Friendships change across the lifespan because the social tasks, opportunities, capacities, and commitments surrounding relationships change. Childhood friendship develops through play, companionship, and increasingly mutual trust. Adolescence adds intimacy, identity exploration, and complex peer-group dynamics. Young adulthood often expands social opportunities while introducing disruptive moves and new responsibilities. In midlife, competing demands can reduce contact without necessarily reducing affection. In later life, people may prioritize emotionally meaningful ties, maintain longstanding friendships, and establish new ones under changing conditions of health and access. These are population-level developmental tendencies, not a universal sequence or a required number of friends. Wrzus et al., 2013; Antonucci et al., 2014; Augustsson et al., 2025.
The most important developmental distinction is between the structure of a friendship network and the experience of friendship. People can have fewer contacts yet retain dependable, satisfying relationships; they can also have many contacts and feel disconnected. Understanding the lifespan therefore requires examining who becomes a friend, how mutuality and closeness are expressed, which relationships persist, and how historical circumstances shape opportunities to connect. This article traces those processes from childhood to advanced old age while distinguishing them from the independent psychology of loneliness, friendship conflict, and clinical mental health.
What Changes When Friendships Develop Across Life?
A friendship is generally a voluntary, personally meaningful relationship involving some degree of mutual recognition, goodwill, shared activity, or emotional investment. Its expression is age-sensitive. A child's friend may be someone with whom play is repeated and reciprocated; an adolescent's friend may become a central confidant; an adult's friend may be someone who remains reliably available despite infrequent meetings. Researchers operationalize these relationships in different ways, including friendship nominations, mutual nominations, closeness ratings, interaction frequency, network mapping, and qualitative accounts. These indicators describe related but separate dimensions of social life. Bukowski et al., 1994; Antonucci et al., 2014.
Network size describes how many people someone names or contacts. Network composition describes the proportions of friends, relatives, partners, coworkers, neighbors, and other ties. Friendship quality concerns trust, mutual responsiveness, security, companionship, conflict, and meaningful support. Stability refers to the continuation of an identified tie across observations; a stable friendship can change in intimacy or frequency, while a relationship can end without implying developmental failure. A complete account of friendship development examines all four dimensions rather than treating the number of contacts as an index of human worth.
Friendship, loneliness, isolation, and chosen solitude
Loneliness is a subjective distressing discrepancy between desired and experienced social connection. Social isolation refers to objectively limited contacts or participation. Solitude can be chosen, restorative, and compatible with excellent friendships. Belonging is an experience of accepted membership and meaningful connection; support concerns resources perceived or received from others. These concepts can vary independently. A highly social adolescent may feel lonely in an unwelcoming peer group, and an older adult who enjoys a small circle may feel deeply connected. The WHO Commission on Social Connection, 2025 explicitly treats loneliness, isolation, and social connection as distinct population concerns.
The Main Scientific Models of Friendship Development
The social convoy model
The convoy model represents social relationships as an evolving set of close and more peripheral ties moving with individuals through life. Its central insight is that the meaning and function of a social network depend on personal characteristics, situational conditions, and relationships accumulated across time. Inner-circle bonds may remain highly significant for decades, while neighbors, colleagues, classmates, and activity partners move into and out of the convoy as circumstances change. The model makes room for continuity and replacement, stability and shifting roles, and the different kinds of help relationships provide. It is especially useful when age alone cannot explain why two people have very different friendship histories. Antonucci, Ajrouch, and Birditt, 2014.
Socioemotional selectivity and perceived time
Socioemotional selectivity theory proposes that perceived future time influences social goals. When time feels expansive, people may invest in exploration, information, and new social possibilities. When time feels constrained, emotionally meaningful encounters may take greater priority. Longitudinal evidence finds that reductions in adult network size disproportionately involve peripheral partners, whereas close partners can remain comparatively stable. This pattern supports selective investment as one mechanism, but declining network size can also arise from bereavement, disability, mobility barriers, or unequal access to social spaces. Motivation and circumstance must be assessed separately. English and Carstensen, 2014.
Life-course transitions and developmental context
Life-course analysis focuses on how institutions and transitions organize contact. Children encounter peers through schools, neighborhoods, and caregiving arrangements; young adults through education, work, housing, and community activities; older adults through neighborhoods, family, organizations, and changing care environments. A major meta-analysis found that network patterns vary in association with life events such as entering employment, parenthood, and widowhood. These transitions often occur at recognizable ages, but their timing and social meaning vary across societies and generations. Wrzus et al., 2013.
These frameworks explain different levels of the same process. The convoy model describes relationship organization; socioemotional selectivity proposes a motivational pathway; life-course approaches identify historical and institutional opportunities and constraints. Each gains explanatory precision when researchers measure the relationships and conditions it actually predicts.
Early Childhood: From Shared Play to Mutual Friendship
Repeated contact and early affiliation
Early childhood offers a foundation for peer relationships through repeated play, proximity, shared routines, and developing cooperation. A young child may describe a friend as a preferred playmate and show friendship in invitations, turn-taking, imitation, and enjoyment of common activities. Caregivers, teachers, neighborhoods, and access to inclusive play environments influence which children meet repeatedly. Opportunities matter: two children may get along well but rarely become friends if their families or institutions cannot sustain contact. Friendship formation is therefore both an interpersonal process and an environmental achievement.
As language, emotional understanding, and self-regulation develop, children become more capable of discussing preferences, negotiating rules, and recognizing that friends have perspectives separate from their own. Early friendships can bring comfort and practice with disagreement. Their variability is expected, particularly when children change classrooms or activities. Short-lived connections are not automatically poor-quality relationships; a developmental interpretation asks whether the child has chances for safe, mutually enjoyable interaction.
Reciprocity and the meaning of a mutual nomination
Researchers often ask children whom they consider friends and compare their nominations. A reciprocated nomination indicates that two children identify each other as friends; a one-sided nomination records an important perception without confirming mutual recognition. This method is informative in school settings but cannot capture the entire quality of the relationship or friendships outside a measured group. Friendship quality scales include companionship, conflict, help, security, and closeness, demonstrating that reciprocal identification and experienced relationship quality are related but conceptually different. Bukowski, Hoza, and Boivin, 1994.
Adults supporting younger children can facilitate repeated play, accessible group experiences, and opportunities to recover from ordinary disagreement. They can also notice patterns of exclusion or coercion that require attention. A child's preference for fewer companions, independent play, or quieter settings should be understood through their comfort and opportunities rather than measured against an arbitrary popularity target.
Middle Childhood: Cooperation, Loyalty, and Group Membership
During the school years, friendship can acquire stronger expectations of reliability, fairness, shared interests, and loyalty. Children begin to compare the meaning of being a friend with the broader status of being liked by classmates. Popularity and friendship are different: being frequently chosen or admired in a peer group does not guarantee having a reciprocal confidant, and a child with one secure friendship may be quite satisfied without high status. Earlier developmental work specifically distinguished popularity, friendship, and emotional adjustment. Bukowski, Hoza, and Boivin, 1993.
Classroom assignments, after-school programs, transportation, disability accessibility, and household resources influence the consistency of contact. Friendships may form around games and hobbies and deepen through shared difficulties, mutual help, or repaired disagreements. Exclusion from peer groups and bullying represent distinct concerns from ordinary differences in friendship preferences. Observers should attend to who has access to participation, who experiences mistreatment, and how the child describes their relationships, rather than estimating adjustment solely from visible group size.
Adolescence: Intimacy, Identity, and Peer Relationships
Disclosure and emotional importance
Adolescence commonly brings a greater emphasis on being understood by peers, managing privacy, exploring values, and finding groups that recognize emerging identities. Close friendships may support self-disclosure and reciprocal emotional understanding. Adolescents also become more capable of experiencing differences in closeness across friendships: a long-standing activity companion, a trusted confidant, and a wider group member need not serve the same function. Friendship quality remains multidimensional, and supportive relationships can coexist with conflict. Bukowski et al., 1994.
Peer relationships can contribute to belonging and identity development, but the desire for affiliation also creates sensitivity to rejection or status threats. These developmental processes vary with school climate, local norms, family circumstances, migration, discrimination, neurodevelopmental differences, and online access. No single communication style or preferred group size constitutes the norm for every adolescent. A meaningful assessment considers personal safety, consent, reciprocity, emotional experience, and the options that are realistically available.
Why teenage friendships frequently change
School years combine increasing social experimentation with repeated transitions in classmates, interests, and identity. Friends may grow apart without betrayal when schedules, aspirations, or values diverge. A 2025 theoretical review synthesized evidence that friendship dissolution is common from childhood through young adulthood and emphasized multiple pathways, including changing circumstances and interpersonal processes. The evidence base for dissolution is stronger for younger ages than for many periods of later adulthood; extrapolation should be cautious. Santucci, Dirks, and Lydon, 2025.
An ending can produce grief or relief, and sometimes both. Whether it becomes disruptive depends on the relationship's importance, the circumstances of the ending, available alternatives, and whether the adolescent experienced aggression or exclusion. The separate Friendship Breakups: Causes, Grief, and Recovery examines friendship loss as a primary topic; here the relevant finding is that changing ties are part of developmental mobility rather than a standardized sign of individual deficiency.
The Transition From High School to Young Adulthood
A natural experiment in changing opportunity
Graduation often reorganizes friendship networks because shared classrooms, lunch periods, extracurricular routines, and neighborhood proximity may disappear almost immediately. Contact then depends more heavily on deliberate coordination, transportation, available time, compatible activities, and sometimes remote communication. This transition provides a particularly revealing case for separating interpersonal preference from structural opportunity. Friends may remain affectionate even when their everyday relationship becomes less frequent.
A 2025 network study following approximately 2,000 respondents around the high-school transition reported that only 35 percent of senior-year friendships identified in its sample remained intact one year after graduation. The result describes the study's operational definition, cohort, and follow-up; it is not a universal probability that any friendship will survive. Previously reciprocated friendships and friendships with mutual connections were more likely to persist after graduation, suggesting that relationship structure matters when institutional contact falls away. McMillan et al., 2025.
Friendship continuity and new social worlds
College, apprenticeships, employment, relocation, and new communities create opportunities to meet people with compatible interests. They also introduce economic and emotional pressures. Students or workers navigating unstable housing, financial constraints, discrimination, or mobility limitations may have less capacity for spontaneous socializing. Someone who remains in their hometown can undergo equally significant changes as friends leave or daily routines change. Developmental analysis therefore asks who experiences the transition, under which conditions, and with what social resources.
New friendships frequently develop through repeated low-pressure interaction and shared activity. Hall's studies of recent movers and first-year students found associations between time spent together and increasing friendship closeness, while everyday conversation and leisure activities also mattered. These are observed tendencies, not a stopwatch that specifies how many hours every person must spend to achieve intimacy. Hall, 2019.
Early Adulthood: Expansion, Selective Investment, and Uncertainty
The social world of early adulthood is often unusually diverse: friends from childhood coexist with educational contacts, coworkers, neighbors, partners' friends, and community groups. Some relationships become more intimate through shared personal decisions; others remain enjoyable but situational. Research aggregating cross-sectional and longitudinal studies finds that overall social networks commonly expand into young adulthood before contracting, although definitions and measurement methods influence the size of the estimated change. Wrzus et al., 2013.
The expansion of social opportunities does not guarantee satisfaction. Some adults are entering environments where they can freely choose friends for the first time; others encounter competitive work demands, unsafe settings, or language barriers. Friendship may be central to practical support and identity, but it should not be forced into a single psychological function. Adults may rely on a friend for humor and recreation, another for intellectual conversation, and a third for a history of mutual assistance.
Dating, partnerships, and friendship networks
Forming an intimate partnership may reorganize schedules, shared social circles, and household responsibilities. Friendships can gain new pathways through a partner's connections or become harder to sustain as time is redistributed. Neither outcome is inevitable. Likewise, remaining single can shape opportunities without determining the quality of friendship. The appropriate developmental question is how people allocate limited time and preserve reciprocal recognition, not whether one relationship category ought to replace another. When loneliness is specifically located inside a marriage, that problem has a distinct canonical owner outside this article.
Reciprocity through life changes
Reciprocity is a long-term orientation toward mutual care and recognition rather than an exact exchange of equal messages, money, or availability every week. Friends may alternate between giving and receiving support during exams, job searches, illness, pregnancy, unemployment, or caregiving. A relationship becomes difficult when asymmetry remains unacknowledged or a person's needs are chronically disregarded. Even then, the underlying reason may be capacity, access, or incompatible expectations rather than indifference. The detailed practices belong to How to Maintain Friendships: Reciprocity, Contact, and Care.
Midlife: Friendship Under Competing Responsibilities
Midlife combines diverse family structures, work arrangements, caregiving roles, and evolving health conditions. Many adults have fewer opportunities for spontaneous gatherings than they did at school or in early adulthood. Contact may become scheduled, geographically dispersed, or integrated into family and community routines. This change in frequency may be experienced as loss by some people and as a manageable reorganization by others. It is essential to assess desired rather than idealized contact levels.
The Wrzus meta-analysis found that friendship-network size declines across adulthood, while network change is associated with life events. A 2025 Swedish population study of reported friend-contact frequency, spanning surveys from 1968 to 2021, identified a decline after young adulthood, a midlife plateau, and a further decline in later ages. Different measures, national contexts, and cohorts limit direct comparison between these findings. Network size and frequency of visiting are not interchangeable. Wrzus et al., 2013; Augustsson et al., 2025.
Why long friendships sometimes become less visible
A person may remain deeply attached to a friend while making fewer calls or visits. Shared history can preserve familiarity, but affection alone cannot guarantee continuous availability or mutual understanding. Over time, priorities can diverge, and an old friendship may need a different conversational rhythm or a revised set of expectations. Some ties withstand infrequent contact because trust has accumulated; others weaken because opportunities to share meaningful experiences vanish. Cross-sectional snapshots cannot determine which mechanism is operating.
Midlife friendship and unequal social opportunity
Caregiving, disability, shift work, rural residence, travel costs, family responsibilities, and social exclusion affect who has time and access to maintain friendships. Economic circumstances may determine whether people can join clubs, travel to meet friends, or enjoy quiet private spaces. A useful practical recommendation should therefore consider accessible and affordable arrangements rather than assuming free evenings and flexible transportation. Community infrastructure, workplace organization, and neighborhood safety all shape what seems at first to be an individual friendship decision.
Later Adulthood: Continuity, Selection, and New Bonds
The pattern of smaller networks
On average, social networks often become smaller in later adulthood, but the nature of the change matters. Longitudinal work found that peripheral partners account for much of the reduction, while close partners are comparatively stable. The result is compatible with socioemotional selectivity, but the same period can involve retirement, relocation, bereavement, changes in transportation, or reduced accessibility. Researchers must distinguish a preferred smaller circle from involuntary social loss. English and Carstensen, 2014.
An older adult's network may include friends made at school, at work, through parenting, in a neighborhood, through faith or civic organizations, or recently through volunteering and shared activities. Long duration is one possible source of intimacy; it is not the only one. New friendships can offer intellectual exchange, shared enjoyment, and a recognition of present identity that differs from friendships grounded in distant history. A specialist analysis appears in Friendships and Social Networks in Older Adulthood: Change and Continuity.
Health, accessibility, and the possibility of new friendship
Hearing, vision, mobility, chronic illness, transportation, financial resources, and neighborhood design can change the practical conditions of meeting others. A person who enjoys close contact may nonetheless encounter obstacles in sustaining it. Accessible spaces, predictable schedules, and opportunities for reciprocal participation can support continued connection. These conditions should be understood as social and structural determinants as well as personal circumstances. Maintaining agency and respecting preferences remain central even when assistance is needed.
Cohort differences and changing norms
Birth cohorts grow up with different communications technologies, educational opportunities, gender roles, and expectations about friendship. Augustsson and colleagues' 2025 analysis found that later-born Swedish cohorts reported frequent contact with friends more often at many ages, although differences converged in advanced age. Its historical survey data are valuable for distinguishing age from period and cohort, but the contact measure emphasizes visiting and cannot represent all digital, emotional, or transnational connections. Findings from one country require testing elsewhere. Augustsson et al., 2025.
The Shape of Friendship Networks Across Adulthood
The quantitative synthesis of 277 studies involving 177,635 people offers a landmark account of broad social network development. Wrzus and colleagues reported that the overall network increased into young adulthood and decreased thereafter; personal and friendship networks declined across adulthood, while family-network size was more stable in the aggregate. The study combined cross-sectional and longitudinal evidence and identified methodological moderators. These patterns summarize averages, not prescriptions for any individual's social network. Wrzus et al., 2013.
A central methodological distinction is that the size of a named network depends on the question used to elicit it. Asking about emotionally important people generates a different boundary than asking whom someone spoke with last week. Studies that ask for three names impose different limits than studies permitting unlimited nominations. An apparent developmental decline may partly reflect changing opportunity, motivation, recall, or survey design. Researchers should state exactly which kind of network was measured before offering an explanation.
Friendship Quality Versus Friendship Quantity
Quantity can increase the range of potential companionship and practical help, but friendship quality concerns whether relationships are mutually meaningful and safe. In childhood, validated multidimensional scales distinguish companionship, security, help, closeness, and conflict. Across adulthood, social convoy analyses likewise emphasize tie quality and differentiated relational functions. Counting friends alone omits whether someone feels able to confide, whether contact is welcome, and whether the relationship can endure disagreement. Bukowski et al., 1994; Antonucci et al., 2014.
Friendship quality may shift as a relationship moves from shared activity to emotional support, from daily companionship to occasional exchange, or from practical dependence to greater independence. These transformations are not necessarily declines. Some friends rediscover closeness after years of distant contact; others remain cordial but no longer share central parts of life. A lifespan view tracks change in meaning as well as change in frequency.
Do Friendships Really Last a Lifetime?
Some friendships persist for decades, but no particular duration is required for a relationship to have been valuable. The capacity to survive transition depends partly on mutual recognition, shared social connections, geography, opportunities to interact, and whether the friendship remains meaningful to both people. A study spanning the transition out of high school found that reciprocal friendships and shared friendship circles were associated with greater persistence after graduation. This identifies correlates of continuity in a specific setting; it does not establish guarantees or a universal algorithm for lasting friendship. McMillan et al., 2025.
Research into friendship dissolution adds another caution: the same observable ending may reflect ordinary developmental separation, conflict, personal growth, environmental barriers, or protection from a harmful relationship. Consequently, lifelong persistence should not become the standard against which every meaningful friendship is judged. Santucci et al., 2025.
Gender, Culture, and Individual Variation
Friends and friendship expectations are embedded in culture. Norms about disclosure, emotional language, shared activities, family duties, gender roles, and independence affect how relationships are expressed and reported. Some people prefer frequent verbal contact; others demonstrate loyalty through reliable action or shared pursuits. Generalizations about women, men, or cultural groups are particularly vulnerable to confounding by social role, measurement choice, and unequal opportunity. A robust lifespan account compares contexts and asks how social norms interact with access and personal priorities.
Individual differences in temperament, sensory needs, energy, health, disability, migration histories, sexuality, family structure, and work conditions also shape social opportunity and preference. These variations should not be interpreted as symptoms solely because a person maintains an atypical number of friendships or communicates in an unconventional way. When someone describes distress, relational coercion, or persistent unwanted isolation, those concerns deserve attention on their own terms. Friendship diversity is an empirical feature of social life.
Digital Communication and Friendship Over Time
Text messaging, video calls, gaming communities, social networking services, and online interest groups can sustain or initiate friendships across geographic distance. Their roles differ according to whether the relationship is reciprocal, how people communicate, whether encounters feel meaningful, and whether participation complements or displaces desired experiences elsewhere. The existence of online communication does not itself determine relationship quality. The distinction between friends, followers, audience members, and anonymous encounters is crucial when interpreting digital network counts.
Digital opportunities also vary with age, accessibility, broadband availability, privacy, disability accommodations, and digital literacy. An older adult may preserve a long-distance friendship through video calls; an adolescent may find a supportive peer group outside school; a migrating adult may stay connected to friends across borders. These possibilities coexist with risks of harassment, exposure, pressure, and exclusion. The specialized evidence concerning platforms and loneliness belongs to Social Media and Loneliness: Connection, Displacement, and Individual Differences.
AI companions and conversational systems are a different relationship category from reciprocal human friendship. Psychological attachment to a conversational system can be studied as an experience without assuming a human social partner, mutual obligations, or proven machine subjective experience. Findings about human friendship trajectories cannot simply be transferred to AI interactions. Likewise, evidence for clinical digital interventions should be evaluated separately from everyday messaging and online friendships.
Friendship, Well-Being, and Health Across the Life Course
Friendship can contribute companionship, belonging, practical assistance, enjoyment, emotional recognition, and opportunities for shared activity. Studies of social relations frequently find associations between positive relationships and well-being, while negative interaction and relational strain can also be consequential. A longitudinal friendship-trajectory study followed adults across several waves and found diverse patterns in friend presence and positive and negative quality; its selection of respondents who reported a best friend at every wave limits conclusions about people whose friendships ceased or who lacked a best friend. Ajrouch et al., 2024.
Population research also links social isolation and loneliness to health outcomes, as summarized in the WHO Commission on Social Connection report, 2025. Those findings should not be recast as a causal estimate of friendship itself. Observational associations may involve reverse causation, preexisting health, material deprivation, mobility, personality, or broader social environments. Whether a change in friendship quality causes a particular medical outcome requires research designs capable of testing causal pathways. Loneliness is not a psychiatric diagnosis, and a modest network is not evidence of a disorder.
How Researchers Study Developmental Change
Longitudinal versus cross-sectional evidence
Cross-sectional research compares individuals of different ages at a given point or within a short window. It can describe age differences but cannot fully distinguish aging from generational differences. Longitudinal research follows the same people over time and reveals within-person transitions, although attrition, selective survival, and repeated measurement may bias conclusions. Cohort-sequential designs combine age groups observed over multiple historical periods, helping identify age, period, and cohort effects under explicit assumptions. Wrzus et al., 2013; Augustsson et al., 2025.
Reciprocity, quality, and contact measurement
Researchers may ask participants to list friends, nominate classmates, rate closeness, report contact frequency, or keep interaction diaries. Reciprocal school nominations reveal mutual recognition within a bounded setting; quality questionnaires examine particular relational dimensions; network inventories record composition and role. Each method answers a specific question, and no single score fully captures friendship experience. Reports from two friends can differ, and those differences can themselves provide information about perceived mutuality. Bukowski et al., 1994.
Biases that matter for lifespan conclusions
Older participants who remain in a long-running study may differ from those lost to follow-up. People with severe illness or unstable housing may be less available to participate. Surveys may define a friend differently across cultures, age groups, or historical periods. Retrospective accounts of childhood friendship are subject to memory reconstruction. Network measures that cap nominations may hide changes in wider networks. These limitations demand careful interpretation of numerical age profiles and prevent treating a single chart as the definitive life course of friendship.
What the Latest Evidence Adds
The 2025 Swedish cohort analysis directly challenges simple statements that older generations necessarily maintain fewer friendships than younger ones. It finds increases in reported frequent friend contact for later-born cohorts at many ages while also finding convergence in older age. Because its measure concerned visiting friends and its setting was Sweden, the result is strongest as evidence about reported contact in that society rather than global friendship quality. Augustsson et al., 2025.
The 2025 high-school transition study identifies reciprocity and shared social connections as correlates of persistence when a shared institutional setting ends. A separate 2025 review emphasizes how friendship loss differs by circumstances and by developmental stage, and points to limitations in adult dissolution research. Together these studies reinforce a central developmental finding: friendship changes are partly relational processes and partly consequences of changing environments. McMillan et al., 2025; Santucci et al., 2025.
Research published during 2026 should be assessed for relevance and methodological quality rather than included simply for recency. The enduring evidence base of social network development, relationship measurement, and long-term observation remains necessary for interpreting new findings. Published evidence establishes heterogeneous developmental trajectories and meaningful associations; many mechanisms and intervention effects remain uncertain.
Supporting Friendship Through Changing Life Stages
The practical implications follow from observed mechanisms rather than a universal formula. Children benefit from opportunities for safe repeated interaction and age-appropriate cooperation. Adolescents may benefit from accessible groups, respectful privacy, protection from bullying, and room to renegotiate ties. Young adults often need ways to convert new encounters into repeated meaningful contact. Midlife friendships may be supported by predictable routines and realistic expectations of availability. In later life, transportation, accessibility, safe local places, and opportunities for reciprocal participation can matter as much as motivation.
One practical principle is to distinguish a relationship that has changed from one that has ended. Contact may be less frequent because someone has new responsibilities or reduced mobility, while mutual trust remains. Another is to allow different friendships to serve different functions without demanding identical intensity from each. People can ask what kinds of contact each friend can realistically sustain, share activities that fit their circumstances, and communicate when changing expectations create uncertainty.
The effectiveness of any specific strategy depends on context and evidence. Hall's observational studies associate shared time and meaningful interaction with growing friendship closeness, but they do not prove that a fixed schedule will work for everyone. Community programs may increase opportunity yet fail to provide reciprocal, personally valued relationships for some participants. Programs directed at loneliness should be evaluated using appropriate outcome measures, comparison groups, and follow-up; reduced contact and subjective loneliness should never be conflated. Hall, 2019; WHO, 2025.
Open Questions in Friendship Development
Important scientific questions remain. Which features of reciprocity best predict the survival of friendships across transitions? How do mutual understanding and closeness change when communication becomes predominantly digital? What protects friendships during migration, illness, and long periods of geographic separation? Which interventions expand equitable access to meaningful friendship without prescribing an ideal level of sociability? How do different cultural meanings of friendship alter the interpretation of questionnaires developed in particular societies? Which apparent age trends reflect historical cohorts rather than aging itself?
A further challenge is to study friendship across the entire life course within the same populations. Many studies concentrate on one stage, such as adolescence or old age, and use incompatible definitions of a friend. Better harmonized longitudinal measures, reciprocal reports, diverse sampling, and careful treatment of attrition can sharpen developmental conclusions. Research should also give sustained attention to people with few friends, changing friendship preferences, or restricted opportunities who are frequently excluded from long follow-up studies.
Frequently Asked Questions
Do people normally have fewer friends as they get older?
Many studies find that named friendship networks become smaller on average across adulthood, especially in more peripheral ties. The size and timing of the change depend on how friendship is measured, life events, historical cohort, and personal circumstances. A smaller circle can be satisfying or unwanted. Wrzus et al., 2013; English and Carstensen, 2014.
At what age are friendships most important?
Friendships can matter at every age, though their tasks change. Play and cooperation often feature in childhood, identity and intimacy become salient in adolescence, companionship and support intersect with adult roles, and emotionally meaningful continuity can be prominent later in life. Research does not establish a universal age at which friendship stops mattering. Antonucci et al., 2014.
Is it normal to lose friends after high school?
Friendship turnover is common when daily institutional contact ends. One 2025 study found substantial turnover in its cohort during the year after graduation, with reciprocity and friends in common associated with persistence. Individual experiences vary, and meaningful friendships can continue or form in new contexts. McMillan et al., 2025.
Can people make close friends in middle or older adulthood?
Yes. Friendship formation remains possible as contexts and opportunities change, although work schedules, access, health, and transportation may shape opportunities. Repeated shared activity and meaningful exchange can contribute to closeness, but no predetermined number of hours produces the same outcome for everyone. Hall, 2019; Antonucci et al., 2014.
Does seeing friends less often mean the friendship is weaker?
Frequency is one indicator of contact, while trust, mutual recognition, security, and satisfaction are others. A friendship can remain meaningful with less frequent meetings; a frequently contacted relationship can be conflictual or emotionally distant. Understanding the change requires considering quality, access, preference, and the circumstances of both people. Bukowski et al., 1994.
Is a person with few friends necessarily lonely?
No. The number of friends does not define subjective loneliness, chosen solitude, or belonging. People differ in desired connection and the quality of their relationships. The WHO treats loneliness and social isolation as distinguishable concerns, and neither is established by simply counting a person's friends. WHO, 2025.
Can online friendships be meaningful across different ages?
Digital communication can provide repeated contact, shared activities, and ways to sustain distant relationships, although reciprocity, safety, access, and individual preference remain important. Evidence about online friendships should be interpreted separately from follower counts, AI companions, and clinical digital interventions. General claims that all online contact either helps or harms connection exceed what the available evidence supports.
Developmental Synthesis
Friendship development is an ongoing reorganization of opportunities, commitments, and personally meaningful ties. Childhood introduces mutual play and emerging trust; adolescence broadens intimacy and identity-based affiliation; young adulthood exposes networks to expansion and transition; midlife reorganizes contact around competing responsibilities; and later adulthood can bring selective investment alongside barriers, new ties, and loss. These patterns arise from changing motivations, social institutions, and historical conditions. Strong conclusions require measuring both who is connected and how those connections are experienced. Wrzus et al., 2013; Antonucci et al., 2014; Augustsson et al., 2025.
Related Articles
Friendship Psychology: How Friendships Form and Last — The formation, functions, and enduring qualities of friendship as a relationship.
How to Maintain Friendships: Reciprocity, Contact, and Care — Practical and psychological mechanisms of friendship maintenance.
Friendship Breakups: Causes, Grief, and Recovery — Why friendships end and how people experience and adapt to their loss.
Friendships and Social Networks in Older Adulthood: Change and Continuity — Specialist analysis of friendships and social networks in later life.
Social Support: Types, Sources, and Psychological Benefits — The forms and perceived functions of support across relationships.
Social Media and Loneliness: Connection, Displacement, and Individual Differences — What evidence distinguishes digital social connection from loneliness.
Friendship Quality: Trust, Intimacy, Support, and Satisfaction
Loneliness in Adolescents: Peer Belonging, Exclusion, and Development
Loneliness in Young Adults and College Students: Transitions and Social Networks
References
Ajrouch, K. J., Hu, R. X., Webster, N. J., and Antonucci, T. C. (2024). Friendship trajectories and health across the lifespan. Developmental Psychology, 60(1), 94–107. https://doi.org/10.1037/dev0001589.
Antonucci, T. C., Ajrouch, K. J., and Birditt, K. S. (2014). The convoy model: Explaining social relations from a multidisciplinary perspective. The Gerontologist, 54(1), 82–92. https://doi.org/10.1093/geront/gnt118.
Augustsson, E., Celeste, R. K., Fors, S., Rehnberg, J., Lennartsson, C., and Agahi, N. (2025). Friends and trends: Friendship across life phases and cohorts. Archives of Gerontology and Geriatrics, 135, 105872. https://doi.org/10.1016/j.archger.2025.105872.
Bukowski, W. M., Hoza, B., and Boivin, M. (1993). Popularity, friendship, and emotional adjustment during early adolescence. New Directions for Child and Adolescent Development, 1993(60), 23–37. https://doi.org/10.1002/cd.23219936004.
Bukowski, W. M., Hoza, B., and Boivin, M. (1994). Measuring friendship quality during pre- and early adolescence: The development and psychometric properties of the Friendship Qualities Scale. Journal of Social and Personal Relationships, 11(3), 471–484. https://doi.org/10.1177/0265407594113011.
English, T., and Carstensen, L. L. (2014). Selective narrowing of social networks across adulthood is associated with improved emotional experience in daily life. International Journal of Behavioral Development, 38(2), 195–202. https://doi.org/10.1177/0165025413515404.
Hall, J. A. (2019). How many hours does it take to make a friend? Journal of Social and Personal Relationships, 36(4), 1278–1296. https://doi.org/10.1177/0265407518761225.
McMillan, C., Jones, K. A., Jacobsen, W. C., Ramirez, N. G., and Feinberg, M. E. (2025). Friends forever? Correlates of high school friendship (in)stability from adolescence to young adulthood. Social Networks, 82, 27–38. https://doi.org/10.1016/j.socnet.2025.02.002.
Santucci, K., Dirks, M. A., and Lydon, J. E. (2025). With or without you: Understanding friendship dissolution from childhood through young adulthood. Journal of Social and Personal Relationships, 42(9), 2675–2699. https://doi.org/10.1177/02654075251334226.
World Health Organization. (2025). From loneliness to social connection: Charting a path to healthier societies. Report of the WHO Commission on Social Connection. https://www.who.int/publications/i/item/978240112360.
Wrzus, C., Hänel, M., Wagner, J., and Neyer, F. J. (2013). Social network changes and life events across the life span: A meta-analysis. Psychological Bulletin, 139(1), 53–80. https://doi.org/10.1037/a0028601.