never formally qualified as an expert, Boles had some experience dealing with the mentally retarded: at the time of his testimony, he was part of a program for teaching sports to mentally challenged children.
The State also called Monica Zepeda to the stand. Zepeda testified that she worked as a waitress at a restaurant visited by appellant and Neville. Appellant ordered his own meal (fried catfish) and appeared to utilize eating utensils in a normal manner. Finally, Detective Richard Nutt testified that he had read appellant warnings mandated by Article 38.22, and appellant appeared to understand those warnings.
Several witnesses discussed appellant’s head injuries. Karen Hall testified to several head injuries sustained by appellant during his childhood. She admitted that, in several of those incidents, appellant was not taken to a hospital. She alleged that three incidents caused him to be taken to a hospital. In one, a trailer fell on appellant’s head and the doctors diagnosed appellant as probably having a slight concussion. In a second incident, he cut his eye open and had to have stitches. In the third incident, he suffered a whiplash injury in an auto accident. Karen admitted that a concussion was the most serious injury appellant had sustained.
While Dr. Cunningham reported obtaining information about head injuries from appellant’s family members, he acknowledged that only one incident — involving a laceration of appellant’s nose — could be independently verified in medical records. Dr. Cunningham also admitted that there were no medical records to verify that appellant ever suffered a concussion. Dr. [32]*32Cunningham further testified that head injuries played a relatively minor role in his evaluation.
Dr. Price did not find the materials regarding head injuries to be particularly significant. In his view, the incidents did not represent serious enough injuries to cause any brain injury.
We note several other types of evidence that might be relevant to the question of whether appellant is or is not mentally retarded. Karen Hall and Dr. Cunningham testified that appellant was given Ritalin for attention deficit disorder. Karen also testified that appellant was physically abused by one of her boyfriends. Dr. Cunningham also testified that appellant was in a family atmosphere that included maternal alcoholism, observed domestic violence, physical abuse, traumatic sexual experience, and parental neglect.
Appellant’s own words and actions are also relevant to this inquiry. At trial, appellant wrote a pro se motion to remové his attorneys. The motion was discussed in an ex parte hearing, outside the presence of the prosecuting attorneys. Although the motion contained only facts and no law, Judge Cooke, the judge hearing the motion, commented that it was “well drafted” and that he could “name some attorneys that can’t draw an instrument any better than this right now.” Judge Cooke further commented that this motion, if discovered and used by the State, could damage appellant’s attorneys’ “diminished capacity” strategy by showing that appellant was actually “pretty bright,” to be capable of drafting such a complicated document. After some discussion, appellant decided to withdraw his motion, and the judge ordered that the transcript of the hearing be sealed.16
Appellant also participated in an interview with Fox 4 News. This interview was videotaped, and the videotape was later played to the jury. On the videotape, appellant’s speech was smooth and fluid, and his thought processes appeared to be coherent and logical.. He stated that the victim was chosen because “she trusted us; it was easy.” He bragged about his role in the murder: Neville could not have accomplished it on his own because it was appellant who got the victim to trust him. Without appellant, the victim would have escaped. Appellant acknowledged that he and Neville would be punished for killing the victim, and he expressed the hope that he would receive the death penalty. He wanted the death penalty because he had a “sueky-ass life.” He indicated that he had tried to take his own life before but had failed: “You can’t take your own life but you can take someone else’s.” He stated that the current, method of execution, which allowed one to die sleeping, was better than the gas chamber or the electric chair. He thought killing the victim would be a good way to get the death penalty because that is what would happen to someone who killed a person with a family that cares about her. When asked whether he was forced to talk to the media, appellant responded that he was told, and understood, that he did not have to.
2. Habeas corpus
Appellant submitted the affidavits of two additional psychologists who opined that appellant was mentally retarded. Sally Church, a licensed professional counselor, licensed marriage and family therapist, and nationally certified school psychologist, licensed in Oklahoma, evaluated appellant’s condition based upon his family history; school, medical, employment, jail, [33]*33and prison records; previous evaluations; investigator’s reports; court testimony and affidavits; and a psychodiagnostic evaluation of appellant conducted by her at the prison. Church testified that appellant’s mother drank alcohol while she was pregnant, had difficulty with depression, was served in special education during public school, and was an overwhelmed, frustrated, abusive, and neglectful parent. Church further testified that appellant was identified in kindergarten as being behind his age norm developmentally, that he was served in special education throughout his public school attendance, that he dropped out in eleventh grade, that the school recommended that he be identified as Mentally Handicapped but his mother declined— wanting him identified as Learning Disabled — and that he was never able to learn reading, math, spelling, and writing beyond the third grade level. She noted that he had vision problems that could be largely corrected with glasses and that he was assessed as having a hearing problem in elementary school, but there was no follow-up on the extent of his hearing loss. She stated that, by the age of ten, appellant had “experienced eleven head injuries, several of which resulted in concussions.” She also noted that appellant had been prescribed Ritalin for inattention, impul-sivity, and hyperactivity. In her opinion appellant is mentally retarded, being less intelligent than 98 percent of the human population, with an IQ score of 67. She stated that appellant has critical deficits in adaptive skills and behaviors and that it is “highly doubtful that he alone could meet the needs of his day to day life.”
She also stated that appellant’s physical appearance was unusual. She characterized his appearance as typical of Fetal Alcohol Syndrome or Fetal Alcohol Effect. She stated that he also exhibited characteristics that resemble other genetic disorders such as XXY, Kleinfelter Syndrome; YYX, Extra Y Chromosome; or Fragile X Syndrome. She noted that all of these disorders are related to mental retardation and are present at birth.
Dr. George Denkowski, a psychologist licensed by the State of Texas and certified by the Texas Department of Mental Health and Mental Retardation to conduct evaluations for the purpose of diagnosing mental retardation, also expressed the opinion that appellant is mentally retarded. He noted that a WISC-R test administered to appellant in 1991 (at age twelve) resulted in a full-scale IQ score of 71, a TONI-2 test administered at age 16 resulted in a score of 84, and a TONI-8 test administered at age 20 resulted in a score of 77. However, he considered the TONI tests to be “unreliable indices of general intellectual functioning.” Dr. Denkowski also relied upon Dr. Cunningham’s adaptive deficits evaluation based upon interviews with Conner and Bybee. And he relied upon the WRAT-III administered by Dr. Cunningham and the K-FAST17 test administered by Dr. Price to show that appellant’s academic skills were seriously deficient. The K-FAST test results indicated that appellant’s arithmetic skills and reading skills were less proficient than those of 99 percent and 95 percent of same-aged persons, respectively. Finally, Dr. Denkowski criticized the Street Skills Survival Questionnaire as being an inaccurate tool for diagnosing mental retardation in an adult. Dr. Denkowski did not personally interview appellant.
Appellant also submitted affidavits from his trial attorneys, a fellow death-row inmate, two defense private investigators, and a former teacher. All of these people opined that appellant was mentally re[34]*34tarded. William Harris, one of the trial attorneys, stated that he had numerous opportunities to observe appellant’s mental limitations. Harris found that, even after multiple explanations, appellant could not grasp the law of parties and why the law of parties made him liable for capital murder. Appellant also had little or no sense of what things cost or what may or may not be available to a jail inmate. Harris also commented that appellant “was fairly adept at masking his mental retardation if you only deal with him superficially.”
Paul Conner, appellant’s other trial attorney, stated that appellant would ask the same questions day after day- — showing an inability to remember or grasp the answers. Appellant would mask his retardation by not asking questions. He would sometimes sound more informed on a subject than expected, but this was because he would parrot words and phrases that he heard, without grasping their meanings.
Bill Coble, a fellow death row inmate, occupied a cell immediately adjacent to appellant’s. Coble remarked that appellant was called “Half Deck” by the prison guards and other inmates on death row. He characterized appellant as slow in his thinking and “worse off than Johnny Pen-ry” whom he also knew about. Appellant could not learn the meanings of new words, even when told their meanings over and over again. Due to his lack of understanding of money, he could not handle buying items from the prison commissary. Appellant had to be reminded every day to wash himself, shave, and clean the toilet. He also had to be reminded how to keep his food container and his bedding clean. Appellant would often “space out” and forget what he had just done. He listened to cartoons on the radio (on a television sound setting) and could parrot what he heard, if it was something he has heard over and over again. Coble stated that appellant was very upset to learn about a civil suit against himself and Kroger regarding Robinson’s death. Appellant thought he could get another death penalty from the civil suit.
Joseph Ward, a private investigator appointed to assist the defense in habeas corpus proceedings, stated that appellant’s ability to understand the subject of conversation appeared to be significantly less than that of others his age. Appellant’s demeanor was childlike, and he acted inappropriately for someone his age. He could not recall or meaningfully relate events surrounding his arrest. He had difficulty remembering the other private investigator’s name even though he had just been told the name. Appellant did not know the meanings of many words, would use words inappropriately, and would mispronounce them. The other private investigator, John Ladd, stated that appellant had an attention span of five minutes and talking to him was like talking to a six- or seven-year-old child.
Stephen Dollar, a licensed attorney at the time of the habeas action, was appellant’s world history teacher during the 1995-96 school year. Dollar testified that appellant could not follow instructions or complete assignments as distributed. Appellant displayed no cognitive ability in the classroom and demonstrated behaviors similar to those of a child with a diagnosis of mental retardation. He would not respond to his own name when called upon and would stare out the window and drool most of the class hour. Appellant could not accomplish the most menial of tasks, even when they were simplified to accommodate his special education needs. He appeared to have no friends, and he was an object of ridicule by his fellow students.
The State presented controverting affidavits. Five prison guards — Brandon [35]*35Daniel, Julie Perego, Suzanne Prosperie, Todd Tatum, and Darrell White — averred that appellant is not mentally retarded. Daniel, who was assigned to appellant’s area about one day a week, said that appellant “acts as normal as anyone in his pod.” He further stated that he “had been around people who were slow mentally” but had “not seen that in” appellant. Further, there was no indication that appellant did not understand how to obey orders and follow prison rules. Under Daniel’s observations, appellant was “just a normal inmate.”
Since Perego arrived at the prison unit in June of 2001, she saw “nothing unusual” in appellant’s conduct that was “different from [that of] the other inmates.” She characterized his behavior as “normal” and said that she never had to tell him what to do more than once. She further stated, “I have never seen anything that would make me think Michael Hall is mentally retarded and he seems pretty normal to me in my observations of him.”
Prosperie saw appellant, on average, two days a week. Hall seemed like “just a normal inmate.” He socialized with other inmates, and there was nothing unusual about his conduct or attitude. She remarked that his cell was filthy, but it was so by his choice. She stated that she had not seen any signs of mental retardation or illness since she had been around him, and she had some experience with mental retardation because her neighbor’s daughter was mentally retarded. She had never heard appellant referred to as “Half Deck” by anyone.
In his three months around appellant, Tatum “observed nothing unusual” about him. Although appellant’s hygiene was not the best, he had “learned the system here” and understood what he was doing. Tatum knew some children in school with Down’s syndrome, but he had not seen anything in appellant to indicate that he is mentally retarded.
White had also worked around appellant for three months. He stated that he had not observed anything unusual about appellant’s conduct and that appellant appeared to be “just a normal inmate.” He had no problems with appellant failing to understand anything he told appellant to do. Appellant took showers, brushed his teeth, and did other normal activities. White had an uncle who was mentally retarded and had to be told repeatedly what to do, and appellant was nothing like his uncle. White further stated, “I have seen nothing in my observations of Michael Hall that makes me think he is mentally retarded.”
The State also presented affidavits of Dr. Price. Referring to his opinion given at trial, Dr. Price stated that appellant’s “intelligence fell either in the borderline range of intellectual functioning (IQ = 70-84) or in the upper end of mild mental retardation (IQ = 50-55 to 70).” Based upon his review of various tests and records, Dr. Price opined that appellant was a poor student but does not have significant adaptive deficits. He concluded: “My review of this case does not clearly indicate that Michael Hall is mentally retarded.” In response to Dr. Church’s affidavit, Dr. Price stated that appellant did not exhibit signs or symptoms associated with genetic disorders such as Fetal Alcohol Syndrome, Kleinfelter’s Syndrome, Extra Y Chromosome, or Fragile X Syndrome.
II. ANALYSIS
A. General considerations
In Atkins, the Supreme Court held that the Eighth Amendment’s “cruel and unusual punishments” clause prohibits the [36]*36execution of mentally retarded persons.18 The Supreme Court left to the States the task of fashioning appropriate procedures for determining who is in fact mentally retarded:
To the extent there is serious disagreement about the execution of mentally retarded offenders, it is in determining which offenders are in faqt retarded. In this case, for instance, the Commonwealth of Virginia disputes that Atkins suffers from mental retardation. Not all people who claim to be mentally retarded will be so impaired as to fall within the range of mentally retarded offenders about whom there is a national consensus. As was our approach in Ford v. Wainwright [477 U.S. 399, 106 S.Ct. 2595, 91 L.Ed.2d 335], with regard to insanity, “we leave to the State the task of developing appropriate ways to enforce the constitutional restriction upon its execution of sentences.”19
In the habeas context, we recently adopted temporary guidelines, to be used during the legislative interregnum, for determining whether a defendant is mentally retarded.20 Under these guidelines, a person is considered mentally retarded if he has these three characteristics: (1) “significantly subaverage general intellectual functioning” (an IQ of about 70 or below), (2) “related limitations in adaptive functioning,” and (3) onset of the above two characteristics before age eighteen.21 Expert testimony is relevant to a fáct-finder’s determination of these factors but is not necessarily conclusive.22 A separate jury determination of mental retardation is not required,23 and at least on habeas corpus, the defendant shoulders the burden of showing his mental retardation by a preponderance of the evidence.24 And we afford almost total deference to the trial judge’s findings of fact, especially when those findings of fact are based upon credibility and demeanor.25
Unlike Briseno, the present case is still pending on direct appeal. Issues are often evaluated differently in the habeas corpus and direct- appeal contexts. Some claims fail on direct appeal because the record is not sufficiently developed to evaluate the claims. This occurs most often with ineffective assistance of counsel claims.26 Other claims may be evaluated under a different (more onerous) standard in the habeas context than in the direct appeal context.27 Neither of these differences are material to the case at hand.
[37]*37B. The habeas record and judicial notice
Had Atkins been handed down before our original opinion on direct appeal, a serious question about the adequacy of the direct appeal record would have been before us. While the parties introduced a significant amount of evidence regarding whether appellant was mentally retarded, mental retardation was not considered as a discrete issue by the trial judge or the jury. Although the parties certainly had incentive to litigate the question of appellant’s intelligence, the litigation occurred as a question of degree: defense counsel could contend that appellant’s low intelligence mitigated his moral culpability even if it did not amount to mental retardation, while the State could contend that, even if appellant were in the mental retardation range, he appreciated the consequences of his actions to a sufficient degree to deserve the death penalty. Had mental retardation been an ultimate issue, the parties may well have litigated the issue even more robustly than they did, as the issue would be a question of kind (which side of the mental divide appellant was on) rather than degree (how much did appellant appreciate the immorality of his conduct).
We did not address the adequacy of the direct appeal record on original submission because, at that time, we did not perceive mental retardation as a bar to execution. But the trial court and this Court did have the benefit of Atkins during the habeas proceedings. The parties had ample opportunity to present evidence at that time on the specific issue of mental retardation.
And we can consider the habeas proceedings and evidence in the current posture of this appeal. Over three decades ago, we addressed one of those rare instances in which we considered a particular issue on direct appeal after evidence had been obtained on that issue in habeas proceedings.28 In Huffman v. State, the defendant filed a motion for new trial, alleging newly discovered evidence.29 This motion was not accompanied by affidavits supporting the allegation.30 While his appeal was pending, the defendant was released from custody due to an oversight.31 This Court dismissed the appeal on the ground that the defendant’s absence from custody was the equivalent of an escape.32 The defendant was subsequently granted an out-of-time appeal by the Fifth Circuit.33 After the right to appeal was granted but before the appeal was filed, a hearing was held on a state application for writ of habeas corpus challenging the same conviction.34 Evidence not contained in the appellate record was presented at this hearing, including the alleged newly discovered evidence and evidence that the defendant was not represented by counsel at the time the motion for new trial was filed.35
We held that this Court could properly take judicial notice on direct appeal of the habeas proceedings and consider the evidence developed at the habeas hearing.36 Finding that the defendant was without counsel when the motion for new trial was [38]*38filed, we considered the merits of his newly discovered evidence claim in light of the habeas record and overruled his points of error.37
The present case is on all fours with Huffman: as in that ease, we are faced in a direct appeal with an issue that has already been presented to us on habeas corpus. Consequently, we address appellant’s mental retardation in light of both the direct appeal and the habeas records. In this vein, we reject any notion that the direct appeal record in this case must be viewed in isolation. The additional, habeas evidence is before us; taking it into account is necessary for a complete and accurate view of appellant’s intellectual capabilities.38
C. Burdens of proof and standards of review
That standards for evaluating a claim sometimes differ from appeal to ha-beas is also of no concern here. While direct appeal standards may apply to a claim on direct appeal even though the claim is supported by evidence obtained in judicially noticed habeas corpus proceedings,39 if a substantive component of a particular type of claim already contains a standard that equals or exceeds the applicable standard on habeas corpus, then the inquiry collapses to analyzing the claim by its components.40 In the present case, we decide that mental retardation is comparable to an affirmative defense, and thus, the burden is always upon the defendant to prove that condition by a preponderance of the evidence. So, the standards for proving mental retardation at trial and on ha-beas are the same, and the resulting standard of review of a trial court’s findings against the defendant are also the same in both contexts.
The groundwork for this conclusion has already been laid in Briseno. There we observed that the issue of mental retardation is similar to the issues of insanity, competency to stand trial, and competency to be executed — all of which impose upon the defendant a preponderance of the evidence burden of proof.41 We now also observe that the mental retardation issue could be described as a confession-and-avoidance type of punishment-mitigating factor, which would make it like several statutory punishment-mitigating factors that carry the same proof standard as affirmative defenses: sudden passion in the murder context,42 release in a safe place for the offense of aggravated kidnapping,43 imperfect renunciation for the offense of organized criminal activity,44 and, at least arguably, imperfect renunciation [39]*39for inchoate offenses.45 The counter-example would be the mitigation special issue in capital cases, for which the Legislature has prescribed no burden of proof.46 But the mitigation special issue is a very broad issue, capable of taking into account a wide range of circumstances. By contrast, mental retardation is a single, discrete fact that by itself mitigates the penalty for capital murder, just as sudden passion, release in a safe place, and renunciation all are discrete facts that by themselves mitigate the penalty for their respective crimes. Given the legislative backdrop for similar affirmative defenses and analogous punishment mitigating factors, we find, absent further legislative guidance, that mental retardation is the type of issue that must be proven by the defendant by a preponderance of the evidence—regardless of when the claim is presented.
From that conclusion, it follows that the standard of review is also the same. Giving “almost total deference to a trial judge’s determination of the historical facts supported by the record”47 in the habeas context is essentially the same as Jackson v. Virginia’s requirement that the evidence be viewed “in the light most favorable to the prosecution” when the findings are adverse to the defendant.48 Given the same burden, the standard of deference will also be the same, whether the Court conducts a sufficiency review of a mental retardation claim decided at trial or a legal review of a trial court’s recommendation on habeas corpus.
D. The mental retardation claim
We now turn to the merits of appellant’s claim. His claim has already been addressed on habeas corpus and found to be without merit. The trial court considered the entire trial record and the additional evidence presented in the habe-as proceedings. The trial court made its findings after Atkins was decided and after the Supreme Court had remanded the direct appeal in this case. The trial court’s consideration of the mental retardation claim and our review in the habeas proceedings satisfied the mandate of Atkins, and we believe that taking judicial notice of the habeas proceeding and its outcome satisfies the Supreme Court’s remand order in the present case. Because the burden of proof and the standard of review are the same on this issue on direct appeal and in habeas proceedings and because we can properly notice the habeas evidence on direct appeal, our conclusion on direct appeal is necessarily the same as our conclusion in the habeas proceedings.
In an abundance of caution, however, we have re-reviewed the evidence, and the result of that review is that our conclusion has not changed. While appellant supported his claim of mental retardation with the testimony of three psycholo[40]*40gists, his mother, his brother, his trial attorneys, two private investigators, four teachers, and a fellow death row inmate, the State controverted his claim with the testimony of a psychologist, five prison guards, a waitress, appellant’s former work supervisor, a former co-worker, and a police detective. The State also controverted the claim through cross-examination, pointing to school records indicating that appellant was not mentally retarded. There is also the videotaped interview, the hearing on the defendant’s pro se motion to remove his attorneys, and of course, the circumstances of the crime itself. While there was significant evidence in favor of a finding of mental retardation, there was also significant evidence against such a finding. The trial judge, who presided over the trial and the habeas proceedings, was in the best position to evaluate the conflicting evidence. Her findings, which we have judicially noticed in the current direct appeal, deserve great deference. Because the trial court’s conclusion that appellant is not mentally retarded is supported by the record, we should and do defer to that conclusion.49
The judgment of the trial court is affirmed.
PRICE, J., filed a concurring opinion in which COCHRAN, J., joined.
JOHNSON, J., filed a dissenting opinion in which HOLCOMB, J., joined.
HOLCOMB, J., filed a dissenting opinion.