*1188Petitioner Robert Lewis, Jr., seeks habeas corpus relief, urging that he is ineligible for execution because he is intellectually disabled1 and that his counsel failed to investigate **758and present mitigating evidence as to penalty. We issued an order to show cause and subsequently ordered a reference hearing in the superior court. The referee found that (1) petitioner is intellectually disabled and (2) his trial counsel did not perform deficiently at the penalty phase.
The referee's findings regarding intellectual disability are supported by substantial evidence and are adopted. Because petitioner is entitled to relief from the *116death judgment under Atkins v. Virginia (2002) 536 U.S. 304, 321, 122 S.Ct. 2242, 153 L.Ed.2d 335 ( Atkins ) and In re Hawthorne (2005) 35 Cal.4th 40, 24 Cal.Rptr.3d 189, 105 P.3d 552 ( Hawthorne ), we need not reach the penalty phase ineffective assistance of counsel claims.
I. FACTUAL BACKGROUND
This evidentiary summary is drawn from the opinion in petitioner's first automatic appeal. (See People v. Lewis (1990) 50 Cal.3d 262, 271-274, 266 Cal.Rptr. 834, 786 P.2d 892 ( Lewis I ).)
A. Guilt Phase
In October 1983, victim Milton Estell had been trying to sell his car. He displayed it in a parking lot and ran a newspaper advertisement. On October 27, neighbors saw him standing outside his home, looking at the car and speaking with petitioner.
*1189On October 28, after Estell's ex-wife, Jacqueline, had been unable to contact him and neighbors expressed concern, an officer went to Estell's home to check on him. He found Estell's body in a closet, bound and gagged. Estell had been stabbed repeatedly and shot in the back. Toilet paper had been stuffed in his mouth. His wallet lay nearby, containing credit cards but no cash. The car was missing.
There were no signs of forced entry. Petitioner's fingerprints were found in three locations in the house, including the bathroom near the toilet paper dispenser. Jacqueline confirmed a number of items were missing, including a gold chain. She identified a gold chain petitioner had worn during the preliminary hearing as the missing item.
On November 1, officers saw Estell's car parked on the street. Petitioner and a woman got in the car and drove off. The officers stopped the car, impounded it, and arrested petitioner. He was carrying $400 in cash and gave officers a false name. When interviewed, petitioner initially claimed he bought the car on October 24. He paid $11,000 in cash, which he had won in Las Vegas and carried in a brown paper bag. He said that the entire transaction took place on Estell's porch and that he never entered the house. He had Estell make out the bill of sale to his girlfriend because he did not want the car in his name. Reinterviewed the next day, petitioner provided conflicting details. He said he had won $17,000 in Las Vegas and carried the money in a white bag. He maintained the sale took place on October 24, even when told neighbors had seen the car at Estell's on the 27th. A search of the impounded car revealed a bill of sale bearing Estell's forged signature. A door opener to Estell's garage was also recovered.
Petitioner's father testified for the defense that he had registered petitioner at a motel on October 24. The father entered the license plate number for Estell's car on the registration card. Petitioner's sister testified she had given him a gold chain that looked like the one he had worn at the preliminary hearing. Petitioner did not testify.
Petitioner was convicted of first degree murder and robbery. The jury found true allegations that he used a deadly weapon and a firearm, and the special circumstance of murder during the commission of robbery.
B. Penalty Phase
At the penalty phase, the defense stipulated petitioner had been convicted of robbery in 1982, 1972, and twice in 1977. In mitigation, the defense presented evidence that petitioner's mother had died in 1967, his father had been sentenced to prison several times, and his brother was currently *117serving **759*1190a prison term. His sister testified that she loved him and cared about "what happens to him." The jury returned a death verdict, which the court imposed.
II. PROCEDURAL BACKGROUND
On direct appeal, we affirmed the judgment in all respects, but vacated the death sentence because the trial court erroneously considered a probation report in ruling on petitioner's application to modify the penalty. ( Lewis I, supra , 50 Cal.3d at pp. 286-287, 192, 266 Cal.Rptr. 834, 786 P.2d 892 ; see Pen. Code, former § 190.4, subd. (e)2 .) On remand for the new modification hearing, the trial court denied the application to modify the penalty and reinstated the judgment of death. We affirmed. (See People v. Lewis (2004) 33 Cal.4th 214, 14 Cal.Rptr.3d 566, 91 P.3d 928.)
When petitioner filed a habeas corpus petition, the Director of the Department of Corrections and Rehabilitation was ordered to show cause why relief should not be granted. After the respondent Attorney General's return and petitioner's reply, we ordered a reference hearing and directed the referee to address several questions. The first was whether petitioner is intellectually disabled within the meaning of Atkins, supra, 536 U.S. 304, 122 S.Ct. 2242 and Hawthorne, supra, 35 Cal.4th 40, 24 Cal.Rptr.3d 189, 105 P.3d 552. The remaining questions, each with several subparts, concerned the penalty phase ineffective assistance of counsel claims.3
Los Angeles County Superior Court Judge Robert J. Perry served as referee.4 Some 15 witnesses testified at the evidentiary hearing over the course of 14 days. At least 12 witnesses were questioned about petitioner's intellectual functioning. The referee filed a 42-page report with findings and *1191recommendations. Petitioner and the Attorney General filed exceptions to the report and provided supplemental briefing.
III. DISCUSSION
A. Legal Framework
A habeas corpus petition is a collateral attack on a presumptively valid judgment, thus " 'the petitioner bears a heavy burden initially to plead sufficient grounds for relief, and then later to prove them.' " ( In re Price (2011) 51 Cal.4th 547, 559, 121 Cal.Rptr.3d 572, 247 P.3d 929.)
*118The standard of proof is preponderance of evidence. ( In re Cudjo (1999) 20 Cal.4th 673, 687, 85 Cal.Rptr.2d 436, 977 P.2d 66 ; see also In re Bacigalupo (2012) 55 Cal.4th 312, 333, 145 Cal.Rptr.3d 832, 283 P.3d 613.) "Because the referee observes the demeanor of the witnesses as they testify, we generally defer to the referee's factual findings" and give them great weight if supported by substantial evidence.5 ( **760Bacigalupo , at p. 333, 145 Cal.Rptr.3d 832, 283 P.3d 613.) These findings are not binding, however. Ultimately, this court must make the findings necessary to resolve petitioner's claim. ( In re Thomas (2006) 37 Cal.4th 1249, 1256-1257, 39 Cal.Rptr.3d 845, 129 P.3d 49.)
Evaluation of intellectual disability involves examination of two primary factors: 1) significantly subaverage general intellectual functioning, and 2) deficits in adaptive behavior. Both must manifest before age 18. (See § 1376, subd. (a).) In Hawthorne we recognized that " 'an IQ between 70 and 75 or lower ... is typically considered the cutoff IQ score' " for the intellectual functioning prong. ( Hawthorne , supra , 35 Cal.4th at p. 48, 24 Cal.Rptr.3d 189, 105 P.3d 552.) We declined, however, to adopt a fixed numerical cutoff, concluding that such a requirement would be "inconsistent with established clinical definitions" and undesirable in light of the imprecisions of IQ test scores, which have a five-point measurement error. ( Ibid . ) This approach remains sound. (See Hall v. Florida (2014) --- U.S. ----, 134 S.Ct. 1986, 2000-2001, 188 L.Ed.2d 1007 ( Hall ).) As to adaptive behavior deficits, Hawthorne quoted from *1192Atkins and the clinical definitions from the American Association on Mental Retardation (AAMR)6 and the American Psychiatric Association (APA). ( Hawthorne , at pp. 47-48, 24 Cal.Rptr.3d 189, 105 P.3d 552.) The most current diagnostic guidance on adaptive behavior deficits is discussed post 233 Cal.Rptr.3t at pages 126-127, 417 P.3d at pages 766-767.
The professional psychological community continues to refine its understanding of these concepts. For our purposes here, intellectual disability is "a question of fact ... 'not measured according to a fixed intelligence test score or a specific adaptive behavior deficiency, but rather constitutes an assessment of the individual's overall capacity based on a consideration of all the relevant evidence.' " ( People v. Superior Court (Vidal ) (2007) 40 Cal.4th 999, 1012, 56 Cal.Rptr.3d 851, 155 P.3d 259 ; see id. at p. 1012, fn. 6, 56 Cal.Rptr.3d 851, 155 P.3d 259.)
B. Hearing Testimony and the Referee's Report
1. The Expert Witnesses
Natasha Khazanov, Ph.D., testified for petitioner.7 She had evaluated over 50 defendants *119for intellectual disability as a defense expert, determining that three of them suffered from that condition. She examined petitioner three times at San Quentin State Prison and was the only testifying expert to have contact with him in the 24 years since his conviction. Relying on her interactions with petitioner and other information in the case, Khazanov concluded that he is intellectually disabled. With respect to risk factors, she noted petitioner's mother may have been intellectually disabled, drank heavily when pregnant with him, and struck him frequently. Petitioner may have suffered from malnutrition and he fathered a son with Down syndrome.
Respondent called three experts. Michael Maloney, Ph.D.,8 had been consulted by the **761defense before trial in 1984. Although Maloney had no independent recollection of petitioner, he testified based on his records and on his custom and practice in similar cases. Retained to administer tests and *1193present a psychological profile, Maloney was asked specifically to determine if petitioner was legally sane and competent to stand trial. He met petitioner twice and had him tested by his registered psychological assistant, Nancy Kaser-Boyd, Ph.D.9 Maloney reviewed case materials, met with counsel and participated in family interviews. He did not write a report because he believed his input would not assist petitioner's case.
Maloney testified he would have considered "significant" mental retardation a potentially mitigating factor. If petitioner had been "clearly mentally retarded," he would have informed defense counsel. Maloney had not considered petitioner mentally retarded by 1984 standards but emphasized that the question was not central to his inquiry into petitioner's potential mental illness or capacity to stand trial. Under current standards, Maloney considered the question of intellectual disability a "closer case." He could not say whether petitioner was so impaired.
Kaushal Sharma, M.D.,10 also examined petitioner in 1984. He had no independent recollection of the case before reviewing his report. Defense counsel had asked him to determine whether petitioner was competent for trial and sane at the time of the *120offense. Dr. Sharma believed he followed his usual practice of interviewing the subject twice and reviewing documents. He administered no tests and could not recall if petitioner could read and write.
Dr. Sharma wrote a three-page report concluding petitioner was sane and competent. He found no indication of psychosis, organic brain disorder, depression or other major defect. He agreed with a previous diagnosis of antisocial personality disorder starting at an early age. It is likely he told counsel he could not assist the defense. He was not asked to evaluate intellectual disability and focused on mental illness and behavioral problems. He and Maloney often worked together. Dr. Sharma understood Maloney was evaluating petitioner's intellectual functioning and had "better tools" to make such an assessment.
Charles Hinkin, Ph.D.,11 was respondent's principal expert witness. He never met with petitioner. He did review medical and legal records and *1194watched Khazanov and Maloney testify. He acknowledged, "It's always better to examine a patient than to offer a diagnosis based on a review of medical records." Hinkin opined that petitioner is not intellectually disabled, although he is certainly a person with mental health issues and is functionally illiterate. Even as to the intellectual disability question, Hinkin admitted there is "room for honest ... disagreement," and it is not "slam-dunk, clear-cut [and] incontrovertible one way or the other." In Hinkin's view, the evidence of petitioner's adaptive behavior weighs disproportionately against a finding of intellectual disability. **7622. Intellectual Functioning
The referee made the following findings with respect to petitioner's general intellectual functioning. Beginning at age six, petitioner was given multiple IQ tests: first by school authorities, then while in custody as a juvenile and young adult, later in 1984 in preparation for trial, and finally in 2003 in connection with this petition. The results are summarized below.12
*121*1195Test Date Age IQ Score Kuhlmann-Anderson Form A 4/14/59 6 89 Stanford Binet Form L 11/20/59 7 83 Lorge-Thorndike Form 1A 1/8/60 7 78 Kuhlmann-Anderson Form C 1/8/62 9 82 Lorge-Thorndike Form 3A-V 5/16/63 10 57 WISC 5/21/63 10 70 Revised Beta IQ 8/68 16 83 Thurstone Primary Mental Abilities SRA IQ 9/3/68 16 67 (verbal) (Thurstone Primary) 99 (non-verbal) Revised Beta IQ 12/8/72 20 80 Quick Test 6/16/84 32 75 Wechsler Adult Intelligence Scale - Revised 7/5/84 32 72 (verbal) (WAIS-R) 76 (performance) 73 (full scale) Wechsler Adult Intelligence Scale - III 6/10/03 51 66 (verbal) (WAIS-III) 75 (performance) 67 (full scale)
The referee noted the experts' agreement that such frequent testing in petitioner's youth "reflected an obvious concern by school authorities." Based on the early testing, Khazanov testified, "There was something fundamentally wrong with [petitioner] from [an] early age." Maloney opined petitioner "was probably singled out as being a kid who needed to be looked at and evaluated for some reason at that time." And Hinkin conceded that "something about this kid got the attention of the school teachers and authorities. They're not giving [IQ] tests [to] every single kid, you know, 6, 8, 10 times throughout their school [years]. ... So there was something about him that made him stand out...." However, the experts disagreed as to whether the test results showed subaverage intellectual functioning.
The referee found petitioner's WISC score of 70 in 1963 the most reliable assessment of petitioner's IQ before age 18, noting the Wechsler tests are the standard for assessing intellectual functioning (see, e.g., Atkins, supra, 536 U.S. at p. 309, fn. 5, 122 S.Ct. 2242 ) and that the experts agreed the WISC was "the most reliable test." The referee found the other tests given to petitioner before age 18 were "of questionable reliability." For example, Maloney called the Stanford-Binet **763"the most biased" of the tests because it "was normed on white young people." Khazanov testified the Revised Beta IQ test is considered to have "low reliability" because of similar norming bias. Maloney opined group tests like the Kuhlman-Anderson and the Lorge-Thorndike tests were used in the 1950's and 1960's but were "not good at all" in measuring IQ. *122The person who administered the Thurstone Primary while petitioner was in the California Youth Authority described him as a "non-reading, non-bookish boy whose cultural set is so diverse from the major cultural patterns that he can not be adequately tested. His scores as listed are meaningless[,] for subject is not academic or vocationally oriented. He may be able to function as a dull normal, but that surmise is a projection based on the non-verbal SRA score ...." Maloney testified "dull normal" is usually above the cutoff for intellectual disability, but is "kind of like borderline."
The referee found petitioner's 1963 WISC score to be "remarkably consistent" with his later scores on the WAIS-R in 1984 and the WAIS-III in 2003. His scores on the WAIS-R and WAIS-III subtests were also "very similar," presenting "a consistent profile which appears to be reliable." The consistency in petitioner's Wechsler test scores over 41 years led the referee to reject respondent's argument that petitioner's WISC score reflected a lack of *1196motivation or poor attitude toward school. The experts concluded, and the referee agreed, that petitioner was not malingering on any of the tests. The consistency of scores also led the referee to find that petitioner's WAIS-III score of 67 at age 51 was not caused by dementia.
Petitioner's Wechsler test scores are reflected below and include the WAIS-R as rescored by Khazanov. The referee found that the rescored WAIS-R scores "more accurately represent petitioner's intellectual functioning in 1984."13
Test Date Age Verbal Performance Full IQ IQ Scale IQ WISC 5/21/63 10 70 WAIS-R 7/5/84 32 72 76 73 WAIS-R (rescored) 7/5/84 32 68 76 71 WAIS-III 6/10/03 51 66 75 67
The experts opined on what effect, if any, socioeconomic factors played in petitioner's scores. Maloney testified that persons with petitioner's background "will always score low on these kind of tests, lower than probably their level of intelligence ... [b]ecause he's got less than average level of education for sure" and experienced "a lot of disadvantageous factors," including being placed in custody at age 12. Hinkin explained the more a person diverges from the mainstream increases the chance that the person's score will differ from his actual level of intelligence. "There's [a] whole host of reasons why African-Americans have gotten the short end of the stick when it comes to educational opportunities, occupational opportunities, the kind of things that would ... affect [IQ] test performance." Hinkin testified the "gap" in IQ performance between African-Americans and the mainstream is shrinking, "but it still remains to this day."
*123Khazanov believed the WAIS-III is properly normed and reliable and that adjusting scores based on race or socioeconomic factors is inappropriate. She testified the most recent publication of the AAIDD, Intellectual Disability: Definition, Classification, and Systems of Supports (11th **764ed. 2010) (AAIDD-11) supports using standardized tests without adjustments based on race or socioeconomic status. The referee declined to adjust petitioner's IQ scores, citing the difficulty in *1197quantifying any adjustment and the AAIDD's pronouncement that these adjustments should not be made.
The referee concluded, "[P]etitioner meets the first prong of the mental retardation definition" in that he "clearly has significant subaverage general intellectual functioning which was manifested before age 18." He found the consistency in petitioner's Wechsler tests scores particularly persuasive and noted all these scores "fall well within the cutoff score for mental retardation of 75." The report noted even Hinkin conceded that "if you focus just on the better tests, which are the [Wechsler] tests, then ... it would not be unreasonable to conclude that [petitioner's IQ scores] meet[ ] that [first] prong...."
3. Adaptive Behavior
Expert testimony differed as to petitioner's adaptive behavior. Khazanov, petitioner's expert, found he "demonstrated severe deficits in the conceptual skills area[,] which in her view satisfied the adaptive behavior prong for mental retardation." She described petitioner as "basically illiterate" and observed he was "the only person actually at San Quentin that I met who didn't know how to read and write." Additionally, Khazanov found he "had serious difficulties in understanding spoken words and concepts and exhibited mental and cognitive deficits." Khazanov found she had to repeat instructions for a test "many, many times" and that petitioner had "extreme difficulty understanding what I wanted from him." Khazanov noted that past probation reports had observed that petitioner didn't "make sense." Khazanov opined petitioner's inability to read and write is not the result of poor schooling. She felt that "he attended school enough to be able to learn to read if he could." Petitioner's lifelong friend, Larry Cleveland, attended school with petitioner and his sisters. All of them learned to read and write while petitioner did not. Khazanov considered this strong evidence of petitioner's adaptive behavior deficits.14
Khazanov did not interview or test third parties who had interacted with petitioner in his youth. Her tests and interviews of petitioner revealed such severe deficits in reading, writing, language, and money concepts that further steps were unnecessary to satisfy the adaptive behavior prong. Khazanov concluded petitioner was not malingering but instead was trying to please and *1198favorably impress her. He also told her that he did not want her to find him "crazy." The referee found petitioner's resistance to this label "consistent with being mentally retarded."
Maloney's assistant, Nancy Kaser-Boyd, gave the WAIS-R test to petitioner in July 1984 when he was 32 years old. She noted he "might have a hearing prob[lem]-he seemed to mishear a number of questions." Maloney testified petitioner might *124have aphasia, "an expressive language difficulty," which "might be associated with mental retardation." Kaser-Boyd also noted petitioner's "talk comes out all confused" when dealing with the test section on comprehension. Petitioner had "real trouble disting[uishing] the concept" involved in the section on similarities. Maloney agreed that such a difficulty could indicate intellectual disability.
Testimony from childhood friends and family also pointed to adaptive deficits. Larry Cleveland testified petitioner could not read street signs or spell words. He was "slow" and did not learn how to shoot pool or gamble with dice as well as Cleveland and others did. Steven Harris, also a childhood friend, could not remember petitioner ever reading anything. He did not think petitioner was "retarded," but petitioner was not as smart as Harris or Cleveland. "He just wasn't as quick as the rest of us." Tommie McGlothin, petitioner's first cousin, testified, "[S]omething [was] wrong with him. He had a problem." McGlothin said petitioner kept getting **765into trouble by "doing the same thing, going out and stealing bicycles."
Respondent's primary expert, Hinkin, acknowledged that "everybody agrees" petitioner is functionally illiterate and that the "reading and writing piece" of the adaptive deficit "is clearly there." He also agreed petitioner was not malingering. He questioned whether petitioner has brain damage in the absence of any neurological examination. He conceded, however, it was "quite possible that there is some sort of lateralized disease process that has been undetected through [petitioner's] 51-odd years of life." Hinkin further agreed that "we know something is wrong with" petitioner and that "there's a reason why he was sent to see the school psychologist" so many times. Even so, Hinkin felt the record was inadequate to satisfy the adaptive behavior prong.
Applying the AAIDD's current diagnostic definition (discussed post, at 233 Cal.Rptr.3d at pages 126-127, 417 P.3d at pages 766-767), the referee found "[p]etitioner clearly exhibited significant adaptive behavior deficits before the age of 18." Friends and family considered him mentally "slow." To this day he remains unable to read and write or effectively understand and communicate. The referee rejected respondent's arguments "that petitioner was a successful street hustler and his ability to survive on the streets shows he did not suffer adaptive behavior deficits" and *1199that "petitioner's interview by detectives on November 21, 1983 ... shows that petitioner did not suffer communication difficulties." Instead, the referee noted, "An ability to survive on the street does not necessarily refute adaptive behavior deficits. The evidence does not suggest that petitioner was a successful street hustler. He was arrested and convicted numerous times. His apparent failure to learn from his experience, provides additional support that he is intellectually disabled. Petitioner's interview by detectives similarly fails to provide persuasive evidence rebutting his weakened mental condition. In the interview, petitioner acknowledged being a 'pimp,' a 'gambler,' and a 'speed hustler' who had been 'hustling all [his] life,' but he also confidently denied his true identity and asserted he had never before been to prison. ... Petitioner's adamant assertion of such easily disprovable lies, displays unclear thinking and an obvious lack of cunning. [Fns. omitted.]"
4. The Referee's Conclusion
The referee found petitioner is intellectually disabled within the meaning of Atkins, supra, 536 U.S. 304, 122 S.Ct. 2242 and Hawthorne, supra, 35 Cal.4th 40, 24 Cal.Rptr.3d 189, 105 P.3d 552. At the outset, *125he noted, "The difficulties in attempting to determine whether petitioner is mentally retarded are obvious. Mental retardation requires an assessment of intellectual functioning and adaptive behavior prior to the age of 18. Petitioner, who is now 59, committed the murder in this case in 1983 at age 31. The psychologist and psychiatrist who examined petitioner in 1984 were not focused on mental retardation because the Atkins decision exempting mentally retarded defendants from execution did not occur until 2002, more than 18 years later. To further complicate matters, the standard for mental retardation has changed over the years and is today more lenient than in 1983 and 1984."
As to subaverage intellectual functioning, the referee noted petitioner's difficulties in school; his repeat of first grade; the frequent IQ testing in his youth; and the consistency of the Wechsler test results, all of which fell within the intellectual disability range. Addressing adaptive behavior deficits, the referee cited petitioner's illiteracy, his difficulties with understanding, and the testimony of his friends and cousin. The referee noted that petitioner's expert concluded that he is intellectually disabled, while "Maloney testified that under today's more lenient standards, this is a 'closer case,' and he cannot say whether petitioner is today mentally retarded." "Dr. Sharma did not assess petitioner for mental retardation, and relied on ... Maloney to test petitioner's intellectual functioning." The referee considered it significant that respondent's primary expert, Hinkin, did not examine petitioner, yet acknowledged it is "always better to examine a patient than to offer a diagnosis based on review of medical records." The referee concluded Hinkin's **766testimony is entitled to lesser weight than that offered by the other experts. *1200C. Analysis
Our own review of the evidence demonstrates petitioner has met his burden of proving he is intellectually disabled. As noted, an intellectual disability assessment examines two related criteria, both of which are evident before age 18: (1) significantly subaverage general intellectual functioning, and (2) deficits in adaptive behavior. Substantial evidence demonstrates petitioner experienced deficient intellectual functioning before he turned 18. The reliable Wechsler scores all fall within the widely accepted range for intellectual disability. These scores remained consistent over decades. Both Khazanov and Maloney opined it would have been extremely difficult, if not impossible, for someone to falsely produce such similarities. Even Hinkin, respondent's primary expert, conceded that, based on these scores, "it would not be unreasonable to conclude that" petitioner satisfies the first prong.
Respondent challenges the referee's failure to consider other intelligence tests, his lack of motivation, and other socioeconomic factors. He urges that petitioner's scores underestimate his general intellectual functioning. He alleges Khazanov had limited experience and ignored other possible explanations for petitioner's low scores. Yet there were sound reasons for the referee to credit Khazanov's testimony. She was the only testifying expert to interact with petitioner since 1984, which enabled her to assess petitioner's current functioning alongside his historical records. She could presently recall petitioner and his performance, unlike Maloney and Dr. Sharma. She specifically focused on the question of petitioner's intellectual disability, which Maloney and Dr. Sharma did not do when consulted by the defense. The consistency between petitioner's score on the WAIS-III (which Khazanov administered) and his earlier test scores supports *126the conclusion that petitioner's low IQ scores are the result of a long-standing intellectual disability, rather than a lack of motivation or dementia.
Additionally, the referee reasonably concluded that Dr. Hinkin's testimony was "entitled to lesser weight than that offered by the other experts." The referee was not bound by the opinion of any particular expert witness, but " 'may weigh and consider all evidence bearing on the issue of mental retardation.' " ( Hawthorne , supra , 35 Cal.4th at p. 46, 24 Cal.Rptr.3d 189, 105 P.3d 552.) Hinkin never met with petitioner personally, and he conceded it was better to examine a patient rather than offer a diagnosis based on a record review. The referee had discretion to weigh his testimony accordingly. (See People v. Bassett (1968) 69 Cal.2d 122, 146, fn. 22, 70 Cal.Rptr. 193, 443 P.2d 777 [" 'the extent of an expert's knowledge goes to the weight of his testimony' "].) Hinkin himself acknowledged it would not be unreasonable to conclude petitioner's test scores meet the first prong, and Maloney agreed the question was a close one.
*1201We further note that in Moore v. Texas (2017) --- U.S. ----, 137 S.Ct. 1039, 197 L.Ed.2d 416 ( Moore ), the United States Supreme Court held that courts must "continue the inquiry and consider other evidence of intellectual disability where an individual's IQ score, adjusted for the test's standard error, falls within the clinically established range for intellectual-functioning deficits." ( Id. at p. 1050.) Petitioner has two IQ scores that fall at or below 70 without adjustment for the standard error of measurement: his score of 70 on the WISC and his score of 67 on the WAIS-III. Petitioner also has a third score with the lower range falling at or below 70 when adjusted. His score of 73 on the WAIS-R yields a range of 68 to 78 or, if rescored as 71, produces a range of 66 to 76. Moore thus dictates that we must also consider petitioner's adaptive functioning.
On this topic, respondent complains the referee evaluated adaptive behavior according to "the most current authority on this subject" instead of the authority specifically endorsed by Atkins or Hawthorne . However, those cases do not preclude reliance on more current AAIDD criteria. The legal determination of intellectual disability is, of course, distinct from a medical diagnosis, but both our decisions and those of the United States Supreme Court contemplate that the determination must be an individualized one, informed **767by the views of medical experts. (See, e.g., Moore, supra, 137 S.Ct. at pp. 1051, 1053 ; Hall, supra, 134 S.Ct. at p. 2000 ; Hawthorne, supra, 35 Cal.4th at p. 49, 24 Cal.Rptr.3d 189, 105 P.3d 552.) As such, the referee had discretion to consider current medical standards in assessing petitioner's adaptive behavior.
The AAIDD's diagnostic manual defines adaptive behavior as "the collection of conceptual, social, and practical skills that have been learned and are performed by people in their everyday lives." (AAIDD-11, supra , p. 45.) Conceptual skills include language and literacy; concepts regarding money, time, and numbers; and self-direction. Social skills include interpersonal skills, social responsibility, self-esteem, gullibility, naïveté, problem solving, and the ability to follow rules and laws and to avoid victimization. Practical skills include personal care, occupational skills, healthcare, travel and transportation, schedules and routine, safety, and use of money and the telephone.15 (AAIDD-11, p. 44.) The *127description of adaptive functioning in the APA's Diagnostic and Statistical Manual V (5th ed. 2013) (DSM-5) is similar. *1202Substantial evidence shows petitioner had significant adaptive difficulties before age 18. Deficits included his functional illiteracy and poor progress in school. His friends and siblings grew up in similar circumstances and attended the same schools, yet appeared to function at a significantly higher level. These facts suggest petitioner was unable, as opposed to unmotivated, to learn. The evidence suggests petitioner may suffer from aphasia, a language difficulty that can be associated with intellectual disability. The professionals who tested petitioner as an adult documented difficulties in understanding and following directions. His verbal IQ scores indicate impairment in problem solving, comprehension, judgment, information processing, and logical thinking.
Respondent urges a focus on petitioner's adaptive strengths, including successful gambling in Las Vegas, supporting himself though illegal activities, maintenance of lasting relationships, and ability to banter with police when questioned. Respondent criticizes Khazanov's assessment as non-standardized and as failing to acknowledge petitioner's adaptive strengths or "street smarts." Respondent argues, "[A]n individual's strengths can cancel out the deficits in other areas." It is well settled, however, that adaptive strengths and weaknesses can coexist with intellectual disability. (See Brumfield v. Cain (2015) 576 U.S. ----, 135 S.Ct. 2269, 2281, 192 L.Ed.2d 356.) The United States Supreme Court has cautioned against overemphasizing perceived adaptive strengths to counter evidence of intellectual disability. The Moore court rejected the view that adaptive strengths constitute evidence adequate to overcome considerable objective evidence of adaptive deficits, noting the medical community focuses the adaptive-functioning inquiry on adaptive deficits . The Moore court observed: " 'significant limitations in conceptual, social, or practical adaptive skills [are] not outweighed by the potential strengths in some adaptive skills.' " ( Moore, supra, 137 S.Ct. at p. 1050 ; see also, AAIDD-11, supra , p. 47; DSM-5, supra , p. 33.) Moreover, as the referee noted, petitioner was not a successful street hustler, having been arrested and convicted numerous times, nor was he adept at deceiving police during his interview.
In sum, the referee's findings as to both intellectual functioning and adaptive behavior are substantially supported and we adopt them. Because he is intellectually disabled, petitioner is ineligible for execution.
IV. DISPOSITION
The petition for a writ of habeas corpus is granted insofar as claim XVIII alleges that petitioner is ineligible for execution because he is intellectually disabled. The judgment in **768People v. Robert Lewis, Jr. (Super. Ct. Los Angeles County, 1984, No. A027897), is vacated to the extent it imposes a sentence of death. *1203In light of the above, we do not address the petition's allegations of ineffective assistance of counsel in claims XIV, XV, and XVI. The order to show cause on those claims is discharged. *128The petition's remaining claims will be resolved by later order to be filed separately, as is our practice. (See In re Scott (2003) 29 Cal.4th 783, 829, 129 Cal.Rptr.2d 605, 61 P.3d 402.)
WE CONCUR:
CANTIL-SAKAUYE, C. J.
CHIN, J.
LIU, J.
CUÉLLAR, J.
KRUGER, J.
CHAVEZ, J.*