OPINION
delivered the opinion of the Court
in which MEYERS, PRICE, WOMACK, JOHNSON, KEASLER and COCHRAN, JJ., joined.
In May 1998, applicant was convicted of capital murder and sentenced to death. *589He claims in a successive habeas corpus application filed two days before his scheduled execution that this Court should remand this case to the trial court a second time to consider the merits of applicant’s previously rejected claim that his execution is barred under Atkins v. Virginia1 because he is mentally retarded. We dismiss applicant’s successive habeas corpus application because this pleading shows that a rational finder of fact could find that applicant is not mentally retarded.
In Atkins v. Virginia, 536 U.S. at 321, 122 S.Ct. 2242, the United States Supreme Court decided that it violates the Eighth Amendment for a state to execute a mentally retarded murderer. The Atkins decision left to the states “the task of developing appropriate ways to enforce th[is] constitutional restriction upon its execution of sentences.” See Atkins, 536 U.S. at 317, 122 S.Ct. 2242 (internal quotes omitted). Because our Legislature has not enacted legislation to carry out the Atkins mandate, we continue to follow an American Association on Mental Retardation (AAMR) definition of mental retardation, adopted by this Court in Ex parte Briseno, for Atkins claims presented in Texas death-penalty cases. See Ex parte Briseno, 135 S.W.3d 1, 5-8 (Tex.Cr.App.2004). This AAMR definition defines mental retardation as a disability characterized by: (1) “significantly subaverage” general intellectual functioning, which is usually evidenced by an IQ “of about 70” or below,2 (2) accompanied by “related” limitations in adaptive functioning,3 (3) the onset of which occurs prior to the age of 18. See id,4 Under Briseno, some “other eviden-tiary factors which factfinders in the criminal trial context might also focus upon in weighing evidence as indicative of mental retardation” include:
Did those who knew the person best during the developmental stage — his family, friends, teachers, employers, authorities — think he was mentally retarded at that time, and, if so, act in accordance with that determination?
Has the person formulated plans and carried them through or is his conduct impulsive?
Does his conduct show leadership or does it show that he is led around by others?
Is his conduct in response to external stimuli rational and appropriate, regardless of whether it is socially acceptable?
Does he respond coherently, rationally, and on point to oral or written questions or do his responses wander from subject to subject?
Can the person hide facts or he effectively in his own or others’ interests?
Putting aside any heinousness or gruesomeness surrounding the capital offense, did the commission of that offense *590require forethought, planning, and complex execution of purpose?
See Briseno, 135 S.W.3d at 8.
The evidence from applicant’s 1998 trial shows that in the spring of 1997 applicant abducted a nine-year-old boy and his eleven-year-old sister from their home in the middle of the night.5 Applicant engaged in “sexual activity” with the girl before abducting her and her brother.6 Applicant drove the children to a remotely located cemetery where he severely beat and choked the boy until the boy lost consciousness. Applicant left the unconscious and seriously injured boy at the cemetery.7 Applicant drove the girl to another secluded location where he cut her throat “almost all the way around” with a knife, resulting in her death. The girl had also been severely beaten.
Applicant became a suspect in the children’s disappearance soon after they were reported missing. Several police officers questioned applicant at different times while the police were searching for the children. Applicant initially denied any involvement in the children’s disappearance. When the police told applicant that the boy had been found alive, applicant told the police, “You will not find [the girl] alive. I cut her throat.” Applicant led the police to the dead girl and asked them “if this was going to help him showing [the police] where the body is?” Applicant gave the police a signed, type-written statement in which applicant claimed that he accidentally cut the girl’s throat when she suddenly “jerked real hard” into .the knife that he was holding against her throat.8
Applicant testified at trial that he did not kill the girl. He testified that he and the boy got out of the car at the cemetery and that he asked the boy if his mother was seeing someone else. He claimed that he “popped” the boy in the head “pretty hard” with the palm of his hand about three times. The boy was knocked unconscious when he fell back and hit his head on a fence post. Applicant testified that he became scared and that he and the girl left the unconscious boy at the cemetery and drove to applicant’s house where his cousin was because his cousin “would know what to do.” The cousin and the girl left together in applicant’s car without applicant. The next day the cousin told applicant that he thought the girl was dead. Applicant went to where his cousin said the girl was and saw that her throat had been cut. Applicant’s cousin was dead at the time of applicant’s trial.
Applicant also denied telling the police, “I cut her throat,” when he learned that the boy was alive. Applicant claimed that he told the police, “Her throat’s been cut.” Applicant also claimed that his signed, type-written statements did not accurately reflect what he told the police. Applicant’s alleged deficiency in reading skills was a component of this claim, apparently to refute some police testimony that applicant *591read the type-written statements before signing them.9
Applicant testified on cross-examination that “I’m not saying that I can’t read, I’m just saying there are words I can’t read” when the State was cross-examining applicant on the number of books that applicant “looked at” while he was in jail.
Q. [STATE]: Okay. How many books have you looked at in our jail, would you estimate?
A. [APPLICANT]: I’m not — I’m not saying that I can’t read, I’m just saying there are words I can’t read.
Q. Would it surprise you to know, [applicant], that you have looked at over a hundred books in our book cart?
A. No, I haven’t looked at over a hundred books.
Q. How many?
A. I don’t know.
Q. Twenty?
A. (No response.)
Q. Fifty? I’m asking?
A. Look, I — I asked for books so I could see if I could find any I can read.
Q. It’s just real convenient that you can’t read, see, because you can sit and look at our books in our jail for a year, but you try to sale [sic] to this jury that you can’t read, so you can avoid these confessions; wasn’t that the point?
A. No, sir.
At the punishment phase of applicant’s pre-Atkins 1998 trial, applicant presented expert testimony that he is mentally retarded as mitigating evidence in support of a life sentence.10 A defense psychologist (Landrum), who did not personally interview or test applicant, testified that applicant “is, and always has been, always will be, a mentally retarded person.” Land-rum based this opinion on a review of various materials, including portions of a report of a non-testifying defense psychologist (Pita).
Pita’s report indicates that the defense requested that Pita personally evaluate applicant “to determine whether [applicant] meets the criteria for mental retardation, mental illness and if he is competent to stand trial.” Pita’s report concluded that applicant is not mentally retarded. Pita’s testing of applicant showed that applicant had a full scale IQ score of 70 on the Wechsler Adult Intelligence Scale-Third Edition (WAIS-III).11 Based on the Wide Range Achievement Test-3 (WRAT-III), Pita’s report indicates that applicant was functioning at a first-grade level in spelling and at a second-grade level in reading and arithmetic. Pita used the Street Survival Skills Questionnaire (SSSQ) during an interview with applicant to assess applicant’s adaptive skills. Pita’s report indicates that the SSSQ “was constructed for a developmentally disabled population” and is used to “evaluate an individual’s functional knowledge, skills important for independent living in the community and relevant adaptive skills.”12 Pita’s report states that *592applicant’s overall score of 95 on the SSSQ “indicates that [applicant’s] knowledge of adaptive behaviors which facilitate living and working in the community is within the normal range when compared to the normative group.” Pita’s report concludes that applicant is not mentally retarded because “applicant is capable of functioning independently in the community” although “testing and [applicant’s] history of formal education strongly suggests the presence of learning disabilities.”
Pita’s report also assessed applicant’s competency to stand trial using the “Competence Assessment for Standing Tidal for Defendants with Mental Retardation (CAST*MR).” Pita’s report indicates that applicant’s perfect score (50 out of 50) on this test was higher than a mentally retarded person would be expected to score. Pita’s report states:
Cast*MR norms indicate that approximately 99% of defendants who are not mentally retarded will obtain a score of at least 39 out of a possible 50, corresponding to 78% correct. Moreover, a mean score of 74% correct was obtained for a sample of defendants who were mentally retarded, but found or presumed to be competent to stand trial. Finally, a mean score of 54% correct was obtained for a sample of defendants who were mentally retarded, but found or presumed to be incompetent to stand trial
[Applicant’s] score of 100% (i.e. 50 out of 50) is consistent with those scores obtained for defendants who are not mentally retarded and found to be competent to stand trial and well above the score for those who were found to be incompetent to stand trial. [Applicant] appears to be capable of understanding basic legal concepts and proceedings and of assisting his attorneys in preparing a defense. However, information should be presented to [applicant] in simple, concrete terms to ensure his understanding.
Landrum testified at applicant’s 1998 trial about the “types of things” that he reviewed (including Pita’s report) in concluding that applicant is mentally retarded.
Q. [DEFENSE]: What types of things have you had an opportunity to review?
A. [LANDRUM]: Well, I reviewed a recent psychological evaluation by Dr. Pita, which included IQ testing. I asked for and received a variety of data, including offense reports, autopsy reports. I reviewed a number of pictures. I looked at jail records. I studied pretty closely the early part of [applicant’s] testimony in this case, and then ran out of time, and kinda skimmed over the last parts of that. I saw the statements, three different statements, that [applicant] made. And I interviewed his grandmother for information about conduct disorder.
Q. You relied on a report from Dr. Pita; is that correct?
A. I used that information in my evaluation, yes.
Q. What else have you relied on in addition to the report?
*593A. Well, certainly information from his grandmother. There was a sample of the [applicant’s] handwriting which was highly consistent with the IQ testing, and educational level assessed by Dr. Pita.
Q. Dr. Landrum, you talked about the data that you have looked at in this particular ease?
A. Yes.
Q. And as part of that data was there a report from a mental health expert?
A. Yes.
Q. And is it of the type that’s reasonably relied on by experts in your particular field?
A. Certainly.
Q. And based on all the data that you have looked at and used, do you have an opinion or determination as to whether [applicant] is mentally retarded?
A. [Applicant] is, and always has been, always will be, a mentally retarded person.
On cross-examination, Landrum testified that it was applicant’s lawyer’s decision for him not to personally evaluate applicant. The State also cross-examined Landrum on the SSSQ that Pita used to assess applicant’s adaptive skills. Landrum made no claim that the SSSQ was an invalid instrument or rating scale for assessing adaptive skills in a mental-retardation determination.
Q. [STATE]: Dr. Landrum, I want to speak with you for just a few minutes about mental retardation coupled with adaptive behavior. I know you’re familiar with that?
A. [LANDRUM]: Yes.
Q. Now adaptive behavior has a very direct and distinct influence on whether or not a person is regarded as mentally retarded; is that correct?
A. It’s part of the criteria that you use.
Q. Okay. And do you have a copy of this report that we have been referring to?
A. Yes.
Q. Okay, thank you. Now would you look on page 4?
A. (Complying)
Q. Down to the part where it talks about adaptive behavior, do you see that on the report?
A. The listing?
Q. Yes, sir.
A. Yes.
Q. And below the listing, this is called Street Survival Scales Questionnaire. I guess that’s a test that’s given?
A. It’s a rating scale, yes.
Well, this scale, I will kinda paraphrase it or read it precisely. “SSSQ was constructed for a developmentally disabled population.”
Q. Okay, now what does that mean?
A. Developmentally in a variety of ways.
Q. All right. And what was [applicant’s] score on that, do you recall?
A. [Applicant’s] overall score was 95.
Q. And if you will look down — -“[Applicant’s] overall score SSQ equals 95 — ”, and read what follows there?
A. “Indicates that [applicant’s] knowledge of adaptive behaviors which facilitate living and working in the community is within the normal range when compared to the normative group.” That is the disabled persons.
The State also cross-examined Landrum on the portion of Pita’s report assessing applicant’s competency to stand trial using the CAST*MR. Landrum acknowledged *594that applicant’s perfect score (50 out of 50) on this test was higher than a mentally-retarded person would be expected to score. Landrum made no claim that such a high score on the CAST* MR would have no bearing on whether a person is mentally retarded.
Q. [STATE]: Okay, we’ll move down to that. If you will go down to the second paragraph beneath the raw scores and percentage of correct, and read that to the jury?
A. [LANDRUM]: “Cast MR norms indicate that approximately 99 percent of defendants who are not mentally retarded will obtain a score of at least 39 out of a possible 50, corresponding to 78 percent correct.”
Q. All right. So a mentally retarded person would score 34 out of 50, is that a fair paraphrase of that?
A. More or less.
Q. And if you will, then go down to the next paragraph and read that to—
A. —Oh, I’m sorry, back up. You said 34.
Q. 39, I’m sorry.
A. The average is 54 percent correct, if — you didn’t mix your words there.
* * *
Q. If you’ll now go down to the next paragraph and read that paragraph to this jury?
A. “[Applicant’s] score of 100 percent, i.e., 50 out of 50, is consistent with these scores obtained for defendants who are not mentally retarded and found to be competent to stand trial and well above the score for those who were found to be incompetent to stand trial.”
On re-direct examination, Landrum more specifically described the basis upon which he concluded that applicant is mentally retarded.
Q. [DEFENSE]: And was this [Pita] report, Dr. Landrum, the only thing you looked at in making your opinion, coming to your conclusion that [applicant] is mentally retarded?
A. [LANDRUM]: Oh, heavens, no. No. Based on a lot more than this.
Q. What else was it based on?
A. Well, certainly you have his handwritten statement [to the police],13 it’s easy to recognize as grossly deficient in thinking. And presenting ideas and spelling and practice and handwriting even. His general lifestyle, regardless of this showing he’s normal or average, his lifestyle is believably one of grossly impaired ability to function; living with his grandmother, doesn’t have a car, doesn’t have a drivers license, can’t keep a job. Has had job training so he has some skills but still can’t work.
On recross-examination, Landrum made no claim that the methodology that Pita used to assess applicant’s adaptive skills was questionable. Rather, he just seemed to disagree with Pita’s opinion that applicant is not mentally retarded.
Q. [STATE]: All of the data that you looked at, it does include this report from Dr. Pita; is that right?
A. [LANDRUM]: I looked — I considered the IQ scoring, and I would have done that if it hadn’t already been done.
Q. And you just discounted the rest of this report from another clinical psychologist who had personally and actually performed the test?
A. For the reason that he saw the man in the first place, pretty much so, yes.
*595Q. But, you did take — you didn’t discount some of the findings, but you did discount others, is that what you’re telling this jury?
A. I used almost exclusively the IQ rating, mainly to determine whether I needed to do it.
Q. Okay. But what I’m asking you is, you said okay, I’m going to look at this report, I’m going to read this report, I’m never going to talk to this defendant personally and form an opinion personally that I can relate to this jury, I am going to take a part of this that I do like and I’m going to discount the rest of it, is that right?
A. No.
Q. Three times in this report Dr. Pita, who did talk to this defendant, who did sit down with him, who did administer the different tests, the clinical interview, the Wechsler Adult Intelligence Scale, the Wide Range Achievement Test, the Competency Assessment For Standard Trial For Defendants With Mental Retardation, Adaptive Behaviors, Street Survival Skills questionnaire, the Rorschach, and Thematic Apperception Test, he did all of these tests. They were done under his direction. Did a historical background on him. Three times in Dr. Pita’s report, he says that this defendant is not mentally retarded?
A. That’s his opinion, yes, sir.
The State presented rebuttal testimony of a psychiatrist (Gripon) who also did not question the methodology that Pita used in assessing whether applicant is mentally retarded. Gripon testified that Pita “found [applicant] to be of borderline intelligence, and that’s what he is is borderline intelligence [sic].”
This Court affirmed applicant’s conviction and sentence on direct appeal in June 2000, and denied relief on his initial habeas corpus application in September 2000. The Supreme Court decided Atkins in June 2002. In April 2003, applicant filed a successive habeas corpus application that presented an Atkins claim. Finding that applicant’s Atkins claim was not available to applicant when he filed his initial habeas corpus application in September 1999,14 we remanded the case to the district court in May 2008 for an evidentiary hearing and to make fact-findings on applicant’s Atkins claim. In September 2004, the district court held a two-day hearing, during which applicant was represented by court-appointed counsel.
At this 2004 writ hearing, applicant’s court-appointed counsel presented the testimony of a clinical psychologist (Schmitt), whom the district court had appointed to personally assess applicant for mental retardation. Schmitt concluded that applicant is mentally retarded under the following definition of mental retardation.
Q. [DEFENSE]: Okay. And can you give us an idea of the definition of mental retardation which you used in preparing this report, the—
A. [SCHMITT]: Basically that a person should have significantly subaverage intelligence as measured by a standardized IQ test. Normally a 70 or below would be the cutoff point or the standard. Secondly, that a person would show serious impairment in at least 2 of 10 categories of adaptive behavior.15 *596And third, that the onset would be before age 18 or during the developmental years.
Schmitt testified that he reviewed numerous materials including Pita’s 1997 report. Schmitt used the WAIS-III IQ test (the same test that Pita used) to measure applicant’s IQ. This time applicant’s Full Scale IQ score on the WAIS-III was 68 which was an adjusted score for what Schmitt described as the “Flynn Effect.”
Q. [THE COURT]: All right. Now, I want you to go into what I understand is called the Flynn effect. My understanding of that is that over the years that the standard and norm for a person’s intelligence quotient has continued to raise (sic) three points per year.
A. [SCHMITT]: Actually that would be three-tenths of a point per year, or one point every three years.
And basically what that effect is saying is that education is getting better, nutrition is getting better, for a variety of reasons, and, also, a lot of items in the norms are — become dated, that people just generally show a slight improvement over time, and for that reason we need to be somewhat careful in using a dated or an older IQ test, because at that three-tenths per year, if we’re looking at giving a person a test based on 10-year-old norms, then their IQ score might be three points higher than it would have been had the person been tested when those norms were current 10 years previously. The norms for this test were developed in 2002 and, therefore, there’s a 2-year lag. Applying Flynn’s criteria of three-tenths of a point per year, six-tenths of a point might be the average expected improvement.
[THE COURT]: So that’s something you took into account when you looked at this, is that right?
A. Yes, sir.
Schmitt testified on cross-examination that “there’s a possible 5 point deviation on either side” of applicant’s Full Scale IQ score of 68 for a range of 63 to 73.
Q. [STATE]: No, I’m talking about,— maybe I didn’t ask this right. That if you say 68, there’s a possible 5 point deviation on either side of this?
A. [SCHMITT]: Yes, a range of competence or competence interval, yes.
Q. So you’re saying that your 68 shows 63 to 73, and given that possibility?
A. Yes.
Schmitt further testified that applicant scored a 63 on the verbal portion of the WAIS-III and an 80 on the performance portion. Schmitt considered this difference to be significant and indicative that applicant’s “perception motor skills are considerably better than his verbal skills.”
[COURT]: Is there a big significance between a 63 and the 80?
A. [SCHMITT]: Yes, sir, that would be considered significant.
[COURT]: Tell me about that.
A. Well, his perception motor skills are considerably better than his verbal skills, and that was borne out in some of the other testing that was done. When it comes to doing things with his hands, [applicant] is — is pretty good.
[COURT]: In regular life, before, I understand, he was a cook?
A. Yes, uh-huh.
[COURT]: He was able to do that well?
A. Yes, sir.
Schmitt used the Scales of Independent Behavior Revised (SIB-R) test, which was done by interview with applicant’s grandmother (Ruby Woods), to assess applicant’s adaptive skills. He testified, consistently with his report, that the results of this test indicated that applicant “is func*597tioning at or near his age level of 38 years in every category and in every scale except a scale measuring Money and Value.” Schmitt’s report states:
[Applicant’s] adaptive functioning is another matter. The Scales of Independent Behavior-Revised Test was administered using his Grandmother, Ruby Woods, as the respondent. This test was administered by interview with Ruby on September 25, 2004. She was [applicant’s] primary caretaker from the time he was approximately six months of age until approximately 1996 when he began residing with the mother of the children who were the victims of his offense. Despite [applicant’s] continuous incarceration since 1997, Ruby has maintained regular contact with [applicant] through visits during the term of his incarceration. She is a person who [applicant] claims and whom she agrees knows him best. On most items comprising the test, Ruby was inclined to give [applicant] the benefit of the doubt, even to the point of assuming that he might know how to do certain things that she has not actually observed him do. Clarifications were asked to determine whether her responses were based on direct observations or on what she believed but had not observed. Ruby was judged to be a reliable informant as to what she had observed although her responses tended to give [applicant] the benefit of any doubt and thus tended to make him appear as functional as possible.
Basically the test purports to measure functional independence and adaptive behavior in an individual’s ability to effectively meet social and community expectations for establishing personal independence, maintaining physical needs, conforming to social norms, and sustaining interpersonal relationships. The test is normed for children, adolescents, and adults. Individual items define various scales which measure areas of functioning involving Gross Motor, Fine Motor, Social Interaction, Language Comprehension, Language Expression, Eating, Toileting, Dressing, Self-Care, Domestic Skills, Time and Punctuality, Money and Value, Work Skills, and Home-Community Skills. These scales are then incorporated into categories of Motor Skills, Social Interaction and Communication Skills, Personal Living Skills, and Community Living Skills. There are additional scales which measure Broad Independence, Maladaptive Behavior, and Support. The results indicate that [applicant] is functioning at or near his age level of 38 years in every category and in every scale except a scale measuring Money and Value 16 where he had an age equivalent of age 16-6.17 He also functioned at or above his chronological age in a category that measures Broad Independence. In the measure of Maladaptive Behavior, he scored marginally serious on only one of four measures, that being the Internalized Maladaptive Index which consists of problem behaviors that are inwardly directed. Primarily, difficulties being around others and paying attention contribute to most of this index, which was in the Marginally Serious range. A scale measuring his need for Support from other people was in a range indicating that such support is needed only on an infrequent basis.
*598Schmitt testified that applicant is not a functional reader and is not capable of reading anything more than a comic book, but that he might check out more sophisticated books from a library to look at the pictures or to get other people to explain the books to him. Schmitt agreed with the portion of Pita’s 1997 report finding that applicant functions at a first-grade level in spelling and at a second-grade level in reading and arithmetic.
Schmitt also testified that applicant’s school records indicate that he was a “slow learner” with problems in “memory and repetitive academic skills,” all of which contributed to him leaving school in the eighth grade.
Q. [DEFENSE]: Now, these same school records, though, do — do indicate that he was a slow learner.
A. [SCHMITT]: Yes.
Q. That he had problems with memory and repetitive academic skills.
A. Yes.
Q. That, as a result, he had poor grades.
A. Yes.
Q. He also had problems of an emotional immaturity nature, and also just getting along with others while in school.
A. Yes.
Q. All somewhat interrelated, is that correct?
A. That is correct.
Q. And all of which contributed to his leaving school while he was in the eighth grade, before completing it, is that correct?
A. Yes, that is correct.
Schmitt also testified that applicant’s job experience as a short-order cook at the Round the Clock Grill for a few weeks “would be in keeping with a mildly retarded person.” According to Schmitt, applicant’s duties were not of a “particularly taxing intellectual nature” but were something of a “repetitive nature” such as responding appropriately if someone said, “Give me two fried eggs and some bacon.” Schmitt considered it significant that applicant’s duties as a short-order cook did not involve dealing with money since he would not usually be able to figure the change for a $10.50 purchase with a $20 bill.
Schmitt concluded that applicant had deficits in two areas (functional academic skills and work)18 “sufficient for applicant to be classified as mentally retarded.”
Q. [DEFENSE]: And I may have covered this but I just want to be absolutely sure. There are ten areas that we look at for deficits. There have to be at least two in order for him, assuming that he has got an IQ of under 70 and has had an onset under or before age 18.
A. [SCHMITT]: Yes, sir.
Q. The two areas in which you found a deficit sufficient for him to be classified as [a] mentally retarded person were what?
A. His functional academic skills. Those have been significantly subaver-age, poor, marginally failing, basically since his entry in public education, and extending every year that he was in public education, until he withdrew.
Q. Okay. So we have academic skills, and the other would be?
A. Work.
Q. Because he’s been unable to hold a steady job for any length of time or perform at more than a basic level?
A. Yes, sir. And he had 14 years between the age of 18, when he began adulthood, until the age of 32 when he was arrested, to establish something for himself that was stable and consistent in the world of work. He was never able to do that.
*599Schmitt also testified that he was aware that Landrum had made a very similar determination at applicant’s 1998 trial and that there was nothing in Landrum’s testimony that he would “take issue at with regards to the results of [Schmitt’s] examination and evaluations.”
Q. [DEFENSE]: You were aware that Dr. Landrum made a determination very similar to yours of [applicant] being a mentally retarded person?
A. [SCHMITT]: Yes, sir.
Q. Is there anything, in reviewing his testimony, that you would take issue at with regards to the results of your examination and evaluations?
A. No, sir, there’s not.
Schmitt criticized Pita’s 1997 opinion that applicant is not mentally retarded as having been based on Pita using the “half-full” approach in assessing applicant’s adaptive skills which Schmitt claimed was contrary to the Supreme Court’s decision in Atkins and “accepted methodology” which, according to Schmitt, requires the “half-empty” approach. According to Schmitt, the “half-full” approach looks at “positive attributes” while the “half-empty” approach looks at deficits.
Q. [DEFENSE]: Okay. One of the things I want to look at, and it was noted in, I believe, both Briseno and in Atkins versus Virginia, is that there’s basically two ways to score adaptive functioning. One of them is the half full, one of them is the — you know what I’m talking about?
A. [SCHMITT]: Yes.
Q. The other one is a half empty. In other words, in one respect, experts look at the deficits in adaptive functioning—
A. Uh-Huh.
Q. —in other words, the half empty?
A. Yes.
Q. And other experts will look only at the positive functioning, in other words, the half full?
A. Yes.
Q. When we’re looking at your report, I’m assuming that you are looking at the deficits as opposed to the positive functioning for purpose of scoring, or am I incorrect there?
A. No, that is correct.
Q. Okay. Now, the case law that we have, which is scant in this area, indicates that if you take these two different opposing half full versus half empty, you will have experts looking at the same data but achieving completely different results.
A. Yes.
Q. Now, that given — given the fact, when we look at the three areas of — for the definition of mental retardation of IQ score, obviously that is testable and that’s not going to be subjected to a subjective analysis.
A. No, it isn’t.
Q. When we look at the onset of the events giving rise to the symptoms of mental retardation before age 18, once again, that is something that’s going to be very objective and not subjective.
A. Yes.
Q. But when we look at adaptive functioning, this is a completely and totally subjective analysis, is that correct?
A. Well, and it — it is to a degree. Now, this is an objective test, that it, it is a scored and normed test.
Q. That’s what I wanted to find out about.
A. Yes, uh-huh.
Q. So there are objective standards, but then there’s still a subjective approach to analysis, in other words, a half full versus half empty, even though you have concrete data?
A. Yes.
Q. Okay. So if a person were to meet the two deficits that the Supreme Court *600indicated in Atkins, then we would be looking at the half empty kind of approach, in other words, look at the deficits rather than we’re looking at positive approach or positive adaptive ability to determine if the person’s mentally retarded, is that right?
A. That is correct.
Q. In other words, if we were to take the opposite approach and say, “Well, he may have two deficits but he has these eight positives, those two deficits should be averaged out or wiped out,” we would take the half full approach, is that correct?
A. Yes.
Q. Which is not the one used by the Supreme Court in Atkins versus Virginia.
A. Yes.19
Q. Okay. I’ve been having you look at Dr. Pita’s report and discuss various things that were in it. You’ve read the entire report, is that correct?
A. I have.
Q. And you’ve made some conclusions with respect to your evaluation and testing, and that [sic] was performed in 1997. What was that?
A. Well, basically that there’s really very little difference in the results of his IQ testing and mine. I think he tested [applicant] perhaps at a couple of points higher, at IQ 70. My full scale IQ was 68. That would be very close. I don’t think we disagree as to whether or not [applicant’s] intellectual difficulties arose during the developmental period. His report seems to make clear that they did. I think the only difference is how he construes [applicant’s] adaptive behavior and how I construe [applicant’s] adaptive behavior.20
Schmitt testified on cross-examination that the SIB-R test that he used and the SSSQ that Pita used yielded very similar results.
Q. [STATE]: So the tests that you gave him which is normed for children, adolescents and adults, shows that he is within the normal range in every category but one,—
A. [SCHMITT]: Yes.
* ⅜ ⅞
Q. Very similar to Dr. Pita’s finding, is that right?
A. Yes.
Q. The only difference, I believe, is that the SSSQ that Dr. Pita’s (sic) gave found that [applicant] fell within the normal range for the money and value. Do you remember that?
A. Yes.
Schmitt also testified on cross-examination that he knew that applicant had checked out over 115 books from the prison library while he was on death row and that some of these books were “complex novels” with no pictures in them. Schmitt *601testified that he did not know whether applicant read these books.
Q. [STATE]: Did you — did you discuss his reading of these books?
A. [SCHMITT]: I asked him what he did with his time on death row, and one of the things he told me he did with his time was go to the library and check out books that mostly were of western themes.
Q. You don’t deny that he read them, do you?
A. I don’t know whether he read them or not.
With respect to applicant’s work history (an area in which Schmitt found a deficit), Schmitt testified on cross-examination that applicant’s job as a short-order cook took “some skill, some level of organization and thinking” and that applicant was considered a good and reliable employee. He also testified that his knowledge of applicant’s job history was not complete and that “being at a job for a short period of time doesn’t mean that a person is mentally retarded.”21
Q. [STATE]: And we’ve talked about the short-order cook and — and that type of thing. He had a number of other jobs, too, is that your understanding? A. [SCHMITT]: Yes.
Q. Do you — do you know what the other jobs were?
A. He was a cook for the Granbury Independent School District, according to some records. He had worked in a Waffle House, according to what he told me. There were indications in Dr. Pita’s report, I believe, that he had done some type of construction work at times in the past. But primarily, it’s my understanding, and based on the records I reviewed, he’s worked in food preparation.
Q. I — I don’t have a full knowledge and list of every place that [applicant] worked and how long he worked there from the time he began working until the time that he didn’t work anymore. Do you have that?
A. No, sir, I don’t.
Q. Okay. So we’re dealing on [sic] something with some facts here that are not complete,-
A. Yes.
Q. —is that fair to say?
A. It is.
Q. Now, having — being at a job for a short period of time doesn’t mean that a person is mentally retarded, does it?
A. No.
Q. That is not — you couldn’t make that leap anywhere, could you?
A. Well, not just based on that information alone, no, sir.
With respect to applicant’s functional academic skills (the other area in which Schmitt found a deficit), Schmitt testified on cross-examination that he did not talk to any of applicant’s teachers and principals. Schmitt also testified that none of applicant’s school records indicated that applicant was ever considered to be mentally retarded and that this is an important consideration in determining whether a person is mentally retarded.
Q. [STATE]: Now, do you know from your experience and — and the way things were done in these — in these schools back then, that people were recognized to be mentally retarded?
A. [SCHMITT]: Yes.
Q. And — [applicant] wasn’t, was he?
A. No.
*602Q. If he had been diagnosed as mentally retarded that probably — that would have been noted a lot, wouldn’t it?
A. Well, it would have been — he probably would have been in a totally different class, for one thing.
Q. And what would that class have been?
A. A class for mentally retarded people.
Q. Okay. But he wasn’t in that class, was he?
A. No.
Q. He was not classified as mentally retarded, was he?
A. No, he wasn’t.
Q. And when we first started talking, you — you agreed with me that it would be important evidence for this court to hear the people who were with him at that time, his educators, his teachers. You still believe that to be true?
A. Yes.
Q. And why is that?
A. Because they would have something to contribute about his behavior, well, his adaptive behaviors even during those years. You know, they’re not going to be able to tell you precisely whether he’s mentally retarded or not because teachers don’t give IQ tests, but they can certainly tell you whether he was academically on the same footing as other children in the class, or whether he was behind, and something about his level of motivation, whether he gave his best effort or whether he tended to give up or rushed his work or whatever.
Q. They can tell you whether or not a person had been diagnosed as mentally retarded and were treated as such in the school, couldn’t they?
A. In the school, yes.
Schmitt also testified on cross-examination that somebody else could look “at the adaptive behavior test and say that [applicant] is not mentally retarded.”
Q. [STATE]: But what I’m saying is, somebody else could look at the very same — these very same tests and say [applicant] is not mentally retarded.
A. [SCHMITT]: Well, they’re not going to look at the IQ test and say that [applicant] is not mentally retarded. They may look at the adaptive behavior test and say that [applicant] is not mentally retarded.
Applicant’s school records indicate that applicant was in special education classes for most of the time that he was in school and that his grades were very poor. Applicant’s school records also indicate that applicant had a Full Scale IQ score of 80 when he took the Wechsler Intelligence Scale for Children (WISC) in 1972 in the first grade. Applicant had a Full Scale IQ score of 78 when he took the same test in 1975 in the fourth grade.22
Several of applicant’s educators testified at the 2004 writ hearing that they considered applicant to be learning-disabled and a “slow learner,” but not mentally retarded. They attributed much of applicant’s poor academic performance to applicant’s lack of effort.23
*603An educator of the boy whom applicant severely beat and choked into unconsciousness testified about her interactions with applicant at the boy’s school while applicant was living with the boy’s mother. This educator testified that she had dealt with hundreds of mentally retarded people and that applicant seemed “perfectly normal.”
The State also presented the testimony of a prison psychologist (Gilhousen) who conducted a “mental status” review of applicant when he arrived on death row. Gilhousen testified that he administered the short-form Wechsler Adult Intelligence Scale Revised (WAIS-R) IQ test to applicant. This short-form test consisted of a vocabulary subtest and a block design sub-test from which Gilhousen claimed that he could arrive at an estimate of applicant’s Full Scale IQ.24 Gilhousen testified that applicant’s estimated Full Scale IQ was 83 based on his performance on the two sub-tests. Gilhousen testified that the short-form WAIS-R is “more than adequate” for ruling out mental retardation.
Q. [STATE]: What test did you do?
A. [GILHOUSEN]: I used the — one of the short forms of the Wechsler Adult Intelligence Scale Revised, where I administered the vocabulary subtest and the block design subtest. And from that, then, using published tables, I arrived at an estimate of what would’ve been his full scale IQ, had he taken the whole test.
Q. In your testing, did you find any indications that [applicant] suffered any degree of mental retardation?
A. No, I did not.
Q. The test that you used or the portions of the test that you used and the way you applied it, in your opinion, was that an effective measure of whether he had, for that purpose, mental retardation?
A. I think for screening purposes, to rule out mental retardation, I think it’s more than adequate. If I wanted to rule in mental retardation, if I was actually testing him to see if I thought he was mentally retarded and I was wanting to make that determination, well, then, I would probably apply the full test on him, but that wasn’t the purpose of my — of my encounter with him, was to rule out mental retardation. And I — I think I effectively did that.25
Gilhousen also testified that applicant had reported to him that applicant “had had no prior contact with the mental health profession, and no history of treatment.” Gilhousen testified that he did not see anything during his interview with applicant or in applicant’s prior history that would have changed his “screening opinion” that applicant is not mentally retarded.
Q. [STATE]: Did you go beyond the screening test to include other information obtained from him about his past history, whether he had ever used mental health or mental retardation services in the past and so forth?
A. [GILHOUSEN]: Yes.
Q. What did you find out or what did he tell you about that?
A. He indicated to me that there — he had had no prior contact with the mental health profession, and no history of treatment. He also revealed to me that at his trial, the punishment phase of his *604trial, that he was found to have borderline intellectual functioning.
Q. From the interview, did you get any indications, subtle things you might know to look for as a professional, or anything of that nature that would have changed your screening opinion that he did not have mental retardation?
A. No. I don’t think there was any indication at all. I think his communication skills were — were adequate. He— his mood was fíne, his affect was fíne. I just didn’t really see anything there that would indicate that, you know, this was a man that was mentally retarded that would require us to provide some sort of service for him.
On cross-examination, Gilhousen testified that in this day and age it is “pretty unusual for somebody to pop up all of a sudden at age 32 and they’re mentally retarded.”
Q. [DEFENSE]: All right. Okay. So even if the person wasn’t diagnosed as mentally retarded before age 18, if all of [the] factors of the definition are present, then he would meet the definition of a mentally retarded person, is that right?
A. [GILHOUSEN]: That’s possible. But —
Q. Okay.
A. —I think in this day and age, that’s — that would — that’s pretty unusual for somebody to pop up all of a sudden at age 32 and they’re mentally retarded. You know, I think that the school systems are pretty good, and obviously that he — the records that I have reviewed indicated that he was tested and — and repeatedly tested and offered special education services, and so they were on top of it, you know, at the time, and, you know, he wasn’t retarded then, and all of a sudden, you know, he’s retarded now, I don’t think so.
Gripon also testified at the 2004 writ hearing that he listened to all of the evidence presented at this hearing and that it was still his opinion that applicant is not mentally retarded. He testified that applicant “scored well enough on [the SSSQ by Pita] not to demonstrate substantial adaptive deficits.” He also testified that Schmitt’s adaptive skills test “pretty much correlated] with the findings of Dr. Pita,” neither of which identified “substantial deficits in [applicant], they didn’t in — six years ago and they don’t now.”
Gripon also disagreed with Schmitt’s opinion that applicant is deficient in work and functional academic skills. He testified that applicant “functioned well in the job of a short-order cook.” He also testified that merely because a person might have a number of jobs over a short period of time is not necessarily indicative of mental retardation (Gripon testified that he “examined a man once who was a college graduate that had 57 jobs in 20 years, and he certainly was not mentally retarded but he certainly could not sustain a job”). Gripon also testified that there were multiple reasons for applicant’s academic issues, “mental retardation not being one of them.” He testified that applicant’s trial testimony claiming that his dead cousin murdered the girl was coherent and indicated that applicant “understood the difficulty that he was in.”
Q. [STATE]: And your (sic) remember who he blamed for having committed the murder?
A [GRIPON]: It was a person who was deceased at that point in time, who had committed suicide, and I believe he was related to him, as I understand it, I think he was a cousin.
Q. Does that tell you anything about whether or not you — you can form an opinion as to whether or not [applicant] *605can hide facts or lie effectively in his own self-interest?
A. Well, that would indicate that he understood the difficulty that he was in and that he was offering an alternative explanation for that, placing the blame someplace else, and attempting, obviously, to divert the attention from himself, so it indicated that he could give at least a coherent explanation for that, even if it happened to be one that the jury didn’t believe. It was not something that was out of the realm of plausibility, but they didn’t buy it, but it was certainly coherent.
In February 2005, the district court signed detailed findings of fact and conclusions of law rejecting applicant’s Atkins claim.26 These findings and conclusions were based on the evidence presented by the parties at the 2004 writ hearing and the evidence presented at applicant’s 1998 trial, including Pita’s report. On April 27, 2005, this Court adopted the district court’s findings of fact and conclusions of law and denied relief on the merits of applicant’s Atkins claim.
Applicant subsequently filed a federal habeas corpus petition presenting the same Atkins claim. In July 2007, the Fifth Circuit Court of Appeals decided, among other things, that it could not conclude “that the state court applied Atkins in an objectively unreasonable manner” given “the substantial evidence presented [in the state court proceedings] that [applicant] is not mentally retarded.” See Woods v. Quarterman, 493 F.3d 580, 587 (5th Cir.2007) and at 585-87 (summarizing the evidence presented by the parties during the 2004 writ hearing in state court on the issue of applicant’s mental retardation).
Applicant’s execution date was set for October 23, 2008. Just two days before this scheduled execution, applicant filed this successive habeas corpus application presenting the same Atkins claim that this Court had rejected on the merits more than three years before on April 27, 2005. Applicant relies on some additional evidence that was not presented at the 2004 writ hearing or at his 1998 trial. We granted applicant a stay of execution and filed and set this case to address whether the merits of applicant’s Atkins claim can be considered again.
We do not understand applicant to claim that any of the additional evidence that he relies on in this proceeding is “new” for purposes of meeting the requirements of Article 11.071, § 5(a)(1).27 We, therefore, address whether this Court may consider the merits of applicant’s current successive habeas corpus application under Article 11.071, § 5(a)(3), Tex.Code CRim. Proc., which prohibits this Court from considering the merits of a successive habeas corpus application unless this application establishes “by clear and convincing evidence, but for a violation of the United States Constitution, no rational ju*606ror would have answered in the state’s favor one or more of the special issues that were submitted to the jury in the applicant’s trial.” In Ex parte Blue,28 we construed this to mean that an applicant, raising an Atkins claim for the first time in a successive habeas corpus application that he could have raised in an earlier one, must “demonstrate to this Court that there is evidence that could reasonably show, to a level of confidence by clear and convincing evidence, that no rational finder of fact would fail to find he is mentally retarded.”
The State claims that, since the issue of applicant’s mental retardation has previously been litigated with a rational finder of fact (the district court) having found that applicant is not mentally retarded and with this Court having adopted that finding, applicant’s current successive habeas corpus application should be held to an even higher standard than that set out in Blue under common-law principles of collateral estoppel and res judicata. Applicant argues that his successive habeas corpus application should not be held to this higher standard and that the “threshold question under Section 5 is whether all the facts presented by [applicant], taken as true, meet the applicable burden.” Applicant further argues that the “prior findings, and the State’s evidence, are not relevant to this determination.”29
We do not agree with any of these contentions. Article 11.071, § 5(a)(3), as construed in Blue, clearly controls the determination of whether applicant’s successive habeas corpus application may be considered on the merits. This, however, does not mean that the prior evidence and findings have no bearing on whether applicant’s successive habeas corpus application meets the requirements of Article 11.071, § 5(a)(3). We agree with the State that the prior evidence and findings “are not floating in space” with no bearing on the disposition of applicant’s current successive habeas corpus application. The prior evidence and findings are relevant to a determination of whether applicant’s current pleading meets the requirements of Article 11.071, § 5(a)(3), as construed in Blue. See Blue, 230 S.W.3d at 162-63 (this Court reviews adequacy of the pleading in determining whether it meets the requirements of Article 11.071, § 5(a)(3)).30
The issue then is whether, considering the prior evidence and findings, applicant’s additional evidence reasonably shows, by clear and convincing evidence, that no rational finder of fact would fail to find that he is mentally retarded. See Blue, 230 S.W.3d at 154, 163. Stated another way, the issue is whether a rational finder of fact could still find that applicant is not mentally retarded based on the prior evidence and the additional evidence set out in applicant’s successive habeas corpus application.31
Attached to applicant’s current successive habeas corpus application is the previously unconsidered affidavit of another psychologist (Seay) with another opinion to express. Seay states in this affidavit that, *607while she would have preferred “additional information, such as meeting with applicant,” the materials that she has reviewed are sufficient for her to reach an “initial conclusion” consistent “with a diagnosis of mental retardation.” Seay’s “initial” conclusion is based on materials previously considered by this Court and the district court except for previously unconsidered affidavits from applicant’s “family members, childhood friends and co-workers” and applicant’s previously unconsidered short-form IQ test score of 60 from 1980.32
Seay also criticizes the methodology used by some of the prior experts to measure applicant’s IQ. Seay claims that applicant’s “short-form” IQ tests (this would include applicant’s previously unconsidered short-form IQ test score of 60 from 1980) are invalid and some of the others were not adjusted for the Flynn Effect. Seay explains the Flynn Effect and its application to applicant’s IQ scores:
16. The Flynn Effect was discovered by James Flynn, an American professor who was the first to document the fact that IQ scores increase over time. (Footnote omitted). The Flynn Effect describes the phenomenon that different IQ tests using the same instrument will result in increased scores the farther in time the test is given from the time it was last “normed” for the general population — or, more generally, that the intelligence of the population increases over time.
26. The primary measure of intelligence given to [applicant] after he entered public school was the Wechsler Intelligence Scale for Children (WISC) administered in 1972 when he was 6 years old and in 1975 when he was ten years of age. On this test, he obtained a Full Scale IQ of 80 in 1972 and 78 in 1975. At the time, he was not diagnosed as having mental retardation. At that time, scores from the Wechsler scales were generally trusted, and it was the practice in the assessment reports to use the obtained Wechsler score for classification. This was reflected in the assessment reports and the 2004 testimony of teachers and [applicant’s] school principal.
27. Flynn’s research, of course, was not available in the mid-1970’s. In fact, the WISC test was last normed in 1947/48 — decades before it was used to test [applicant] in the 1970’s. To correct for the Flynn Effect, or IQ gains over time, [applicant’s] scores must be adjusted down at the rate of .30 points per year from the time the test was normed until the test was administered. (Footnote omitted). It is particularly important to adjust for the Flynn Effect when, as here, there is such a dramatic lapse of time between norming and administration.
According to Seay, “removing the scores resulting from short form tests, and adjusting the WISC scores for the Flynn Effect, valid, accurately scored results for [applicant’s] intellectual functioning” are: (1) 1972 WISC (1st grade), Full Scale IQ 72/73, (2) 1975 WISC (4th grade), Full Scale IQ 69/70,33 (3) 1997 WAIS-III (Pita), Full Scale IQ 70, and (4) 2004 WAIS-III (Schmitt), Full Scale IQ 68. Relying on several academic studies, the State claims *608that the “Flynn Effect, as applied to reduce IQ scores, is an unexamined, highly-criticized, and controversial scientific theory not properly used by clinicians for diagnosis.”
We find it unnecessary in this case to attempt to resolve any controversy over the application of the Flynn Effect to IQ scores. Even accepting Seay’s analysis of applicant’s IQ scores, and taking into account Schmitt’s testimony concerning a “possible 5 point deviation on either side,”34 applicant’s Full Scale IQ scores range anywhere from 63 to 78. Seay’s adjusted IQ scores for applicant, therefore, establish that applicant’s IQ may fall below 70. This, however, is consistent with the previously considered reports of Pita and Schmitt. Seay’s IQ analysis adds little to the previously considered evidence of applicant’s IQ scores. Even under Seay’s analysis, a rational finder of fact could find that applicant’s Full Scale IQ falls above 70.35 And, assuming that such a finding would not be rational, there still remains the issue of applicant’s adaptive skills on which previous expert opinions have differed.36
Seay addresses this by criticizing the methodology used by Pita and Schmitt in assessing applicant’s adaptive skills. Seay claims:
9. As stated by the American Association of Mental Retardation (AAMR) in 1992, “mental retardation refers to substantial limitations in present functioning. It is characterized by significantly subaverage intellectual functioning, existing concurrently with related limitations in two or more of the following applicable adaptive skill areas: communication, self-care, home living, social skills, community use, self-direction, health and safety, functional academics, leisure, and work. Mental retardation manifests before age 18.”
11. In 2002, AAMR updated its classification system and defined mental retardation as “a disability characterized by significant limitations in both intellectual functioning and in adaptive behavior as expressed in conceptual, social, and practical adaptive skills. This disability originates before age 18.”
*60939. In his 2004 evaluation, Dr. Schmitt utilized the Scales of Independent Behavior-Revised (SIB-R) to assess adaptive behavior. Mr. Woods’ grandmother, Ruby Woods, served as the only informant. Dr. Schmitt found [applicant] to have a limitation on a scale of Money and Value. The remaining scores were reportedly near his age level of 38. To conduct such an assessment on the basis of one informant— particularly someone who, as here, may herself have been compromised — is patently inadequate. In addition, Dr. Schmitt described behaviors [sic] of having difficulties in being around others and in paying attention. Dr. Schmitt went on to say that [applicant] met the criteria for the diagnosis of mental retardation. If he used these described deficits in social skills, it could have provided the necessary second criterion for the assessment of adaptive functioning limitations for the 1992 AAMR definition. Had he used the 2002 AAMR definition, the limitations in Money and Value would have been sufficient.
40. Dr. Pita used the [SSSQ] to assess adaptive behavior. On this test, the ex-aminee is required to point to the correct answer when given a choice of 4 pictures. [Applicant’s] Street Survival Quotient (SSQ) of 95 was reported to place his adaptive behaviors within the normal range. However, the SSSQ has been criticized for the lack of inclusion in the normative sample of adults who did not have mental retardation. This test was intended to be used along with other assessments to determine readiness of persons with developmental disabilities for community employment, so the difficulty level of the items was set to accommodate persons with developmental disabilities. At this level of difficulty, the test is easy for persons who can perform basic community work as those with mild mental retardation can. There were 400 persons with developmental disabilities included in the original normative sample. It was also given to three groups of adolescents with no developmental disability before it was published as a way to increase its usefulness with the adolescent population, though the test developers referred to this group as “average adults.” The tables in the manual used for conversion of raw scores to scaled scores and total scores to standard scores using the data from the adolescent administrations may have led to the erroneous belief that the SSSQ was normed for adults who did not have mental retardation and was therefore an adequate measure of adaptive behavior. (Footnote omitted) Thus, [applicant’s] “normal” score places his skills squarely within those who are, like him, persons with mental retardation.
We note that Landrum did not question Pita’s use of the SSSQ at applicant’s trial in 1998. On the contrary, Landrum testified that Pita’s report is “of the type that’s reasonably relied on by experts” in the field. At the 2004 writ hearing, Schmitt also did not seem to question Pita’s use of the SSSQ. Rather, Schmitt criticized Pita for taking the “half-full” approach in assessing applicant’s adaptive skills by looking at applicant’s positive attributes. Seay, on the other hand, criticizes Pita for even using the SSSQ. We believe that these differing approaches and these criticisms of psychologists by other psychologists are for the fact finder to resolve and that a rational finder of fact could rely on Pita’s SSSQ test results to find that applicant has no significant deficits in adaptive skills.37
*610It is also noteworthy that Pita’s SSSQ scale and Schmitt’s SIB-R scale to assess applicant’s adaptive skills yielded “very similar results.” Seay does not criticize Schmitt’s use of the SIB-R, claiming, instead, that Schmitt’s use of applicant’s grandmother as the only informant was “patently inadequate.” A rational finder of fact, however, could find that Schmitt’s use of applicant’s grandmother as the only informant was not “patently inadequate” since the evidence shows that she was applicant’s primary caretaker and knew him best. See Briseno, 135 S.W.3d at 8 (evidentiary factor which factfinder could consider as indicative of a person’s mental retardation is whether those who knew the person best during the developmental stage thought he was mentally retarded at the time).38
As far as Seay suggesting that Schmitt may have used the “wrong” definition of mental retardation (the AAMR 1992 definition instead of the AAMR 2002 definition),39 we note that the unobjected-to definition of mental retardation used at applicant’s 1998 trial and at the 2004 writ hearing is constitutionally adequate and not inconsistent with this Court’s decision in Briseno. See Briseno, 135 S.W.3d at 5-8. We further note that, contrary to what Seay seems to suggest in her affidavit, Schmitt did testify that applicant has significant adaptive deficits in two (not one) areas: work and functional academic skills. In any event, we have noted that a rational finder of fact could find that applicant has no significant deficits in adaptive skills.
Seay’s affidavit also relies on the previously unconsidered affidavits of “family members, childhood friends and co-workers of applicant” to “belie the test results obtained by Drs. Schmitt and Pita.” Seay’s affidavit states:
41. Furthermore, recent data gathered, including affidavits from numerous family members and friends recounting [applicant’s] behavior and characteristics during the developmental period, belie the test results obtained by Drs. Schmitt and Pita.
(Emphasis in original).
We note that, while these affidavits generally attest to applicant’s borderline intelligence and could be used as evidence that applicant has deficits in various adaptive skills, none of these affidavits indicate that any of the affiants thought that applicant was mentally retarded during his developmental period. See Briseno, 135 S.W.3d at 8 (evidence of mental retardation might include whether those who knew the person best during his developmental years thought at the time that he was mentally *611retarded). We also agree with the State that a rational finder of fact could find that the additional affidavits from applicant’s friends and family submitted for the first time just hours before applicant’s scheduled execution “should never be considered the type of compelling evidence” for overturning a previous determination that applicant is not mentally retarded. A rational finder of fact could find that these additional affidavits do not “belie the test results obtained by Drs. Schmitt and Pita.”
Seay also takes issue with any reliance on Pita’s use of the CAST*MR in assessing whether applicant is mentally retarded because this test is used only to assess whether mentally retarded adult criminal defendants are competent to stand trial and is not relevant to a determination of mental retardation. Seay states:
43. However, the CAST*MR is not a measure of intellectual or adaptive functioning. It is designed for use by forensic evaluators to assist in the evaluation of an adult defendant with mental retardation for competence to stand trial in connection with a current legal proceeding. Even the authors caution that, “Generally, IQ tests provide good discrimination between people who do and do not have mental retardation but less information for discriminating among different groups of people with mental retardation. In contrast, the CAST*MR provides good discrimination between different groups of people with mental retardation.” (Footnote omitted) Thus, the results of this test are not relevant to a determination of mental retardation.
(Emphasis in original).
Pita’s report, however, acknowledges that, in addition to evaluating applicant for mental retardation, he was also evaluating applicant’s competency to stand trial. Pita’s report also acknowledges that the CAST*MR is used to determine whether a mentally retarded person is competent to stand trial. We do not read Pita’s report as necessarily using the CAST*MR as a measure of applicant’s intellectual or adaptive functioning for purposes of making a mental-retardation diagnosis. Nevertheless, we believe that a rational finder of fact could find that applicant’s perfect score (50 out of 50) on this test, which is higher than a mentally retarded person would be expected to score, would, with all of the other evidence, have some bearing on the question of whether applicant is mentally retarded.
Seay also speculates that applicant was not classified in school as mentally retarded because the school systems may have been “loathe to classify a child as having mental retardation, or without funds to do the proper evaluation.” Seay states:
34. It is also important to understand that people who have mental retardation can be, and often are, diagnosed as having a learning disability. In other words, the two diagnoses can and often do co-exist. In school systems loathe to classify a child as having mental retardation, or without funds to do the proper evaluation, a learning disability classification may be assigned instead of a diagnosis of mental retardation.
This speculation is not evidence that applicant’s school systems may have been “loathe to classify a child as having mental retardation.” A rational finder of fact could consider it significant that applicant’s teachers did not consider applicant to be mentally retarded. See Briseno, 135 S.W.3d at 8 (evidentiary factor which fact-finders in criminal context may consider as indicative of a person’s mental retardation is whether those who knew the person best, such as teachers, considered the person to be mentally retarded).
Seay also claims that the trial court relied on “misconceptions” about mental *612retardation in making its findings. She states:
57. The facts that [applicant] could drive, hold down a job as a short order cook, and checked out 115 novels while in jail were suggested to be contraindi-cative of mental retardation. People who are mentally retarded can often drive and possess driver’s licenses. In fact, [applicant’s] relatives have stated that he was provided an oral version of the driver’s test,40 which is often the case for persons with mental retardation. Their intellectual disability does not necessarily impact them ability to drive. Holding a job is also not unusual for a person with mental retardation who has found a good match for his skills. Checking out books does not mean that a person can read them. This could indicate a desire to read or have another person read to them. In some cases, people with mental retardation desire to mask their disability so that others do not see them as different. They in essence attempt to wear a “cloak of competence.” (Footnote omitted) Having books that one cannot read is one way this masking phenomenon may be expressed.
The district court’s findings from the 2004 writ hearing, rejecting applicant’s mental retardation claim, do not turn solely on evidence that applicant could drive, hold down a job as a short-order cook, and check out books from the library. Moreover, we believe that this evidence, together with all of the other evidence, is relevant to a determination of whether applicant is mentally retarded. We also believe that a rational finder of fact could find that applicant has the ability to read more than a comic book and that he actually read the 115 novels that he checked out of the prison library. A rational finder of fact could find it relevant that applicant testified at his 1998 trial, “I’m not saying that I can’t read, I’m just saying there are words I can’t read.” A rational finder of fact could also find it relevant that neither Schmitt nor any one else with the burden of proving that applicant is mentally retarded apparently never bothered to ask applicant whether he read these books or conducted any tests to determine whether he read or could read these books.
We also agree with the State that a rational finder of fact could consider it relevant that applicant “skillfully testified in his own defense” at the guilt phase of his trial that his dead cousin murdered the girl. See Briseno, 135 S.W.3d at 8 (other factors that fact finder in criminal context may consider in weighing evidence as indicative of a person’s mental retardation is whether the person hides facts or lies effectively in his own interests). The State presents this supported-by-the-trial-recórd argument in its response to applicant’s current successive habeas corpus application:
In addition to the testimony of [applicant’s] own experts and his former employers, the record from trial also supports the conclusion that [applicant] does not possess any other significant deficits in adaptive behavior. For instance, [applicant] skillfully testified in his own defense during the guilt proceedings. Although the physical evidence against him was overwhelming, [applicant] carefully constructed a story which accounted for much of the State’s most incriminating proof. [Applicant] *613asserted that, despite appearances, his cousin had murdered [the girl], [Applicant] was able to explain [the boy’s] eyewitness identification by testifying that he did not turn [the girl] over to his cousin until after [the boy] was knocked unconscious. He also accounted for the matching tire tracks at both the [children’s home] and the crime scene by stating that his cousin took the ear from [applicant] and then drove [the girl] to her death. [Applicant] further stated that his footprints at the crime scene and the blood on his shirt could be explained because he later went to see [the girl’s] body after his cousin returned. Finally, [applicant] claimed that the scratches on his face had come from running through the woods after he found [the girl’s] body. [Applicant’s] ability to counter the State’s evidence with such an elaborate lie speaks greatly to his adaptive abilities.
Based on the foregoing, we decide that applicant’s additional evidence does not compellingly or dramatically undermine the previously considered substantial evidence that supports a finding that applicant is not mentally retarded. Even with a consideration of applicant’s additional evidence, a rational finder of fact could still find that applicant is not mentally retarded and that, applicant manufactured a mental-retardation claim in an attempt to escape the ultimate punishment for the brutal murder of an eleven-year-old girl. Applicant’s current successive habeas corpus application, therefore, does not meet the requirements of Article 11.071, § 5(a)(3).41 See Blue, 230 S.W.3d at 154, 162-63 (Article 11.071, § 5(a)(3), requires an Atkins claimant to present a pleading demonstrating “to this Court that there is evidence that could reasonably show, to a level of confidence by clear and convincing evidence, that no rational trier of fact would fail to find that he is mentally retarded”).
Applicant’s current successive habeas corpus application is dismissed, and the stay of execution is lifted.
*614KELLER, P.J., filed a concurring opinion.
HOLCOMB, J., dissented.