Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Family Systems, Roles, and Relational Boundaries

Abstract. Family Systems, Roles, and Relational Boundaries investigates how recurrent patterns of participation, responsibility, authority, communication, and care organize family life across dyads, triads, and wider kinship arrangements. Through a critical narrative synthesis of historical family-systems theory, observational research, psychometric validation, prospective studies, meta-analyses, and intervention trials, this monograph examines the empirical meanings of subsystems, boundary permeability, enmeshment, parentification, triangulation, conflict, and role allocation. It evaluates measurement validity, reciprocal influence, individual and cultural variation, developmental transitions, material constraints, and methodological limitations in causal inference. The evidence supports studying differentiated family processes through multiple informants and repeated observations, while emphasizing the limits of universal family labels and of causal claims based on observational association. The synthesis identifies research priorities for precise operationalization, replication, member-specific outcomes, safety, and technology-mediated coordination.

Keywords: family systems; family roles; relational boundaries; family subsystems; structural family therapy; family systems theory; triangulation; family coalitions; enmeshment; parentification; boundary dissolution; coparenting; interparental conflict; family assessment; family functioning; family observation; intergenerational relations; family stress; developmental transitions

Introduction: Families as Organized Relational Systems

A family system is an enduring arrangement of relationships in which members coordinate care, resources, authority, belonging, and individual development through recurring interaction. The research object is the patterned organization of these relationships across people and time. Family structure refers to who participates and the positions they occupy; family process refers to how members act, interpret, respond, and revise their conduct; family functioning concerns performance of activities and the consequences for participants. These constructs become scientifically informative when translated into observable indicators such as caregiving allocation, conversational turn taking, conflict sequences, decision authority, or reports from different family members. Classical systems approaches proposed that reciprocal influence and patterned interaction deserve attention alongside individual attributes Minuchin (1974) Bowen (1978).

This monograph examines systemic organization, roles, subsystems, coalitions, feedback, and relational boundaries. Its central questions are how family members coordinate differentiated responsibilities; how intergenerational and interpersonal boundaries organize autonomy and care; which configurations predict functioning or difficulties; how roles change after stress and transition; and how researchers distinguish plausible interactional mechanisms from interpretive labels. It integrates historical theoretical texts, psychometric validation, behavioral observation, prospective and cross-sectional evidence, contemporary meta-analyses, and selected intervention trials. It is a critical narrative synthesis rather than a registered systematic review or new experimental report. The analysis distinguishes empirically measured interaction patterns from constructs inherited from the language of family therapy Epstein et al. (1983) Thompson et al. (2024).

Family outcomes are multidimensional. One household may sustain stable membership while exposing a young person to frightening conflict; another may be geographically distributed yet provide reliable care, effective decision making, and recognizable individual space. Family health cannot be inferred from household composition, marital status, or whether members resemble a culturally familiar arrangement. Assessment instead asks who has access to support and authority, how responsibility is distributed, what happens when expectations diverge, and whether relational arrangements permit adaptation and safety. This analytical stance follows ecological approaches to parenting and coparenting, in which individual resources, material pressures, and relationships jointly shape developmental opportunities Belsky (1984) Feinberg (2003).

The System Concept and Its Explanatory Commitments

A system description identifies elements, relations among elements, recurrent exchanges, and processes that change those relations. In family research the elements are interacting persons, but the unit of explanation may be a dyad, a triad, a household network, or a sequence of events. Reciprocal influence means that one person's conduct can alter another's opportunities and responses, which may feed into subsequent behavior. It does not imply equal power or equal responsibility. A child and caregiver can affect one another while legal authority, physical dependence, and access to income remain asymmetrical. Family-systems hypotheses become testable when they specify a temporal ordering, a measurable response, and an alternative account Minuchin (1974) Feinberg (2003).

Feedback is a way of describing how the consequences of one interaction enter later exchanges. If criticism produces withdrawal, and withdrawal prompts more criticism, investigators can test the ordering and recurrence of those behaviors with repeated observation. A feedback claim requires several observations and evidence about timing; a single interview or memorable anecdote cannot establish a stable loop. Processes that preserve habitual patterns sometimes accompany the capacity to coordinate everyday life, and sometimes maintain escalating disputes. Processes that produce change can promote flexibility or destabilize reliable caregiving. The evaluation therefore depends on specific outcomes, context, and the interests of participants rather than on assigning moral value to stability or change Steinhauer et al. (1984) Walker et al. (1984).

Systemic language poses a special inferential risk when relational metaphors are treated as independently established causes. Terms such as homeostasis, permeability, triangulation, and differentiation can guide observation, but their meanings shift across clinical traditions and research measures. Demonstrating that a family member experiences divided loyalty does not by itself prove an unconscious system-level intention to preserve a particular balance. Scientific reasoning proceeds from defined variables to observations, comparisons, predictive tests, and causal evaluation. It asks which actors perform which actions under identifiable conditions, whether the pattern repeats, and whether the same findings could result from economic strain, individual vulnerability, or institutional constraints Grych & Fincham (1990) Conger et al. (1992).

Historical Formation of Family Systems Thought

Mid-twentieth-century family therapy developed at the intersection of psychiatry, communication research, social theory, and clinical work with households facing significant difficulties. Murray Bowen's collection Family Therapy in Clinical Practice, published in 1978, assembled writings on differentiation, triangulation, multigenerational patterns, and relational anxiety. Salvador Minuchin's Families and Family Therapy, published in 1974, advanced structural concepts involving subsystems, boundaries, coalitions, and reorganizing interaction during treatment. These works were influential conceptual programs and sources of clinical hypotheses. Their original cases and theoretical commitments constitute historical evidence about the field's development, while contemporary claims about causation and clinical effectiveness require independent empirical tests Bowen (1978) Minuchin (1974).

A subsequent movement toward assessable family dimensions made systemic thinking more explicit. The McMaster model articulated problem solving, communication, roles, affective responsiveness, affective involvement, and behavior control. Epstein, Baldwin, and Bishop developed the Family Assessment Device as a questionnaire with seven scales, including a general functioning measure. Its 1983 development report presented observations from 503 individuals. Other models, including the Process Model of Family Functioning, emphasized dynamic links among relational dimensions. These contributions made broad clinical judgments susceptible to psychometric examination, while also exposing disagreement over whether scales reflect individual perceptions, shared family attributes, or behavior directly observed across members Epstein et al. (1983) Steinhauer et al. (1984).

The later research history is characterized by methodological differentiation rather than a single linear succession of schools. Investigators developed structured triadic tasks, multi-informant longitudinal designs, sibling comparisons, family observation protocols, and quantitative syntheses of boundary phenomena. Contemporary scholarship can test some structural predictions by coding who includes or excludes whom in interaction, and can evaluate broader outcome claims using prospective cohorts. The durability of family-systems concepts partly reflects their capacity to organize complicated interpersonal observations, yet the historical survival of a concept never substitutes for construct validity. The evidential standing of each claim depends on the measured pattern and the quality of the relevant research design Fivaz-Depeursinge et al. (1996) Thompson et al. (2024).

Family Membership, Households, and Relational Networks

Family membership is defined through biological, legal, adoptive, marital, caregiving, social, and personally recognized ties. Household residence is one important arrangement, yet it does not capture extended kin, separated coparents, shared custody, migration, stepfamily relations, or persistent nonresidential caregiving. A study that recruits two married parents and their children samples a particular institutional form; conclusions about families in general require evidence from additional configurations. The appropriate research unit depends on the question. Investigations of household finances may require household data; studies of support networks need relationships extending beyond residence; analyses of child safety must identify actual caregivers and their access to decision power Belsky (1984) Feinberg (2003).

The distinction between formal roles and enacted relationships matters. A person named as a parent may provide little daily support, while another relative undertakes transportation, monitoring, emotional assistance, and school coordination. Formal status influences legal rights and obligations, but actual family functioning emerges from repeated exchanges. Researchers should identify who performs tasks, who can initiate decisions, whose preferences are acknowledged, and who absorbs the costs of disagreement. Measures limited to family composition can obscure the organization of care. Conversely, measuring only affectionate reports can miss unequal workloads, restricted autonomy, or exposure to controlling behavior Epstein et al. (1983) Thompson et al. (2024).

A relational network also has a time dimension. Membership and practical access change with birth, death, separation, migration, illness, school transitions, employment, and aging. The presence of more caregivers may create redundancy and support, while conflicting instructions or unpredictable availability can complicate coordination. A credible systems study therefore records who belongs to the network at each wave and how the investigator defines a stable relationship. Ideally, it compares what different members report with observable exchanges and the external resources governing available choices. Such a design can detect change without assuming that every alteration in residence or kinship entails relational dysfunction Conger et al. (1992) Walker et al. (1984).

Subsystems, Dyads, and Triadic Organization

A subsystem is a recurring grouping within a family whose members carry out distinguishable interactional functions. Parent-child relationships, partnerships, sibling ties, and coparenting collaborations may overlap in membership and vary in autonomy. The structural model assigns particular importance to how subsystems coordinate while maintaining appropriate generational responsibilities. Yet the descriptive fact that a child talks with one caregiver more often than another does not establish a pathological coalition. Interpretation requires evidence about exclusion, instrumental use of the child, decision pressure, conflict, and consequences for the child's autonomy and security Minuchin (1974) Feinberg (2003).

Triadic research provides a useful extension beyond dyads. In the Lausanne Trilogue Play, investigators observe two caregivers and a child in structured interactive configurations, coding participation and coordination rather than relying solely on a questionnaire about family quality. Early work described assessment of a father-mother-infant alliance, and later longitudinal studies examined different interactional patterns in families classified as alliances or coalitions. These procedures offer observable indicators for structural constructs, but their validity depends on the task, reliability of coding, sampling, and prediction outside the laboratory. An interaction observed during play is a situated sample of family behavior, not a complete portrait of private life Fivaz-Depeursinge et al. (1996) Fivaz-Depeursinge et al. (2009).

Dyadic and triadic questions cannot always be answered by the same analysis. A caregiver may be responsive when alone with a child but become withdrawn when the second caregiver enters; the family pattern is then conditional on the presence of a third person. Conversely, a child's discomfort in a three-person task could reflect novelty, fatigue, developmental abilities, or temporary stress. Researchers need repeated contexts, comparison tasks, and information about member-specific characteristics. They must also distinguish a child being included respectfully in a family discussion from the child being recruited to manage adult hostility. This distinction anticipates the evidence on triangulation and parent-child boundary dissolution Fosco & Grych (2010) Thompson et al. (2024).

Relational Boundaries as Empirical Constructs

A relational boundary specifies patterned access to information, emotional involvement, authority, obligations, and influence between members or subsystems. Boundary clarity is therefore a collection of operational questions: who may disclose a private experience, who decides about schooling, who is expected to soothe adult conflict, and who can decline a demand without disproportionate retaliation? A boundary can be coherent in one functional domain and diffuse in another. Privacy in an adolescent's friendships, for example, can coexist with shared responsibility for physical safety. The conceptual distinction between rigid, diffuse, and flexible boundaries gained prominence in structural theory, but a rigorous evaluation requires situational coding or measures with demonstrated predictive validity Minuchin (1974) Olson (2011).

Boundary permeability changes across development and circumstances. A young child reasonably depends on adult assistance with many tasks; increased competence may permit independent decisions while reliable support remains available. Additional constraints arise from disability, illness, migration status, economic dependence, and legal custody. A single uniform ideal of distance between relatives cannot capture these demands. The most useful measures specify whether information flows with consent, whether responsibility is appropriate to available capacity, and whether an individual can exercise age-appropriate agency. Across family structures, the relevant question is how obligations and authority are coordinated under real conditions, with particular attention to coercive conduct and unequal power Belsky (1984) Thompson et al. (2024).

Contemporary evidence gives empirical content to some boundary formulations. Thompson, Platts, and Davies synthesized 478 studies concerning parent-child boundary dissolution and children's psychological difficulties. Their distinction among enmeshment, disorganization, caregiving, and coerciveness makes clear that the field contains multiple patterns with different associations. The aggregate evidence concerns associations with internalizing and externalizing outcomes, not a diagnostic test for a particular household. Boundary labels should consequently be tied to identifiable behaviors, developmental expectations, and outcome measures. Researchers must avoid treating the same observed act as intrinsically harmful without considering context, frequency, consent, and alternatives available to family members Thompson et al. (2024).

Enmeshment, Disengagement, and Differentiation

Enmeshment describes a pattern of excessive intrusion or involvement in which individuality and generational distinctions become difficult to maintain. Disengagement describes insufficient accessibility or coordination relative to the demands of a relationship. The circumplex approach locates family cohesion along a continuum and treats extreme positions as potentially problematic under its theoretical model. FACES IV was designed to assess balanced cohesion and flexibility alongside hypothesized unbalanced configurations such as enmeshment, disengagement, rigidity, and chaos. Validation evidence supports the usefulness of specified scales while leaving the precise interpretation and cross-context generality of factor scores open to further testing Olson (2011) Alexander et al. (1984).

The two extremes should not be inferred simply from frequency of contact. Frequent consultation can be adaptive in households with substantial caregiving needs, and limited contact can reflect a deliberate, mutually respected adult arrangement. Clinically meaningful impairment depends on the consequences of intrusion, unavailability, coercion, or unmet responsibilities. The same level of shared activity may be associated with distinct outcomes across ages and cultures. Historical differentiation language emphasizes a person's ability to maintain thoughtful functioning during relational stress, yet empirical measurement must identify independent cognitive, emotional, and behavioral indicators rather than treating therapist impressions as a validated score Bowen (1978) Olson (2011).

The debate also reveals a limitation of global family judgments. Some members may experience a relationship as supportive while another experiences its demands as intrusive. Aggregating these voices into a single family score can conceal meaningful asymmetries. Research should estimate shared variance and person-specific disagreement, use self-reports alongside direct observation, and compare findings across situations. Psychometric investigations of early FACES scales found weak agreement among family members and limited correspondence with therapist ratings in their samples. These historical critiques are consequential because a measurement instrument can appear internally consistent while measuring an individual's perception rather than a stable family-level property Alexander et al. (1984) Walker et al. (1984).

Role Allocation, Responsibility, and Role Conflict

Family roles are recurrent expectations concerning care, income, protection, education, emotional support, household labor, decision making, and coordination. An enacted role consists of observable tasks and decisions, whereas a perceived role consists of expectations and judgments about who ought to perform them. Role ambiguity arises when members hold conflicting assumptions or tasks lack a reliable owner. Role overload arises when requirements exceed available time, capacity, or resources. These conditions can be studied through task diaries, structured interviews, observational coding, and reports from multiple members. The McMaster model places roles within a wider set of functioning domains because their effectiveness depends on communication and problem solving Epstein et al. (1983) Miller et al. (1985).

Role distribution is shaped by institutions as well as personality. Employment conditions, access to paid leave, family law, disability services, and social expectations influence the work that relatives can undertake. Financial pressure may increase parental emotional distress and conflict while narrowing practical choices about care. In the 1992 Iowa study of early adolescent boys, Conger and colleagues tested a proposed process linking economic hardship, perceived pressure, parental distress, marital conflict, disrupted parenting, and youth adjustment. Their evidence combined trained observations and family reports, highlighting pathways through which external conditions affect intimate processes. Such findings make a purely intrafamilial interpretation incomplete Conger et al. (1992) Belsky (1984).

Roles must also be distinguished from character judgments. Describing one relative as the responsible child or the difficult parent can compress years of changing interactions into a fixed identity and invite confirmation bias. An investigation instead records tasks by context, opportunities to refuse, negotiation procedures, changes following stress, and whether the role is recognized by other members. A child may competently contribute to household chores without being burdened with adult emotional regulation. A parent may delegate decision making while retaining ultimate responsibility. The criterion of interest is a situated distribution of responsibility with measurable consequences for autonomy, care quality, and adjustment Thompson et al. (2024) Epstein et al. (1983).

Parentification and Intergenerational Role Reversal

Parentification refers to forms of role reversal in which a child assumes caregiving or emotional responsibilities ordinarily expected of adults. Research distinguishes instrumental assistance from emotional responsibility for a parent's wellbeing, and distinguishes occasional contributions from chronic obligations that interrupt development or leave the child unable to seek support. Any measure requires attention to age, duration, perceived fairness, resources, cultural expectations, and whether adults retain accountability. Treating every act of filial help as parentification would confound cooperative family life with burdensome boundary dissolution. Contemporary synthesis explicitly isolates the caregiving pattern from enmeshment, disorganization, and coerciveness in the parent-child subsystem Thompson et al. (2024) Feinberg (2003).

The developmental pathways associated with these arrangements are varied. A child asked repeatedly to mediate adult quarrels may become vigilant to changes in tone and anticipate conflict even in unrelated settings. Another child caring for an ill family member may acquire skills and solidarity while also experiencing exhaustion or restricted opportunities. An observational and longitudinal design should measure practical demands, emotional obligation, adult availability, and the child's own sense of agency. Outcomes ought to include wellbeing, schooling, peer participation, and family security instead of presuming that apparent competence always indicates positive adjustment. Evidence of statistical associations does not establish that the role assignment alone caused a later symptom Thompson et al. (2024) Grych & Fincham (1990).

There are important ethical implications. Children cannot negotiate obligations with the same authority as adults, and reports of willing participation require interpretation in light of dependence and loyalty. Qualitative interviews may reveal obligations hidden in household-level scales, while repeated quantitative measures can test how responsibilities and outcomes co-develop. Family-level explanations must not distribute accountability for coercive or abusive adult behavior onto children. Interventions, where warranted, should assess environmental supports and redistribute responsibility through adults and institutions. The relevant scientific question concerns the conditions under which a developmentally appropriate contribution becomes an enduring, harmful role, and whether changing those conditions improves outcomes Belsky (1984) Thompson et al. (2024).

Coalitions, Alliances, and Triangulation

An alliance is cooperation between family members toward an agreed task; a coalition denotes collaboration that may exclude or oppose another member, particularly when a child becomes enlisted in parental conflict. Triangulation is used across clinical and empirical literatures for involvement of a third person in tension between two others. Its operational forms include pressure to take sides, relaying hostile messages, seeking reassurance from a child against another caregiver, and assigning the child responsibility for adult reconciliation. The mere presence of three people in a discussion is insufficient. Investigators should identify conflict context, demand characteristics, power relations, and the consequences of the third person's involvement Minuchin (1974) Fosco & Grych (2010).

Fosco and Grych studied 171 adolescents ages fourteen to nineteen at two time points. Their prospective analysis connected perceived threat, later triangulation, self-blame, and the quality of parent-adolescent relations. This design supports directional hypotheses while remaining vulnerable to unmeasured time-varying factors and the limits of self-report. The Lausanne work offers a complementary observational approach: a longitudinal study with 38 families identified 15 coalition cases during structured interaction at eighteen months and differentiated several coalition styles. Both literatures show why triangulation deserves definition at the behavioral level and why findings from one age group or setting cannot be automatically generalized to another Fosco & Grych (2010) Fivaz-Depeursinge et al. (2009).

Triangulation studies also raise questions about responsibility and interpretation. An adult's request that a child explain their feelings about a household decision is not equivalent to requiring the child to adjudicate an adult dispute. Family members may recruit allies because of immediate danger, separation, contested custody, or chronic hostility; patterns must be understood with attention to safety. Coding systems should distinguish spontaneous support seeking, pressured side taking, and deliberate protective disclosure. An ethical analysis keeps authority and accountability attached to the adults who create or control the conflict. The broader empirical task is to identify which third-party positions forecast distress or adaptive coordination over time Davies & Cummings (1994) van Eldik et al. (2020).

Coparenting as a Coordinating Subsystem

Coparenting denotes the ways adults coordinate their responsibilities for raising and protecting a child, including agreement about goals, support for each other's efforts, division of labor, and management of disagreements about caregiving. Its membership is determined by actual parental or caregiving functions rather than a necessary romantic partnership. Coparenting research distinguishes the quality of adult partnership from the cooperation involved in caregiving. Parents may be emotionally distant as partners while maintaining effective arrangements for a child; adults satisfied with their partnership may nevertheless undermine one another's parenting. Feinberg's ecological framework articulates multiple coparenting dimensions and hypothesizes their relations to parental adjustment, parenting, and child outcomes Feinberg (2003) van Dijk et al. (2020).

A systems investigation of coparenting must distinguish agreement about abstract rules from coordinated implementation. Two caregivers can endorse identical standards yet alternate unpredictably in responding to distress or discipline. Conversely, different styles can be negotiated explicitly and implemented coherently. Observational assessments can code joint attention, task transitions, mutual support, interruptions, and repair after disagreement. Reports from both adults help clarify whether observed coordination is representative of their experience. The prenatal Lausanne Trilogue Play extends this inquiry before a child is born by asking prospective caregivers to enact a triadic encounter with a doll; its predictive associations with later observed alliance suggest possibilities and limits for early assessment Carneiro et al. (2006) Fivaz-Depeursinge et al. (1996).

The coparenting subsystem can be reorganized after separation, repartnering, migration, and changes in work or health. These events alter schedules, authority arrangements, and communication channels. Causal interpretation is difficult because preexisting conflict, socioeconomic conditions, and legal constraints may influence both coparenting and child adjustment. Meta-analytic work in divorced families investigates associations among interparental conflict, parenting, and child outcomes through structural equation modeling, but the estimated mediation in aggregated observational studies does not alone prove causal pathways. An appropriate design tracks the timing of changes, includes multiple respondents, and recognizes the contribution of institutional conditions van Dijk et al. (2020) Conger et al. (1992).

Conflict, Appraisal, and Emotional Security

Conflict is a family process involving incompatible goals, expectations, interpretations, or interests. Its developmental relevance depends on how disagreement is expressed and resolved, who is exposed, and how participants understand the event. Grych and Fincham developed a cognitive-contextual framework in which children's appraisals of threat and self-blame help explain reactions to interparental disagreement. Davies and Cummings subsequently articulated an emotional security hypothesis that directs attention to the child's attempts to preserve a sense of safety in the family. The explanatory content of these models lies in testable links among specific conflict features, cognitive and emotional responses, and later functioning Grych & Fincham (1990) Davies & Cummings (1994).

Research increasingly differentiates conflict frequency from hostility, disengagement, ineffective resolution, child-related disagreement, and broader relationship quality. A meta-analysis by van Eldik and colleagues synthesized 169 studies concerning child maladjustment and 61 concerning children's responses to interparental conflict, and found that distinct dimensions deserve separate analysis. This matters because a high count of openly discussed disagreements may carry different implications from recurring intimidation or unresolved hostility. Studies that average diverse behaviors into a single conflict score risk obscuring the processes most relevant to safety and adjustment. Researchers should assess observers' reports, children's reactions, the duration of conflicts, and actual repair van Eldik et al. (2020).

Evidence for appraisal pathways is informative but conditional. A prospective study of Welsh children linked interparental conflict with later threat and self-blame appraisals and distinct patterns of adjustment, after accounting for prior measures. This provides temporal information unavailable from a single interview, although unmeasured exposures and reciprocal child effects remain possible. In applied settings, the presence of a child at a disagreement is not enough to characterize danger; adult behavior and the child's experienced fear are central. When coercive control or violence occurs, safety takes precedence over a symmetric description of mutual conflict. Systems theory must retain responsibility with the actors exercising coercive power Grych et al. (2003) van Eldik et al. (2020).

Family Stress, Economic Conditions, and Resource Flows

Family organization is embedded in material conditions. Changes in employment, housing stability, health costs, and access to transportation can alter role allocation, time for care, and the perceived fairness of daily responsibilities. Conger and colleagues' family stress model linked objective hardship to subjective financial pressure, parental distress, marital conflict, parenting practices, and adolescent adjustment. Their original study included 205 early adolescent boys in rural Midwestern families and used both observed behavior and family reports. The work is historically important because it specifies an interconnected pathway that can be tested, while the narrow sampling frame limits direct generalization to all family forms Conger et al. (1992).

Economic conditions influence behavior through constraints as well as psychological processes. A caregiver working multiple unpredictable shifts may miss appointments or become less responsive because of limited time, independently of commitment or affection. A parent may delegate responsibilities to another adult because adequate public services are unavailable. Financial hardship can also intensify conflict about obligations and authority. An observational study without adequate measurement of resources might misclassify a practical adaptation as emotional disengagement. Ecological theory addresses these concerns by treating stress and support as determinants of parenting alongside caregiver resources and child characteristics Belsky (1984) Conger et al. (1992).

The strongest explanatory designs compare plausible pathways rather than assuming that every correlation reflects family-level dysfunction. Longitudinal studies can test whether changes in income precede changes in conflict and later adjustment; natural experiments or policy variation may offer further leverage when assumptions are justified. Repeated interviews reveal how responsibilities are negotiated under new constraints. Family-role analysis should examine who gains access to resources, who sacrifices time, and whether caregiving is concentrated on a participant with limited bargaining power. The same boundary arrangement can produce different consequences when social support and access to safe alternatives differ Feinberg (2003) Thompson et al. (2024).

Sibling Subsystems, Differential Treatment, and Reciprocity

Siblings share some family circumstances while occupying different developmental positions, relationships, and histories. Differences in age, disability, need, temperament, and caregiving access may generate legitimate variation in adult response. A sibling subsystem also includes cooperation, rivalry, mutual protection, conflict, and shared interpretations of parental conduct. It provides an important test of systems claims because family-level conditions and child-specific experiences can be estimated simultaneously. Evaluating fairness requires distinguishing relative treatment, in which one sibling is favored compared with another, from absolute differences in the amount of parental behavior received by siblings Jensen & Thomsen (2024) Jiang et al. (2025).

Jensen and Thomsen's 2024 meta-analysis summarized 88 sources nested in 43 samples, representing 26,451 participants, and associated parental differential treatment with children's internalizing and externalizing behavior. A subsequent systematic network meta-analysis integrating 26 studies and 37,025 participants examined a range of symptom domains and reported generally small positive associations for differential treatment. These findings are compatible with concerns about unequal family experiences, but effect estimates vary by informant, operational definition, and population. The data do not establish a universal causal effect of unequal treatment, because child characteristics and family circumstances can affect how adults distribute care Jensen & Thomsen (2024) Jiang et al. (2025).

A complete analysis should observe relationships among siblings as well as connections from each sibling to caregivers. One child may receive additional support because of illness while another interprets that support as favoritism; both the material reasons and subjective appraisals may matter. Interviewing only one sibling can misrepresent household dynamics. Multi-informant longitudinal designs permit assessment of whether perceived unfairness predicts subsequent interaction and whether changes in child behavior influence subsequent caregiver actions. The broader systems implication is that a family can contain several distinct relational environments at the same time, and a global label can conceal those differences Jensen & Thomsen (2024) Alexander et al. (1984).

Developmental Transitions and Reorganization of Authority

Family patterns change across developmental time. Infancy requires intensive adult coordination; childhood expands responsibilities for schooling and social relationships; adolescence typically increases negotiation over privacy, peers, and independence; adulthood can reorganize relationships around economic assistance, elder care, and reciprocal support. A boundary suited to one period may become restrictive or inadequate in another. Longitudinal observation is therefore necessary to distinguish continuity in family processes from age-linked shifts in visible behavior. The developmental meaning of authority depends on who can make decisions, what competence is relevant, and which forms of protection remain essential Belsky (1984) Thompson et al. (2024).

The transition to parenthood offers a particularly revealing reorganization because two adults must establish caregiving routines while adapting to disrupted schedules, increased responsibilities, and altered partner interaction. Prenatal assessments of collaborative family alliance attempt to characterize these processes before daily caregiving begins. Longitudinal Lausanne work linked prospective parents' coordinated task performance with later family interaction, suggesting potential continuity at the triadic level while leaving important questions about effect sizes and sample representativeness. These studies support the feasibility of observing a developing family subsystem rather than reducing a transition to personal stress scores Carneiro et al. (2006) Fivaz-Depeursinge et al. (1996).

Transitions also expose unequal resources and historical expectations. A separation may create two households with continued coparenting obligations; an older child may become a caregiver during parental illness; migration may redistribute emotional and financial responsibilities across borders. Research must distinguish temporary accommodations from chronic role reversal, and transition-related strain from sustained impairments. Changes in family structure do not automatically determine psychological outcomes. The mechanisms to investigate include predictability, access to safe care, negotiation of responsibility, and ability to revise arrangements. Comparative studies should trace adaptation over time and attend to the surrounding legal and economic systems Feinberg (2003) Conger et al. (1992).

Operationalization and Construct Validity

Scientific measurement of family functioning begins with an explicit target. A research claim about role allocation needs variables recording actual duties, recognition of responsibility, and the consequences of unfulfilled tasks. A claim about boundaries needs measures of privacy, unwanted intrusion, dependency, control of information, and authority. A claim about coalition formation needs observation of repeated inclusion or exclusion during conflict. Combining these into a global family health score may be convenient, but it sacrifices explanatory discrimination. Construct validity requires evidence that an instrument measures the proposed construct, distinguishes neighboring constructs, and predicts relevant outcomes without simply redescribing the same judgments Epstein et al. (1983) Olson (2011).

The Family Assessment Device illustrates the value and challenges of standardized measurement. Its developers created seven scales corresponding to problem solving, communication, roles, affective responsiveness, affective involvement, behavior control, and general functioning. Subsequent reliability and validity studies examined its performance beyond initial construction. The McMaster Clinical Rating Scale introduced trained-rater assessment following in-depth interviews, offering a method distinct from member questionnaires. A family considered dysfunctional on a self-report scale may receive different observational judgments, and discordance is itself relevant evidence about perspective and method. A responsible synthesis compares the measures, respondent population, and purposes of each instrument Epstein et al. (1983) Miller et al. (1985) Miller et al. (1994).

Validity must be studied anew when an instrument moves across languages, family forms, ages, or service settings. A scale that functions in a clinical sample of nuclear households may not demonstrate the same factor structure in extended kin networks or among adults who live separately but coordinate parenting. Translation equivalence, measurement invariance, response style, and access to different kinds of family information should be investigated. Respondents may also interpret emotionally loaded words differently. The evidence base includes early psychometric studies reporting modest correspondence among family members and across methods, which caution against treating an internally reliable score as a transparent measure of the family as a whole Alexander et al. (1984) Walker et al. (1984).

Family Observation and Ecological Measurement

Direct observation records events that family members may recall differently. Researchers can use structured discussions, problem-solving tasks, everyday routines, meal observations, and standardized triadic play. Coding may include who initiates an interaction, who responds, the duration of interruption, affective tone, physical proximity, shifts in attention, and how disagreements are repaired. Observation provides a time-stamped view of interaction; its interpretation depends on sampling and coding decisions. An apparently harmonious laboratory exchange may mask conflict at home, and a difficult experimental task can create disagreement that seldom occurs elsewhere. Researchers should therefore assess both coding reliability and ecological transportability Walker et al. (1984) Fivaz-Depeursinge et al. (1996).

Walker, Thomson, and Lindsay compared interview-based ratings with observations in mealtime and play contexts, reporting limited generalization for most scales across methods and settings. This finding demonstrates that method variance cannot be dismissed as error without investigation. The Lausanne Trilogue Play exemplifies a focused protocol for studying three-person coordination through structured changes in participation. Prenatal versions attempt to observe emerging coparenting organization before a child can join the interaction. These methods translate abstract hypotheses into reproducible tasks, but conclusions are bounded by task demands, developmental suitability, coder training, and the composition of participating families Walker et al. (1984) Carneiro et al. (2006).

Ecological momentary methods and household diaries can complement laboratory designs by recording role negotiation, support seeking, and conflict in daily contexts. Repeated reports help estimate within-family variation and the degree to which particular events precede others. Practical barriers include reporting burden, unequal device access, the possibility that monitoring changes behavior, and the exclusion of members unable or unwilling to report. Confidentiality is especially important when family members may face retaliation for disclosing experiences. Observational protocols should distinguish a participant's personal account from directly recorded behavior while retaining both as legitimate dimensions of the research object Epstein et al. (1983) Thompson et al. (2024).

Multi-Informant, Dyadic, and Family-Level Statistical Models

Family data violate the assumption that all observations are independent. Responses from relatives share exposure, history, and interaction, while differences between members may be substantive. Multilevel modeling can represent members nested within families, and repeated measures nested within participants. Dyadic models can distinguish an actor's characteristic or behavior from a partner's associated outcome. Family-level analyses require additional care because one person's report is not automatically a common family property. Estimation should specify the unit of inference and account for dependency in standard errors and uncertainty intervals Miller et al. (1994) Jensen & Thomsen (2024).

The social relations approach decomposes interpersonal ratings into perceiver, target, and relationship-specific components when the design provides enough observations. In principle this helps separate a person who views everyone critically from a family member widely viewed as difficult and from a particular dyadic mismatch. Round-robin assessment can expose relational patterns that disappear in one global questionnaire. The method requires adequate participants, identifiable relational direction, and attention to missing observations. Its results are descriptive decompositions unless embedded in designs that support stronger causal inference. The methodological lesson is that a family-wide description should be earned through the design rather than imposed by terminology Cook & Kenny (2004) Alexander et al. (1984).

Triadic relations pose additional analytic questions. A change in one caregiver-child exchange may occur only when another adult is present, and the association cannot be reduced to two independent dyads. Structured tasks can estimate participation, coordination, and exclusion, whereas network measures can identify whether information or support travels through a third party. Multiple observations permit testing whether an apparent coalition persists across situations. Researchers should predefine behavioral indicators, register primary comparisons when conducting confirmatory tests, and consider alternative models including shared stress, temperament, and social context. Statistical complexity is useful only when the sampling design actually identifies the hypothesized relations Fivaz-Depeursinge et al. (2009) Feinberg (2003).

Longitudinal Evidence and Causal Inference

A systems explanation often claims that interactional patterns persist, amplify, or redirect development. Establishing this temporal structure requires measurement at relevant intervals. A cross-sectional association between conflict and adolescent distress cannot identify which came first or whether another condition shaped both. Prospective designs can adjust for earlier outcomes and compare change across waves. Yet cross-lagged models still rely on assumptions about omitted variables, stable individual differences, and measurement quality. Causal inference becomes stronger when independently manipulated interventions alter the predicted mechanism, when plausible competing explanations are evaluated, or when design features provide credible quasi-experimental variation Fosco & Grych (2010) Grych et al. (2003).

Empirical family research offers instructive examples of both achievement and limitation. The Iowa economic hardship study specified an ordered family stress process and incorporated observer reports, making it more informative than an unstructured correlation. The prospective triangulation study examined adolescent appraisals and later relational outcomes, but its two-time-point design could not resolve every reciprocal pathway. Meta-analytic mediation studies of conflict and parenting in divorced families integrate broader evidence, while the input studies' observational designs constrain causal interpretation. A synthesis must identify these differences instead of placing all evidence under a single claim that family systems cause adjustment Conger et al. (1992) Fosco & Grych (2010) van Dijk et al. (2020).

Developmental transitions offer opportunities to study reorganization within families, but preexisting trajectories and selection into transitions matter. Families who separate, change residence, or enter treatment may differ from comparison families in ways that influence outcomes. Sensitivity analyses, detailed baseline measures, multiple reporters, and explicit causal diagrams can clarify what remains unidentified. Repeated measures also distinguish a stable between-family difference from a temporary within-family response. Strong claims should name the exposure, comparator, outcome, follow-up interval, and principal confounders. Such specification gives systemic explanations a predictive form that can survive independent scrutiny van Eldik et al. (2020) Thompson et al. (2024).

Intervention Research and Clinical Interpretation

Family-based interventions attempt to change relational processes, practical routines, or access to resources by involving more than one person or by explicitly addressing relationships. These approaches vary substantially in treatment targets, session structure, therapist training, populations, and outcome measures. The phrase family therapy is therefore insufficient to identify an evidence-based intervention. Structural family therapy, functional family therapy, multisystemic interventions, and disorder-specific family-based treatment have distinct models and empirical literatures. Evidence concerning one approach or diagnosis cannot be transferred wholesale to another. Evaluation requires independently controlled trials, defined outcomes, treatment fidelity, adverse-event monitoring where relevant, and comparison with plausible alternatives Minuchin (1974) Lock et al. (2010).

A randomized trial by Lock and colleagues compared family-based treatment with adolescent-focused individual therapy for adolescents with anorexia nervosa, demonstrating how a specified family intervention can be tested against an active treatment. The result concerns a particular clinical population, manualized approach, comparator, and follow-up schedule; it neither validates every historical concept in family systems theory nor establishes a universal treatment for family conflict. Intervention studies should distinguish improvement in the diagnosed condition from changes in family communication or boundaries. Without mediational evidence, a beneficial clinical result does not reveal precisely which family process produced the outcome Lock et al. (2010).

Clinical assessment must also consider unequal power, safety, and consent. Bringing several relatives together may be inappropriate when one uses coercion or intimidation against another, because joint participation could expose vulnerable people to risk. A family-level formulation does not imply shared accountability for abuse. Research and practice should examine confidentiality, willingness to participate, the capacity to speak freely, and referral needs. Claims about mechanisms remain hypotheses until investigated with credible process measures. Educational descriptions of the evidence are distinct from individualized recommendations; meaningful clinical guidance requires evaluation of the actual people and safety conditions involved Thompson et al. (2024) van Eldik et al. (2020).

Culture, Kinship Norms, and Measurement Invariance

Families coordinate obligations within culturally and historically situated institutions. Norms concerning elder respect, financial reciprocity, private disclosure, living arrangements, and children's household participation vary within and across populations. These differences change both the meaning of an action and the options available to individuals. The same observed contact frequency may indicate freely chosen closeness, practical caregiving necessity, or unwanted supervision depending on the relationship. Research must separate behavior, interpretation, and reported consequences instead of using a single cultural ideal as the normative benchmark. Structural categories remain useful as research questions when their operational meaning is tested in the populations studied Minuchin (1974) Thompson et al. (2024).

Cross-cultural generalization requires representative recruitment and assessment of equivalence. English-language family questionnaires may encode assumptions about household composition, individual privacy, and the division of caregiving work. Translation alone does not guarantee that respondents understand an item similarly or that the underlying factor structure is stable. Investigators should test measurement invariance, evaluate item functioning, include locally relevant role categories, and report the limits of sample composition. Recent sibling research illustrates this issue: meta-analytic samples have substantial geographical and demographic concentration, and some estimates differ across contextual characteristics. General statements about universal family patterns require convergent results from more diverse settings Jensen & Thomsen (2024) Jiang et al. (2025).

A culturally informed approach also preserves the agency of individual family members. An expectation described as customary can be experienced as meaningful solidarity by one person and as coercive pressure by another. Neither a culturally distant researcher's assumption nor the view of a single spokesperson can stand for all members. Multi-informant interviews and observation help recover disagreement, while institutional analysis identifies constraints on dissent or exit. Claims about healthy boundaries should specify consent, safety, fairness, developmental capacity, and access to support. These criteria allow inquiry into family organization without equating cultural difference with dysfunction Belsky (1984) Epstein et al. (1983).

Technology, Distributed Care, and Family Interaction

Contemporary families coordinate through messaging platforms, shared calendars, video calls, location services, school portals, and automated reminders. These tools can distribute information across households and sustain nonresidential caregiving, while also affecting privacy, monitoring, expectations of availability, and responsibility for decisions. Their relational effects depend on who controls the technology, who bears the coordination work, and how disagreements about access are resolved. A messaging channel may support transparent coparenting or become a medium for repeated pressure. A systems analysis identifies the participants, technological affordances, and consequences rather than assuming that increased connectivity produces improved family functioning Feinberg (2003) Minuchin (1974).

Artificial systems add further research questions about task delegation and the interpretation of mediated support. Automated schedules may redistribute administrative labor; conversational tools may offer information used in family decisions; synthetic companions may become relevant to discussions of emotional support and responsibility. Whether these technologies alter family roles requires evidence about actual use, reliance, privacy, and the human relationships affected. An artificial agent's apparent responsiveness is not by itself evidence that it has human attachment needs or moral authority over a child. Any claim about psychological organization in an Artificial Era must state its operational definition and distinguish family members' psychological attributions from independently demonstrated properties of a system Epstein et al. (1983) Feinberg (2003).

The most informative designs compare family routines before and after a specific technology is introduced while tracking competing changes, such as relocation, work demands, and developmental transitions. Researchers can examine whether a tool improves task completion, reduces conflict, creates surveillance concerns, or shifts authority to whoever administers it. Equity matters because devices, literacy, language access, and data control vary across households. Ethical research must minimize exposure of private family conversations and avoid collecting children's information without appropriate protections. The general systems insight is that tools may become durable parts of coordination loops, while their effects must still be established empirically Conger et al. (1992) Thompson et al. (2024).

Critical Scientific Discussion: What the Evidence Supports

The strongest cumulative findings concern narrower, operationally defined family processes rather than the validity of every proposition associated with historical family systems schools. Validated questionnaires describe domains of perceived functioning, yet measures differ and reports from different members can disagree. Observational triadic studies show that family configurations can be coded, but small and selected samples restrict generalization. Prospective investigations connect some conflict-related appraisals and triangulation patterns with later adjustment, while residual confounding remains. Large meta-analyses associate specific boundary dissolution patterns and interparental relationship dimensions with child outcomes. Their contribution is to identify substantive phenomena and heterogeneity, not to furnish a universal family health diagnosis Alexander et al. (1984) Fivaz-Depeursinge et al. (2009) van Eldik et al. (2020) Thompson et al. (2024).

Several recurrent explanatory alternatives deserve direct testing. Individual temperament and symptoms can influence participation and caregiver reactions; shared economic hardship can affect both conflict and child adjustment; family composition and transitions can change practical opportunities for care; respondents' general distress can influence their ratings of everyone else. Reverse causation may occur when a distressed child elicits changed parental behavior. Selection effects enter research when treatment-seeking families are compared with families recruited in the community. A systems hypothesis gains scientific strength when it specifies predictions distinguishable from these explanations and survives comparison with adequately measured rivals Conger et al. (1992) Jensen & Thomsen (2024) Miller et al. (1994).

The evidence base also contains clear conceptual and methodological tensions. Flexible involvement is often valued, but its appropriate degree varies with age, disability, and material conditions. A family aggregate can summarize useful shared experience while erasing the individual most at risk. Intervention success can demonstrate benefit within a defined clinical setting without identifying the precise interactional mechanism. Systemic language sometimes encourages a neutral description of reciprocity even where coercive power is unequal. Scientific synthesis requires transparent construct definitions, disaggregated outcomes, recognition of sampling limits, and explicit differentiation between theory, association, experiment, and clinical efficacy Lock et al. (2010) Olson (2011) Thompson et al. (2024).

Unresolved Problems and a Cumulative Research Agenda

An important priority is to develop measures that connect directly observed behavior with longitudinal outcomes and member-specific experience. Boundary scales should distinguish privacy, information sharing, coercion, role obligation, and generational responsibility rather than folding them into a single continuum. Role research should record the demands, resources, and voluntariness attached to each task. Triadic work would benefit from preregistered coding definitions, independent coder agreement, repeated ecological settings, and replication across family forms. Comparing these measures in a common design would help determine when different theoretical traditions are describing the same construct under new names Epstein et al. (1983) Fivaz-Depeursinge et al. (1996) Thompson et al. (2024).

Causal research should examine the timing and mechanisms of role reorganization. Intensive longitudinal designs could test whether specific changes in adult conflict precede child intervention in disputes, or whether resource shocks precede shifts in caregiving allocation. Randomized interventions could experimentally modify coordination routines where ethically and clinically appropriate, with measures of both intended outcomes and potential harms. Natural experiments involving access to services or policy changes may distinguish systemic pathways from household selection if design assumptions can be defended. A priority is independent replication in geographically and culturally diverse populations, including extended and nonresidential family networks Conger et al. (1992) van Dijk et al. (2020).

Future syntheses should report effects separately for children, caregivers, and other relatives, and should examine satisfaction, autonomy, safety, care quality, and developmental adjustment as distinct outcomes. More attention is needed to longitudinal role changes under aging, illness, disability, migration, and technology-mediated coordination. In Artificial Era research, hypotheses should address concrete changes in family decision processes and reliance on artificial systems, with appropriate safeguards for privacy and children. A cumulative science of family systems will depend on comparing coherent mechanistic predictions against observable alternatives, linking theories to repeatable measures, and documenting the conditions under which findings generalize Belsky (1984) Feinberg (2003) Jensen & Thomsen (2024).

Conclusion

Family systems, roles, and relational boundaries constitute a coherent research object when they are defined as patterned relationships of participation, authority, care, and influence observed across people and time. The historical systems traditions made reciprocal interaction and family subsystems central objects of psychological explanation. Subsequent measurement programs and longitudinal research translated parts of that inheritance into operational questions. Current evidence supports studying family-level coordination through multiple informants, structured observation, developmental follow-up, and analyses that preserve dependency among participants. The central scientific achievement is the ability to describe how responsibilities and experiences differ within a shared relational environment Minuchin (1974) Epstein et al. (1983) Miller et al. (1994).

The literature establishes meaningful associations between distinct forms of parent-child boundary dissolution and children's difficulties, between specific interparental conflict dimensions and adjustment, and between differential sibling treatment and reported outcomes. Studies of triadic interaction and prospective triangulation provide narrower mechanistic observations. Family stress work demonstrates why material resources and external institutions must be part of any adequate explanation. Controlled trials show that specified family-based interventions can be evaluated for defined populations, while clinical effects do not validate all theoretical premises. These findings justify a research architecture organized by constructs, populations, measurement methods, temporal evidence, and explicit alternative hypotheses Thompson et al. (2024) van Eldik et al. (2020) Jensen & Thomsen (2024) Lock et al. (2010).

The next stage of inquiry concerns the conditions under which family roles remain adaptable, relational boundaries protect both care and autonomy, and organized interaction permits safe development across changing circumstances. Progress depends on transparent definitions, independent replication, attention to power, and empirical comparison across household and kinship forms. Artificially mediated communication and coordination extend the empirical field while retaining the same demands for evidence. A mature science of family systems treats each member's experience and each observable exchange as part of a structured relational history whose causes and consequences can be investigated, disputed, and revised through research Feinberg (2003) Olson (2011) Thompson et al. (2024).

References

  • Alexander, B. B., Johnson, S. B., & Carter, R. L. (1984). A psychometric study of the family adaptability and cohesion evaluation scales. Journal of Abnormal Child Psychology, 12(2), 199–207. https://doi.org/10.1007/BF00910663
  • Belsky, J. (1984). The determinants of parenting: A process model. Child Development, 55(1), 83–96. https://doi.org/10.1111/j.1467-8624.1984.tb00275.x
  • Bowen, M. (1978). Family therapy in clinical practice. Jason Aronson. https://murraybowenarchives.org/books/family-therapy-in-clinical-practice/
  • Carneiro, C., Corboz-Warnery, A., & Fivaz-Depeursinge, E. (2006). The prenatal Lausanne Trilogue Play: A new observational assessment tool of the prenatal co-parenting alliance. Infant Mental Health Journal, 27(2), 207–228. https://doi.org/10.1002/imhj.20089
  • Conger, R. D., Conger, K. J., Elder, G. H., Jr., Lorenz, F. O., Simons, R. L., & Whitbeck, L. B. (1992). A family process model of economic hardship and adjustment of early adolescent boys. Child Development, 63(3), 526–541. https://doi.org/10.1111/j.1467-8624.1992.tb01644.x
  • Cook, W. L., & Kenny, D. A. (2004). Application of the social relations model to family assessment. Journal of Family Psychology, 18(2), 361–371. https://doi.org/10.1037/0893-3200.18.2.361
  • Davies, P. T., & Cummings, E. M. (1994). Marital conflict and child adjustment: An emotional security hypothesis. Psychological Bulletin, 116(3), 387–411. https://doi.org/10.1037/0033-2909.116.3.387
  • Epstein, N. B., Baldwin, L. M., & Bishop, D. S. (1983). The McMaster family assessment device. Journal of Marital and Family Therapy, 9(2), 171–180. https://doi.org/10.1111/j.1752-0606.1983.tb01497.x
  • Feinberg, M. E. (2003). The internal structure and ecological context of coparenting: A framework for research and intervention. Parenting: Science and Practice, 3(2), 95–131. https://doi.org/10.1207/S15327922PAR0302_01
  • Fivaz-Depeursinge, E., Frascarolo, F., & Corboz-Warnery, A. (1996). Assessing the triadic alliance between fathers, mothers, and infants at play. New Directions for Child Development, 1996(74), 27–44. https://doi.org/10.1002/cd.23219967404
  • Fivaz-Depeursinge, E., Lopes, F., Python, M., & Favez, N. (2009). Coparenting and toddler's interactive styles in family coalitions. Family Process, 48(4), 500–516. https://doi.org/10.1111/j.1545-5300.2009.01298.x
  • Fosco, G. M., & Grych, J. H. (2010). Adolescent triangulation into parental conflicts: Longitudinal implications for appraisals and adolescent-parent relations. Journal of Marriage and Family, 72(2), 254–266. https://doi.org/10.1111/j.1741-3737.2010.00697.x
  • Grych, J. H., & Fincham, F. D. (1990). Marital conflict and children's adjustment: A cognitive-contextual framework. Psychological Bulletin, 108(2), 267–290. https://doi.org/10.1037/0033-2909.108.2.267
  • Grych, J. H., Harold, G. T., & Miles, C. J. (2003). A prospective investigation of appraisals as mediators of the link between interparental conflict and child adjustment. Child Development, 74(4), 1176–1193. https://doi.org/10.1111/1467-8624.00600
  • Jensen, A. C., & Thomsen, A. E. (2024). Parental differential treatment of siblings linked with internalizing and externalizing behavior: A meta-analysis. Child Development, 95(4), 1384–1405. https://doi.org/10.1111/cdev.14091
  • Jiang, Z., Yang, Y., & Chen, B.-B. (2025). Parental differential treatment of siblings and child psychopathology: A network meta-analysis. Clinical Child and Family Psychology Review, 28(3), 555–572. https://doi.org/10.1007/s10567-025-00530-w
  • Lock, J., Le Grange, D., Agras, W. S., Moye, A., Bryson, S. W., & Jo, B. (2010). Randomized clinical trial comparing family-based treatment with adolescent-focused individual therapy for adolescents with anorexia nervosa. Archives of General Psychiatry, 67(10), 1025–1032. https://doi.org/10.1001/archgenpsychiatry.2010.128
  • Miller, I. W., Kabacoff, R. I., Epstein, N. B., Bishop, D. S., Keitner, G. I., Baldwin, L. M., & van der Spuy, H. I. J. (1994). The development of a clinical rating scale for the McMaster model of family functioning. Family Process, 33(1), 53–69. https://doi.org/10.1111/j.1545-5300.1994.00053.x
  • Miller, I. W., Epstein, N. B., Bishop, D. S., & Keitner, G. I. (1985). The McMaster family assessment device: Reliability and validity. Journal of Marital and Family Therapy, 11(4), 345–356. https://doi.org/10.1111/j.1752-0606.1985.tb00028.x
  • Minuchin, S. (1974). Families and family therapy. Harvard University Press. https://books.google.com/books/about/FAMILIES_AND_FAMILY_THERAPY.html?id=N3dHAAAAMAAJ
  • Olson, D. H. (2011). FACES IV and the Circumplex Model: Validation study. Journal of Marital and Family Therapy, 37(1), 64–80. https://doi.org/10.1111/j.1752-0606.2009.00175.x
  • Steinhauer, P. D., Santa-Barbara, J., & Skinner, H. (1984). The Process Model of Family Functioning. The Canadian Journal of Psychiatry, 29(2), 77–88. https://doi.org/10.1177/070674378402900202
  • Thompson, M. J., Platts, C. R., & Davies, P. T. (2024). Parent–child boundary dissolution and children's psychological difficulties: A meta-analytic review. Psychological Bulletin, 150(7), 873–919. https://doi.org/10.1037/bul0000440
  • van Dijk, R., van der Valk, I. E., Deković, M., & Branje, S. (2020). A meta-analysis on interparental conflict, parenting, and child adjustment in divorced families: Examining mediation using meta-analytic structural equation models. Clinical Psychology Review, 79, 101861. https://doi.org/10.1016/j.cpr.2020.101861
  • van Eldik, W. M., de Haan, A. D., Parry, L. Q., Davies, P. T., Luijk, M. P. C. M., Arends, L. R., & Prinzie, P. (2020). The interparental relationship: Meta-analytic associations with children's maladjustment and responses to interparental conflict. Psychological Bulletin, 146(7), 553–594. https://doi.org/10.1037/bul0000233
  • Walker, L. G., Thomson, N., & Lindsay, W. R. (1984). Assessing family relationships: A multi-method, multi-situational approach. The British Journal of Psychiatry, 144(4), 387–394. https://doi.org/10.1192/bjp.144.4.387