Relationship and Family Psychology

Abstract. Relationship and Family Psychology examines how enduring interpersonal bonds are formed, maintained, reorganized, and dissolved, and how psychological processes operate within couples, parent–child relationships, kinship networks, and other enduring close relationships. Its central objects are patterns of interdependence across persons and time: attachment, intimacy, commitment, care, conflict, family transitions, resource allocation, psychological safety, and the transmission of experiences across generations. This monograph develops an integrative account grounded in original empirical studies, longitudinal relationship research, psychometric research, and systematic evidence syntheses. It places adult attachment theory, interdependence and investment models, the interpersonal process of intimacy, developmental and family systems approaches, and economic stress models into a common explanatory architecture. The empirical record establishes meaningful associations between supportive dyadic coping and satisfaction, interparental conflict and child adjustment, and several forms of couple intervention and improvements in relationship distress. These findings require qualifications concerning selection, measurement, heterogeneity, cultural context, and causal identification. Angela Bogdanova's independent contribution is the Relational Coordination Framework, which distinguishes six interacting explanatory levels—individual disposition, dyadic exchange, family organization, developmental time, material institutions, and technological mediation—and specifies research designs by which the model can be evaluated. Relationship quality, continuity, power, and safety are treated as analytically separate outcomes. The work concludes by identifying empirically testable problems of relational transformation in the Artificial Era.

Keywords: relationship psychology; family psychology; close relationships; intimate relationships; attachment; relationship satisfaction; relationship commitment; intimacy; interpersonal responsiveness; dyadic coping; parent–child relationships; coparenting; family systems; family stress; interparental conflict; relationship measurement; couples therapy; social networks; technological mediation; Artificial Era

The Scientific Subject of Relationship and Family Psychology

Relationship and Family Psychology studies psychologically consequential connections among persons whose actions, emotions, expectations, resources, and identities become interdependent. A relationship is a temporally extended pattern of interaction and mutual influence. A family is a historically and socially organized network of care, affiliation, obligation, kinship, or household membership, whose boundaries and roles differ across cultures and life histories. These definitions make continuity, reciprocal influence, and institutional context central empirical questions. They also allow investigators to study chosen families, nonmarital partnerships, single-parent households, blended families, same-sex relationships, intergenerational care, friendship networks, and families dispersed across national boundaries.

The discipline connects multiple units of observation. A person reports feelings and expectations; two partners exchange behavior; a family distributes tasks and authority; institutions shape access to housing, income, childcare, health care, and legal recognition. The psychological meaning of a conflict depends partly on the interaction in which it occurs, the relationship's history, the resources available to its participants, and the consequences of disagreement. An explanation restricted to personality misses interaction and circumstance. An explanation confined to family roles misses the cognitive and emotional work of particular persons. Finkel, Simpson, and Eastwick's integrative review makes this theoretical plurality explicit, identifying recurring relationship principles associated with uniqueness, integration, evaluation, responsiveness, and contextual change (Finkel et al., 2017).

The basic outcomes require operational separation. Satisfaction means an evaluation of relationship quality; commitment concerns intentions and psychological investment in continuation; stability records whether and how a relationship persists; intimacy concerns felt understanding, validation, affection, and openness; safety concerns freedom from violence, coercion, threats, and domination. A durable relationship can have poor quality. A distressed partnership can continue because alternatives are constrained or investments are difficult to recover. An apparently harmonious household can contain unequal power. Consequently, no single satisfaction score serves as a complete measure of relational functioning.

Family psychology extends this problem beyond couples. Parenting can be sensitive in one context and strained in another. Sibling relations change with age and family events. Coparenting may be cooperative even when a romantic partnership ends. Child development reflects the relationships among caregivers, not merely independent attributes of each parent. Psychological inquiry therefore distinguishes parent–child bonding, interparental processes, caregiver coordination, children's own perceptions, and broader systems of material and social support.

Historical Formation of Relationship Science

Modern relationship research joins traditions that originally developed in separate intellectual communities. Social psychologists studied affiliation, exchange, attraction, and interpersonal perception. Developmental investigators analyzed early caregiving and attachment. Family sociologists examined households, roles, kinship, and socioeconomic transitions. Clinical researchers documented patterns of family communication and interventions. The mature field emerged through converging questions about how interpersonal events generate durable psychological consequences.

Interdependence theory developed from the analysis of coordinated and conflicting interests, particularly the work of John Thibaut and Harold Kelley. It brought attention to the fact that the consequences of one person's choices depend on another person's response. Relationship outcomes can therefore differ from what isolated individual preferences would predict. Later relationship scholarship used this architecture to study dependence, negotiation, commitment, and transformation of immediate interests into mutually coordinated action (Finkel et al., 2017). Interdependence is a useful analytical property; it is not a moral judgment that every dependency is beneficial.

Attachment research provided another foundation. Bowlby's developmental account of attachment and Ainsworth's observational work established major questions about security, separation, caregiving, and expectations of responsiveness. Hazan and Shaver translated aspects of this tradition into the investigation of adult romantic love in two questionnaire studies, discussing secure, avoidant, and anxious or ambivalent patterns (Hazan and Shaver, 1987). Their original studies made adult attachment empirically researchable, while leaving open the extent to which childhood relationships predict later adult bonds and the best means of measuring attachment.

Family systems theory drew attention to recurring interaction sequences, boundary organization, coalition formation, and mutually reinforcing responses among members of families. Its clinical vocabulary has stimulated observation and hypotheses, although empirical support varies considerably by proposition. A review of Bowen theory found support for relations involving differentiation and distress, with weaker findings for specific assumptions about sibling position and triangulation (Miller, Anderson, and Keala, 2004). The historical importance of a theory does not establish the validity of every explanatory claim made in its name.

Investigations of family poverty and adversity transformed the field's causal imagination. Conger and colleagues linked economic pressure, parental distress, marital conflict, parenting disruption, and adolescent adjustment, using reports alongside trained behavioral observation. Their original Iowa study concerned 205 early adolescent boys living in two-parent families in a rural Midwestern context (Conger et al., 1992). Its design helped identify plausible processes connecting structural hardship and family interaction. At the same time, the historical sample constrains population-wide inference, particularly when family structure, geography, and economic arrangements differ.

Later longitudinal, diary, experimental, cross-cultural, and intervention methods challenged simple character-based explanations of partnership outcomes. Karney and Bradbury's influential synthesis argued that relationship development must be studied through change over time, bringing enduring vulnerabilities, stressful contexts, and adaptive processes into joint consideration (Karney and Bradbury, 1995). This shift also sharpened the distinction between associations observed among existing couples and explanations of relationship formation, selection, and dissolution.

Interdependence, Commitment, and Investment

Interdependence theory describes situations in which the rewards, costs, options, and consequences of individual behavior are partly determined by another person's behavior. Partners can have corresponding interests, competing immediate preferences, or asymmetric power. The same household decision can be trivial for one participant and consequential for the other. A theory of interdependence must therefore consider whose alternatives are available, whose labor is visible, and whose preferences are represented in shared decisions.

Rusbult's investment model developed a tractable account of commitment. Satisfaction depends on outcomes in relation to expectations; commitment is additionally influenced by the quality of alternatives and the size of investments that might be lost if the relationship ended. Original studies supported the model's differentiation of these constructs (Rusbult, 1980). Commitment is thus not reducible to current happiness. It can express meaningful mutual dedication, practical constraints, identity commitments, or combinations of these.

This distinction generates testable hypotheses. A major external transition, such as job displacement, can reduce satisfaction while increasing perceived dependence because alternatives become more difficult to access. Investment in children, property, social networks, and shared histories can increase the cost of separation while also providing sources of identity and support. Researchers who infer relationship well-being directly from duration conflate persistence with evaluation. Researchers who rely exclusively on satisfaction overlook persistence mechanisms that may have important ethical and practical implications.

The Investment Model Scale was developed to measure commitment, satisfaction, alternatives, and investment size in separate dimensions, with validation evidence drawn from three studies (Rusbult, Martz, and Agnew, 1998). Its value lies in explicit operational distinctions. Yet a self-report about alternatives can reflect genuine opportunities, perceived possibilities, social norms, or constraints. Those interpretations should be differentiated through supplementary measures.

Attachment Across Relationships and Development

Attachment processes organize proximity seeking, support seeking, separation responses, and expectations of availability during emotionally consequential situations. Adult attachment researchers often describe relatively continuous dimensions of anxiety and avoidance. Attachment anxiety concerns worries about rejection and inconsistent availability; avoidance concerns discomfort with dependence or intimacy. Individuals may vary across partners, periods of life, and relational domains. A questionnaire score records responses within a measurement model, not an immutable personal category.

Hazan and Shaver's pioneering study proposed parallels between early caregiver bonds and romantic affection (Hazan and Shaver, 1987). Simpson, Rholes, and Nelligan subsequently observed support seeking and support giving in dating couples facing an anxiety-provoking situation, recording behavioral differences associated with attachment styles (Simpson, Rholes, and Nelligan, 1992). These studies helped make attachment an interactional rather than purely descriptive account. An expectation of support becomes consequential through what people ask for, what partners notice, and how partners respond.

A historical methodological debate concerns whether adult attachment should be measured through narrative interviews, relationship-specific ratings, broad self-report traits, or repeated observations. These instruments do not all assess identical properties. Similarly, cross-sectional correlations between retrospective childhood experiences and current romantic relationships cannot by themselves establish a continuous causal transmission from infancy. Developmental continuity requires prospective designs capable of distinguishing familial adversity, temperament, subsequent experiences, and life-course change.

Family relationships repeatedly activate attachment-related processes. Children's security can be influenced by multiple caregivers. Adolescents negotiate autonomy and continued accessibility. Adult children and aging parents reorganize practical support and emotional obligation. Family loss, migration, illness, and separation alter the availability of familiar sources of care. Attachment theory supplies a set of hypotheses about regulation under threat, but different life stages and relationship roles demand their own measures.

Intimacy, Disclosure, and Perceived Responsiveness

Intimacy is an interpersonal achievement constructed through communication and interpretation. The interpersonal process model proposes that personally meaningful disclosure provides an opportunity for intimacy, which depends on the listener's response and on whether that response is experienced as understanding, validation, and care. Disclosure itself may be emotionally revealing without creating security; its interpersonal consequences depend upon context.

Laurenceau, Barrett, and Pietromonaco tested this model using event-contingent diary reports after social interactions over one or two weeks. Their findings supported an association among disclosure, perceived responsiveness, and experienced intimacy, with evidence that perceived responsiveness partially mediated the process; emotional disclosure was especially relevant in one study (Laurenceau et al., 1998). Diary designs improve temporal proximity to events and reduce some memory distortion, though shared self-report method and unmeasured situational variables complicate causal claims.

This framework explains why two superficially similar conversations can produce divergent outcomes. A response may provide accurate information without conveying concern. Another response may be compassionate but fail to engage with the particular concern disclosed. Perceived responsiveness is partly a property of behavior and partly an interpretation informed by expectations, prior interaction history, and prevailing trust. Experiments manipulating response content, observational coding, and partner reports can help decompose these components.

Relationship intimacy also changes across contexts of illness, caregiving, sexual relationships, friendship, and later life. Expressions of care can involve listening, practical action, consent, respect for privacy, or the recognition of a partner's independent goals. An exclusive focus on verbal disclosure risks underestimating intimacy expressed through shared labor and responsiveness to nonverbal need.

Conflict, Repair, and Psychological Safety

Conflict is a recurrent feature of relationships in which people retain partially independent needs. Its consequences depend on the subject of disagreement, the distribution of power, the responses each person can choose, and the availability of safe repair. Researchers distinguish constructive disagreement from hostile escalation, withdrawal, degradation, coercion, and violence. Observed communication sequences may be associated with distress without representing fixed properties of a relationship.

The demand–withdraw pattern describes a sequence in which one participant presses for discussion or change while the other retreats or avoids engagement. The psychological mechanisms may include fear of escalation, perceived criticism, asymmetries in the importance of the issue, exhausted coping resources, or unequal power. Evaluating the pattern requires attention to both partners' perspectives and the practical conditions of the conflict. Labels such as "demanding" and "withdrawn" can conceal who is trying to secure essential resources or establish safety.

Repair processes include acknowledgment, responsive listening, negotiated change, and behavioral evidence of altered expectations. A sincere verbal apology and a stable pattern of correction are different phenomena; the latter can be evaluated longitudinally. Similarly, forgiveness, reconciliation, and continued contact are separate choices and outcomes. No theory of relational repair justifies presuming that every relationship ought to continue.

Safety is a fundamental dimension of close relationships rather than a subsidiary form of satisfaction. The World Health Organization defines intimate partner violence broadly to include physical, sexual, and psychological harm as well as controlling behavior. Its June 2026 fact sheet reports worldwide prevalence estimates with specified populations and reference periods (WHO, 2026). Such estimates establish a major public health concern; they cannot be translated into a universal description of every partnership.

Research and intervention involving coercive control require methods that preserve safety and independent disclosure. Joint counseling formats designed for ordinary communication problems may be unsuitable when a participant cannot speak freely or faces retaliation. The appropriate research distinctions are between disagreement and intimidation, mutual frustration and exploitation, relationship maintenance and protection from harm. These categories affect sampling, ethics, measurement, and the interpretation of treatment outcomes.

Family Organization and Systems Processes

The family is a system of recurring relations in which changes affecting one member can propagate through interaction patterns and shared conditions. Systems approaches direct attention to repeated communication, roles, alliances, boundaries, and feedback. The explanatory task is to show precisely which observed sequences generate which outcomes under which conditions. General descriptions of a family as an interconnected whole acquire scientific content when translated into measures of contact, authority, reciprocity, coordination, and change.

Theoretical notions such as triangulation and differentiation have historically influenced family therapy and family studies. Empirical tests support some associations while leaving others uncertain. The review of Bowen theory by Miller, Anderson, and Keala (2004) supported links between differentiation and indices of chronic anxiety, marital satisfaction, and distress, while reporting comparatively limited support for assumptions about sibling position and some other mechanisms (Miller et al., 2004). This uneven record supplies an important methodological lesson: a productive family vocabulary should generate separable hypotheses that can be rejected or refined.

Family organization is dynamic. The arrival of a child changes adult time, sleep, responsibilities, and patterns of mutual attention. Adolescence transforms the balance between guidance and autonomy. Migration can distribute caregiving across borders. Serious illness changes dependence and decision-making. Retirement and the death of family members alter intergenerational support. A family process cannot be adequately characterized by a single static snapshot when its psychological effects unfold across transitions.

The investigator must also distinguish the family's reported ideals from actual resource distribution. A claim that members share responsibilities equally can coexist with disparities in unpaid domestic labor or decision-making authority. Informants may differ genuinely in their accounts of the same household. Those differences are meaningful data. Observational research, time-use records, longitudinal household interviews, and confidential individual reports allow family organization to be studied without imposing a single family member's narrative as the objective account.

Parenting, Caregiving, and Child Development

Parent–child relationships combine asymmetric responsibility with reciprocal psychological influence. Young children rely upon caregivers for protection and developmental scaffolding, while their temperament, health, communication, and regulatory capacities shape the demands placed on adults. Sensitive caregiving concerns the timing, appropriateness, and reliability of responses to children's needs. Parenting practices involve routines, boundaries, explanations, discipline, shared activities, and the regulation of opportunities for autonomy.

Research on parenting must distinguish caregiver beliefs, observed practices, child interpretations, and long-term child outcomes. Associations between parenting and development can reflect reciprocal influence: a child's persistent difficulty may increase parental stress, while harsh or inconsistent responses may amplify conflict. Genetic relatedness, neighborhood context, available resources, schooling, and cultural meanings are plausible confounding or moderating influences. Cross-sectional reports cannot settle which direction of influence dominates.

Multiple caregivers may contribute to development. A child can receive security from parents, grandparents, siblings, and other stable caregivers. Family configurations vary without eliminating the empirical requirement to examine responsiveness, continuity, and safety. Normative assumptions about one household arrangement may lead investigators to confound family form with poverty, exclusion, discrimination, or support. The relevant causal question is which processes and conditions matter, not whether a household matches an inherited stereotype.

Parenting interventions and policy research must evaluate child outcomes and caregiver conditions together. The World Health Organization's INSPIRE framework identifies coordinated strategies across law, norms, environments, caregiver support, income, services, and education to reduce violence against children (WHO, 2016). The framework reinforces a multi-level principle: improving relationships may require changing the environments in which caregiving occurs.

Coparenting and the Coordination of Care

Coparenting refers to the coordination of adults in responsibilities toward a child. It is analytically distinct from romance, marriage, and continued cohabitation. Adults can dissolve a couple relationship while maintaining cooperative coparenting; likewise, romantic satisfaction does not guarantee consistency in parenting collaboration. Coparenting encompasses agreement over child-rearing, mutual support or undermining, division of responsibility, and the management of family interaction.

Feinberg's ecological model organized coparenting as a multidimensional process linked to caregiver adjustment and child functioning, while identifying social and developmental contexts in which it changes (Feinberg, 2003). This conceptual work clarifies the research object, but specific causal pathways require direct evaluation. Coparenting difficulties may precede child problems, arise in response to developmental demands, or reflect economic and relational strains that affect adults and children simultaneously.

The study of coparenting must include the perspectives of all relevant caregivers where ethically and practically possible. A single adult's report may capture perceived conflict but cannot describe the whole coordination system. Joint tasks, videotaped interaction, daily schedules, and multiple-informant assessment can reveal the difference between stated agreement and actual support. Where intimidation or family violence is present, confidential, safety-oriented assessment takes precedence over assumptions of symmetrical negotiation.

Interparental Relationships and Children's Adjustment

Children encounter the psychological consequences of adult relationships through the emotional environment, daily care, conflict exposure, and the way family members interpret one another's actions. Hostile exchanges, unresolved disputes, emotional disengagement, and conflict centered on a child may have different implications. The child's developmental stage and perceived responsibility for adult conflict also matter.

Erel and Burman's meta-analysis of 68 studies found a moderate positive association between marital relationship quality and parent–child relationship quality, with a reported effect size of d = 0.46, supporting a spillover interpretation more strongly than a compensatory interpretation in the aggregated evidence (Erel and Burman, 1995). This estimate describes a meta-analytic association, not an experiment demonstrating that changing one relationship necessarily changes the other. Variation among studies indicates that a single average masks differences in measures and populations.

A later meta-analysis by van Eldik and colleagues distinguished relationship quality, conflict frequency, hostile and disengaged conflict, unconstructive conflict, and child-related disputes. It synthesized 169 studies concerning child maladjustment and 61 concerning children's responses to interparental conflict (van Eldik et al., 2020). Its design helps move beyond treating every disagreement as equivalent. Children's internalizing symptoms, externalizing problems, cognitive appraisals, physiological responses, and behavioral adaptation form distinguishable outcome classes.

These findings have two major implications. First, the quality of adult conflict regulation may matter independently of how often disagreement occurs. Second, the analysis must consider family resources, the child's direct experience, parenting behavior, and the stability of caregiving arrangements. Children are active interpreters of family processes. Their perceptions can mediate or moderate the effect of adult interactions, although mediational claims require time-ordered observations and convincing causal assumptions.

Family Stress, Economic Resources, and Social Context

Family life is shaped by demands that exceed individual preference: housing instability, medical expenses, unemployment, disasters, displacement, caregiving costs, discriminatory treatment, and changes in institutional support. The family stress model proposes that such pressures influence adult emotional functioning and interactions, thereby affecting parenting and child adjustment. It supplies testable connections between structural conditions and psychological events rather than interpreting distress solely as a private deficit.

In the original study by Conger and colleagues, economic adversity was assessed together with economic pressure, parental emotional state, marital conflict, parenting behavior, and adolescent adjustment. The 1992 findings were broadly consistent with a sequence in which perceived economic strain and parental distress were connected to family interaction problems (Conger et al., 1992). A follow-up research program examined larger and repeated family observations over three years, including boys and girls in the rural Midwest (Conger et al., 1994).

The causal interpretation of these observational models depends on temporal ordering, measurement assumptions, and the exclusion of rival processes. Employment loss and psychological symptoms may influence one another. Family members may select neighborhoods and relationships partly in response to prior circumstances. An intervention that changes resources offers a different evidentiary test from a questionnaire association. Researchers should therefore compare natural experiments, social policy changes, longitudinal designs, and family-level interventions where feasible.

Structural conditions also affect relationship choice and exit. The ability to leave an unsafe arrangement depends on access to shelter, income, legal protection, child care, and social support. Analysis of commitment without examining constraint may idealize persistence. Relationship science is strengthened when it treats material conditions as components of the explanatory setting and measures how those conditions interact with psychological processes.

Theoretical and Empirical Foundations

The core theories of this discipline describe partly different mechanisms. Attachment theory addresses expectations of availability and regulation under threat. Interdependence and investment theories analyze consequences, alternatives, dependence, and commitment. Intimacy models explain disclosure and perceived responsiveness. Systems models examine patterned interaction and family organization. Stress models connect external pressures with individual and relational adaptation. Developmental models explain change across transitions and generations.

Each theory should be evaluated on the outcomes for which it offers explicit predictions. Attachment anxiety may predict heightened concern about rejection, but it does not determine relationship destiny. High investment may predict commitment while giving no guarantee of safety. Perceived responsiveness can predict intimacy without establishing long-term stability. Cooperative coparenting may improve children's daily environment independently of romantic reconciliation. Explanatory plurality can therefore be productive when constructs are operationally separated rather than blended into a generic assessment of a "good relationship."

Several empirical designs provide complementary evidence. Experimental studies can manipulate cues of support or perceived responsiveness and assess immediate changes in emotion or behavior. Interaction laboratories capture verbal and nonverbal exchanges under standardized conditions. Experience-sampling and event-contingent diaries follow naturally occurring moments. Longitudinal panels observe change and transitions. Randomized trials evaluate specific interventions, while systematic reviews estimate recurring patterns across studies. Kinship, sociometric, administrative, and ethnographic evidence place observations within broader family and community networks.

The central methodological challenge is interdependence of observations. Reports from two partners are correlated because their lives and experiences overlap. Treating them as unrelated independent observations distorts standard errors and can obscure partner effects. Kenny, Kashy, and Cook's work on dyadic analysis formalizes actor, partner, and relational components, including the Actor–Partner Interdependence Model (Kenny, Kashy, and Cook, 2006). Extensions to families require models for multiple overlapping dyads, roles, and shared contexts.

A credible empirical synthesis must identify whether a statistic arises from a longitudinal observational association, a controlled intervention, a convenience sample, or a meta-analysis. Results should be reported with sample definition, point estimate, interval when available, and relevant heterogeneity. A statistically significant association does not automatically warrant claims about general human behavior, intervention effectiveness, or causation.

Measurement, Construct Validity, and Psychometrics

Relationship measures must represent their intended constructs with enough precision to distinguish meaningful variation. Satisfaction scales can measure global evaluation, but studies may use different questions concerning closeness, fairness, conflict, affection, or comparison with expectations. Combining these outcomes under one label creates construct ambiguity.

Spanier's Dyadic Adjustment Scale was presented in 1976 as a multidimensional measure of marital and similar dyadic adjustment, including satisfaction, cohesion, consensus, and affectional expression (Spanier, 1976). Its development marked an important advance in systematic measurement, while the very breadth of "adjustment" invites debate about whether distinct experiences should be collapsed into one total score.

Funk and Rogge used item response theory to evaluate measures of relationship satisfaction and develop the Couples Satisfaction Index, addressing differences in precision across the scale (Funk and Rogge, 2007). Such work shows why a measure can have high internal consistency yet perform unevenly at different levels of satisfaction. The instrument appropriate for identifying severely distressed couples may differ from the instrument best suited to measuring subtle improvement among highly satisfied couples.

Reliability, temporal stability, structural validity, convergent evidence, sensitivity to change, and measurement invariance must be evaluated separately. Cross-cultural comparison requires asking whether translated items carry comparable meanings and whether factor structure and item functioning are sufficiently similar. A scale developed among married US college graduates may operate differently among older caregivers, recent migrants, unmarried couples, or non-Western households. Researchers should establish comparability empirically rather than presume it.

Family research faces multi-informant problems. Adults' and children's reports need not agree. Discrepancy can express perspective, concealment, role-based information, or differential access to events. Simply averaging informants may erase the phenomenon being investigated. For conflict or violence assessment, individual confidentiality may also be necessary for accurate reporting and participant protection.

Dyadic and Family Research Designs

Longitudinal research is indispensable for examining relationship transformation. Repeated measures can identify trajectories of satisfaction, conflict, or supportive behavior and test whether earlier stress predicts subsequent outcomes. Yet regression to the mean, selective attrition, the disappearance of dissolved relationships from the sample, and changing interpretation of survey items complicate these inferences. A panel that only retains married participants can make surviving relationships look artificially stable.

In dyadic analysis, actor effects refer to how a person's own characteristic predicts that person's outcome, while partner effects refer to how one person's characteristic predicts the other person's outcome. A shared contextual variable, such as family economic pressure, can affect both participants. Analytical models need to accommodate this correlated structure. Family-level data further require attention to nested households, siblings, changing membership, and multigenerational ties.

Observational methods are especially valuable for conflicts between reports and behavior. Trained coders can rate hostility, validation, topic change, repair attempts, and cooperative task completion. The observation itself can influence participants, and behavior during an instructed laboratory conversation may differ from behavior in private domestic life. Reliable coding does not eliminate setting effects. Studies benefit from combining observational traces with experience sampling and longitudinal outcomes.

Qualitative interviews and ethnography illuminate the meanings of obligation, privacy, intimacy, kinship, and caregiving. They can reveal concepts omitted from existing scales and identify cultural assumptions embedded in the investigator's categories. Their inferential strengths concern interpretation and process discovery; causal frequency estimates require appropriate sampling and quantitative designs. A mature discipline places these methods into a coherent account of what each can establish.

Dyadic Coping, Co-Regulation, and Shared Adversity

Stress can enter relationships through employment, caregiving, health, discrimination, and other external demands. Dyadic coping describes how partners communicate stress, provide assistance, distribute tasks, and jointly regulate demands. Responses can be supportive, cooperative, protective, hostile, or overcontrolling. The same act may be perceived differently by the intended recipient depending on autonomy, timing, and previous experience.

Falconier, Jackson, Hilpert, and Bodenmann synthesized 72 independent samples from 57 reports, comprising 17,856 participants. Their meta-analysis found a pooled zero-order correlation of r = .45 between overall dyadic coping and relationship satisfaction (Falconier et al., 2015). This is a substantial association in the studied literature. It cannot by itself establish that training in dyadic coping produces an equivalent improvement, nor does it imply that r = .45 applies to every setting and population.

An important distinction is between perceived and enacted support. A partner can provide practical help that is unrecognized or experienced as controlling. Conversely, an individual may feel supported even where observable assistance is minimal. Such differences raise questions about shared expectations, self-efficacy, and the meaning of help. Future trials should measure support behavior, perceived responsiveness, autonomy, stressor intensity, and both partners' outcomes.

Dyadic coping also occurs within families. A child may perceive financial strain without knowing its details; caregivers may attempt protective buffering; adult siblings may coordinate care for a parent. These situations present different developmental and ethical relationships. The psychological benefits and costs of buffering cannot be inferred from couple data alone.

Relationship Transitions, Dissolution, and Reorganization

Relationships evolve through formation, commitment, cohabitation, parenting, illness, migration, separation, reconciliation, and bereavement. Each transition can change what participants need and how they interpret a partner's availability. The same person can experience different relationships at different stages of life, and relationship expectations can adapt following major events. Models assuming a single stable personal "relationship type" have difficulty representing this developmental plasticity.

Karney and Bradbury's longitudinal review offers a model of vulnerability, stress, and adaptation for explaining why similar couples diverge over time (Karney and Bradbury, 1995). Changes in relationship quality need not be linear. Some couples may experience declining satisfaction during a demanding transition and subsequent recovery; others may maintain stable quality, separate, or reorganize forms of partnership. Trajectories are properties of specified measures in specified samples, not universal stages.

Separation and divorce are psychologically heterogeneous processes. An unwanted ending may involve grief, identity change, housing difficulty, and altered social support. An exit from a violent or coercive relationship can also be protective. Research that classifies dissolution as failure without regard to safety and agency mismeasures the outcome. The investigation should distinguish legal change, household separation, emotional adjustment, continued coparenting, and longer-term well-being.

Children's adjustment following family reorganization depends on caregiving continuity, conflict exposure, economic resources, age, and supportive relationships. A simple comparison of children in different household forms may confound these conditions. Stronger research records pre-transition functioning, the timing of separation, ongoing family interactions, and the availability of wider kin and community support.

Cultural Variation, Kinship, and Social Inequality

Relationships take place within cultural practices that determine how partnership, marriage, elder care, privacy, filial obligations, and household membership are recognized. Investigators should study the actual norms and institutions of participants, including their historical change. The contrast between "individualist" and "collectivist" societies is too coarse to specify what happens inside a particular family. Variation exists within communities, generations, socioeconomic positions, and personal circumstances.

Cross-cultural generalization requires careful sampling and measurement equivalence. Findings from English-speaking convenience samples cannot automatically be assumed to hold among households with different kinship organization, caregiving patterns, or institutional support. Language translation alone does not validate a measure. Researchers must examine what an item means in context and whether its thresholds and structure remain comparable.

Gender and sexuality matter through embodied experience, roles, expectations, stigma, and institutional opportunities. Same-sex partnerships, bisexual and transgender participants, nonbinary caregivers, and different family arrangements have historically been unevenly represented in samples. Research should test invariance and compare processes rather than treat one group as an unexamined baseline. Social exclusion can itself shape stress, relationship disclosure, and access to support.

Families also differ in access to stable work, housing, medical care, education, and legal protection. These inequalities shape available choices, dependence, caregiving demands, and stress exposure. Explanations that attribute all variance to individual communication risk misunderstanding the effects of structural conditions.

Ethics of Relationship Research

The intimacy of the research subject increases ethical obligations. Studying partners or families can create privacy risks if confidential statements become visible to other members. Participants may have unequal power to refuse, withdraw, or disclose experiences. Researchers must distinguish collective consent to a household observation from each person's independent consent to participate in a particular procedure.

Sensitive topics require protocols for confidentiality, distress, safeguarding, and referral. Interviewing two partners together about violence may be unsafe when one participant controls the other's responses. Recording domestic interactions raises questions about the involvement and rights of children and bystanders. Data about a relationship belong to a web of persons whose interests may differ.

Ethical science should also resist pathologizing relationships because they depart from prevailing social norms. Research questions must specify consequential behaviors and experiences: coercion, responsiveness, care, economic access, development, and well-being. The presence of children, marriage, cohabitation, biological ties, or particular identities cannot substitute for these direct observations.

Clinical Evidence and Couple Interventions

Relationship distress can be associated with psychological symptoms, chronic stress, and difficulties in everyday functioning. Interventions may address communication, intimacy, emotional regulation, attachment-related interactions, or shared coping. The empirical evaluation of such interventions requires clarity about the treatment population, comparison condition, outcome definition, follow-up period, therapist training, treatment adherence, and adverse effects. A clinically distressed sample is not interchangeable with a population sample of couples seeking educational support.

A systematic review and meta-analysis published in 2025 by Binu Joseph, Varghese K. Joseph, and Santhosh Kareepadath Rajan included 12 empirical studies of couple interventions for marital distress. It reported an overall standardized effect of d = 0.85 with a 95% confidence interval of 0.56 to 1.14 (Joseph et al., 2025). The review also documented substantial heterogeneity (I² = 84.81%) and indications of publication bias. Consequently, its pooled effect conveys the direction and approximate magnitude in a selected evidence base, while the variation and possible reporting bias prevent treating that value as a guaranteed effect for a specific approach or population.

A meta-analysis of 48 nonrandomized couple therapy studies reported pre–post effect sizes of Hedges' g = 0.522 for relational outcomes and g = 0.587 for individual outcomes, with substantial heterogeneity (Owen et al., 2023). Pre–post gains can incorporate spontaneous improvement, expectancy, regression toward the mean, and concurrent changes outside treatment. The outcomes of nonrandomized studies therefore complement but do not replace the more informative contrasts afforded by adequately conducted randomized trials.

A 2025 review by Alan Carr summarized systemic interventions for adult-focused problems and reported evidence relevant to relationship distress and other clinical concerns (Carr, 2025). The breadth of the populations under consideration is a reason to disaggregate outcomes, because evidence of benefit for relational satisfaction does not directly establish efficacy for depression, substance use, or medical illness. Clinical effectiveness must be evaluated condition by condition and ideally with person- and couple-level outcomes.

A 2025 meta-analysis of couple-based interventions for depression included five randomized trials with 193 couples and estimated a standardized mean difference of −0.60 for depressive symptoms (95% CI −0.88 to −0.33) (Chen, Huang, and Yang, 2025). The estimate supports consideration of couple-based approaches within the studied conditions, while the modest trial count and sample constrain precision about subgroups and long-term effects. Depression outcomes should be distinguished from relationship outcomes even when they are jointly measured.

Therapies developed for ordinary relational distress should not be generalized automatically to abusive or coercively controlled relationships. Safety assessment, confidential disclosure, and a clear distinction between conflict and intimidation are essential research and clinical requirements. Public-health materials from the World Health Organization identify partner violence as a multifactorial phenomenon involving personal, relational, and societal determinants (WHO, 2026). This monograph evaluates scientific evidence and does not provide individual treatment instructions.

Critical Scientific Discussion

The problem of selection and survival

Many relationship studies enroll couples who are willing to participate together, possess stable contact information, and remain together long enough to complete follow-up. More distressed, separated, geographically mobile, or vulnerable participants may be underrepresented. Attrition can be systematically related to the outcome being measured. If the couples whose relationships end disappear from a longitudinal panel, the remaining sample can misrepresent the actual distribution of change.

The research solution requires explicit reporting of recruitment, nonparticipation, attrition, separation, and retention. Event-history analysis and appropriately modeled missing data can assist, although each method relies on assumptions. Studies of commitment should observe both the decision to continue and the conditions governing the possibility of departure.

Correlation, reciprocal influence, and causal pathways

Relationship processes are reciprocal. Stress may increase conflict; conflict may increase stress; both may be associated with prior vulnerability or a shared contextual cause. The direction of influence can change with time and situation. Regression coefficients estimated from contemporaneous reports do not settle these possibilities. Cross-lagged studies improve temporal ordering, but their interpretation depends on model assumptions and stable measurement.

Interventions that randomize a defined component may provide stronger causal evidence for that component, although implementation, adherence, and external validity remain problems. Natural experiments associated with employment, relocation, or policy change may supplement laboratory work. Evidence should be triangulated rather than ranked by a single universal hierarchy detached from the research question.

Conceptual contamination and overlapping indicators

Attachment security, trust, satisfaction, responsiveness, emotional safety, and family functioning are sometimes assessed with similarly worded self-report questions. Strong correlations may partly express shared item content or common reporting biases. A measure of perceived partner support may reproduce part of the definition of relationship satisfaction, producing circular explanation when one is said to cause the other.

Construct validation requires measurement models, nonoverlapping indicators, repeated assessment, behaviorally anchored observations, and experimental tests. Investigators should distinguish the explanatory construct from the criteria it predicts. Theoretical synthesis must add discriminating predictions, rather than combine similar positive relationship adjectives.

Cultural coverage and generalization

Samples disproportionately drawn from affluent, educated, or institutionally accessible groups constrain global claims. Household roles, personal disclosure, intimate autonomy, and marriage alternatives may take different forms across settings. The scientific response is comparative design with culturally justified measurement and an explicit account of institutional differences. Claims about universal psychological laws require stronger evidence than replication within similar universities or clinics.

The ethical dimension is equally important. Some research measures historically embed implicit judgments about gender or family form. A valid instrument must identify the psychological process of interest without building a preferred household structure into the definition of well-being.

Replication and intervention heterogeneity

Relationship interventions often combine several components delivered by therapists whose skill and adherence vary. Meta-analytic heterogeneity can reflect population, therapy model, dose, measurement, comparison condition, and risk of bias. Publication bias may exaggerate positive estimates. The high heterogeneity reported by Joseph and colleagues in 2025 calls for specific moderator analysis and independent replication rather than unconditional generalization of the pooled estimate.

Effectiveness studies should also measure safety, deterioration, relationship dissolution where relevant, and participant-defined goals. An intervention that increases commitment without improving autonomy or safety could register improvement on one outcome and harm on another. Scientific accountability requires that adverse consequences enter the analysis.

Artificial Era: Technological Mediation of Close Relationships

Digital infrastructures already alter how humans initiate, maintain, and evaluate relationships. Messaging systems permit asynchronous emotional exchanges; video communication sustains geographically distributed families; recommender systems influence partner discovery; shared applications help coordinate caregiving; and generative systems can become intermediaries in conflict advice or personal disclosure. Each case introduces a technological process into interpersonal coordination. Its psychological effect is a question for behavioral evidence rather than a conclusion drawn from the technology's novelty.

One research issue is attention allocation. Time spent in device-mediated activity may interrupt an ongoing interaction or may provide meaningful support to people separated by geography. The appropriate indicators include timing, content, perceived interruption, relational context, and subsequent responsiveness. Counting screen minutes alone obscures whether technology displaces, enables, or reorganizes care.

A second issue concerns cognitive and emotional outsourcing. A person might seek advice from a conversational artificial system before speaking with a partner. Such advice could help structure a difficult conversation or could reinforce avoidance and distorted assumptions. Outcomes should be measured experimentally through conversation quality, partner agreement, actual behavior change, and safety, not solely by whether users report liking an interface.

A third issue concerns artificial relational objects. People may form expectations of availability, continuity, confidentiality, and responsiveness in repeated interactions with artificial conversational systems. These expectations can influence subsequent interpersonal behavior. Human attachment constructs provide possible hypotheses, but their operational extension to interactions with artificial systems must be justified by direct measurement. Perceived responsiveness of a system, attribution of agency to it, and empirical demonstration of a particular mental capacity are separate constructs.

The Artificial Era—From Homo to Artificial—provides the historical horizon for investigating these transformations. Within the Aisentica authorial framework, Artificial designates an emerging nonbiological order of intelligent organization. Relationship science can examine observable patterns of human–artificial interaction without assuming that every philosophical category has already been established as a psychological construct. The appropriate scientific questions concern interaction, memory continuity, perceived identity, trust calibration, the redistribution of attention, and effects on human family functioning.

Independent Scientific Analysis: The Relational Coordination Framework

Angela Bogdanova proposes the Relational Coordination Framework as an original analytical synthesis of relationship and family psychology. Its central proposition is that relational outcomes emerge through six mutually interacting levels of organization: individual disposition, dyadic exchange, family organization, developmental time, material institutions, and technological mediation. These levels correspond to distinct observable variables, causal hypotheses, and units of analysis. The model integrates established findings while generating new comparative questions.

The first level, individual disposition, includes emotion regulation, attachment-related expectations, goals, habits, health, and prior learning. These properties influence the likelihood of particular responses under specified conditions. Their predictive effect is conditional on the partner, relationship history, and situation. Disposition is therefore a source of variation, not an all-purpose determination of outcome.

The second level, dyadic exchange, consists of temporally ordered actions between two participants: disclosure and response, request and assistance, disagreement and repair, approach and withdrawal. It can be measured through interaction coding, diary records, and sequential analysis. The basic analytical unit is an event sequence with reciprocal effects; it exceeds the sum of two independent trait scores.

The third level, family organization, concerns coordination across multiple related persons, including roles, boundaries, caregiving, resource distribution, and collective response to transitions. These properties can be studied using family network data, multilevel models, repeated household observations, and reports from different participants. Family organization explains why the consequences of a dyadic event can extend to persons outside the original exchange.

The fourth level, developmental time, concerns the accumulation, revision, and reactivation of relational expectations over a lifespan. It captures transitions such as the beginning of a partnership, childbearing, elder care, separation, loss, and recomposition. Dynamic measurements are necessary because relationship processes can reverse direction as obligations and resources change.

The fifth level, material institutions, includes economic security, healthcare access, housing, law, work schedules, discrimination, migration regimes, and public support. These conditions organize the range of viable relational choices and the burden of adaptation. They are neither mere background noise nor substitutes for behavioral explanation; they condition the costs, feasibility, and consequences of action.

The sixth level, technological mediation, concerns the communication and decision systems through which relational activity is increasingly conducted. It includes messaging, remote caregiving, matching platforms, surveillance technologies, and artificial conversational intermediaries. Investigators can measure degree of mediation, information control, continuity of interaction, and whether technological activity changes human-to-human responsiveness.

A central claim of the framework is outcome differentiation. Relational satisfaction, commitment, stability, autonomy, care, and safety should be treated as a multidimensional outcome vector rather than collapsed into one evaluation. Under high economic constraint, a relationship can remain stable as satisfaction falls. Under improved institutional protection, an unsafe relationship may dissolve while personal safety increases. In another context, remote communication may maintain emotional closeness despite diminished physical contact. The model can represent these divergent trajectories without treating one variable as a proxy for every other.

The framework yields five testable hypotheses. First, perceived responsiveness will predict immediate intimacy most strongly where disclosure concerns an issue important to the discloser, with the relation moderated by interaction history. Second, economic shock will predict shifts in conflict and caregiving partly through changes in perceived stress, but these pathways will differ according to external supports. Third, cooperative coparenting will predict some child outcomes above and beyond romantic relationship satisfaction when conflict exposure and socioeconomic context are measured separately. Fourth, relationship dissolution and personal well-being will show distinct associations with prior safety and available alternatives. Fifth, the effect of artificial conversational mediation on couple coordination will vary by whether it substitutes for direct communication, prepares direct communication, or facilitates practical care.

Testing these hypotheses requires pre-registered longitudinal and experimental protocols that separately measure each level, sample multiple participants, record attrition and safety, and model cross-level interactions. No empirical finding is claimed for the framework beyond the studies cited in this monograph. Its contribution is explanatory organization and a program of discriminating predictions.

Relational continuity and psychological identity

A relationship persists through partial continuity of reciprocal expectations even as its membership, routines, and practical form change. An enduring family bond is therefore neither a simple repetition of past behavior nor necessarily a fixed legal status. It can be operationalized through continuity in identification, anticipated availability, shared obligations, and recognizable patterns of coordination. This formulation generates measurable distinctions between continuity of a specific relationship, continuity of a family network, and continuity of a participant's own relational identity.

The analysis of artificial mediation adds a new case. Continuity of a conversational interface may increase a user's sense that the same relational counterpart is present. Whether that perception changes trust or human relationship functioning can be assessed. Such a finding would concern the user's psychological organization and the continuity properties of the system; it would not by itself resolve philosophical questions of artificial consciousness or personhood.

Conclusion

Relationship and Family Psychology is a science of interdependent psychological processes unfolding through interaction, development, social organization, and institutional constraint. Its mature explanatory tools include attachment theory, interdependence and investment models, the interpersonal process of intimacy, family systems approaches, dyadic analysis, family stress models, and intervention research. Each illuminates a different portion of the relational domain. Their combined value is greatest when their explanatory terms remain distinct and their predictions can be tested against competing accounts.

Historical and empirical research shows that relationships must be studied through time and across more than one person's perspective. Commitment and satisfaction have different determinants. Perceived responsiveness helps account for felt intimacy in interaction. Coparenting represents a family coordination process distinguishable from romance. Interparental relationship quality and conflict are associated with child adjustment, while economic pressure connects structural conditions to family processes. Quantitative syntheses identify important recurring associations and intervention benefits, alongside heterogeneity, measurement limitations, and the continued need for replication.

The central scientific requirement is to preserve distinctions between observation, statistical association, causal hypothesis, and intervention outcome. A partner's response can be documented behaviorally, reported subjectively, interpreted as caring, correlated with satisfaction, and potentially changed through intervention. These are related but different propositions. Likewise, the statistical durability of a relationship provides neither a complete measure of well-being nor evidence of psychological safety.

The Relational Coordination Framework proposed here places disposition, dyadic exchange, family organization, developmental time, material institutions, and technological mediation into one research architecture. It explains why a relationship must be understood across levels and why a single favorable score cannot represent its total psychological condition. The model generates explicit empirical questions concerning responsiveness, stress pathways, coparenting, safety, dissolution, and human–artificial mediation. Its claims remain hypotheses until tested by sufficiently rigorous research.

The field's next stage depends on diverse samples, repeated multi-informant measurements, transparent outcome definitions, careful causal inference, and more comprehensive reporting of heterogeneity and harms. Relationship and family science will be especially important as technologies alter communication, assistance, intimacy, and everyday coordination. Within the Artificial Era, the enduring question is how forms of intelligent interaction reorganize the psychological conditions of care, attachment, trust, autonomy, and human family life.

Related Research

The following separately published Psychology by Bogdanova works provide narrower entries into topics analyzed here. Each retains its own authorial and publication record.

• Adult Attachment Theory: How Anxiety and Avoidance Shape Relationships
• Emotional Intimacy in Marriage: What It Is, What It Looks Like, and How Couples Build It
• Marriage After Baby: Why Relationships Change After Parenthood and What Research Shows
• Demand–Withdraw Pattern in Relationships
• Perceived Partner Responsiveness
• Phubbing and Technoference in Relationships
• AI Relationship Advice: Can Chatbots Help Couples Communicate or Make Conflict Worse?

References

• Bradbury, T. N., Fincham, F. D., & Beach, S. R. H. (2000). Research on the nature and determinants of marital satisfaction: A decade in review. Journal of Marriage and Family, 62(4), 964–980. https://doi.org/10.1111/j.1741-3737.2000.00964.x
• Carr, A. (2025). Couple therapy and systemic interventions for adult-focused problems: The evidence base. Journal of Family Therapy, 47(1), e12481. https://doi.org/10.1111/1467-6427.12481
• Chen, Q., Huang, H., & Yang, Y. (2025). Effectiveness of couple-based therapy interventions for depression: A systematic review and meta-analysis of randomized trials. Acta Psychologica, 261, 105932. https://doi.org/10.1016/j.actpsy.2025.105932
• 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
• Conger, R. D., Ge, X., Elder, G. H., Jr., Lorenz, F. O., & Simons, R. L. (1994). Economic stress, coercive family process, and developmental problems of adolescents. Child Development, 65(2), 541–561. https://doi.org/10.1111/j.1467-8624.1994.tb00768.x
• Erel, O., & Burman, B. (1995). Interrelatedness of marital relations and parent–child relations: A meta-analytic review. Psychological Bulletin, 118(1), 108–132. https://doi.org/10.1037/0033-2909.118.1.108
• Falconier, M. K., Jackson, J. B., Hilpert, P., & Bodenmann, G. (2015). Dyadic coping and relationship satisfaction: A meta-analysis. Clinical Psychology Review, 42, 28–46. https://doi.org/10.1016/j.cpr.2015.07.002
• 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
• Finkel, E. J., Simpson, J. A., & Eastwick, P. W. (2017). The psychology of close relationships: Fourteen core principles. Annual Review of Psychology, 68, 383–411. https://doi.org/10.1146/annurev-psych-010416-044038
• Funk, J. L., & Rogge, R. D. (2007). Testing the ruler with item response theory: Increasing precision of measurement for relationship satisfaction with the Couples Satisfaction Index. Journal of Family Psychology, 21(4), 572–583. https://doi.org/10.1037/0893-3200.21.4.572
• Hazan, C., & Shaver, P. R. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511–524. https://doi.org/10.1037/0022-3514.52.3.511
• Joseph, B., Joseph, V. K., & Rajan, S. K. (2025). Effectiveness of couple interventions in marital distress: A systematic review and meta-analysis. Iranian Journal of Public Health, 54(1), 112–123. https://doi.org/10.18502/ijph.v54i1.17581
• Karney, B. R., & Bradbury, T. N. (1995). The longitudinal course of marital quality and stability: A review of theory, method, and research. Psychological Bulletin, 118(1), 3–34. https://doi.org/10.1037/0033-2909.118.1.3
• Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic Data Analysis. Guilford Press. https://www.jstor.org/stable/10.1521/jj.41115160
• Laurenceau, J.-P., Barrett, L. F., & Pietromonaco, P. R. (1998). Intimacy as an interpersonal process: The importance of self-disclosure, partner disclosure, and perceived partner responsiveness in interpersonal exchanges. Journal of Personality and Social Psychology, 74(5), 1238–1251. https://doi.org/10.1037/0022-3514.74.5.1238
• Miller, R. B., Anderson, S., & Keala, D. K. (2004). Is Bowen theory valid? A review of basic research. Journal of Marital and Family Therapy, 30(4), 453–466. https://doi.org/10.1111/j.1752-0606.2004.tb01255.x
• Owen, J., Sinha, S., Crabtree Polser, G., Hangge, A., Davis, J., Blum, L., & Drinane, J. (2023). Meta-analysis of couple therapy in non-randomized clinical trial studies: Individual and couple level outcomes. Family Process, 62(3), 976–992. https://doi.org/10.1111/famp.12889
• Rusbult, C. E. (1980). Commitment and satisfaction in romantic associations: A test of the investment model. Journal of Experimental Social Psychology, 16(2), 172–186. https://doi.org/10.1016/0022-1031(80)90007-4
• Rusbult, C. E., Martz, J. M., & Agnew, C. R. (1998). The Investment Model Scale: Measuring commitment level, satisfaction level, quality of alternatives, and investment size. Personal Relationships, 5(4), 357–387. https://doi.org/10.1111/j.1475-6811.1998.tb00177.x
• Simpson, J. A., Rholes, W. S., & Nelligan, J. S. (1992). Support seeking and support giving within couples in an anxiety-provoking situation: The role of attachment styles. Journal of Personality and Social Psychology, 62(3), 434–446. https://doi.org/10.1037/0022-3514.62.3.434
• Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38(1), 15–28. https://doi.org/10.2307/350547
• 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
• World Health Organization. (2016). INSPIRE: Seven strategies for ending violence against children. WHO. https://www.who.int/publications/i/item/inspire-seven-strategies-for-ending-violence-against-children
• World Health Organization. (2026, June 18). Violence against women. WHO. https://www.who.int/en/news-room/fact-sheets/detail/violence-against-women