Abstract
Decades of research on motivation shows that students’ beliefs about ability shape how they interpret difficulty and whether they persist after setbacks. Much of this work, and its translation into education policy, has focused on adolescents and on one belief in particular: whether ability is fixed or malleable. Recent research challenges both emphases. Ability beliefs take shape well before adolescence: Children as young as 4 already reason in systematic, sensible ways about success and failure, and by the early elementary years they hold a coherent, multidimensional “ecosystem” of beliefs about ability. This ecosystem includes not only beliefs about malleability but also beliefs about the necessity of “brilliance” for success and the universality of high-level potential. These early beliefs are already linked to children's goals (e.g., to learn vs. to look competent), willingness to take on challenges, and concerns about being judged. They are also shaped by the cues children encounter in classrooms, and they intersect with societal stereotypes to shape early interests relevant to later career pursuit. The present review discusses implications for education policy, including teacher preparation and institutional practices that send signals about ability. Research investments are needed to support evidence-based policy decisions.
Keywords
Social Media
As early as preschool, children are already learning what ability is, who has it, and what success requires. Education policy should attend to the classroom cues and school structures that shape this motivationally consequential “ecosystem” of beliefs about ability.
Key Points
Children's beliefs about ability matter earlier than education policy often assumes.
Children's beliefs about ability are broader than fixed vs. growth mindsets; they also include beliefs about whether success requires “brilliance” and whether high ability is broadly attainable or reserved for a few.
By early elementary school, children's beliefs about ability form a coherent, motivationally consequential “ecosystem,” with different beliefs linked to different motivational outcomes.
The levers that shape these beliefs are wide-ranging. Everyday classroom cues, placement systems, and gifted labels all communicate what ability is, who has it, and what success requires.
Education policy should focus on designing school environments that send more adaptive messages about ability.
Achievement is not shaped by skills alone (e.g., Eccles & Wigfield, 2020). Learners’ interpretations of their successes and setbacks along the way—especially their setbacks—can make a big difference. A first-grader who struggles with subtraction, a middle-schooler who fails a science test, and a college student who feels lost in introductory computer science must each make sense of that experience. Does it mean they need more practice or a better approach, or that they are not smart enough to succeed in this subject? Learners’ answers shape whether they persist or withdraw from an academic path (Dweck, 2006).
These interpretations are often rooted in what motivation researchers call “beliefs about ability.” Perhaps the most influential examples are fixed and growth mindsets: beliefs about whether ability is relatively stable (fixed mindset) or can grow through effort, strategies, and support (growth mindset; Dweck, 2006). Students who view ability as malleable rather than fixed engage more constructively with difficulty: They are more likely to choose challenging tasks (which usually lead to greater learning), persist after setbacks, and treat mistakes as information rather than as evidence of low ability (e.g., Blackwell et al., 2007; Yeager et al., 2019). This work has been influential, including in education policy (e.g., West et al., 2018). It is one of the more successful translations of psychological science into educational practice in recent memory.
Yet the policy conversation has often been narrower than the science. In particular, two assumptions that have shaped educational policy are increasingly difficult to reconcile with the current evidence. The first is developmental: Beliefs about ability are often treated as more relevant for adolescents’ motivation and achievement than for young children's (for reviews, see Cimpian, 2017; Muradoglu et al., 2025). The second assumption is conceptual: Growth and fixed mindsets have come to stand in for beliefs about ability writ large in policy conversations. This article reviews recent evidence that calls both assumptions into question: Children's beliefs about ability develop earlier than current educational policy assumes, and the beliefs that shape students’ motivation and achievement extend well beyond beliefs about the malleability of ability. Broadening the scope of policymaking in both directions (toward younger ages and toward more beliefs) opens new possibilities for supporting children's motivation and achievement. The present review assesses these two limiting assumptions and concludes by considering policy implications. The term “policy” here spans formal levers such as curricula, teacher preparation, and assessment, as well as the everyday classroom practices they shape.
Limiting Assumption 1: Beliefs About Ability Do Not Matter Until Adolescence
Young children often appear optimistic. They insist that they can master nearly anything if they keep at it, and they seem less concerned than older children with comparisons and failure. These observations have encouraged the idea that the early years are, in some sense, protected from the motivational vulnerabilities associated with older children and adolescents. This idea was also reinforced by an influential developmental theory according to which young children do not yet clearly distinguish ability from effort (for reviews, see Butler, 2005; Cimpian, 2017). Without the ability to make this distinction, failure has a different meaning for young children. A poor outcome could never imply low ability; it would simply signal insufficient effort. And if effort is the only ingredient in performance, doing better next time is just a matter of trying harder, so there is no point in getting discouraged. More complex beliefs about ability (such as whether it is fixed or malleable) could only be formulated once children developed a differentiated concept of ability in the first place.
This view had a clear corollary: If the ability beliefs that shape achievement motivation are not yet in place, or not yet coherently organized, in early childhood, then efforts to address them are better directed at older students. This canonical developmental picture also helps explain why the field's most rigorous intervention science has focused on older students, especially adolescents and school transitions. The largest and most influential test of a belief-focused intervention to date, the National Study of Learning Mindsets, was conducted with a nationally representative sample of ninth graders and timed deliberately to the transition into high school (Yeager et al., 2019).
None of this is to suggest that beliefs about ability are absent from early-childhood and elementary practice. On the contrary, the vocabulary of growth mindset (e.g., praising effort, treating struggle as productive) is by now familiar to many early educators. But this grassroots adoption has outrun the science it draws on: Much of the rigorous evidence base for growth-mindset interventions was built with older students, and early-childhood practice has often imported this single belief construct without a detailed developmental account of how young children understand ability, how their beliefs about ability are structured, or which everyday classroom cues shape them.
Yet recent evidence is increasingly difficult to square with the canonical view. Even in the preschool and early elementary years, children distinguish ability from effort (e.g., Heyman & Compton, 2006; Muradoglu & Cimpian, 2020), hold structured and coherent beliefs about the nature of ability (e.g., Gunderson et al., 2017; Muradoglu et al., 2026b), and act on those beliefs in motivationally consequential ways—pursuing different goals, seeking or avoiding challenging tasks, and worrying about how their abilities will be judged (e.g., Muradoglu et al., 2026b). Three lines of evidence are particularly revealing.
First, children differentiate ability from effort earlier than assumed. In an influential study supporting the canonical view, children were shown two students working side by side on the same test, with one working steadily throughout and the other visibly off-task for part of the time (Nicholls, 1978). When children were told that both students scored the same, most judged the harder-working student to be smarter, as if equating ability with the effort that produced the score. This pattern, replicated across studies (e.g., Nicholls & Miller, 1984), was taken as evidence that young children do not yet understand ability as a separate cause of performance.
However, Muradoglu and Cimpian (2020) recently showed that this conclusion depends on two methodological features (see also Heyman & Compton, 2006). First, they changed what children were asked immediately before judging ability. In the canonical task, children were asked about effort immediately before ability—a task demand that may have made effort especially salient and anchored children's ability judgments to the harder-working student. Muradoglu and Cimpian therefore added a condition in which children were asked instead about how difficult the task was, so that their ability judgments were not preceded by an explicit focus on effort. Second, Muradoglu and Cimpian made the off-task behavior in the experiment a clearer signal of needing less time to complete the task by having the off-task student complete the task in one continuous block at the beginning before going off task, rather than working intermittently. These changes reversed the results: Even 4-year-olds judged the student who reached the same outcome with less effort to be the more able of the two.
This is not to claim that young children think about ability in exactly the same way adults do. For instance, children may still be less likely to compare themselves with peers (Dijkstra et al., 2008) and more likely to expect improvement over time (Lockhart et al., 2002). The point is more circumscribed: From early in development, children have the conceptual tools to reason about ability as distinct from effort; whether and how those tools are activated depends less on developmental availability and more on what children's social environments call for (Cimpian, 2017).
Second, young children's beliefs about ability are not just present but also measurable. Once children can reason about ability as distinct from effort, they can also begin to form specific beliefs about what ability is. Until recently, however, a major obstacle to studying these early beliefs has been measurement. Students at different developmental stages interpret key terms such as “intelligence” differently (Limeri et al., 2020). Recent work has begun to develop measures targeted to specific ages. For instance, Limeri and colleagues (2023) developed and validated a measure of beliefs about ability designed specifically for undergraduates, and Muradoglu and colleagues (2024) developed and validated a Growth Mindset Scale for Young Children (GM-C; see also Gunderson et al., 2013; Ruzek et al., 2020). The GM-C uses concrete, age-appropriate language focused on familiar academic abilities (such as being good at math or drawing) rather than abstract questions about intelligence, along with child-friendly response options designed to capture gradations in children's beliefs. This scale showed evidence of reliability and validity across samples in the United States and South Africa.
Third, these early beliefs about ability are motivationally active. For example, in a study on 5- to 11-year-olds, children with stronger growth mindsets oriented toward learning rather than performance goals and preferred more challenging tasks (Muradoglu et al., 2026b). Other early-emerging beliefs about ability converge with this picture. By first grade, children in many parts of the world show signs of acquiring cultural stereotypes about ability and gender: Girls are less likely than boys to attribute being “really, really smart” to their own gender, and this gendered view of brilliance is associated with gendered interests, with girls showing less interest than boys in unfamiliar activities described as being for children who are “really, really smart” (e.g., Bian et al., 2017, 2018; Kim et al., 2024; Zhao et al., 2022).
The cumulative picture is that children's beliefs about ability emerge, cohere, and begin to shape motivation far earlier than the canonical view has assumed. By the early elementary-school years, these beliefs are organized into a multidimensional structure and are already linked to key motivational processes such as learning versus performance goals and challenge-seeking. This evidence suggests that early-childhood and elementary settings are not merely places where children acquire academic skills; they are also environments in which children learn what ability is. Placing students on an educational track based on a single test in first grade sends a strong message to children about the fixedness of abilities. A second-grade math curriculum that, in an effort to stimulate children's interest, introduces role models whose success is attributed to their genius communicates messages about what it takes to be successful in math (Dasgupta & Stout, 2014; Gladstone & Cimpian, 2021). Young children's social environments are already teaching them, implicitly or explicitly, what ability is and how much of it they need. Once that is recognized, the policy-design question becomes not whether to address beliefs about ability in early childhood, but how to shape the everyday cues surrounding young learners so that these beliefs develop in directions that support, rather than constrain, what children are willing to try.
Limiting Assumption 2: Beliefs About Ability Are Basically Mindsets
A second assumption is that beliefs about ability are essentially the same thing as mindsets. Mindset theory has been foundational for motivation science and unusually influential in education. It gave educators and policymakers a clear way to think about how students interpret difficulty: Do students see struggle as a sign that their ability is limited, or as part of the process by which ability develops? This question has generated a large body of evidence and has inspired many efforts to make classrooms more supportive of challenge and learning (e.g., Blackwell et al., 2007; Paunesku et al., 2015; Porter et al., 2022; Yeager et al., 2019). Mindset research also made a broader idea visible for the general public: Students’ assumptions about ability can shape their educational trajectories.
But this success has also had an unintended side effect. In many policy conversations, “beliefs about ability” has come to be synonymous with “mindsets.” The result is that a much broader belief system is often collapsed into a single question: “Can ability change?” That question is important, but it is not the only question learners are trying to answer. When students enter a classroom, they are also trying to figure out, for example, what is required for success there and who has what it takes. “Is this just for super smart kids?” and “Can anyone do well here, or only a few special kids?” are also top-of-the-mind questions for students in many academic settings.
A more complete account treats mindsets as one part of an ecosystem of beliefs about ability. Learners also form beliefs about whether success in a subject requires exceptional ability (“brilliance beliefs”; e.g., Bauer et al., 2025; Cimpian & Leslie, 2017; Jenifer et al., 2024; Limeri et al., 2023; Muradoglu et al., 2023, 2026b; Rutten et al., 2024), and about whether everyone (or only some people) can reach the highest levels of ability in a subject (“universality beliefs”; e.g., Limeri et al., 2023; Muradoglu et al., 2026b; Rattan et al., 2012, 2018). Brilliance beliefs concern the importance of intellectual ability for success: How smart does one have to be to succeed here? Universality beliefs concern the distribution of high ability: Can everyone reach the highest levels, or only some people? Although related to mindsets, these beliefs are conceptually distinct. A student may believe that mathematical ability can improve and still believe that being a mathematician ultimately requires a rare kind of brilliance that only some people possess. A classroom may communicate that mistakes are useful for learning while also celebrating the historical geniuses who came up with theorems and proofs seemingly out of thin air. In each case, a growth-oriented belief or message about improvement coexists with messages that imply scarcity, exclusivity, or the need for exceptional talent.
These beliefs are also empirically distinguishable. Among university students (Limeri et al., 2023), mindsets, brilliance beliefs, and universality beliefs were correlated in sensible ways (students who endorsed a growth mindset also tended to hold more universal beliefs and weaker brilliance beliefs) but did not reduce to a single underlying dimension. Each explained unique variance in students’ motivational and academic outcomes (see also Porter & Cimpian, 2023; Rutten et al., 2024). In other words, mindsets were incomplete as a stand-in for the entire belief system.
The same multidimensional structure emerges early in development. Among 5- to 11-year-olds, beliefs about the malleability of ability, its importance for success (i.e., whether brilliance is required), and its universality were best captured by separate dimensions rather than by a single general ability-belief factor (Muradoglu et al., 2026b). These dimensions were also coherently related: Children who viewed ability as fixed tended to view it as less universal and more necessary for success, whereas children who viewed ability as malleable showed the opposite pattern. Finally, these beliefs had distinct motivational correlates. Growth mindsets were linked to learning goals (a preference for learning even at the risk of making mistakes) and willingness to choose challenging tasks, whereas brilliance beliefs were linked, especially among younger children, to greater concern that mistakes would lead others to judge their ability negatively.
This developmental picture is, however, still emerging. One reason is the assumption examined in the previous section: When beliefs about ability are treated as unlikely to matter early in life, the full set of beliefs in young children (mindset and the rest) receives less attention than it should, and the early-childhood evidence on every belief in this ecosystem remains less mapped and less replicated than the corresponding evidence among adults. Even so, what is known so far points consistently in one direction: By the early elementary years, children hold a multidimensional set of beliefs about ability that are linked to their motivation. For policy, the central point is that mindset messages cannot do all the work. An exclusive policy focus on malleability may miss other cues that shape whether learners feel able and willing to engage with challenging work.
This broader set of questions also highlights how beliefs about ability intersect with structural concerns about equity: When children believe that exceptional ability is required for success in a subject, for example, the stereotypes already circulating about who is brilliant (by gender, race, or social class) may begin to shape children's interests and aspirations (e.g., Bian et al., 2017; Goudeau et al., 2025; Storage et al., 2016, 2020). Education policies that shape these early belief environments are, by extension, also shaping patterns of representation in prestigious and high-earning fields that emphasize intellectual ability.
From Science to Policy
The evidence is not yet precise enough to dictate one preferred policy package, and the right approach is likely to vary across ages and subjects. What the evidence does do is widen the field of options. If children reason about ability from early in development, and if their ability beliefs extend beyond mindset, then the settings and practices that shape those beliefs become possible points of intervention. Below are four policy directions that follow from this broader view.
Address Beliefs Early Through Everyday Environments
One option is to begin addressing beliefs about ability in preschool and early elementary school. Early classrooms are already transmitting to children, often unintentionally, implicit “recipes for success.” Some recipes feature ingredients such as practice, strategies, help-seeking, and curiosity; others feature inborn talent, effortless insight, or being identified early as “gifted.” Children pick up on which recipe the adults around them seem to operate with through ordinary classroom interactions (e.g., how difficulty is explained, which examples of achievement are highlighted). Policy could support teachers in recognizing the messages their classrooms already send about what it takes to succeed—and in making those messages accurate and motivating.
Where districts and researchers want to pilot more intentional approaches, evidence from intervention research and the cognitive science of learning suggests several design principles. First, alternative recipes are most credible when they are conveyed by adults who define the local norms of the setting (teachers, principals, and other trusted school leaders) because their endorsement signals what the setting itself takes to be true (e.g., Tankard & Paluck, 2016). Second, interventions should do more than tell children that an existing belief is wrong; they should make a more compelling alternative available (Lewandowsky et al., 2012; Yeager et al., 2016). Third, the alternative recipes themselves need to be substantively rich to be credible. Slogans such as “you can do anything” are easy but, on their own, can ring hollow; they may also imply that children who struggle simply have not tried hard enough (Ahn & Sheu, 2026). The richer accounts that researchers have built over the past several decades—for example, about effective study strategies (e.g., Bjork et al., 2013; Dunlosky et al., 2013)—are what gives alternative recipes their credibility. A central scientific task in the near term is to translate this evidence into accessible, age-appropriate messages that are embedded in the classroom routines young children actually experience.
To reiterate, the goal is not necessarily to talk young children into a particular belief about ability, though some change in beliefs may follow. The broader goal is to design early environments whose ordinary routines make a fuller, more adaptive recipe for success available to every child.
Broaden Mindset Programs to Cover More of the Ecosystem
A second option is to broaden existing mindset programs to cover more of the ecosystem of beliefs discussed here. Growth-mindset interventions have accumulated a stronger evidence base than interventions targeting brilliance or universality beliefs, and it would be unwise to discard them. But the present review suggests that a narrow focus on malleability may miss other beliefs that matter for motivation. Future interventions could retain core growth-mindset content while also addressing the idea that success in some fields requires brilliance or that only some students have the potential to reach high levels of ability. In practice, this might mean pairing messages about growth with examples of how expertise in a domain is actually acquired (combating beliefs about the role of brilliance) and of how high-level achievement is not reserved for a small number of students (combating beliefs about the non-universality of ability). Such interventions may be especially relevant in subjects such as engineering, computer science, and the physical sciences, where brilliance beliefs are strong (e.g., Leslie et al., 2015; Meyer et al., 2015) and demographic representation remains especially uneven (e.g., Cimpian et al., 2020; Cimpian & King, 2024). At the same time, the intervention-relevant evidence for beliefs beyond mindsets is still much thinner than the evidence for mindsets. These broader “ecosystem” interventions should therefore be treated as a research priority, not as an off-the-shelf solution for immediate deployment.
Examine Structural Practices That Signal Brilliance is Key or Ability is Scarce
A third option is to examine structural practices that signal that brilliance is necessary for success or that high-level ability is scarce. Many educational practices serve legitimate administrative or instructional functions: Schools need to place students in courses, identify students who need additional support, recognize achievement, and sometimes offer accelerated opportunities. But these same practices may also communicate messages about ability. Gifted-and-talented labels, early tracking, high-stakes placement tests, curved grading, and weed-out courses can imply both that high ability is rare and that possessing it is essential for success. Sometimes these implicit messages are at odds with the explicit motivational messaging a school is trying to convey through its curriculum or its educator training: A child can be told that ability grows in their morning lesson and then, in the afternoon, watch a small group of classmates be pulled aside for an accelerated math program for which they were selected on the basis of a test described as measuring aptitude.
The science reviewed here does not by itself dictate whether such practices should be retained or replaced. It does suggest, however, that policymakers should weigh the ability messages these practices communicate alongside their instructional functions. A placement system that is efficient from an administrative standpoint may still carry motivational costs if it teaches young children that some of them have the “right” kind of potential and others do not (e.g., Muradoglu et al., 2026a). Similarly, programs designed to challenge advanced learners may carry different motivational consequences depending on whether they are framed around fixed identification (that is, a small group of children identified early as having the right kind of mind) or around open access, where any student who wants to take on harder work can do so, with the support to succeed at it.
Invest in Measurement and Implementation Research
A fourth option is to invest in measurement and implementation research. If the belief ecosystem is multidimensional, then policy-relevant research needs tools that can capture this complexity without overburdening children, teachers, or schools. Recent measurement work has made it possible to assess young children's mindsets using developmentally appropriate language (Gunderson et al., 2013; Muradoglu et al., 2026b; Ruzek et al., 2020), and newer work has begun to assess multiple ability beliefs in childhood (Muradoglu et al., 2026b). But much remains to be done: longitudinal studies that track how children's beliefs about ability develop from the early years onward; studies that identify which classroom, family, and institutional cues shape these beliefs; and intervention studies that test whether changing brilliance and universality messages improves outcomes beyond standard mindset content. Also needed is implementation research that examines how belief-related policies work in real classrooms. A message that is effective in a short experimental intervention may not operate the same way when embedded in a curriculum or a teacher-training program.
Across these options, one guiding principle is that beliefs should be treated as interactions between students and their environment, not merely as traits inside children. This shift matters for policy. If a child avoids a challenging math activity, the explanation may not be that the child lacks something (e.g., the right belief, the right level of willpower). Instead, the child may be responding reasonably to an environment that has made difficulty feel diagnostic, success feel reserved for a select few, or brilliance seem necessary for success. The goal of policymaking should thus be widened from “How do we change children's beliefs?” to “What beliefs do our educational environments communicate, and how can those environments be designed more wisely?”
Conclusion
Taken together, this evidence broadens the picture of motivation relevant to schooling. Children's beliefs about ability take shape much earlier than adolescence; by the time children enter formal schooling, they are already interpreting their successes and failures through these beliefs. Nor are these beliefs reducible to growth and fixed mindsets; they form a broader ecosystem of ideas that operate alongside, and sometimes at cross-purposes with, mindset messages.
The policy implication is that the levers available to policymakers are broader than the current conversation suggests. Curricula, educator-preparation programs, placement and assessment systems, and the everyday routines of early classrooms are all already communicating something to children about what ability is and how much of it they need to do well. Those messages can be audited and improved. At the same time, the science is not yet precise enough to prescribe a single new approach to replace the mindset framework, and one-size-fits-all messaging is unlikely to translate across the varied cultural and institutional contexts in which children actually grow up. Nor should education policymakers turn beliefs about ability into one more thing that schools and children are ranked on. Measures of these beliefs were developed to help researchers understand what environments are teaching, not to evaluate children and teachers. The promise of this work lies in designing environments whose everyday practices make it more likely that children come to believe that ability is developable, that high potential is broadly distributed, and that success does not require being marked early as brilliant.
Footnotes
Acknowledgments
We are grateful to the following friends and colleagues for their helpful feedback on previous versions of this manuscript: Elizabeth Canning, Madalin Deliu, Kama Einhorn, Andy Elliot, Nadine Knab, and Nathaniel Woznicki.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
