Defendant, who is intellectually disabled, makes an as-applied challenge to his 75-month mandatory minimum prison sentence for first-degree sexual abuse, ORS 163.427, on the ground that it violates Article I, section 16, of the Oregon Constitution, and the Eighth Amendment to the United States Constitution, which prohibit sentences that are disproportionate to the offense for which they are imposed.
Defendant pleaded guilty to one count of first-degree sexual abuse and three counts of third-degree sexual abuse, for over-the-clothes touching of the sexually intimate parts of nine- and fourteen-year-old victims. At his sentencing, defendant argued that the 75-month minimum sentence for first-degree sexual abuse, which was mandated by ORS 137.700 (2)(a)(P) (Measure 11), would be disproportionate as applied to him. As part of the proportionality analysis, defendant argued that the trial court should take into account his intellectual disability, as well as the availability of residential rehabilitative treatment for him as part of an alternative probationary sentence.
The trial court noted that defendant was intellectually disabled, but the court did not indicate that it had considered that factor in its proportionality analysis, and the court ruled that it lacked authority to consider the availability of rehabilitative treatment for defendant in a non-incarcerative setting, unless it first could conclude that the Measure 11 prison term was disproportionate. The court then concluded that the Measure 11 sentence was not disproportionate. The Court of Appeals affirmed without opinion. State v. Ryan, 275 Or App 22, 364 P3d 1012 (2015). For the reasons explained below, we conclude that the trial court erred when it compared the gravity of defendant’s offense and the severity of the Measure 11 sentence, because the court failed to consider evidence of defendant’s intellectual disability when that evidence, if credited, would establish that the sentence would be arguably unconstitutional because it shows that defendant’s age-specific intellectual capacity fell below the minimum age level of criminal responsibility for a child. However, we decline to consider defendant’s argument *605on review that the availability of rehabilitative treatment is relevant to the gravity of his offense, because defendant failed to adequately develop that argument within the context of this court’s analytical framework for proportionality challenges under Article I, section 16. Accordingly, we remand defendant’s conviction for first-degree sexual abuse for resentencing, and otherwise affirm.
I. FACTS AND PROCEDURAL HISTORY
As we will explain, defendant has intellectual disabilities, as well as attention deficit hyperactivity disorder (ADHD). When he committed the offenses at issue, defendant was on probation for second-degree criminal mischief for having masturbated into an item of clothing in a department store dressing room.
In his plea petition in this case, defendant acknowledged that, on July 15, 2013, he subjected a nine-year-old child, AS, to sexual contact by touching her genital area. The incident occurred at a sleepover birthday party involving a group of defendant’s adult disabled friends. One friend brought his sisters, AS (age nine) and CS (age 14). On the evening of the sleepover, defendant flirted with CS and sent her a text message asking her to join him in the bathroom to kiss. Inside the bathroom, defendant slapped and grabbed CS’s buttocks, ground his penis against her, attempted to expose her breasts by pulling on her dress, and kissed her on the mouth.
The next morning, while most of the party guests were outside, AS went into the house to retrieve her shoes. While inside, she encountered defendant, who was the only other person in the house. Defendant pushed AS to the floor, got on top of her, grabbed her genital area outside her clothing, and ran his hand down her leg. AS told defendant to get off her, and defendant complied, but he then chased her as she tried to run away. AS ended the pursuit by kicking defendant. Afterwards, AS was upset and crying, and she stated at sentencing that she had been very frightened during the incident.
The state charged defendant with three counts of third-degree sexual abuse based on the incident involving *606CS, and one count of first-degree sexual abuse for his conduct with respect to AS. Defendant pleaded guilty to all four charges. However, defendant argued at sentencing that, in view of his intellectual disability, the imposition of a 75-month prison term on the first-degree sexual abuse conviction under Measure 11 would be unconstitutionally disproportionate. In support of that argument, defendant provided the court with written reports from four mental health evaluations performed between 2008 and 2013. All the evaluators diagnosed defendant with intellectual disabilities. The first evaluator reported an IQ score of 50 for defendant, the most recent IQ test scored defendant at 60, and each evaluator found significant impairment in his adaptive functioning.1 Defendant represented to the court that he functioned at an approximate mental age of 10, and the state did not dispute that representation.
More specifically, the first evaluation—performed when defendant was 17 and living in an adolescent group home—was part of an effort to secure services for defendant based on his developmental delay. The evaluator, Dr. Sacks, noted that defendant had a history of striking out at others and that, between 2001 and 2006, he had engaged in misconduct that “seemed to increase in severity.” Sacks diagnosed defendant with Conduct Disorder and Reactive Attachment Disorder and stated that defendant needed a residential setting with highly developed structure to avoid impulsive and dangerous acts.
The second evaluation was performed in 2012, when defendant was 21, to determine whether he was able to aid and assist in his defense on the criminal mischief charge. The evaluator, Dr. Stoltzfus, diagnosed defendant with low cognitive functioning, attention deficit hyperactivity disorder (ADHD), and Conduct Disorder. Stoltzfus reported that *607defendant had been placed in foster care at age 12 for kissing a seven-year-old girl and that he primarily had lived in group home settings between the ages of 12 and 21. In his interview with Stoltzfus, defendant made repeated references to aggression toward people who made him angry. Stoltzfus opined that defendant had a high degree of impul-sivity and reactive hostility that could be ameliorated to some extent with psychotropic medication, but that “[h]is low cognitive and low adaptive functioning are not amenable to treatment and will never change.” Stoltzfus concluded that defendant was not then capable of aiding and assisting his defense. As a consequence, defendant was placed in the Oregon State Hospital for further evaluation and treatment.
In December 2012, Dr. Corbett evaluated defendant at the state hospital. He noted that defendant had been placed in nonrelative foster care for extended intervals between 2005 and 2008, and that he had received services for his developmental delay in Marion County from 2006 to 2009. Defendant stated that he had been so angry at the hospital that he wanted to hit people, but Corbett noted that defendant had made some progress in “competency restoration education.” Corbett opined that defendant had made sufficient progress that he was then able to aid and assist in his defense.
Finally, in December 2013, Dr. Nance evaluated defendant for his sentencing in this case. Nance noted the allegation that defendant had violated his probation on the criminal mischief charge by possessing pornography and engaging in improper internet use. Defendant was in jail at the time of his evaluation and told Nance that he did not feel safe there. He described suicidal and homicidal thoughts, but denied that he would act on them.
Nance diagnosed defendant with limited intellectual functioning and as being immature, paranoid, and depressed, and having questionable judgment. On account of his intellectual deficiencies, defendant was unable to take a useful polygraph examination, which was a concern to Nance, because mandated polygraphs are a primary tool of community supervision. According to Nance, defendant was at high risk for re-offending, because he had a sense *608of sexual entitlement with “rape attitudes,” some hostility toward women, and a lack of concern for others. So far, Nance opined, defendant had expressed an attitude that did not support probation. In Nance’s view, defendant posed a high risk to commit a similar or more serious crime and was not a suitable candidate for community supervision. Defendant’s prognosis for “full rehabilitation” was poor to fair, Nance opined, but good for “some benefit from supportive therapy.” Nance recommended a lengthy course of court-mandated sex offender treatment and stated that the court “may consider” a group home that would administer psychotropic medications and ultimately support defendant’s community supervision. Nance opined that the antisocial aspect of defendant’s disorder would be exacerbated if he were to be incarcerated.
In addition to the evaluators’ reports, defendant presented testimony at sentencing from the director of a residential facility that specialized in treating intellectually disabled sexual offenders. The director, Watson, testified that he had one residential space available to treat defendant as part of an alternative nonincarcerative sentence. Participants living in his program’s homes were supervised, but not in “closed custody,” meaning that “[fit’s not a locked, secure facility.” Watson had not personally evaluated defendant, but he had reviewed reports from other evaluators. He expressed reservations about defendant’s suitability for the program, because, despite the prior existence of “supports,” defendant had not engaged in sex offender treatment. However, Watson stated that he would accept defendant under “certain conditions,” including that “I need to make sure I have the backing of the court,” because,
“based on the evidence that I’ve heard today, [defendant] certain is a—this moderate high-risk guy. And his risk is high under these unsupervised, unstructured environments. What that looks like underneath highly structured and supervised, I don’t know. I’ve not evaluated him and can’t render an opinion until we know more.”
In opposing defendant’s constitutional challenge, the prosecutor “agree [d] with the [d]efense that one of the things that the Court can also look at is the characteristics of both the [d] efendant and the victim. And the State *609does not dispute that this [d] efendant has mental disabilities.” However, the prosecutor asserted that “you don’t look at just one part of the [d] efendant, but rather, again, we get to look at all of his characteristics.” The prosecutor noted that defendant’s own evaluator, Nance, had “place [d] him in the middle of the high risk to reoffend with similar or more serious types of crimes in our community. He’s also thought to not be a good candidate for community supervision.” Because defendant had abused more than one child and had an ongoing problem with controlling his sexual urges, and because he was “impulsive, emotionally agitated, verbally aggressive and abusive, and in need of structure to avoid impulsive and destructive acts,” the prosecutor argued that community safety required his incarceration.
Defendant countered that, in view of his intellectual disability, imposing a 75-month prison sentence on him would shock the moral sense of reasonable people. He requested, instead, that he be sentenced to probation, conditioned on court-mandated residential treatment in Watson’s program. In response, the prosecutor asserted that the availability of alternatives to incarceration had no bearing on the determination whether a particular term of incarceration was constitutionally disproportionate.
The trial court noted that defendant was intellectually disabled. However, the court did not indicate that it had considered that characteristic in resolving defendant’s constitutional challenge. Consistently with the request of the victim CS at sentencing, the court did state that it would like to provide treatment for defendant. However, the court opined that it could not consider the availability of treatment as part of an alternative sentence without first concluding that the prescribed Measure 11 sentence was constitutionally disproportionate.2 Again, without acknowledging *610or indicating that it had considered defendant’s intellectual disability in reaching its conclusion, the trial court then stated that the sentence was not disproportionate. The court therefore sentenced defendant to 75 months’ imprisonment on the first-degree sex abuse conviction as to AS; the court then sentenced defendant to a single consecutive term of six months in prison for the three third-degree sex abuse convictions as to CS (concurrent with each other, but consecutive to the 75-month sentence).
Defendant appealed, raising an as-applied challenge to the constitutionality of the 75-month prison sentence on his first-degree sexual abuse conviction.3 As noted, the Court of Appeals affirmed without opinion.
On review, defendant asserts, based on this court’s framework for analyzing proportionality challenges under Article I, section 16, that the trial court erred in failing to sufficiently consider his intellectual disability in assessing the proportionality of the Measure 11 sentence. Defendant contends, based on the record here, that he deserved less severe punishment than other defendants who commit similar offenses and, consequently, that the mandatory prison term was not properly proportioned to his offense. Defendant also argues that the prescribed sentence is relatively more severe punishment for him than for a normally-abled offender because intellectually disabled offenders are especially vulnerable to abuse and other adverse effects in *611prison. In addition, defendant argued in his opening brief on review that the trial court erred in its application of the framework that this court set out in State v. Rodriguez/Buck, 347 Or 46, 58, 217 P3d 659 (2009)—specifically, in determining the gravity of his offense—by failing to consider the availability of rehabilitative treatment for his disability.4 In light of those factors, defendant asserts that the trial court erred in failing to properly consider his intellectual disability in determining whether the Measure 11 sentence was disproportionate.
In response, the state again acknowledges that defendant’s individual characteristics—including his intellectual disability—are relevant considerations in evaluating the proportionality of his sentence. However, the state asserts that “the availability of alternative treatment options is completely irrelevant to an assessment of the gravity of the offense; indeed, it has nothing to do with a particular defendant’s culpability.” More generally, the state argues that defendant’s focus on treatment is untenable because the constitutional provisions on which he relies do not require courts to choose the least restrictive means of protecting the public, even when an offender is intellectually disabled. To the contrary, the state argues, it is the legislature’s responsibility to prescribe appropriate criminal sanctions, and the state and federal constitutions afford it broad latitude in doing so.
As the state sees it, defendant’s treatment-based argument fails to take into account the sentencing objectives of retribution and deterrence and, generally speaking, would transform an objection to the constitutionality of a mandatory sentence into a full-blown sentencing hearing. Finally, the state urges, consideration of alternative treatment options could yield inconsistent results for similarly culpable offenders, based solely on whether appropriate treatment happens to be available at a particular time or place—an inconsistency, the state argues, that highlights the fallacy of recognizing such a constitutional requirement. It follows, the state asserts, that the sentencing court in this *612case was not authorized to consider the availability of treatment in determining whether the Measure 11 sentence was constitutionally disproportionate.
Alternatively, the state notes that the residential treatment program that defendant proposed was not cus-todially secure and that it was not clear that the program would be suitable for defendant. Under those circumstances, the state argues, the prescribed sentence was not disproportionate. With the parties arguments thus framed, we turn to defendant’s constitutional challenge under the Oregon Constitution. See State v. Newcomb, 359 Or 756, 764, 375 P3d 434 (2016) (describing first-things-first approach).
II. ANALYSIS
A. Article I, Section 16
1. Analytical framework
Article I, section 16, provides, in part:
“Excessive bail shall not be required, nor excessive fines imposed. Cruel and unusual punishments shall not be inflicted, but all penalties shall be proportioned to the offense.”
“For the most part, this court has analyzed the requirement that penalties shall be proportioned to the offense” separately from the related prohibition on cruel and unusual punishments. State v. Althouse, 359 Or 668, 683, 375 P3d 475 (2016). In considering the proportionality requirement, this court ordinarily has asked whether the length of the sentence would shock the moral sense of reasonable people. Id. (citing, e.g., State v. Rogers, 313 Or 356, 380, 836 P2d 1308 (1992), cert den, 507 US 974 (1993) (death penalty for murder committed during the course of attempted first-degree sex abuse “would not shock the moral sense of reasonable people”)). That standard “reflect[s] the principle that the legislature has primary authority to determine the gravity of an offense and the appropriate length of punishment.” Id. at 683-84. Only in those rare instances when the legislature has exceeded that authority may a court conclude that a particular punishment is unconstitutionally disproportionate. Id. at 684.
*613In Rodriguez/Buck, this court considered (as in this case) the as-applied constitutional challenges of two defendants to Measure 11 sentences on single counts of first-degree sexual abuse. 347 Or at 49. One defendant had touched a 13-year-old boy by bringing the back of his head against her clothed breasts for a minute, and the other defendant had caused his hand to touch the clothed buttocks of a 13-year-old girl. On review, this court identified three factors that bear on the proportionality inquiry:
“(1) a comparison of the severity of the penalty and the gravity of the crime; (2) a comparison of the penalties imposed for other, related crimes; and (3) the criminal history of the defendant.”
Id. at 58.
In applying those factors to the cases before it, this court in Rodriguez I Buck held that the trial court in each case properly had declined to impose statutory minimum sentences of 75 months’ imprisonment on the defendants’ convictions for first-degree sexual abuse. With respect to the first factor, the court emphasized that the first-degree sexual abuse statute, ORS 163.427(1), criminalizes a broad range of conduct,
“including, but not limited to, momentary touching of an intimate part without the victim’s awareness or knowledge, touching that the victim apprehends but does not appreciate as sexual, momentary touching over clothing, prolonged hand to genital contact, prolonged skin to skin genital contact, and, of course, forcing a person under 18 to engage in bestiality.”
Id. at 69 (internal quotation marks omitted).
Because some of that criminalized conduct is especially harmful (the court used the example of a 50-year-old man forcing a 13-year-old girl to engage in prolonged skin-to-skin genital contact with him), a mandatory prison term of 75 months’ imprisonment for first-degree sexual abuse does not, on the face of the statute, violate Article I, section 16. Id. “But because the statute also encompasses conduct that reasonable people would consider far less harmful, [the] defendants [were] entitled * * * to argue that the *614mandatory sentence, as applied to the particular facts of their cases, [was] unconstitutionally disproportionate.” Id. at 69-70 (emphasis omitted). In considering the defendants’ challenges, the court compared the defendants’ conduct with other especially harmful conduct that also could be prosecuted as first-degree sexual abuse. Id. at 71. In tentatively concluding (based on its assessment of the first factor) that the defendants’ sentences were disproportionate, the court remarked that the differences between the defendants’ conduct and that more harmful conduct—also punishable by the same mandatory 75-month sentence—-“is obvious to any reasonable person.” Id.
Regarding the second factor, the court stated that, “ [i] f the penalties for more ‘serious’ crimes than the crime at issue result in less severe sentences, that is an indication that the challenged penalty may be disproportionate.” Id. at 63. The court noted that Oregon’s “elaborate listing of sex offenses” provides a useful basis for comparing the conduct constituting the crime and the penalty to other sex crimes, although the inquiry should be limited to “other crimes that have similar characteristics to the crime at issue.” Id. at 65. The court cautioned that courts are not free to “roam * * * through the criminal code, deciding which crimes are more or less serious than others.” Id. at 64.
Finally, the court examined the criminal history of each defendant. Neither of the defendants had prior convictions, and, in the absence of Measure ll’s mandate, their presumptive sentences for first-degree sexual abuse would have been 16 to 18 months’ imprisonment. Id. at 77-78. Even with a more serious criminal record of three or more person felonies, the defendants would have received presumptive sentences of 41 to 45 months in prison. Id. at 78. Taking their criminal histories into account, as well as the other two factors, the court held that imposing Measure 11 sentences would have been unconstitutional under the circumstances. Id. at 79.
2. Standard of review
We review for legal error the trial court’s conclusion that defendant’s sentence was constitutional under Article I, *615section 16. State v. Rangel, 328 Or 294, 298, 977 P2d 379 (1999). In conducting that review, we are bound by any findings of historical fact that the trial court may have made, if they are supported by evidence in the record. State v. Hall, 339 Or 7, 10, 115 P3d 908 (2005), abrogated on other grounds by State v. Unger, 356 Or 59, 333 P3d 1009 (2014).
3. Defendant’s arguments on review
In this case, defendant does not argue that the mandatory Measure 11 sentence was disproportionate based on a comparison of the penalties imposed for other, related crimes, and, apart from noting that he had only one criminal conviction before committing the charged offenses in this case, he makes no focused argument that his criminal history weighs significantly in the proportionality calculus. Instead, to reiterate, defendant has primarily focused on the first Rodriguez/Buck factor and, indeed, more narrowly on the extent to which—in comparing the gravity of his offense and the severity of the prescribed penalty—the trial court was required to consider his intellectual disability. As discussed above, as part of that argument, defendant urged in his opening brief on review that the trial court erred in its application of the Rodriguez/Buck framework—in determining the gravity of his offense—by failing to consider the availability of rehabilitative treatment for his disability.
a. Intellectual disability
We begin with defendant’s overarching argument that the Measure 11 sentence was disproportionate as applied to him because of his intellectual disability. In addressing that contention, we are guided by the analytical framework set out in Rodriguez /Buck. In that case, this court stated that, in addressing a proportionality challenge a sentencing court may consider, among other case-specific factors, the personal characteristics of the defendant as part of the first Rodriguez /Buck inquiry. The full measure of the court’s statement was this:
“We therefore conclude that a defendant’s ‘offense,’ for purposes of Article I, section 16, is the specific defendant’s particular conduct toward the victim that constituted the *616crime, as well as the general definition of the crime in the statute. In considering a defendant’s claim that a penalty is constitutionally disproportionate as applied to that defendant, then, a court may consider, among other things, the specific circumstances and facts of the defendant’s conduct that come within the statutory definition of the offense, as well as other case-specific factors, such as characteristics of the defendant and the victim, the harm to the victim, and the relationship between the defendant and the victim.”
Id. at 62. In short, the gravity of an “offense” refers to the gravity of the defendant’s particular conduct and the statutorily defined crime itself. To the extent that an offender’s personal characteristics influence his or her conduct, those characteristics can affect the gravity of the offense. Id. at 63 (stating that, to determine whether penalty is proportioned to gravity of offense, it is appropriate to consider gravity of instant conduct in comparison with other criminal conduct in light of relative harm to victims and society and offender’s culpability).5
The question remains whether an offender’s intellectual disability is such a characteristic and, if so, how it should be considered. As the parties note, no Oregon case has specifically addressed the application of Article I, section 16, to a sentence for a crime committed by an intellectually disabled offender. However, the United States Supreme Court has discussed relevant considerations in proportionality challenges by intellectually disabled offenders sentenced to death in capital murder cases, and other courts applying the Eighth Amendment have considered the issue with respect to mandatory prison sentences. Because the test for proportionality under the Eighth Amendment is similar to that under Article I, section 16, at least in its comparison of the gravity of the offense and the severity of *617the penalty,6 we now consider the Supreme Court’s decisions and the decisions of other courts following them insofar as they may shed light on our path.
In 2002, the Supreme Court held that the Eighth Amendment prohibits the execution of an intellectually disabled offender. Atkins v. Virginia, 536 US 304, 122 S Ct 2242, 153 L Ed 2d 335 (2002). The Court initially observed that “an IQ between 70 and 75 or lower *** is typically considered the cutoff IQ score for the intellectual function prong of the mental retardation definition.” Id. at 309 n 5. In the wake of Atkins, the Court has made clear that, even for offenders who test above that cut-off IQ score, deficits in the offender’s adaptive functioning are relevant to a determination of intellectual disability. See Moore v. Texas, _ US _, _, 137 S Ct 1039, 1050, 197 L Ed 2d 416 (2017) (requiring courts to “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”); see also Hall v. Florida, _ US _, _, 134 S Ct 1986, 1990, 188 L Ed 2d 1007 (2014) (holding that it is unconstitutional to foreclose “all further exploration of intellectual disability” simply because a capital defendant is deemed to have an IQ above 70).
The Court in Atkins emphasized that “the American public, legislators, scholars, and judges” had deliberated over the question of the death penalty for the intellectually *618disabled and had come to a consensus that it should be prohibited. Atkins, 536 US at 307. The Court noted that, although some states still imposed the death penalty on intellectually disabled individuals convicted of heinous crimes, the consistency of the direction of change was more important than a simple numerical tally. Id. at 315. The Court stated that the practice of executing the intellectually disabled was “uncommon,” id. at 316, but that evidence of consensus, though important, did not “wholly determine” the matter, insofar as the Court was required to bring its own judgment to bear by asking whether there was reason to disagree with the judgment reached by the citizenry and its legislators, id. at 312-13.
The Court concluded that, for an intellectually disabled offender, the case for retribution was diminished. Id. at 319. Further, it stated, the rationale of deterrence was diminished by the reduced ability of the intellectually disabled “to understand and process information, to learn from experience, to engage in logical reasoning, [and] to control impulses.”7 Id. at 320.
If left to case-by-case determinations, the Court opined, there was “[t]he risk that the death penalty [would] be imposed in spite of factors which may call for a less severe penalty.” Id. at 320 (internal quotation omitted). The Court explained:
“Those [intellectually disabled] persons who meet the law’s requirements for criminal responsibility should be tried and punished when they commit crimes. Because of their disabilities in areas of reasoning, judgment, and control of their impulses, however, they do not act with the level of moral culpability that characterizes the most serious adult criminal conduct. Moreover, their impairments can jeopardize the reliability and fairness of capital proceedings against [so diagnosed] defendants.”
*619Id. at 306-07. Viewing an intellectually disabled offender’s culpability in light of the “penological purposes served by the death penalty,” the Court determined that such defendants “should be categorically excluded from execution.” Id. at 317-18. Concerning retribution, the Court found that, because “severity of the appropriate punishment necessarily depends on the culpability of the offender[,] *** an exclusion for the [intellectually disabled] is appropriate.” Id. at 319. Culpability also was central to the Court’s determination that execution of the intellectually disabled did not serve the penological purpose of deterrence, because “it is the same cognitive and behavioral impairments that make these defendants less morally culpable.” Id. at 320. “Construing and applying the Eighth Amendment in the light of our ‘evolving standards of decency,’ [the Court] therefore conclude [d] that such punishment is excessive and that the Constitution ‘places a substantive restriction on the State’s power to take the life’ of [an intellectually disabled] offender.” Id. at 321 (citation omitted).
Despite the Court’s pronouncements in Atkins about the reduced justification for treating intellectually disabled offenders the same as other offenders, lower courts faced with Atkins-haseá. challenges generally have held that Atkins applies only to offenders otherwise subject to death penalty sentences.8 Some of those decisions have not provided extended explanations for why the Atkins rationale should not apply to true-life or other long-term prison sentences, but, in general, they have relied on the notion that the death penalty is fundamentally different from other sentences. See, e.g., U.S. v. Shields, 480 F App’x 381, 389 (6th Cir), cert den, 568 US 923, 133 S Ct 381 (2012) (so stating).
One commentator has explained the reluctance of courts to extend the Atkins rationale beyond the death penalty context in this way:
*620“The conclusion that Atkins seems to play no role in non-capital sentencing should not, perhaps, be wholly surprising. It may be explained by the notion of incapacitation. After all, a disabled defendant spared death under Atkins is still imprisoned, likely for the rest of his life. Using the Atkins rationale in a non-capital sentencing situation may result in a defendant being imprisoned for a significantly shorter period of time. True, but if one believes what the Court wrote in Atkins about mentally disabled defendants being less culpable than others, such a result should be applauded, not avoided. And, if we are to be serious about the application of Atkins, sentencing statutes and guidelines ought to expressly take account of intellectual disabilities, and sentencing judges should also be required to refer specifically to low intelligence of offenders in passing sentences. Such changes would be a welcome recognition of the wisdom of Atkins beyond the death penalty prosecution.”
Paul Marcus, Does Atkins Make a Difference in Non-Capital Cases? Should It?, 23 Wm & Mary Bill Rts J 431, 465 (2014) (footnote omitted). The author further observes:
“Just about everyone working in the field who speaks to the matter seems to believe that the diminished intelligence of the offender ought to be a major factor in determining appropriate sentences. And I do mean just about everyone[.]”
Id. at 456 (footnote omitted). Among those authorities is the American Bar Association Criminal Justice Section. See ABA Criminal Justice Standards Committee, ABA Criminal Justice Mental Health Standards § 7-9.3, 472 (1989) (“Evidence of mental illness or mental retardation should be considered as a possible mitigating factor in sentencing a convicted offender”). See also OAR 213-008-0002(1)(a)(C) (listing defendant’s reduced mental capacity as a potential mitigating factor in sentencing).
We acknowledge the force of that view. The Supreme Court in Atkins repeatedly emphasized the relevance of intellectual disability in determining both the gravity of an offense and the severity of its penalty, 536 US at 319-20, and this court expressly stated in Rodriguez/Buck that an offender’s personal characteristics are relevant in making a proportionality determination, 347 Or at 62. Evidence of an offender’s intellectual disability therefore is relevant *621to a proportionality determination where sentencing laws require the imposition of a term of imprisonment without consideration of such evidence. Accordingly, we conclude that, where the issue is presented, a sentencing court must consider an offender’s intellectual disability in comparing the gravity of the offense and the severity of a mandatory prison sentence on such an offender in a proportionality analysis under Rodriguez/Buck. See id. at 62-63.9
The question remains how that consideration should affect the proportionality analysis. Because there exists a broad spectrum of intellectual disabilities that may reduce, but not erase, a person’s responsibility for her crimes, see Atkins, 536 US at 318 (a defendant’s mental “deficiencies do not warrant an exemption from criminal sanctions”), a one-size-fits-all approach is not appropriate. For that reason, a sentencing court’s findings, among other factual considerations, as to an intellectually disabled offender’s level of understanding of the nature and consequences of his or her conduct and ability to conform his or her behavior to the law, will be relevant to the ultimate legal conclusion as to the proportionality—as applied to the offender—of a mandatory prison sentence. See id. at 319 (holding that “severity of the appropriate punishment necessarily depends on the culpability of the offender”). The length of the prescribed prison sentence also is relevant in determining the severity of the penalty.10
In so concluding, we recognize the inherent tension between considering the personal characteristics of an offender and other case-specific factors in conducting a comparison of the gravity of an offense and the severity of a penalty, and the principle that a proportionality review must be “informed by objective factors to the maximum possible extent.” Harmelin, 501 US at 1000 (internal quotation marks and citations omitted). Indeed, the metrics that are available for making such comparisons are far from precise. *622See id. at 1001 (recognizing that Court “lack[ed] clear objective standards to distinguish between sentences for different terms of years”); see also Solem v. Helm, 463 US 277, 294, 103 S Ct 3001, 77 L Ed 2d 637 (1983) (“It is clear that a 25-year sentence generally is more severe than a 15-year sentence, but in most cases it would be difficult to decide that the former violates the Eighth Amendment while the latter does not.” (Footnote omitted.)).11
However, those difficulties are inherent in a proportionality test that asks whether a particular sentence for a particular offender would shock the moral sense of reasonable people. The fact that a comparison of the gravity of an offense and the severity of its penalty involves factual considerations does not mean that it is unmoored in principle. Nor do challenges posed by the application of such a test justify rejecting it. The Court in Solem and this court have pointed to various factors that can be assessed relatively objectively. In instructing a court to judge the gravity of the offense, this court
“assumes that courts are competent to judge the gravity of an offense, at least on a relative scale. In a broad sense, this assumption is justified, and courts traditionally have made these judgments-—-just as legislatures must make them in the first instance. Comparisons can be made in light of the harm caused or threatened to the victim or society, and the culpability of the offender.”
Rodriguez/Buck, 347 Or at 63 (quoting Solem, 463 US at 292-93).
b. Evidence of treatment options
We briefly turn to defendant’s more particularized treatment-based argument. Before the Court of Appeals, defendant argued that the trial court erred in failing to consider evidence of “an available treatment option that would be more effective and appropriate because of defendant’s condition,” but defendant did not explain how that argument *623comported with the Rodriguez/Buck framework. Later, as noted, in his opening brief before this court defendant argued that the availability of rehabilitative treatment as part of an alternative sentence was relevant to the gravity of his offense, but he has not sufficiently explained how that argument comports with the Rodriguez/Buck framework to permit a carefully considered analysis of it.12 Accordingly, we decline to address it, and we express no opinion as to whether and, if so, how consideration of treatment options for an intellectually disabled offender as part of an alternative sentence could be relevant to a proportionality challenge under Article I, section 16.
III. APPLICATION
Defendant argued before the trial court, and he has consistently argued on appeal and review, that the prescribed Measure 11 sentence in this case was unconstitutionally disproportionate as applied to him because of his intellectual disability. For the following reasons, we conclude that the trial court failed to sufficiently consider defendant’s intellectual disability in addressing his proportionality challenge.
As discussed, the undisputed evidence at sentencing showed that defendant is an intellectually disabled offender who has an IQ score between 50 and 60, a full 10 to 20 points below the cutoff IQ score for the intellectual function prong of the intellectual disability definition recognized in Hall. See _ US at _, 134 S Ct at 2000. Moreover, it is undisputed that defendant has significantly impaired adaptive functioning, such that he functions—as it pertains to standards of maturation, learning, personal independence, and social responsibility—at an approximate mental age of 10, two years below the minimum age for establishing criminal responsibility of a child under Oregon law. See *624ORS 161.290(1).13 That legislative pronouncement is relevant here because it is objective evidence of a societal standard that eschews treating persons with the attributes of a preteen child as if they were normally abled adult offenders. Cf. Haugen v. Kitzhaber, 353 Or 715, 744, 306 P3d 592 (2013), cert den, _ US _, 134 S Ct 1009 (2014) (citing Graham, 560 US at 61, for proposition that, in proportionality challenge under Eighth Amendment, court considers “objective indicia of society’s standards, as expressed in legislative enactments and state practice to determine whether there is a national consensus against the sentencing practice at issue”).
To be sure, the trial court generally noted defendant’s intellectual disability, but, as discussed, the court did not address its implications in rejecting his proportionality challenge. That missing linkage is problematic, because it suggests that, although the court appeared to grasp the factual foundation of defendant’s argument, it did not fully appreciate its constitutional implications.
In the absence of express findings, we ordinarily would presume that the trial court resolved factual disputes consistently with its ultimate decision. Ball v. Gladden, 250 Or 485, 487, 443 P2d 621 (1968). Thus, in the absence of an indication that the court misapprehended the import of defendant’s argument, we would (as noted above) review its factual findings for sufficient evidence in the record, and then determine whether the court correctly applied legal principles to those facts. However, where, as here, the court did not address defendant’s intellectual disability in comparing the gravity of defendant’s offense with the severity *625of the Measure 11 sentence, we would have to speculate to conclude that the court properly considered that factor and made any related factual findings with respect to it.14
As noted, evidence of defendant’s intellectual disability, if credited, would establish that defendant’s age-specific intellectual capacity, including his level of adaptive functioning, fell below the minimum age level for the imposition of criminal responsibility. Permissible inferences that the sentencing court could draw from that evidence could affect the court’s comparison of the gravity of defendant’s offense with the severity of the prescribed penalty. Accordingly, we must remand to the trial court for resentencing.15 In so concluding, we do not suggest that defendant’s intellectual disability is the sole determinant of whether the Measure 11 sentence would shock the moral sense of all reasonable people. As discussed, other case-specific factors, including the nature of defendant’s conduct, its effect on the victim, and the length of the prescribed sentence, also are relevant considerations in making the proportionality comparison under Rodriguez/Buck. 347 Or at 62. In short, this opinion should not be taken to imply that the proper consideration of defendant’s intellectual disability necessarily would lead to a different sentence. Because different inferences could be drawn from the evidence, that is a decision, in the first instance, for the trial court.
As stated above, we hold only that the trial court erred—in comparing the gravity of defendant’s offense and the severity of the Measure 11 sentence under the first Rodriguez/Buck factor—in failing to consider evidence of *626defendant’s intellectual disability when that evidence, if credited, would establish that the sentence would be arguably unconstitutional because it shows that defendant’s age-specific intellectual capacity fell below the minimum level of criminal responsibility for a child.16
The decision of the Court of Appeals is affirmed in part and reversed in part. The judgments of conviction, are affirmed, but the sentences are vacated, and the case is remanded to the circuit court for resentencing, in a manner consistent with this opinion.
Impairment in “adaptive functioning” refers to “significant limitations in an individual’s effectiveness in meeting the standards of maturation, learning, personal independence, and/or social responsibility that are expected for his or her age level and cultural group, as determined by clinical assessment and, usually, standardized scales.” Kevin P. Weis, Confessions of Mentally Retarded Juveniles and the Validity of Miranda Rights Waiver, 37 Brandeis L J 117, 126 (1998) (quoting American Association on Mental Deficiency, Classification in Mental Retardation 11 (Herbert J. Grossman ed. 1983)).
The court stated:
“This Court, based upon the record before me, would find that [defendant] is of mild to moderate cognitive—or possesses a mild to moderate cognitive disability. That based upon the exhibits presented and the file materials that, [defendant], there is a danger to our community [in] placing you out in the community. I have, again, based upon this record, a great deal of respect for Mr. Watson’s program, and think that would be appropriate but for the Measure 11 sentence—or the Measure 11 guidelines that have been laid out *610by the Legislature as well as by the courts, particularly in Rodriguez/Buck. When considering the Oregon Constitution, specifically Article [I], [s]ection 161,] of the Oregon Constitution!,] as well as the Eighth Amendment of the U.S. Constitution as applied to you, [defendant], I do not find, under this record, that your sentence under Measure 11 shocks the conscience of this Court with regard to the underling factual admissions that have been made. It is a challenge for any court [when] the Legislature decides to remove the Court’s ability to fashion an appropriate remedy for somebody who possesses your background and disability, but I do believe that your—the analysis of Rodriguez /Buck that—and Measure 11, that I would have to find shocking of the conscience before I move into that, though I would hope that that would change, and I may be applying it inaccurately, that seems to be the out—the paradigm that has been laid out by the courts in light of Measure 11.
“Therefore, though I would like to agree with [CS] in her statement that you should have treatment, I can’t order that treatment with regard to Measure 11.”
Defendant does not challenge his sentences on the third-degree sexual abuse convictions.
As noted below, defendant’s arguments concerning the relevance of his treatment evidence evolved later in his reply brief and at oral argument.
This court’s reference in Rodriguez/Buck to “culpability” was taken from the United States Supreme Court’s Eighth Amendment jurisprudence, which we discuss in greater detail below. The term generally is used with reference to “relative,” “diminished,” or “reduced” culpability. See, e.g., Atkins v. Virginia, 536 US 304, 317, 321, 122 S Ct 2242, 153 L Ed 2d 335 (2002) (prohibiting the death penalty for persons with intellectual disabilities because of the reduced “relative culpability of [intellectually disabled] offenders”); Roper v. Simmons, 543 US 551, 578, 125 S Ct 1183, 161 L Ed 2d 1 (2005) (prohibiting the death penalty for juveniles because of their “diminished culpability”).
See Rodriguez/Buck, 347 Or at 59-60 (discussing tests under Eighth Amendment to the United States Constitution and Article I, section 16, of the Oregon Constitution). The Supreme Court has identified three criteria for assessing proportionality: the gravity of the offense as compared to the harshness of the penalty; the sentences imposed on others in the same jurisdiction; and the sentences imposed for the same offense in other jurisdictions. Solem v. Helm, 463 US 277, 292, 103 S Ct 3001, 77 L Ed 2d 637 (1983). The gravity of an offense depends heavily on the nature and circumstances of a particular case, including the harm or risk of harm, magnitude of the crime, degree of culpability of the offender, motive, and any other facts specific to the offense. Id. at 292-94; U.S. v. Young, 766 F3d 621, 626-27 (6th Cir 2014), cert den, _ US _, 135 S Ct 1475 (2015). In most cases under the Eighth Amendment, a gravity-versus-harshness analysis will answer the question; only if a court reaches an initial inference of gross disproportionality will it consider the other criteria. Harmelin v. Michigan, 501 US 957, 1004-05, 111 S Ct 2680, 115 L Ed 2d 836 (1991); Graham v. Florida, 560 US 48, 60, 130 S Ct 2011, 176 L Ed 2d 825 (2010).
By “deterrence,” we understand the Court to have meant the ordinary meaning of the term: “the prevention of criminal behavior by fear of punishment.” See Natalie Pifer, Is Life the Same As Death?: Implications of Graham v. Florida, Roper v. Simmons, and Atkins v. Virginia on Life Without Parole Sentences for Juvenile and Mentally Retarded Offenders, 43 Loy L Rev 1495, 1501 n 27 (2010) (so describing the Court’s use of the term, quoting Black’s Law Dictionary 514 (9th ed 2009)).
See, e.g., U.S. v. Gibbs, 237 F App’x 550, 568 (11th Cir), cert den, 552 US 1005 (2007) (holding that Atkins was inapplicable in the context of a sentence that did not involve the death penalty); Harris v. McAdory, 334 F3d 665, 668 n 1 (7th Cir 2003), cert den, 541 US 992 (2004) (same); People v. Brown, 2012 Ill App 091940, 967 NE 2d 1004, 1022 (Ill App), rev den, 981 NE2d 995 (2012), cert den sub nom Brown v. Illinois, 569 US 1031, 133 S Ct 2795 (2013) (same); Commonwealth v. Yasipour, 957 A2d 734, 744 (Pa Super Ct 2008), rev den, 602 Pa 658, 980 A2d 111 (2009) (same).
We emphasize that our holding applies only to intellectually disabled offenders, not to other categories of offenders.
As discussed, defendant offered without objection expert testimony that his disability would be adversely affected by a prison sentence. The state has not contended that that evidence was irrelevant in determining the severity of the prescribed Measure 11 prison sentence in this case.
For this reason, the Court has noted that successful Eighth Amendment challenges to noncapital punishments are extremely rare. Harmelin, 501 US at 1001.
In his reply brief on review, and later at oral argument, defendant conceded that “evidence of a treatment program is difficult to fit perfectly into either the gravity-of-offense or severity-of-punishment box.” Defendant then suggested that the court should consider modifying the Rodriguez/Buck methodology in cases involving intellectually disabled offenders to add a prong focusing more on the characteristics of the offender than the offender’s conduct. We do not reach that question: That argument came too late, and was not sufficiently developed, to permit careful consideration of it in this case.
That statute provides:
“(1) A person who is tried as an adult in a court of criminal jurisdiction is not criminally responsible for any conduct which occurred when the person was under 12 years of age.”
By referring to ORS 161.290(1) as objective evidence of a societal standard, we do not suggest that that statute directly applies to intellectually disabled adults. Nor do we hold that ORS 161.290(1) precludes the imposition of criminal responsibility on an intellectually disabled adult with a mental age lower than that of a twelve-year-old. For example, on remand the sentencing court, after considering evidence of defendant’s intellectual disability, may impose a lesser sentence than the prescribed Measure 11 sentence if the court properly concludes that the prescribed sentence is constitutionally disproportionate to the offense based on evidence in the record.
As discussed, the evidence pertaining to defendant’s intellectual disability was relevant both to the gravity of his offense and the severity of the penalty for it.
We emphasize that our decision stands for the limited proposition that, in certain circumstances where the record suggests that the trial court misapprehended the import of the defendant’s proportionality challenge, we may vacate and remand for the court to consider the relevant factors in the first instance. Vacating and remanding for a trial court to resentence the defendant in those circumstances is appropriate because we would have to speculate as to whether the court properly considered the relevant case-specific factors and made any necessary factual findings. See State v. Sanderlin, 276 Or App 574, 576-77, 368 P3d 74 (2016) (Court of Appeals vacated and remanded for resentencing because the trial court had concluded that it was not permitted to consider the defendant’s “mental problems” in assessing the proportionality of the sentence).
Our conclusion makes it unnecessary to consider defendant’s Eighth Amendment challenge.