dissenting. The Supreme Court has repeatedly pronounced that a valid waiver of Miranda rights must be “a knowing and intelligent relinquishment or abandonment of a known right or privilege.” Edwards v. Arizona, 451 U.S. 477, 482, 101 S.Ct. 1880, 68 L.Ed.2d 378 (1981). The en banc majority opinion creates an entirely new rule that allows defendants to establish a lack of a knowing and intelligent waiver only if they do so at the moment they hear their Miranda rights. Under the majority’s formulation, the primary focus is on the conduct of police officers and whether the officers had reason to know at the time of the interrogation that the defendant lacked the ability to understand the Miranda warning. While it would be impossible to believe that a deaf defendant could give a knowing and intelligent waiver when he is given only a mumbled Miranda warning, the majority’s approach would compel such a conclusion. Demanding more of those whose mental or physical faculties can do only less would surely defeat the purpose of requiring that defendants understand the waiver when they undertake it. Because I believe the majority’s approach is contrary to Supreme Court precedent and that the totality of the circumstances demonstrates that Garner’s waiver of his Miranda rights was not knowing or intelligent, I respectfully dissent.
I. LEGAL STANDARDS GOVERNING THE VALIDITY OF WAIVERS
The majority contends that the “primary focus” in determining whether a defendant’s waiver was knowing and intelligent is “on what the interrogating officers could have concluded about Garner’s ability to *274understand the warnings.” Maj. Op. at 263. Applying this approach, the majority concludes that “[bjecause police had no reason to believe that Garner misunderstood the warnings, and because it is undisputed that the officers were otherwise reasonable and careful in giving the warnings and obtaining the confession, there is no basis for invalidating Garner’s Miranda waiver.” Id. The majority’s focus on the conduct and knowledge of police officers is at odds with the Supreme Court’s repeated pronouncements that the proper inquiry is whether the defendant had the maturity, intelligence, and competency to make a knowing and intelligent waiver.
The Fifth Amendment states that “[n]o person ... shall be compelled in any criminal case to be a witness against himself.” U.S. Const. amend. V. In Miranda v. Arizona, 384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694 (1966), the Supreme Court determined that the right against self-incrimination “is fully applicable during a period of custodial interrogation.” Id. at 461, 86 S.Ct. 1602. The Miranda Court further determined that “the right to have counsel present at the interrogation is indispensable to the protection of the Fifth Amendment privilege.” Id. at 469, 86 S.Ct. 1602. Moreover, the Court held that, prior to custodial interrogation, a suspect must be informed of these rights, now commonly known as the Miranda rights. Id. at 444, 86 S.Ct. 1602 (“Prior to any questioning, the person must be warned that he has a right to remain silent, that any statement he does make may be used as evidence against him, and that he has a right to the presence of an attorney, either retained or appointed.”). Of special import here, the Miranda Court noted that “[tjhe defendant may waive effectuation of these rights, provided the waiver is made voluntarily, knowingly and intelligently.” Id. (emphasis added).
Subsequent decisions by the Supreme Court have further clarified that the validity of a waiver depends on it being made not only “voluntarily,” but also “knowingly and intelligently.” In Moran v. Burbine, for example, the Court stated:
The inquiry has two distinct dimensions. First, the relinquishment of the right must have been voluntary in the sense that it was the product of a free and deliberate choice rather than intimidation, coercion, or deception. Second, the waiver must have been made with a full awareness of both the nature of the right being abandoned and the consequences of the decision to abandon it. Only if the “totality of the circumstances surrounding the interrogation” reveal both an uncoerced choice and the requisite level of comprehension may a court properly conclude that the Miranda rights have been waived.
Burbine, 475 U.S. at 421, 106 S.Ct. 1135 (quoting Fare v. Michael C., 442 U.S. 707, 725, 99 S.Ct. 2560, 61 L.Ed.2d 197 (1979)) (emphasis added) (citations omitted); see also Colorado v. Spring, 479 U.S. 564, 573-75, 107 S.Ct. 851, 93 L.Ed.2d 954 (1987) (analyzing separately whether a suspect’s waiver of his Miranda rights was voluntary and whether it was knowing and intelligent); Edwards v. Arizona, 451 U.S. 477, 482, 101 S.Ct. 1880, 68 L.Ed.2d 378 (1981) (“It is reasonably clear under our cases that waivers of counsel must not only be voluntary, but must also constitute a knowing and intelligent relinquishment or abandonment of a known right or privilege .... ”). Garner does not argue that he waived his Miranda rights involuntarily, but he does argue that he waived his rights unknowingly and unintelligently.
Whether a suspect’s waiver of Miranda rights is “a knowing and intelligent relinquishment or abandonment of a known right or privilege” is “a matter which de*275pends in each case ‘upon the particular facts and circumstances surrounding that case, including the background, experience, and conduct of the accused.’ ” Edwards, 451 U.S. at 482, 101 S.Ct. 1880 (quoting Johnson v. Zerbst, 304 U.S. 458, 464, 58 S.Ct. 1019, 82 L.Ed. 1461 (1938)). A court must examine the “totality of the circumstances” to determine whether a suspect’s waiver was knowing and intelligent, including inquiries into the suspect’s “age, experience, education, background, and intelligence, and into whether he has the capacity to understand the warnings given him, the nature of his Fifth Amendment rights, and the consequences of waiving those rights.” Michael C., 442 U.S. at 725, 99 S.Ct. 2560. “The Constitution does not require that a criminal suspect know and understand every possible consequence of a waiver of the Fifth Amendment privilege,” but does require “that a suspect know[ ] that he may choose not to talk to law enforcement officers, to talk only with counsel present, or to discontinue talking at any time.” Spring, 479 U.S. at 574, 107 S.Ct. 851; see also Burrbine, 475 U.S. at 421, 106 S.Ct. 1135 (“[T]he waiver must have been made with a full awareness of both the nature of the right being abandoned and the consequences of the decision to abandon it.”).
The majority’s approach conflicts with the Supreme Court’s repeated pronouncements that the proper inquiry is whether the defendant actually had the capability to make a knowing and intelligent waiver, see, e.g., Edwards, 451 U.S. at 482, 101 S.Ct. 1880; Michael C., 442 U.S. at 725, 99 S.Ct. 2560; Zerbst, 304 U.S. at 464, 58 S.Ct. 1019, without any reference to police conduct. To suggest, as the majority does, that the validity of a Miranda waiver depends only on the conduct of the police— or what the police knew or should have known at the time — is to read the requirement that a valid waiver be “a knowing and intelligent relinquishment or abandonment of a known right or privilege,” Edwards, 451 U.S. at 482, 101 S.Ct. 1880, out of the Supreme Court’s Miranda jurisprudence. Indeed, under the majority’s formulation, even a suspect who did not hear his Miranda rights being read somehow could give a knowing and intelligent waiver, so long as the police had no reason to believe that the suspect did not hear.
To support its focus on police conduct, the majority relies heavily upon Rice v. Cooper, 148 F.3d 747 (7th Cir.1998), cert. denied, 526 U.S. 1160, 119 S.Ct. 2052, 144 L.Ed.2d 218 (1999). In Rice, the Seventh Circuit held that, because there was no police abuse, a sixteen-year-old made a knowing and intelligent waiver of his Miranda rights despite the testimony of two psychologists that he had been mentally incompetent to make a valid waiver. Id. at 749-51. The Seventh Circuit read the Supreme Court’s decision in Colorado v. Connelly, 479 U.S. 157, 107 S.Ct. 515, 93 L.Ed.2d 473 (1986), for the proposition that a defendant’s waiver of his Miranda rights cannot be unknowing or unintelligent unless there is coercive police activity or the police had some reason to believe that the defendant was incapable of making a rational waiver.
But although the Supreme Court in Connelly held “that coercive police activity is a necessary predicate to the finding that a confession is not ‘voluntary,’ ” it did not suggest that coercive police activity is a necessary predicate to a conclusion that a waiver of Miranda rights was not knowing or intelligent. Id. at 167, 107 S.Ct. 515; see also United States v. Turner, 157 F.3d 552, 555 (8th Cir.1998); United States v. Bradshaw, 935 F.2d 295, 299 (D.C.Cir.1991) (“We read Connelly ... as holding only that police coercion is a necessary prerequisite to a determination that a waiver was involuntary and not as bearing *276on the separate question whether the waiver was knowing and intelligent.”); Miller v. Dugger, 838 F.2d 1530, 1539 (11th Cir.) (“We do not read the Connelly decision as demonstrating an intent to eliminate this distinction between voluntariness and knowing waivers.”), cert. denied, 486 U.S. 1061, 108 S.Ct. 2832, 100 L.Ed.2d 933 (1988). Indeed, the Connelly Court noted that an expert witness “testified that Con-nelly’s illness did not significantly impair his cognitive abilities. Thus, respondent understood the rights he had when [the police] advised him that he need not speak.” Connelly, 479 U.S. at 161-62, 107 S.Ct. 515. Further, the Seventh Circuit in Rice acknowledged that its focus on police conduct diverged from pre-Connelly Supreme Court precedent, or what it called “the conventional approach to waivers of the Miranda rights — that of asking simply whether the defendant had the maturity, competence, etc. to make a knowing waiver of his rights, without reference to what the police knew or should have known.” Rice, 148 F.3d at 751 (citations omitted).
Put simply, the majority completely fails to account for the clear directive of the Supreme Court’s Miranda jurisprudence that the proper inquiry is whether the defendant had the maturity, intelligence, and mental capacity to make a knowing and intelligent waiver. I recognize that the Supreme Court’s requirement that a Miranda waiver be made knowingly and intelligently may, on occasion, put the police in the difficult position of having to assess a suspect’s understanding and intellectual capacities at the time of interrogation. This difficulty is not wholly unique, however, as courts face similar difficulties, for example, when assessing a defendant’s competency and understanding during a plea colloquy or when a defendant waives the right to counsel. Suspicions that a suspect’s initial Miranda waiver was not made knowingly and intelligently also do not preclude the police from interrogating the suspect later under different circumstances — for example, following evaluation by a mental-health professional, following treatment, or in the presence of a lawyer, see, e.g., In re B.M.B., 264 Kan. 417, 955 P.2d 1302, 1309-13 (1998) — if the police desire greater assurances that the suspect’s statement will be deemed admissible at trial.
II. RELEVANT FACTS
Having determined that the proper focus should be on Garner’s “age, experience, education, background, and intelligence,” Michael C., 442 U.S. at 725, 99 S.Ct. 2560, rather than on the conduct of the police, I now turn to analyzing these factors. Because I believe the majority fails to give adequate consideration to the factual record bearing on these factors, I briefly consider the relevant facts. Garner was nineteen years old at the time of the offense. He was “the product of a very abusive and disorganized family of origin.” 2 Joint Appendix (“J.A.”) at 513 (Schmidt-goessling Report at 3). Garner endured physical abuse at the hands of his mother and more than one of her boyfriends, suffered sexual abuse at the hands of an older brother, was left with his siblings to provide food and clothing for himself, and was repeatedly kicked out of his home. Garner’s mother testified that Garner and his twin brother attended the first few years of school together in the same class, but that they were thereafter separated because Garner’s brother had been doing Garner’s work for him. Thereafter, Garner “didn’t do very well” in school. 3 J.A. at 1028 (Mitigation Hr’g 10/13/92 at 52 (Patricia Garner Test.)). Garner told the police that he could read and had completed the twelfth grade, but his mother testified that the last grade that he completed was the seventh grade, and both his mother and school records indicated that Gar*277ner’s grades were always poor, that he was held back at least once, that he was frequently absent from school, and that he was placed in a variety of correctional or treatment-focused schools. According to his mother, Garner had at least one encounter with the juvenile court system. In 1992, the year of the offense, Garner had a full-scale Wechsler Adult Intelligence Scales-Revised IQ score of 76, placing him in the borderline range of intellectual functioning, as well as signs of a learning disability, attention deficit disorder, and organic brain impairment.1
The circumstances of Garner’s interrogation are also relevant. On January 26, 1992, police executed a search warrant at 3260 Burnet Avenue and arrested Garner. Officer Harry G. Frisby, Jr. (“Frisby”), of the Cincinnati Police Department advised Garner of his Miranda rights, and Garner said that he understood his rights.2 Officer Frisby asked Garner about several items that Officer Frisby believed had been stolen, but Garner said that the items were his. Garner was then taken to the police station.
At the police station, Officer Frisby and Officer David Feldhaus (“Feldhaus”) interrogated Garner. Officer Feldhaus advised Garner of his Miranda rights again, read a waiver-of-rights form to Garner, and Garner, Officer Frisby, and Officer Feldhaus signed the form.3 The two officers pro*278ceeded to interrogate Garner. Officer Fel-dhaus testified that Garner appeared “perfectly normal” and “very coherent” and that Garner answered when questioned that he was not under the influence of drugs or alcohol. 3 J.A. at 944 (Suppression Hr’g at 204 (Feldhaus Test.)). Officer Frisby testified that Garner initially denied any involvement with the crimes and that he, Officer Frisby, repeatedly told Garner that he thought Garner was lying. After approximately forty minutes, the two officer began tape recording the interrogation, and Garner confessed to stealing items from 3250 Burnet Avenue and setting a fire.
I now turn to considering “whether [Garner] ha[d] the capacity to understand the warnings given him, the nature of his Fifth Amendment rights, and the consequences of waiving those rights.” Michael C., 442 U.S. at 725, 99 S.Ct. 2560. On collateral review in state court, Dr. Jeffrey Smalldon (“Smalldon”), a mental-health expert appointed by the state trial court to assist with the defense, submitted an affidavit regarding a number of issues. Dr. Smalldon stated that he had personally interviewed, tested, and assessed Garner in addition to reviewing reports from Dr. Nancy Schmidtgoessling (“Schmidtgoess-ling”), who was appointed by the state trial court to assess Garner’s competency to stand trial, and Dr. Joseph D. Schroeder (“Sehroeder”), a clinical neuropsychologist who further assessed Garner because of concerns raised by Dr. Schmidtgoessling. Regarding the issue at hand, Dr. Smalldon concluded that “Mr. Garner’s borderline intelligence, functional (i.e., organic) brain impairment, abusive and socially deprived background, and long history of impulsivity raise serious questions as to whether he could or did understand the consequences of signing the ‘Waiver of Rights.’ ” 3 J.A. at 921 (Smalldon Aff. at ¶ 10). Dr. Small-don further concluded that “[t]he same assessment findings alluded to above, as well as my own clinical impressions, also raise serious questions about whether he had the ability to understand and appreciate the implications of the language used in the ‘Waiver of Rights’ form that he signed.” 3 J.A. at 921 (Smalldon Aff. at ¶ 11). Dr. Smalldon opined that “[m]ore focused assessment would provide better, and perhaps even conclusive, information on this issue.” 3 J.A. at 922 (Smalldon Aff. at ¶ 13).4
Dr. Everington provided this more focused assessment regarding Garner’s understanding of his waiver of Miranda rights. Dr. Everington administered the Grisso test, specifically designed to “assess! ] a defendant’s comprehension of the Miranda warnings themselves” and “pro-vid[e] a comparison of the defendant’s performance to that of other defendants of various ages and levels of intelligence.” Thomas Grisso, InstRüments for AssessiNG Understanding & Appreciation of Miranda Rights 4 (1998). The Grisso test includes four separate testing instruments. The first instrument, Comprehension of Miranda Rights (“CMR”),
assesses the examinee’s understanding of the Miranda warnings as measured by the examinee’s paraphrased description of the warnings. The procedure involves presentation of each of the four Miranda warnings, one by one, to the examinee. After each warning is pre*279sented, the examinee is invited to tell the examiner “what that means in your own words.”
Id. at 5. Answers are scored two points for “adequate” responses, one point for “questionable” responses, and zero points for “inadequate” responses, producing a total CMR score between zero and eight. Id.
The second instrument, Comprehension of Miranda Rights — Recognition (“CMR-R”),
assesses the examinee’s understanding of the Miranda warnings as measured by the examinee’s ability to identify whether various interpretations provided by the examiner are the same as or different from the warning that was presented.
As with the CMR, the CMR-R requires that each warning be presented to the examinee. After each warning statement, the examiner asks the exami-nee to listen to three other statements, some of which are the same as the warning and some of which are not the same. The examinee simply says “same” or “different” after each alternative statement.
Id. Answers are scored one point for each correct response, producing a total CMR-R score between zero and twelve. Id.
The third instrument, Comprehension of Miranda Vocabulary (CMV), “assesses the examinee’s ability to define six words that appear in the version of the Miranda warnings on which the Miranda instruments are based. The examiner reads each word, uses it in a sentence, and then asks the examinee to define the word.” Id. Answers are scored two points for “adequate” responses, one point for “questionable” responses, and zero points for “inadequate” responses, producing a total CMV score between zero and twelve. Id. at 5-6.
The fourth instrument, Function of Rights in Interrogation (“FRI”),
assesses the examinee’s grasp of the significance of the Miranda rights in the context of interrogation. For example, some defendants may understand the warning that they have the “right to an attorney,” yet they may fail to appreciate its significance because they do not understand what an attorney does. The FRI, therefore, goes beyond understanding of the Miranda warning themselves to explore examinees’ grasp of the significance of the warnings in three areas:
• Nature of Interrogation: jeopardy associated with interrogation
• Right to Counsel: the function of legal counsel
• Right to Silence: protections related to the right to silence, and the role of confessions
The FRI uses four picture stimuli, which are accompanied by brief vignettes (e.g., a story about a suspect who has been arrested, accompanied by a picture of a young man sitting at a table with two police officers). Each picture and vignette are followed by a set of standardized questions (15 in all) that assess the examinee’s grasp of the significance of the three matters noted previously.
Id. at 6. Answers are scored two points for “adequate” responses, one point for “questionable” responses, and zero points for “inadequate” responses, producing a total FRI score between zero and thirty as well as subscale scores between zero and ten regarding recognition of the nature of interrogation, the significance of the right to counsel, and the significance of the right to silence. Id.
Dr. Everington administered the Grisso test in 1998 when Garner was 26 years old, *280approximately six years after Garner’s interrogation. Garner received a CMR score of six, which “was below that of mentally typical adult subjects as well as below persons in his IQ range.” 1 J.A. at 378 (Everington Report at 9). Garner’s score was slightly below the mean score of thirteen-year-old juvenile delinquents of average intelligence but slightly above the mean score of twelve-year-old juvenile delinquents of average intelligence.5 See GRISSO, supra, at 87 tbl.5. On the CMR-R, Garner received a perfect score of twelve, “indicating that he did not have difficulty in recognizing the meaning of the warning when presented in a true-false format.” 1 J.A. at 378 (Everington Report at 9). On the CMV, Garner had difficulty defining five of the six vocabulary words: consult, attorney, appoint, entitled, and right. Garner received a score of seven, which was “below mentally typical peers and persons in his IQ range,” id., and below the mean score of twelve-year-old juvenile delinquents of average intelligence, see GRISSO, supra, at 88 tbl.6. Finally, Garner received a FRI score of twenty-four, “below that of adult offenders and non offenders.” 1 J.A. at 378 (Ever-ington Report at 9). Dr. Everington further noted that “all the items that [Garner] missed [on the FRI] were in one are[a]— the function of the right to silence' — indicating that he still does [not] have a full understanding of this right, even after six years.” Id. Garner’s right-to-silence FRI subscale score of four was below the mean scores of adult offenders (7.48), adult no-noffenders (6.84), and juvenile delinquents (5.52). See GRISSO, supra, at 93 tbl.ll. Dr. Everington concluded that the test results “indicate[d] that [Garner] does not have full comprehension of Miranda warnings or his right to remain silent.” 1 J.A. at 373 (Everington Aff. at ¶ 17).
III. ANALYSIS
Given Garner’s low IQ scores and other mental disabilities, I now turn to considering carefully whether Garner knowingly and intelligently waived his Miranda rights. As the majority notes, along with other courts, we have rejected calls to establish a categorical rule that a low IQ or other significant limitations in intellectual functioning are dispositive and make a suspect with such characteristics categorically unable to give a valid waiver of Miranda rights. Maj. Op. at 264-65. However, we also have not established a categorical rule that an express waiver from a person with a low IQ or other significant limitations similar to Garner’s is always knowing and intelligent. Moreover, other courts have concluded that suspects with similar limitations in intellectual functioning did not knowingly and intelligently waive their Miranda rights in particular circumstances. See, e.g., United States v. Garibay, 143 F.3d 534, 538-39 (9th Cir.1998) (concluding that a suspect with an IQ score that placed him in the borderline range of intellectual functioning did not knowingly and intelligently waive his Miranda rights); Cooper v. Griffin, 455 F.2d 1142, 1144-46 (5th Cir.1972) (concluding that two teenage suspects with IQs between 61 and 67 did not knowingly and intelligently waive their Miranda rights); United States v. Aikens, 13 F.Supp.2d 28, 34 (D.D.C.1998) (concluding *281that a suspect with an IQ of 71 did not knowingly and intelligently waive his Miranda rights); State v. Caldwell, 611 So.2d 1149, 1152 (Ala.Crim.App.1992) (affirming the trial court’s ruling that a suspect with an IQ of 71 did not knowingly and intelligently waive her Miranda rights), cert. denied, 510 U.S. 904, 114 S.Ct. 284, 126 L.Ed.2d 234 (1993); People v. Bernasco, 138 Ill.2d 349, 150 Ill.Dec. 155, 562 N.E.2d 958, 963-66 (1990) (affirming the trial court’s ruling that a 17-year-old suspect with an IQ of 80 did not knowingly and intelligently waive his Miranda rights), cert. denied, 500 U.S. 932, 111 S.Ct. 2052, 114 L.Ed.2d 458 (1991), abrogated on other grounds by People v. G.O. (In re G.O.), 191 Ill.2d 37, 245 Ill.Dec. 269, 727 N.E.2d 1003, 1010 (Ill.2000).
Precedent also provides more specific guidance for our inquiry in this case. Those cases in which a court decided that a suspect with mental disabilities knowingly and intelligently waived his or her Miranda rights generally exhibit one or both of two important characteristics not found in this case. In a number of cases, the suspect produced expert evidence of mental disabilities, but did not produce any expert evidence that those disabilities made him or her incapable of knowingly and intelligently waiving Miranda rights or that he or she did not give a valid waiver in that particular instance. See, e.g., Finley v. Rogers, 116 F. App’x 630, 636-38 (6th Cir.2004) (unpublished opinion); United States v. Turner, 157 F.3d 552, 555 (8th Cir.1998); United States v. Male Juvenile, 121 F.3d 34, 40 (2d Cir.1997); Correll v. Thompson, 63 F.3d 1279, 1288 (4th Cir.1995), cert. denied, 516 U.S. 1035, 116 S.Ct. 688, 133 L.Ed.2d 593 (1996); Dunkins v. Thigpen, 854 F.2d 394, 398-400 (11th Cir.1988), cert. denied, 489 U.S. 1059, 109 S.Ct. 1329, 103 L.Ed.2d 597 (1989).6 In those cases in which the suspect did produce specific expert evidence, typically at least one expert, usually the state’s but sometimes even the suspect’s, countered the assertion that the suspect did not knowingly and intelligently waive his or her Miranda rights. See, e.g., Clark v. Mitchell, 425 F.3d 270, 275 (6th Cir.2005); Taylor v. Rogers, No. 95-3904,1996 WL 515349, at *3 (6th Cir. Sept.10, 1996) (unpublished opinion); Young v. Walls, 311 F.3d 846, 849 (7th Cir.2002); People v. Jenkins, 122 Cal.App.4th 1160, 19 Cal. Rptr.3d 386, 395 (2004). But see Smith v. Mullin, 379 F.3d 919, 932-34 (10th Cir.2004).
In the case at hand, in contrast, Dr. Everington offered her unrebutted expert opinion that Garner “does not have full comprehension of Miranda warnings or his right to remain silent.” 1 J.A. at 373 (Everington Aff. at ¶ 17). Although the state did not counter that evidence with expert evidence to the contrary, the majority concludes that because of the limitations of the Grisso test, Dr. Everington’s affidavit and report “do not provide sufficient evidence that Garner’s waiver was not knowing and intelligent.” Maj. Op. at 270. First, the majority notes that the Grisso test measured Garner’s understanding of the Miranda warnings at the time of the test, in 1998, and not at the .time of his interrogation, in 1992. Maj. Op. at 269-70. However, the Grisso test manual does not indicate that it is reasonable to assume that Garner understood the Miranda warnings better at the time of his *282interrogation than he did at the time of the test. The manual lists a number of factors that Dr. Everington was to take into account in making a retrospective determination, see GRISSO, supra, at 71-72, and Dr. Everington concluded that “[i]n [her] professional opinion, it is reasonable to assume that he would not have comprehended the warnings any better under the highly stressful conditions present during the interrogation prior to trial.” 1 J.A. at 373 (Everington Aff. at ¶ 17). Moreover, study results indicate that scores on the Grisso test are positively correlated with age — that is, one would generally expect Garner’s Grisso test scores to be higher in 1998 than in 1992. See GRISSO, supra, at 83 tbl. 1, 87 tbl. 5, 88 tbl. 6.
Second, the majority makes much of the fact that the CMV subtest of the Grisso test administered to Garner contained different language than the Miranda warnings given to Garner. Maj. Op. at 268. Specifically, in addition to a number of slight differences in language, the Grisso test warnings used, for example, the word “attorney” instead of “lawyer” and “interrogation” instead of “questioning.” GRisso, supra, at 20. However, many of Dr. Ever-ington’s conclusions are unaffected by these differences. First, despite differences in language, “[nevertheless, the comparison of the examinee’s performance to the norms offered in the manual will provide an indication of the examinee’s capacities for understanding relative to other examinees in the research study for which the instruments were developed. Thus comparative interpretations regarding the examinee’s performance relative to people of various ages and levels of intelligence can still be made.” GRISSO, supra, at 7. Garner consistently scored below persons in his age and IQ ranges, indicating that his competence for waiving his Miranda rights as suggested by his general cognitive abilities did not accurately reflect whether he actually knowingly and intelligently did so. Second, although three of the words that Garner could not define as part of the CMV — consult, attorney, and entitled — were not used in the warnings actually given him, Garner could not give a satisfactory definition of two key words common to both the test and the warnings: appoint and right. Third, the Grisso test warnings regarding the right to remain silent were identical in all relevant respects to those given by Officers Frisby and Feldhaus, and Garner’s Grisso test results indicated that Garner had significant difficulties understanding the right to remain silent.7
The majority also contends that Garner’s results on the CMR and FRI sub-tests provide little support for Dr. Ever-ington’s conclusion that Garner did not adequately comprehend the Miranda warnings. Maj. Op. at 266-69. First, the majority dismisses the results of the CMR subtest because Garner gave unsatisfactory responses to only two of the four questions. Dr. Everington, by contrast, credited the CMR results, noting that Garner’s score put him “below that of mentally typical adult subjects as well as below persons in his IQ range.” 1 J.A. at 378 (Evering-ton Report at 9). The majority’s rejection of Dr. Everington’s reliance on the CMR results amounts to second-guessing the *283opinion of an expert in forensic psychology who personally administered the tests to Garner. With respect to the FRI subtest, the majority takes issue with the subtest’s methodology, calling it “subjective and legally questionable.” Maj. Op. at 268-69. Citing no authority for this critique, the majority substitutes its own cursory analysis for expert opinion holding that the FRI subtest can be probative of a defendant’s comprehension of the Miranda warnings. As the creator of the Grisso test has explained, the FRI subtest was developed in consultation with a panel of attorneys and psychologists. See GRISSO, supra, at 12, 45. Moreover, the Grisso test as a whole was subjected to extensive peer review during its development and has been widely accepted in the field of forensic psychology. See id. at 74-76.8
Additionally, the majority gives great weight to evidence tending to show that Garner did knowingly and intelligently waive his Miranda rights. However, this evidence is subject to significant limitations not recognized by the majority. First, the majority credits statements from Dr. Schmidtgoessling that Garner “appeared to be of near average intelligence” and “appeared to be able to understand all questions and material presented to him suggesting that his receptive language is intact.” Maj. Op. at 263-64 (quoting Schmidtgoessling Report at 2). However, Dr. Schmidtgoessling’s report must be read in context. The relevant portion of the report states: “[Garner] appeared to be of near average intelligence by observation. His memory appeared to be intact. He appeared to be able to understand all questions and material presented' to him suggesting that his receptive language is intact.” Schmidtgoessling Report at 2 (emphasis added). In this portion of her report, Dr. Schmidtgoessling was describing only her initial observations, observations later determined to be inaccurate by results from her own tests as well as by tests administered by Dr. Smalldon, Dr. Schroeder, and Dr. Everington, and the majority errs in relying on Dr. Schmidt-goessling’s observations as substantive conclusions. The expert evidence that Garner’s appearance did not accurately reflect his level of intelligence and understanding also undermines any substantial reliance on the police officers’ testimony that Garner appeared to understand the warnings. Cf. Morgan Cloud et al., Words Without Meaning: The Constitution, Confessions, and Mentally Retarded Suspects, 69 U. Chi. L.Rev. 495, 511-14 (2002) (discussing the difficulty in estimating the level of understanding of those with mental disabilities).
Similarly, the majority gives great weight to the fact that the police had no reason to believe that Garner could not understand his Miranda rights and that Garner told the police officers that he understood his Miranda rights. Maj. Op. at 265-66. As I have explained, the majority’s focus on whether the police had reason to believe Garner could not make a knowing and intelligent waiver is at odds with Supreme Court precedent. By focusing on whether Garner exhibited “any outwardly observable indications that he did not understand the warnings,” Maj. Op. at 265-66, the majority’s misguided analysis side*284steps the essential question of whether Garner actually had the intelligence, maturity, etc. to make an intelligent and knowing waiver. With respect to Garner’s statements that he understood his rights and the waiver, Dr. Everington concluded in her report that Garner’s “cognitive and linguistic limitations make the likelihood of misunderstanding and suggestibility to input from others greater than with mentally typical individuals.” 1 J.A. at 379 (Everington Report at 10); see also Cloud et ah, 69 U. Chi. L.Rev. at 511-12 & n. 76 (describing how people with mental disabilities are “unusually susceptible to the perceived wishes of authority figures”). Thus, although Garner’s statements of understanding are evidence that he knowingly and intelligently waived his Miranda rights, see, e.g., Turner, 157 F.3d at 555, the probative value of this evidence is limited by Dr. Everington’s expert evidence. Furthermore, although Garner was advised of his Miranda rights twice, repetition of the warnings was unlikely to be of any value if he did not understand them the first time, and warnings given after a suspect has already spoken once with police are often ineffective regardless of the suspect’s cognitive abilities. See Missouri v. Seibert, 542 U.S. 600, 611-14, 124 S.Ct. 2601, 159 L.Ed.2d 643 (2004) (plurality opinion).
In sum, the evidence shows that Garner was nineteen years old at the time of his interrogation and had a very poor education, an IQ of 76, and other significant limitations in intellectual functioning, including limitations directly related to the understanding and comprehension of his Miranda rights. Specifically, Dr. Ever-ington’s unrebutted expert evidence indicated that Garner could not satisfactorily define the word “right” and did not understand the right to remain silent. Similar evidence has led other courts to conclude that suspects did not knowingly and intelligently waive their Miranda rights. See Aikens, 13 F.Supp.2d at 32, 34 (suppressing a statement from a suspect with an IQ of 71 because he did not understand the right to remain silent or that he was entitled to have a lawyer present during questioning, despite the fact that police officers went over each warning with him one by one); Bernasco, 562 N.E.2d at 962-63 (affirming a trial court’s ruling suppressing a statement from a suspect with an IQ of 80 because he did not understand the word “right” and other words contained in the Miranda warnings, although he did understand the right to remain silent). But see Mullin, 379 F.3d at 932-34 (concluding on habeas review under AEDPA that a suspect with “mild to borderline mental retardation” gave a knowing and intelligent waiver despite contrary results from a Grisso test administered years after the interrogation). Similarly here, Garner’s young age, indeterminate prior experience with the legal system, poor education, significant limitations in intellectual functioning, and the unrebutted expert evidence all tend to show that Garner’s Miranda waiver was not made knowingly and intelligently. Cf. Michael C., 442 U.S. at 725, 99 S.Ct. 2560 (listing factors to be considered). The only significant evidence to the contrary is the fact that Garner told police at the time of his interrogation that he understood his rights and the waiver, but he has introduced unrebutted expert evidence indicating that this evidence should not be given great weight. Accordingly, I believe that the preponderance of the evidence shows that Garner did not knowingly and intelligently waive his Miranda rights.9 Thus, admission of his statement at trial was unconstitutional.
IV. CONCLUSION
For the reasons described above, I believe that the proper inquiry in determining whether a defendant made a knowing *285and intelligent waiver of his Miranda rights is the defendant’s actual maturity, education, intelligence, and mental competency. I believe the majority’s focus on police conduct, and whether the police had reason to know that a defendant lacked the capacity to make a knowing and intelligent waiver, departs from well-established Supreme Court precedent. Furthermore, I believe that Garner did not knowingly and intelligently waive his Miranda rights before his interrogation. Accordingly, I would reverse the judgment of the district court and remand the case with instructions that the district court issue the writ of habeas corpus. I respectfully dissent.
. Dr. Everington's report, though not admitted by the district court for this purpose, confirmed that Gamer had relatively consistent IQ scores between 76 and 81 as well as significant deficits in language abilities. 1 J.A. at 376-77 (Everington Report at 2-3).
. Officer Frisby testified as follows:
A: Before I said, Mr. Gamer, let me advise you of your rights and I had a booklet that had his rights in it — on the front of it. You have the right to remain silent, that anything you say can be used against you in court. You have the right to talk to a lawyer for advice before we ask you any questions and have him with you during questioning. If you decide to answer questions now without a lawyer present, you still have the right to stop answering at any time. You also have the right to talk to a lawyer before any questioning if you wish. And I asked him if he understood those rights and he said yes.
Suppression Hr’g at 68 (Frisby Test.).
. Officer Feldhaus testified as follows:
Q: Carry us through and see, you know, exactly what was said as best you can remember.
A: Each line?
Q: Yeah.
A: You have a right to remain silent. He said he understood that. Anything you say can be used against you in court.
Q: Did he reply to that?
A: Yes. Do you understand that? Yes. You have the right to talk to a lawyer for advice before we ask you any questions and have him with you during questioning. You understand that? Yes. If you cannot afford a lawyer one will be appointed for you before any questioning if you wish. Understand that? Yes. If you decide to answer questions now without a lawyer present you will still have the right to stop answering at any time. You also have the right to stop answering at any time until you talk to a lawyer. You understand that? The reply was yes.
I then said below that we have a waiver of rights. And I told him, I’ll read this for you.
Q: Pardon me. Did you read the whole paragraph?
A: I said, I have read this statement on rights. I understand what my rights are. I am going to make a statement and answer questions. I do not want a lawyer at this time. I understand and know what I am doing. No promises or threats have been made to me and no pressure or coercion of any kind have been used again [sic] me.. I asked him if he understood that. He said he did. I said, you have any questions about your rights? He replied, no. I said, well, if there’s no questions and you understand it, I need you to sign your name and the time it is. At that time he signed his name. He said, what time is it? I held my wrist watch out and he looked at it, signed the time.
*2783 J.A. at 955-57 (Suppression Hr’g at 215-17 (Feldhaus Test.)).
. The majority discounts the value of Dr. Smalldon’s statements based on this qualifying language. See Maj. Op. at 263, 266. But, as I explain, Dr. Everington answered this call for a more focused assessment by using the Grisso test to evaluate Garner's understanding of his Miranda rights.
. Grisso notes that CMR, CMR-R, and CMV scores “may be compared to norms for delinquent youths and adult offenders of various ages and levels of intelligence,” as provided in a series of tables reporting results from earlier studies. GRISSO, supra, at 5-6; see also id. at 68. FRI and FRI subscale results form earlier studies are not delineated by age and IQ score, but still provide “norms for delinquent youths and adult offenders of various ages.” Id. at 6.
. For instance, in Turner, which the majority cites, Maj. Op. at 261-62, 264-65, the Eighth Circuit held that the defendant gave a valid waiver despite evidence of the defendant's low IQ and mental illness. 157 F.3d at 555. But the defendant in Turner did not produce expert evidence that these disabilities made him incapable of knowingly and intelligently making a waiver of Miranda rights.
. The majority also notes two other limitations of Grisso test, although these limitations need not concern us long. First, an individual may feign misunderstanding or otherwise attempt to give inaccurate responses. Maj. Op. at 270. However, the Grisso test includes internal mechanisms by which to determine whether a subject is feigning misunderstanding, see Grisso, supra, at 70-71, and, as the district court determined, there is no indication that Garner’s Grisso test results are in any way inauthentic. Second, the Grisso test does not measure the ultimate validity of a Miranda waiver. Maj. Op. at 269-70. That, of course, is a question for the court.
. The majority also argues that Dr. Evering-ton may have lacked the experience necessary to reliably administer the Grisso test to Garner in 1998. Maj. Op. at 270 n. 9. In fact, Dr. Everington has coauthored two scholarly articles — one of which was published in 1995— that involved administering the Grisso test to dozens of criminal defendants. See Caroline Everington & Solomon M. Fulero, Competence to Confess: Measuring Understanding and Suggestibility of Defendants with Mental Retardation, 37 Mental Retardation 212 (1999); Solomon M. Fulero & Caroline Everington, Assessing Competency to Waive Miranda Rights in Defendants with Mental Retardation, 19 Law & Hum. Behav. 533 (1995).
. To be clear, I do not argue that a person with Garner’s mental disabilities is categori*285cally unable to knowingly and intelligently waive his Miranda rights, only that the preponderance of the evidence shows that Garner did not do so in this case. Cf. United States v. Macklin, 900 F.2d 948, 952 (6th Cir.) (describing the potential disempowering effect of ruling that people with mental disabilities do not have the capacity to waive legal rights), cert. denied, 498 U.S. 840, 111 S.Ct. 116, 112 L.Ed.2d 86 (1990). Garner may very well have been able to do so under different circumstances — for example, if his rights had been explained to him in very simple terms, see Young, 311 F.3d at 849, or if he had the assistance of a lawyer, social worker, or family member, cf. G.O., 727 N.E.2d at 1021-22 & n. 11 (McMorrow, J., dissenting) (stating that no confession given by a suspect under the age of 15 should be admitted into evidence unless the suspect is permitted to consult with a lawyer, family member, or other adult personally interested in the child’s well-being and listing states that have adopted such a rule); B.M.B., 955 P.2d at 1309-13 (adopting a similar rule and discussing decisions from other states that have also done so).