filed a statement concurring in the denial of relief in which PRICE, J., joined.
STATEMENT
In this subsequent writ application, filed pursuant to Article 11.071, § 5, of the Texas Code of Criminal Procedure, applicant presents a claim of mental retardation un-*401del’ Atkins v. Virginia.1 After we remanded the application, the trial court conducted an evidentiary hearing and considered a wealth of documentary information. The trial judge then signed findings of fact and conclusions of law and recommended that relief be denied. We have reviewed the record, and we adopt the trial judge’s findings and conclusions. Therefore, we deny relief.
Because this case presents a close question on the ultimate factual issue of mental retardation, I add the following remarks.
I.
Applicant was indicted for the murder of Agnes Herden committed in the course of a burglary or robbery on July 4, 1990. Against his attorney’s advice, applicant pleaded guilty to the capital murder charge. At his 1992 trial, the jury answered the two “aggravating” special issues affirmatively and the mitigation issue negatively, and the trial judge sentenced applicant to death. This Court affirmed that conviction and sentence on direct appeal. Rodriguez v. State, 899 S.W.2d 658 (Tex.Crim.App.1995).
The evidence at trial showed that applicant had entered Ms. Herden’s home in the early hours of the morning, stabbed the 80-year-old woman in the face, neck, and upper chest, hit her with a blunt object, and left her dead in her bed. He stole two television sets — both of which were left abandoned outside the home— some jewelry, and a pistol. At 4:80 a.m., applicant arrived at his sister’s trailer home where he sprinkled salt and pepper on the steps in an apparent effort to cover his scent. He soaked his blood-stained clothes in alcohol. Applicant told two friends who were in the trailer that he had stabbed a “black guy,” and showed them the jewelry and pistol he had stolen, as well as a knife (later determined to be the murder weapon) that he had taken with him. Applicant gave one of the men a dollar to buy a can of spray paint, and the three friends inhaled paint fumes until noon, when police officers came to the trailer. Applicant told his friends that he needed to hide from the police, but he made no effort to do so. Instead, he lay down on the floor, closed his eyes and pretended to sleep. The officers arrested him.
At first applicant denied committing the murder, but when a police officer confronted him and told applicant that he knew that this was a he, applicant gave a full confession. He explained that
I just had a lot of things on my mind. I have a lot of family problems because I can’t never talk to my family. I did not rape the lady and I did not have any sex with the lady. I just lost my mind. This statement is the complete truth as best I can remember it. I am sorry that I killed the lady and I really didn’t mean to do it.
During the punishment phase, Dr. James Sherman testified for the State concerning applicant’s mental health testing in 1982 when, at the age of fourteen, he was in the Bexar County Juvenile Detention Center for “sniffing glue” and auto theft. At that time, applicant already had a four-year history as a runaway, committing various criminal mischief and theft offenses.2 In 1982, applicant described himself to Dr. Sherman as “being a discipline problem,” but he denied having ever been suspended *402from school or having been in special education classes. He said that he had no idea what his grades in school were because he always gave his report card to his sister and did not look at it. His mother had died when he was five, and he did not know where his father was. He moved around and lived with various relatives. Although he was only fourteen, he said that he got drunk frequently, smoked marijuana, and used inhalants “a lot.”
Dr. Sherman testified that his 1982 testing of applicant’s overall level of functioning was measured on the Wechsler Scale as being in the mild (50-70) range of mental retardation. He had a full-scale IQ of 60, with a verbal score of 57 and a performance score of 70. Although concluding that applicant was legally competent, Dr. Sherman diagnosed him as having “overanxious disorder of adolescence,” “undersocialized nonaggressive conduct disorder of adolescence,” and “borderline intellectual functioning with a repetitive language deficit.” He recommended that applicant be placed in a special education program and a stable home, and that he enter into a “behavioral management contract” with his juvenile probation officer. At the 1991 trial, Dr. Sherman explained that, in his opinion, applicant’s ability to function was “much higher” than his full-scale IQ score of 60 implied. Applicant was “absolutely not” mentally retarded, rather he was functioning in the borderline range.
Dr. John Sparks also testified for the State. In his opinion, applicant was competent to stand trial, was not mentally retarded, and was not insane or suffering from any severe mental disease or defect. He found no sign of organic brain damage which is often associated with inhalant abuse. He stated that even though applicant’s most recent IQ score of 68 was in a “range that would normally be retarded,” he believed that applicant “was much more able to adapt to the community and function” than his IQ score suggested. Dr. Sparks explained that applicant deals with things “on a primitive level,” but “[h]e takes care of himself. He is able to find a place to stay. He makes relationships reasonably well. He’s not an isolated, withdrawn person.”
During closing arguments at the punishment stage, applicant’s attorney argued that applicant was both mentally ill and mentally retarded. Nonetheless, the jury did not find that there were sufficient mitigating circumstances to warrant a sentence of life imprisonment rather than a death sentence.3 Therefore, the trial court imposed a sentence of death.
In his first habeas application, applicant claimed that his trial attorneys failed to request the assistance of a “partisan” psychiatrist to evaluate his mental functioning, and, therefore, they provided ineffective assistance of counsel. At this hearing, Dr. Michael Arambula testified for applicant and stated that, in his opinion, applicant was mildly mentally retarded and functioned “right at” the line that separates mild mental retardation from borderline intellectual functioning. He found that applicant functioned like a six-to-eight-year-old child in some areas and a twelve-to-thirteen-year-old in others. In Dr. Arambula’s opinion, applicant had significant impairments in ten out of eleven specific adaptive behavior categories. He had an “unfortunate upbringing” and a long history of depression and very significant anxiety which made him have trouble concentrating, impaired his decisionmak-ing, and led to low self-esteem and irrita*403bility.4 With his long-term inhalant abuse, applicant acted impulsively, was easily agitated, and could not think clearly. Because of the interaction between a low intellectual level, clinical depression, and inhalant-affected behaviors, he was going to be “more irritable, can’t concentrate, be hyperactive.... He probably would be disruptive in class because he can’t sit still.” However, Dr. Arambula also agreed that applicant’s impulsivity and poor judgment would “be a common characteristic of a 22 or 23-year-old school drop out from a broken home with a history of drug use.”
Dr. Margot Zuelzer, a psychologist who also testified for applicant at this first habeas hearing, tested his IQ and believed that applicant’s verbal IQ score of 68 indicated mental retardation, even though his overall score of 71 was not in the retarded range, and his performance IQ of 77 was outside the range of mental retardation. She noted that applicant is “functionally illiterate” and has a “language positing problem.”
Dr. Sparks testified again at the first habeas hearing and largely agreed with Drs. Arambula and Zuelzer. He noted that everyone concurred that applicant was “at the borderline” or “on the cusp” of mental retardation, but that “[t]here aren’t any exact cutoffs between the various groupings ... they blend sometimes into each other.... The borderline really is a judgmental borderline.... It’s not an exact line separating one from the other.” Dr. Sparks believed that applicant fell on the “not mentally retarded” side of the line because he functioned at a higher level than his IQ scores would otherwise indicate, except in the area of academic achievement.
According to one of applicant’s sisters who testified at the first habeas hearing, applicant was a good student when he first started school, but he began having problems after his mother died. She also knew he had problems with spray paint, but she never saw him inhaling it. Another sister testified that applicant was sometimes a “troublemaker” who used to fight when he was young. The mother of applicant’s child testified that applicant used spray paint “all the time” when they were living together. Applicant worked for his brother-in-law in the construction business, but couldn’t get a good job because he wasn’t very smart. One of applicant’s trial attorneys testified at the habeas hearing that he never noticed any mental retardation disability in his client.
This court denied relief on that first habeas application on April 12, 2000. Applicant then sought habeas relief in federal district court. While this writ was still pending, but after the Supreme Court delivered its opinion in Atkins, applicant filed a subsequent federal writ raising a mental retardation claim. On March 31, 2003, the federal district court dismissed that petition without prejudice so that applicant could return to state court to raise his Atkins claim and then, should it be necessary, he could return to federal court after exhausting state remedies. Rodriguez v. Cockrell, Civil No. SA-00-CA-443EP, 2003 WL 1906154 (W.D.Tex. March 31, 2003). On September 10, 2003, this Court entered an order finding that applicant had met the threshold criteria for raising a subsequent claim under Tex.Code Crim. PROC. Art. 11.071, § 5, and we remanded the case to the trial court for further proceedings.
*404On remand, the trial court heard additional testimony concerning applicant’s mental abilities. Dr. Sherman testified again and stated that the fact that a person has a subaverage IQ score does not necessarily mean that he is mentally retarded. An IQ score is indicative of a lessened capacity for doing well in academics, but environmental factors — a stable home, supportive parents, a stable school situation, the language spoken in the home, peer groups, alcohol and drug abuse — all influence academic achievement and impact IQ scores. Dr. Sherman believed that applicant was severely learning disabled and did not test well. He noted that “the whole verbal spectrum of the IQ test is really culturally based,” and therefore a lack of academic achievement and a low reading ability will “inhibit your performance on the test.” Dr. Sherman also expressed his concern that the adaptive deficiencies prong of mental retardation had no objective or quantifiable component.
Dr. Sparks testified again as well. He, too, expressed concern that the adaptive functioning criteria of mental retardation are not objective: “There’s a great deal of subjective input. The observation of the individual, the history that he gives, the history obtained from other sources like family or work history, if there is work history.” The determination of whether one has significant adaptive skills deficits is really “a judgment call.” Although admitting that it was “a close call,” Dr. Sparks reiterated his opinion that applicant was not mentally retarded. His opinion was based upon the facts that applicant worked; had relationships with other people, including family; and had the “ability to adapt to some level above what I would expect a retarded person to adapt in our society.” Dr. Sparks also explained that applicant “tested badly” because of his high anxiety and low motivation levels. Dr. Sparks noted that applicant could communicate reasonably well while in jail and that his “sick call requests” were written very logically and asked for specific things. Applicant wrote his medical form information “at a level that’s higher than any retarded person could.”
Dr. Jim Patton, an adjunct professor of special education at the University of Texas, testified for applicant at the second writ hearing and he agreed with Dr. Sparks that there is a “moderate” correlation between intellectual ability and adaptive behavior. Dr. Patton interviewed applicant and some of his family members and concluded that applicant had shown evidence of adaptive skill deficits before age eighteen, particularly in the areas of communication and academic achievement. He had no opinion whether applicant was presently mentally retarded.
The trial court also considered a written affidavit by Dr. Ruth Luckasson, a special education professor at the University of New Mexico. She did not evaluate applicant personally, but stated that the opinions of Drs. Sparks and Sherman that applicant was “borderline and did not have deficits in adaptive skill areas,” were not supported by their records. She thought that the clinical data developed by Drs. Sparks and Sherman support a diagnosis of mental retardation.
The State offered applicant’s prison records that show he was a member of the Mexican Mafia and that he had numerous disciplinary violations. A recent prison mental status examination report concluded that applicant’s adaptive skills were “normal” as to coping ability and that he had no skills deficit.
II.
The trial court has now made its findings of fact and conclusions of law based *405upon the entirety of the testimony and records. The trial court summarized its conclusion in the following manner:
The applicant had the burden to establish by a preponderance of the evidence each of the three elements of mental retardation: onset during the developmental period, subaverage intellectual functioning, and significant deficits in adaptive skills. The applicant has failed to establish significant deficits in adaptive skills. The expert evidence is in conflict, but the question is not necessarily controlled by expert testing. There is lay evidence in the form of observations by prison and state school officials that applicant did not have significant adaptive deficits. There is evidence that helps explain why applicant could have poor I.Q. scores and yet not have adaptive skills deficits and thus not be mentally retarded. There is evidence that applicant is not docile or easily manipulated by others, and is in fact aggressive and manipulative. There is evidence that applicant is able to communicate and function in his environment.
Based upon those factual findings and legal conclusions, the trial court recommends that this Court deny relief.
All of the experts agreed that applicant’s numerical IQ level is at the borderline or below the cut-off level for a diagnosis of mental retardation. There is also expert and lay witness evidence in this record which would support a finding that applicant does suffer a deficit in adaptive skills, and there is evidence in this record which supports the trial court’s finding that applicant does not suffer a deficit in adaptive skills. A finding on either side of this question is supportable by the record evidence. But as a court reviewing only the cold record, we must be especially deferential to the trial court’s factual findings as he is “Johnny-on-the-Spot” and able to make credibility and demeanor determinations of the witnesses that we are not capable of making on habeas review.5 There are also statements in the two habe-as hearings that the trial court and advocates were personally familiar with both Drs. Sherman and Sparks and their experience and expertise in the mental health field.
I nonetheless reiterate my concern that “[t]he adaptive behavior criteria are exceedingly subjective, and undoubtedly experts will be found to offer opinions on both sides of the issue in most cases.”6
It seems to me that, with its decision in Atkins, the Supreme Court is moving back toward the pre-Furman days of wholly subjective criteria in determining whether a particular person is or is not subject to the death penalty.7 As experts on both *406sides in this case testified, there are no objectively verifiable standards by which to gauge whether a specific person does or does not suffer the kind of significant “adaptive deficits” that a diagnosis of mental retardation requires. Nor are there any scientifically verifiable standards by which one might measure whether a person’s academic, social, or functional deficits are related to innate mental deficiencies, bad upbringing, impoverished environment, bad moral character, emotional problems, poor habits, lack of motivation, drug or alcohol dependence, or other factors.
As school children we were taught that King Solomon weighed all of the evidence before him and made a reasoned decision; Nero divined merit on a whim and just pointed his thumb up or down. I fear that, under Atkins and the subjective legal definition of the “adaptive deficits” prong of mental retardation, we are moving farther from King Solomon and closer to Nero. If there is evidence in the record to support the factfinder’s conclusion, by a preponderance of the evidence, that a person does or does not suffer from significant “deficits in adaptive behavior” — whatever that may mean to the factfinder- — -that conclusion must be affirmed.
In Atkins, the Supreme Court may have intended to create a “bright-line” rule that those who are mentally retarded are, because of their lesser moral culpability, exempt from the death penalty. But I fear that there is no such bright line. There is, on the contrary, broad agreement among mental health experts that determining whether a person suffers from the type and level of “adaptive deficits” that qualifies for a mental retardation diagnosis is highly subjective and largely a matter of individual judgment. Under Atkins, the ostensible issue for the factfinder in a death penalty case is, “Is this person mentally retarded?” As we see in this case, there may be no clear-cut answer to that question. Absent the possibility of a rigorous, scientifically reliable and verifiable “yes-or-no” determination, the relevant question perhaps ought to be, “Do you believe that the defendant is sufficiently mentally responsible for his conduct such that the death penalty is an appropriate punishment?” This is analogous to the Texas statutory mitigation question which, at bottom, asks “Do you believe that the defendant is sufficiently morally responsible for his conduct such that the death penalty is an appropriate punishment?”
In any event, the “evolving standards of decency” of the Eighth Amendment8 depends upon the collective judgment of the twelve ordinary citizens who sit in judgment of the case before them. Their assessment is the best determinant of fairness in assessing both mental and moral responsibility in cases of capital crimes committed by those whose mental capacity and competency is in doubt. Thomas Jefferson had it right: “I know of no safe depository of the ultimate powers of the society but the people themselves, and if we think them not enlightened enough to exercise that control with a wholesome discretion, the remedy is not to take it from them, but to inform their discre*407tion.”9 Expert testimony and various mental retardation criteria may inform the factfinder’s discretion. But it is nonetheless the factfinder who must ultimately decide whether a particular person “who claim[s] 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”10 that the death penalty is an inappropriate punishment.
Based upon the evidence in the record, I agree that the trial court did not err in concluding that applicant failed to prove, by a preponderance of the evidence, that he is mentally retarded. Therefore, I join the Court in denying applicant relief on his mental retardation claim.
. 536 U.S. 304, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002).
. Evidence at the capital murder trial showed that applicant and his friends accosted other, smaller children, stole their bikes, and threatened to beat them up with chains and sticks.
. Tex.Code Crim. Proc. art. 37.0711, § 3(e).
. Dr. Arambula testified that applicant’s extensive record for prison disturbances, including one incident in which he stabbed a fellow inmate five times, may have been the result of his depression and "because he was so anxious."
.Ex parte Briseno, 135 S.W.3d 1, 12-13 (Tex.Crim.App.2004) (in habeas context, "we afford almost total deference to the trial judge's determination of the historical facts supported by the record, especially when those fact findings are based on an evaluation of credibility and demeanor. However, if the trial court's ruling is not supported by the record, this Court may reject the findings”) (footnote omitted); Ex parte Franklin, 72 S.W.3d at 671, 675, n. 5 (Tex.Crim.App.2002) ("although this Court is not bound by the findings of the trial court in post-conviction habeas corpus proceedings, such findings are considered if supported by the record"); Ex parte Evans, 964 S.W.2d 643, 648 (Tex.Crim.App.1998) ("while this Court is not bound by the findings of a habeas judge in a habeas corpus proceeding, where the findings are supported by the record, they should be accepted by this Court").
. Ex parie Briseno, 135 S.W.3d at 8.
. Furman v. Georgia, 408 U.S. 238, 310, 92 S.Ct. 2726, 33 L.Ed.2d 346 (1972) (Stewart, J., concurring) (state death penalty statutes were unconstitutional because they gave juries "open-ended, unstructured discretion” in deciding whether to impose death penalty); *406compare Jurek v. Texas, 428 U.S. 262, 273-74, 96 S.Ct. 2950, 49 L.Ed.2d 929 (1976) (joint opinion of Stewart, Powell, & Stevens, JJ.) (holding Texas death penalty statute constitutional because it "guided and focused the jury’s objective consideration of the particularized circumstances and the individual offense and the individual offender”).
. Atkins, 536 U.S. at 321, 122 S.Ct. 2242 (holding that execution of those found to be mentally retarded is excessive under the "evolving standards of decency” of the Eighth Amendment).
. Thomas Jefferson, letter to William Charles Jarvis, September 28, 1820.
. Atkins, 536 U.S. at 317, 122 S.Ct. 2242.