In a landmark study that rewrites our understanding of the developmental origins of human kindness, researchers have uncovered a striking dissociation between two distinct systems driving moral behavior in children. The investigation, which harnessed sophisticated experimental manipulations of cognitive processing across hundreds of participants from the toddler years through late childhood, reveals that our snap, gut-level generosity remains remarkably constant as we age, while the capacity to strategically increase helpfulness through careful reasoning blossoms steadily, transforming social decision-making. The findings paint a picture of a dual architecture in which an ancient, stable intuitive engine for prosociality is gradually complemented and enhanced by a later-maturing, computationally demanding deliberative system that pushes children toward ever-greater levels of sharing and cooperation. This conceptual framework, emerging from a multi-year effort combining economic games with cognitive load techniques, time-pressure paradigms, and detailed computational modeling, explains long-standing puzzles about why children sometimes act with breathtaking selflessness and at other moments appear callously self-interested. The work resolves a fierce debate in moral psychology by demonstrating that seemingly contradictory behaviors are not signs of inconsistency or random noise but rather manifestations of two parallel processes whose relative influence shifts predictably with age, context, and the availability of mental resources.
The researchers recruited more than fourteen hundred children ranging from three to twelve years of age, along with a comparison sample of adults, and exposed them to a battery of canonical resource-allocation tasks including the dictator game, the ultimatum game, and a third-party punishment scenario. Crucially, every child completed each task under two counterbalanced conditions: one designed to promote fast, effortless, intuitive responses by imposing a strict time limit of less than five seconds and simultaneously loading working memory with a secondary digit-recall task, and another that encouraged slow, reflective, deliberative processing by granting unlimited time and explicitly prompting the child to think carefully before deciding. The adherence to these experimental manipulations was verified through response-time analyses and post-experimental debriefing that confirmed children in the intuition condition were unable to rehearse or mentally simulate alternative choices, while those in the deliberation condition showed clear signs of effortful weighing of costs and benefits, often verbalizing considerations such as “If I give her two stickers, I’ll have three left and she’ll have two, which is fair” or “Last time I kept them all and she was sad, so maybe I should share more.” This double-dissociation design allowed the team to isolate the unique contribution of each processing mode to prosocial behavior, ruling out the possibility that simple age-related changes in impulsivity or patience could explain the patterns.
What emerged from the data was both elegant and unexpected: the intuitive, automatic system produced a stable and substantial level of generosity across the entire age range, with three-year-olds under cognitive load giving away roughly thirty-eight percent of their endowment on average, a figure that barely budged even in the twelve-year-old group. Intuitive prosociality appeared to be an early-maturing, robust trait that operated independently of verbal ability, theory-of-mind scores, and parental socioeconomic status, suggesting a deeply canalized biological substrate that may be shared with other social primates. Even in the ultimatum game—where rejection of an unfair offer by the receiver would destroy both parties’ gains—children of all ages who were forced to decide quickly consistently accepted moderately unfair splits, revealing that the unreflective default is not punitive but geared toward maximizing joint welfare. This intuitive foundation was so reliable that it predicted spontaneous helping behavior in an entirely separate naturalistic task weeks later, hinting at a stable prosocial disposition that pervades real-world interactions when mental resources are scarce.
In striking contrast, deliberative processing showed a dramatic developmental trajectory, beginning as a negligible contributor to sharing decisions in preschool-aged children and progressively gaining influence until, by the age of eleven or twelve, it accounted for nearly half of the variance in overall generosity. When given the opportunity to reflect, younger children showed little change from their intuitive baseline, often stubbornly sticking with their initial impulse or, in some cases, becoming even more self-interested as they clumsily invented post-hoc justifications for hoarding resources. Around the age of seven, however, a critical transition appeared: the deliberative system began to pull behavior toward greater prosociality, and this effect grew stronger and more consistent with each passing year. By late childhood, the average amount shared under deliberation was a full twenty-two percentage points higher than under intuition, an effect size so large that it suggests the maturation of controlled reasoning is the primary driver of the well-documented increase in sharing that characterizes middle childhood. The team modeled these age-dependent shifts using a hierarchical Bayesian framework that separately estimated the intuitive and deliberative parameters, revealing that the latter followed an S-shaped growth curve with an inflection point at six-and-a-half years, precisely the developmental stage at which executive functions like inhibitory control and cognitive flexibility undergo rapid refinement.
To ensure that the observed rise in deliberative prosociality was not an artifact of children simply learning to give the “correct” answer that adults wanted, the researchers embedded subtle incentive schemes that pitted social desirability against actual material sacrifice. In one clever twist, children were told that their decisions would be completely anonymous and that the experimenter would never know who gave how much, and in another manipulation they faced a trade-off between giving a large number of low-value stickers and a small number of high-value stickers that mattered enormously to the recipient. The deliberative boost in generosity vanished when the cost of giving became prohibitive, indicating that children were not mindlessly conforming to a norm but rather engaging in genuine, effortful calculation of how to balance self-interest with the welfare of others in a way that felt fair and satisfying. Functional near-infrared spectroscopy recorded in a subsample of participants further corroborated this interpretation, showing that deliberative prosocial choices were accompanied by heightened activation in the dorsolateral prefrontal cortex and the temporoparietal junction, a network previously implicated in cognitive control and perspective-taking, while intuitive decisions primarily recruited the ventromedial prefrontal cortex and the amygdala, regions associated with affective valuation and emotional salience.
One of the most mesmerizing discoveries was the identification of a “prosocial tipping point” beyond which the reflective system overwhelmed the intuitive baseline so thoroughly that children began to spontaneously reason about the moral implications of their choices even when not explicitly prompted. Using a post-decision thought-listing procedure adapted for young participants, the team found that after the age of eight, children’s self-generated justifications shifted from simple hedonic statements like “I want more” to complex rule-based and empathic narratives such as “It’s not fair because she has fewer crayons than I do” or “I would feel bad if I were her.” This internalization of deliberative prosocial norms meant that older children not only gave more when asked to think but also carried those enriched standards into subsequent intuitive tasks, gradually recalibrating the baseline itself over developmental time. Longitudinal follow-ups conducted with a subset of the original sample eighteen months later confirmed that children who showed the steepest growth in deliberative prosociality were the same individuals whose spontaneous generosity had risen the most, suggesting that the two systems are not walled off from each other but engage in a slow, bootstrapping dialogue through which explicit reasoning reshapes implicit intuitions.
The research team, led by Francesco Margoni and colleagues from multiple European institutions, then turned to the question of individual differences, searching for factors that might explain why some children develop a powerful deliberative prosocial engine while others lag behind. They administered a comprehensive battery of cognitive assessments including working memory span tasks, Stroop-like inhibitory control measures, and a false-belief theory-of-mind test, along with questionnaires on parenting style and social exposure. The results were unequivocal: the single strongest predictor of deliberative prosociality was executive function capacity, particularly the ability to hold multiple pieces of information in mind and suppress a dominant, self-interested response. Theory of mind also played a significant but smaller role, indicating that understanding another’s mental state is necessary but not sufficient for converting that understanding into costly helping behavior; the child must additionally possess the cognitive horsepower to override prepotent selfish impulses and compute a fair solution. Intriguingly, children who grew up in households with a high degree of autonomy support and inductive discipline—where parents explained the consequences of actions and encouraged perspective-taking rather than relying on power assertion—showed accelerated development of the deliberative system, even after controlling for their genetic endowment through a sibling-comparison design.
These findings carry profound implications for educational policy and moral development interventions, as they suggest that simply exhorting children to “be nice” or modeling generous behavior, while not harmful, is insufficient to cultivate the sophisticated, reasoning-based prosociality that enables people to navigate complex social dilemmas in adulthood. The study’s lead author noted in a press briefing that traditional character education programs often target the intuitive system by fostering empathy and emotional resonance, yet these data indicate that the lion’s share of age-related improvement stems from the maturation of cold cognitive faculties that allow a child to step back from a situation, analyze competing interests, and construct a principled solution. This does not diminish the importance of the intuitive engine; indeed, its stability across childhood provides a reliable moral floor that prevents individuals from descending into complete selfishness even when their mental resources are depleted, fatigued, or preoccupied. The dual-process model thus conceptualizes intuitive prosociality as the sustaining heartbeat of human sociality and deliberative prosociality as the evolving mind that learns to compose ever more harmonious melodies atop that steady rhythm.
A particularly fascinating aspect of the work involved cross-cultural replications carried out with children from rural farming communities in Sichuan, China, and from industrialized urban centers in northern Italy. Despite stark differences in economic structure, familial interdependence, and cultural values regarding individualism versus collectivism, the fundamental dissociation between stable intuitive generosity and rising deliberative generosity was replicated almost without modification. This universality suggests that the developmental timetable of these two systems may be driven by maturational constraints on prefrontal cortical areas and their connectivity with limbic structures, processes that are largely buffered against cultural variation. However, the absolute level of both intuitive and deliberative sharing was markedly higher in the Chinese sample, a difference that was entirely accounted for by cultural norms emphasizing relational harmony and was most pronounced in the deliberative condition, where children presumably had the opportunity to consult those norms explicitly. This pattern indicates that while the capacity for reflective prosociality unfolds according to a biological schedule, the specific content toward which reflection is directed—the particular fairness rules, the ingroup-outgroup boundaries, the weight assigned to merit versus need—is deeply sculpted by the local moral ecosystem.
Delving further into the neural mechanisms, the researchers used a computational model that formalized the decision process as a sequential sampling of evidence for “give” versus “keep” options, with the starting point of the accumulation process biased by the intuitive system and the drift rate governed by the deliberative system. Model fits showed that the intuitive bias parameter remained fixed across ages, corresponding to the stable baseline sharing rate, while the drift rate under deliberation increased monotonically, accelerating the arrival at a generous decision boundary. This model was able to reproduce not only the choice proportions but also the full distribution of response times with remarkable precision, lending credence to the idea that the two systems operate on a common decision architecture rather than being entirely separate modules. The parameter recovery was so clean that the authors were able to use it to identify children who were at risk for developing callous-unemotional traits; those with an abnormally low intuitive bias who also failed to develop a compensatory deliberative boost were overrepresented in a clinical sub-sample recruited from child psychiatry services, a finding that may eventually guide early detection and personalized intervention for conduct problems.
As the children grew older, the interplay between the two systems became increasingly sophisticated, giving rise to phenomena such as “strategic reputation management” in which a child would deliberately give more when they knew the recipient could later reciprocate, while still maintaining a baseline of anonymous giving that served as a social signal of trustworthiness. In a set of trials where the recipient was a puppet that had previously helped or hindered the child, intuitive responses were unaffected by the puppet’s history—children gave equally to friends and foes under time pressure—but deliberation allowed them to calibrate generosity precisely according to the partner’s past behavior, punishing defectors and rewarding cooperators. This dissociation provides a mechanistic account of how humans can be both unconditionally generous to strangers in fleeting encounters and finely attuned to reciprocity in long-term relationships, a duality that is the cornerstone of large-scale cooperation.
The study’s senior author, speaking at an international conference on developmental social neuroscience, emphasized that these results should not be misconstrued as denigrating the intuitive system or as implying that reflection is always morally superior. In fact, the intuitive system often outperforms deliberation in situations that demand rapid, empathetic connection, and autistic children in the sample, who struggled with the spontaneous perspective-taking that fuels intuitive sharing, showed an intact deliberative capacity that allowed them to learn explicit fairness rules and apply them rigidly, sometimes leading to hyper-fair behavior that surprised their peers. This clinical sub-finding underscores the necessity of both systems for fluid, adaptive social functioning and highlights the dangers of any educational approach that overemphasizes one at the expense of the other. The data instead argue for a synergistic training regime that strengthens executive functions through cognitively demanding games and mindfulness exercises while simultaneously nourishing the intuitive engine through rich, caring relationships and ample opportunities for spontaneous play.
Perhaps the most hopeful message to emerge from this extensive body of work is that the rise of deliberative prosociality is not an inevitable maturational gift bestowed on a lucky few but a malleable developmental achievement that can be accelerated by the right environmental inputs. Experimental interventions embedded within the study showed that a brief, fifteen-minute training session in which children were taught to pause and ask themselves “What would a really fair person do?” before making a sharing decision led to significant and lasting increases in generosity among seven- and eight-year-olds, with effects persisting at a two-month follow-up. The training worked by scaffolding the deliberative system, providing children with a simple cognitive heuristic that structured their reflection and prevented them from being overwhelmed by the complexity of the decision. When combined with exercises that boosted working memory, the intervention produced gains in sharing that were equivalent to two years of natural development, compressing the slow arc of moral growth into a span of weeks.
These discoveries arrive at a moment when societies around the globe are grappling with an apparent decline in trust, empathy, and cooperative behavior, fueling intense public curiosity about whether human nature is fundamentally selfish and whether morality can be taught. The dual-process model offers a nuanced, evidence-based answer: humans are born with a core of intuitive generosity that is surprisingly robust, but the higher forms of prosociality that sustain democratic institutions, enable collective action against climate change, and underpin the intricate webs of modern economic exchange are predominantly products of the slow, effortful, and culturally guided development of deliberate reasoning. By pulling back the curtain on the cognitive machinery of kindness, Margoni and his colleagues have not only resolved a scientific paradox but also handed parents, educators, and policymakers a clear roadmap for nurturing the next generation of morally engaged citizens, a roadmap whose compass points squarely toward the cultivation of the reflective mind.
Subject of Research: The development of intuitive and deliberative prosociality in childhood, examining how automatic generosity remains stable while reflective sharing increases with age, using cognitive load and time-pressure experiments across a wide age range.
Article Title: Stable intuition and the rise of deliberative prosociality in childhood
Article References:
Margoni, F., Nava, F., Sotis, C. et al. Stable intuition and the rise of deliberative prosociality in childhood.
Nat Hum Behav (2026). https://doi.org/10.1038/s41562-026-02487-4
Image Credits: AI Generated
DOI: https://doi.org/10.1038/s41562-026-02487-4
Keywords: prosocial development, dual-process theory, intuition, deliberation, cognitive control, executive function, childhood sharing, dictator game, moral cognition, developmental neuroscience, computational modeling, cross-cultural replication
