The way young adults respond to other people’s distress may be shaped by an unexpected combination of early family experiences and present-day friendships, according to a new study of 309 emerging adults. Researchers at the University of Macau report that both supportive and unsupportive reactions from caregivers were associated with higher levels of prosocial behavior across much of the group’s measured behavior. The finding complicates a familiar assumption in developmental psychology: that emotionally dismissive or unsupportive caregiving should simply produce poorer social functioning later in life. Instead, the results suggest that difficult emotional environments may sometimes be transformed into a motivation to help others, although the researchers stress that this does not make adversity beneficial or erase its potential harms. The study, published in the Journal of Adult Development, examines how people learn what emotions mean, how they should be expressed, and how one ought to respond when someone else is suffering. Its central message is that compassion in early adulthood may be built from more than warm parenting alone. It can also reflect learned social rules, adaptation to hardship, and the immediate expectations of close friends.

“Emotion socialization” describes the processes through which caregivers and peers influence another person’s emotional development. These processes include direct teaching, such as explaining that sadness is understandable, as well as indirect signals, such as comforting someone, punishing emotional displays, minimizing distress, or encouraging a person to solve a problem independently. Supportive responses typically validate emotion, offer assistance, and help an individual regulate feelings without denying them. Unsupportive responses may include dismissing, punishing, magnifying, or ignoring negative emotions. Over time, repeated interactions can become templates: mental expectations about whether distress will be met with care, criticism, or silence, and about whether helping another person is safe, worthwhile, or socially expected. The new research distinguishes between caregiver emotion socialization remembered from earlier life and friend emotion socialization experienced in the present. That distinction is important because emerging adulthood—a developmental period spanning the late teens and twenties—is marked by increasing independence from parents and greater reliance on intimate friendships, romantic relationships, and wider social networks. The researchers therefore asked whether family emotional experiences continue to matter after young people have begun constructing adult lives, and whether friends provide an independent social force.

Participants completed measures covering retrospective caregiver emotion socialization, current emotion socialization from close friends, and six dimensions of prosocial behavior. The source article’s abstract does not specify the sample’s demographic composition, the individual questionnaires used for every measure, or whether the participants were followed over time. Those limitations matter because the analysis links reported experiences and reported behavior rather than directly observing people helping in real-world situations. The investigators used regression analyses to estimate the unique associations between caregiver responses, friend responses, and prosocial outcomes. In statistical terms, regression allows researchers to examine whether one predictor remains related to an outcome after accounting for other predictors in the model. They also conducted moderation analyses, which test whether the strength or direction of one association changes depending on a third variable. For example, a moderation effect would occur if the relationship between unsupportive caregiving and helping differed for people whose friends responded warmly compared with those whose friends responded dismissively. The study thus moves beyond a simple question of whether supportive relationships are good, examining how several emotional environments may combine—or fail to combine—to shape behavior.

The clearest pattern concerned caregivers. Supportive caregiver responses were positively associated with most of the six prosocial dimensions, a result consistent with the idea that emotional validation and guidance can cultivate empathy, emotional regulation, and sensitivity to other people’s needs. People who grow up with adults who acknowledge distress may become better able to recognize similar feelings in others and to respond without becoming overwhelmed. But unsupportive caregiver responses were also positively associated with most prosocial dimensions. This counterintuitive result challenges what the authors describe as traditional deficit-based models, in which emotionally unsupportive parenting is expected to predict uniformly negative outcomes. One possible explanation is adversity-based learning. Someone who repeatedly experiences emotional dismissal may develop a strong awareness of what it feels like to be unsupported and may later try to prevent others from experiencing the same isolation. In this interpretation, suffering is not directly producing kindness; rather, the person may construct an other-oriented motivation from memories of unmet emotional needs. Another possibility is that some individuals learn to monitor distress closely in unpredictable environments, making them especially alert to emotional cues. Such vigilance can sometimes be channeled into helping, although it may also carry psychological costs that the study did not establish.

The role of friends was more consistent for supportive behavior and more limited for unsupportive behavior. Supportive responses from close friends showed independent positive associations with five of the six prosocial dimensions. This suggests that friendship may operate through a pathway that is partly separate from family history. Friends do not merely reinforce emotional habits learned in childhood; they can establish current norms about what a group values and how members should treat one another. If a close friend listens, validates distress, and offers practical or emotional support, helping may become a visible and expected part of the relationship. Such interactions can provide immediate feedback: compassionate behavior is met with trust, reciprocity, and stronger connection. Friendships may also create situational opportunities to practice prosociality, from comforting someone after a setback to coordinating assistance during a crisis. By contrast, unsupportive friend responses had few detectable associations with prosocial behavior. The one significant negative relationship involved “dire behavior,” a dimension referring to helping in urgent or severe circumstances. A friend who minimizes emotion or avoids engagement may be particularly unlikely to model action when another person faces an emergency. Yet the restricted pattern means that unsupportive friendships did not broadly predict a lack of kindness across every form of helping measured.

The analysis also produced a finding that runs against a straightforward protective-factor story. Supportive caregiver responses did not buffer, or weaken, the associations between unsupportive peer reactions and any prosocial dimension. In other words, the presence of a supportive family history did not reliably shield participants from the possible influence of emotionally dismissive friends. This does not mean that supportive caregiving is unimportant. It may instead indicate that family and peer emotion socialization operate through different mechanisms or at different developmental stages. Caregiver experiences may establish durable internal templates, while friendships provide immediate social norms and contexts in which those templates are expressed. The researchers did identify one moderation effect involving anonymous prosocial behavior. Participants with unsupportive caregiver histories were more likely to report anonymous helping when their close friends also tended not to respond supportively. Anonymous prosociality is distinctive because it removes much of the social reward that can accompany visible generosity. A person who helps anonymously cannot easily gain praise, reputation, or reciprocal assistance from the recipient. The result may indicate that some emerging adults with difficult caregiving histories develop a private form of compassion that is activated when their current social environment does not offer much emotional encouragement. Alternatively, it may reflect a compensatory strategy: helping without recognition allows someone to act according to personal values while avoiding the vulnerability of openly seeking approval.

The findings fit with a broader view of prosocial behavior as a multidimensional phenomenon rather than a single personality trait. People may donate, share, comfort, cooperate, help in emergencies, or assist anonymously for partly different reasons. A person who is highly responsive to another’s suffering may not seek public recognition; another may be generous mainly when social approval is likely. By examining six dimensions, the researchers could detect associations that a single overall score might obscure. The study also places emerging adulthood at the center of a developmental transition in which social motivations can be reorganized. Young people are not simply carrying childhood patterns unchanged into adulthood, nor are they starting from zero. Instead, early experiences may supply expectations and emotional skills, while current friends influence which behaviors are rewarded, discouraged, or made meaningful. The authors propose that caregiver experiences can establish enduring templates for prosocial engagement and that friend emotion socialization contributes through norm-based and situationally responsive pathways. This framework helps explain why the same history may produce different behaviors in different social settings. A person may be compassionate because they have internalized care, because they know the pain of neglect, because friends expect mutual support, or because a specific situation makes another person’s need impossible to ignore.

The study’s conclusions should nevertheless be interpreted cautiously. Retrospective reports of caregiving depend on adult memories of childhood, which can be shaped by current mood, later experiences, and the meaning people now assign to their past. Self-reported prosocial behavior can also be influenced by social desirability—the tendency to present oneself in a favorable light—despite the researchers’ attention to measurement issues. Because the available report describes regression and moderation analyses from 309 participants rather than a longitudinal experiment, it cannot establish that caregiver reactions caused later helping behavior. It is also impossible to conclude that unsupportive parenting generally produces resilience or compassion. The same emotional environment may be associated with distress, inhibition, distrust, or difficulty regulating emotion, depending on individual characteristics and circumstances. The positive associations observed here may represent one developmental pathway among several, not a universal response to adversity. Even so, the results offer a provocative challenge to simplistic accounts of emotional development. Prosocial behavior can emerge from support, but it may also be shaped by attempts to repair, resist, or reinterpret earlier emotional experiences. In adulthood, the impulse to help may be less like a single inheritance and more like a song assembled from many voices: caregivers, friends, memories, social expectations, and the situations in which compassion is finally given a chance to act.