Sustaining Couple Relationships as Romantic Passion Changes

Abstract

What enables couples to sustain satisfying relationships when romantic passion changes? A biological expiration account cannot explain the diversity of long-term relationships, and inventories of beneficial behaviors do not establish which responses help which couples. This critical integrative review distinguishes passion, sexual desire, relationship satisfaction, commitment, and persistence, then evaluates evidence from quantitative syntheses, dyadic longitudinal studies, and experiments. Responsiveness, shared growth, coping, appreciation, and commitment are associated with relationship functioning, but their effects cannot be treated as interchangeable or uniformly causal. Recent evidence on satisfaction before separation, self-expansion, and household labor further emphasizes timing and context. Building on vulnerability–stress–adaptation, dynamic goal, responsiveness, and maintenance theories, the article proposes that adaptation depends on identifying changing demands, selecting feasible responses, and revising those responses through feedback from both partners. The proposed contribution is a set of testable predictions about response specificity, sequencing, and revision, rather than a new inventory of protective factors. An evidence matrix separates established findings from unresolved causal claims, and a research agenda specifies active comparisons, independent measurement of needs, and outcomes after dissolution. Long-term maintenance is best evaluated through both partners’ relationship quality and voluntary participation, with persistence treated as a distinct outcome.

Keywords: romantic passion, relationship maintenance, perceived partner responsiveness, dyadic coping, self-expansion, relationship satisfaction

Introduction

The central problem of long-term couple relationships is adaptation under interdependence. The person from whom an individual seeks comfort, practical cooperation, and intimacy is also someone whose preferences, resources, and circumstances change. Familiarity may reduce excitement while increasing security. Shared responsibilities may deepen investment while reducing opportunities for rewarding interaction. A scientific account of maintenance must explain how couples negotiate these changes and why similar efforts produce different outcomes.

The common question of what keeps couples together “after romantic chemistry fades” contains two assumptions that require revision. First, it treats the decline of passion as a uniform biological stage. Evidence instead distinguishes intense infatuation from enduring romantic love and documents considerable variation in established relationships (Acevedo & Aron, 2009; Carswell & Impett, 2021). Second, it treats staying together as the outcome to be explained. Commitment and constraints can preserve relationships with very different levels of satisfaction, and interventions can improve functioning without measurably reducing separation (Doss et al., 2020; Rhoades et al., 2010). Relationship quality and persistence therefore require related but separate explanations.

Existing research identifies many plausible maintenance processes. Reviews associate positive interaction, assurances, shared tasks, and other maintenance strategies with relationship functioning (Ogolsky & Bowers, 2013; Ogolsky & Stafford, 2023). Yet a list of favorable correlates leaves a consequential question unanswered: when partners experience reduced connection, which process should change, in what circumstances, and over what interval? Reassurance may address uncertainty about being valued; it may leave an exhausting division of responsibilities intact. An enjoyable activity may restore shared reward; it may also impose additional planning on an already overloaded partner. These examples are theoretical contrasts, not established treatment rules.

This review develops an account of maintenance as the selection and revision of responses to changing relational demands. It integrates established theories without claiming that sensitivity to needs is a new principle. The intended contribution is more specific: changes in passion can initiate interpretations and actions whose consequences depend on what has changed, whether the response addresses it, and whether both partners can participate. The review first separates the relevant outcomes, then evaluates mechanisms and their evidentiary limits, and finally specifies predictions that could distinguish this account from general positivity, stable favorable evaluations, and selective persistence. Midlife is considered an informative developmental context rather than a universal stage of decline.

Review Scope and Evidentiary Approach

This article is a critical integrative narrative review, not a systematic review or meta-analysis. Selection began with the theoretical domains of relationship maintenance, vulnerability–stress–adaptation, attachment and responsiveness, self-expansion, commitment, and dyadic coping. Targeted searches of publicly accessible scholarly records, publisher sites, PubMed, and author or institutional repositories were conducted during preparation on October 9–10, 2026. Search combinations included romantic passion and maintenance; responsiveness and desire; self-expansion and longitudinal relationship quality; coping and stress; commitment and stability; midlife and marital quality; and relationship intervention and randomized trial. Targeted updating included publications from 2025 and 2026, including evidence that qualified the proposed account.

Priority was given to theoretical statements, quantitative syntheses, dyadic longitudinal research, and experiments relevant to mechanisms or boundary conditions. Primary sources were used for specific empirical claims. The selection was purposive, with no exhaustive database retrieval, independent screening, preregistered protocol, or formal risk-of-bias scoring. Accessible full texts informed interpretation where available; claims based on abstracts were restricted to the designs and findings reported there. The evidence matrix in Table 1 is an analytic summary of selected studies, not an exhaustive evidence inventory. This approach supports conceptual synthesis but cannot estimate the prevalence of supportive findings or a pooled comparative effectiveness of maintenance strategies.

Three inferential distinctions guide the synthesis. Associations between couples need not describe what happens when a particular couple changes. Temporal prediction does not, by itself, establish a causal mechanism. Finally, an experimental benefit of a complete intervention does not identify the component responsible for that benefit. These distinctions are especially important when extrapolating from daily interactions to relationship trajectories over years.

Specifying What Changes and What Should Be Sustained

Passion and Related Outcomes

Romantic passion refers to an intense motivational orientation toward union with a particular partner; sexual desire concerns interest in sexual activity, including partner-directed desire. Attachment concerns expectations of availability and security. Satisfaction is an evaluation of the relationship, whereas commitment includes the motivation and intention to maintain it (Carswell & Impett, 2021; Li & Chan, 2012; Tran et al., 2019). These constructs overlap, but neither their measures nor their trajectories are interchangeable. Sexual frequency, for example, cannot establish how desired or satisfying sexual contact is for either partner.

For the present account, maintenance refers to processes that preserve or improve valued relationship functioning under changing conditions. The primary outcomes are each partner’s satisfaction and experienced intimacy, supplemented by sexual well-being where relevant to their relationship. Voluntary participation and freedom from coercion are ethical conditions for interpreting an outcome as beneficial, not assumed correlates of high satisfaction. Persistence is assessed separately. This definition permits a relationship to remain rewarding with less intense passion, and it permits persistent incompatibility to be recognized without classifying all separation as failed adaptation.

These distinctions determine what evidence would answer the research question. A study showing immediate closeness after a joint activity establishes neither sustained desire nor lower divorce risk. A study predicting divorce establishes neither the quality of the years before separation nor the well-being of the people afterward. A comprehensive account should preserve these differences rather than combine them into an undifferentiated index of relationship success.

Developmental Variation and the Timing of Deterioration

Bühler et al. (2021) synthesized 165 independent samples comprising 165,039 participants and found systematic variation in satisfaction with age and relationship duration. Their findings concern satisfaction, not a biological timetable for passion. Average trajectories also leave room for heterogeneous experiences and cannot distinguish all effects of duration, life transitions, cohort, and selective dissolution.

A more recent analysis changes the temporal question. In four longitudinal national samples, Bühler and Orth (2025) found that satisfaction in relationships that ended declined more sharply as separation approached. Estimated transitions into the steeper phase occurred approximately 0.58–2.30 years before separation, and time relative to separation described change better than time since relationship formation. Because these trajectories were identified using knowledge of subsequent separation, they are not a prospective diagnostic rule or evidence that decline becomes irreversible. They nevertheless show why elapsed relationship duration is an inadequate explanation of deterioration. The processes surrounding a particular change may matter more than a couple’s position on a presumed universal relationship clock.

Midlife illustrates the same problem. Gorchoff et al. (2008) followed women across 18 years and linked the transition to an empty nest with increased marital satisfaction and greater enjoyment of time with a partner. This observational, cohort-specific result does not establish a universal midlife improvement or a causal benefit of children leaving home. Conversely, Lin et al. (2018), using the Health and Retirement Study, found that empty-nest and retirement transitions and chronic health conditions did not independently predict gray divorce in their adjusted models; marital quality, duration, and economic resources were more informative. A transition can therefore be relevant to satisfaction without being a general predictor of dissolution.

Cohort comparisons further limit universal developmental narratives. Wright et al. (2023) found differences across several dimensions of midlife marital quality over a quarter century, including more disagreement and instability and less fairness and interaction, while happiness was comparable. These comparisons do not show that an increase in divorce causes poorer marriages. Together, the midlife findings support measuring the demands, opportunities, and meanings associated with transitions rather than assigning a fixed effect to age or family stage.

The Limits of a Chemical Expiration Account

Neurobiological research is relevant to attachment and reward but does not establish that romantic chemicals are exhausted and then replaced. Acevedo et al. (2012) observed partner-related activation in reward-associated and attachment-associated regions among 17 people selected for intense romantic love after an average of 21.4 years of marriage. Selection prevents estimates of prevalence, and functional imaging does not directly demonstrate neurotransmitter depletion or replacement. The study is important because enduring intense love is compatible with the observed neural responses, not because it identifies a recipe for producing that experience.

Berendzen et al. (2023) found partner preference and parental behavior in prairie voles with disrupted oxytocin receptor signaling. This result challenges an unconditional claim that this receptor is necessary for every expression of attachment. Developmental compensation and differences between genetic and pharmacological methods remain relevant, and animal partner preference cannot stand in for human satisfaction or commitment. The appropriate explanatory level for the present review is therefore interpersonal: what partners do, how those actions are experienced, and how their circumstances change. Biological mechanisms remain important questions requiring direct measurement.

Theoretical Basis for Adaptive Maintenance

Stress and Changing Goals Organize Relational Demands

The vulnerability–stress–adaptation model explains relationship development through the interplay of partners’ enduring characteristics, stress, and adaptive processes (Karney & Bradbury, 1995). McNulty et al. (2021) extended its dyadic application by examining both partners’ characteristics, stress, and behavior. The model supplies the contextual architecture for maintenance: the same interaction occurs under different capacities and pressures. It also discourages explanations that attribute every difficulty either to incompatible personalities or to deficient communication.

Enduring characteristics should not be equated with fixed destinies. O’Meara and South (2019) found correlated changes in personality and relationship satisfaction; correlated change does not establish which process drives the other. Shiota and Levenson (2007) found that overall Big Five similarity predicted less favorable satisfaction trajectories in their long-term marriage sample, challenging the blanket assumption that similarity is always protective. These findings justify including individual differences and their development, while leaving response selection as a distinct level of explanation.

Dynamic goal theory provides a necessary complement. Li and Fung (2011) proposed that marital satisfaction depends on achieving prioritized goals, including growth, companionship, and instrumental aims, whose relative importance can change across adulthood and contexts. The present synthesis adopts the importance of changing goals without treating chronological age as sufficient to identify them. One partner may seek renewed exploration while the other seeks relief from responsibilities. What requires coordination is the partners’ actual priorities, including disagreement between them.

A relational demand is defined here as an independently assessed condition that obstructs a valued goal: for example, insufficient opportunities for shared reward, unmet expectations of availability, or an unsustainable workload. This definition separates the condition from its interpretation. A partner may interpret exhaustion as rejection or interpret persistent disregard as temporary stress. Neither interpretation should be assumed accurate because it is optimistic or because it preserves commitment.

Responsiveness Connects Intentions with Experience

Attachment theory and the intimacy process literature explain why enacted behavior and received care can diverge. Attachment anxiety and avoidance are associated with poorer relationship quality, although their patterns differ across outcomes (Li & Chan, 2012). Perceived partner responsiveness concerns feeling understood, validated, and cared for. Laurenceau et al. (1998) showed in diary research that disclosure and responsiveness were associated with experienced intimacy. A disclosure creates an opportunity for connection; its interpersonal consequences depend partly on the response.

Feeney and Collins (2015) distinguish support through adversity from support for pursuing opportunities. Their account already establishes that relationships can provide both security and growth and that effective support must attend to the recipient. Brock and Lawrence (2009) additionally found that both insufficient and excessive support were associated with deteriorating marital satisfaction. These precedents rule out claiming that “support should fit needs” is a novel theory. The unresolved issue pursued here is how partners identify changing demands, select among potentially useful responses, and revise them when feedback indicates that they are ineffective.

Responsiveness is not identical to agreement, compliance, or amount of effort. A response may acknowledge a partner’s concern while declining a particular request. A recipient may feel understood even before a practical problem is solved; conversely, a task may be completed without conveying care. Distinguishing objective changes in demands, the provider’s actions, and the recipient’s experience is necessary to test whether responsiveness explains effects or merely redescribes a favorable evaluation.

Maintenance Includes Preserving and Developing the Relationship

The integrative maintenance review by Ogolsky et al. (2017) organizes a broad literature that encompasses individual and dyadic processes and efforts directed toward preserving or enhancing relationships. Ogolsky and Stafford (2023) further identify limitations in contextual and dyadic coverage. The present account builds within this tradition. Its focus is the decision process connecting a perceived change to a maintenance response, rather than the proposal of additional universally desirable behaviors.

Self-expansion and hedonic adaptation theories help explain why maintenance can involve development as well as repair. Appreciation may counter the loss of attention to familiar benefits, while variety can renew opportunities for positive experience (Jacobs Bao & Lyubomirsky, 2013). Self-expansion concerns growth in perspectives, capacities, and identities through a relationship. These theories do not imply that every difficulty reflects boredom. Their contribution is to identify a plausible reward-related pathway that must be distinguished from responses to threat, overload, or incompatible preferences.

What the Evidence Supports and Where It Stops

Appreciation Is an Interaction Rather Than a Prescribed Quantity

Ogolsky and Bowers (2013) synthesized 35 studies with 12,273 participants and found associations between maintenance strategies and relationship quality. Stronger associations for some partner-perception measures than for reports of one’s own behavior raise a measurement question: studies may capture the meaning of behavior as well as its frequency. Satisfaction can also encourage maintenance and favorable interpretations, complicating causal inference.

Gordon et al. (2012) linked appreciation, perceived care, responsiveness, and maintenance across complementary methods. More decisive for prescription is the experiment by Algoe and Zhaoyang (2016). Couples assigned to express gratitude over a month showed modest benefits relative to a relationally active comparison, with outcomes depending in part on how responsive the partner was perceived to be when expressing gratitude. Because perceived responsiveness was measured rather than randomized, its moderating role is not itself a randomized causal effect. The contrast nevertheless challenges the assumption that reproducing the outward behavior reliably reproduces its naturally occurring benefits.

The theoretical implication is that recognition can reinforce investment when it communicates accurate attention to a partner’s contribution. Repeated gratitude that leaves a persistent burden unchanged may serve a different function. That latter proposition requires direct testing; it should not be inferred from gratitude correlations. Appreciation, workload, and the interpretation of appreciation must be measured separately.

Shared Growth Can Renew Reward Without Guaranteeing Durability

Aron et al. (2000) demonstrated immediate improvements in experienced relationship quality after experimentally assigned shared novel and arousing activities. Graham (2008), using experience sampling, associated activation during shared activities with relationship quality. The experiment supports a proximal causal effect of its activity condition; experience sampling improves temporal resolution but remains observational. Neither design alone demonstrates a durable reduction in relationship deterioration.

Dobson et al. (2024) distinguished self-expanding behavior, current perceptions, and perceived future potential in two preregistered dyadic daily-experience studies. All three were associated with daily satisfaction and commitment, but perceived future potential uniquely predicted those outcomes two months later. The distinction suggests that an activity and its implication for the relationship’s future are separable. It does not establish that changing expectations causes long-term commitment, because favorable expectations can partly reflect prior relationship quality.

Recent reviews emphasize expansion and its integration into the self rather than novelty alone (Emery et al., 2025) and consider shared self-expansion as a possible pathway to broader well-being (Aron & Tomlinson, 2026). These refinements suggest evaluating whether partners find an activity meaningful and mutually wanted. An activity can be new without providing growth, and growth can occur through a familiar activity approached differently. The empirical priority is to isolate these possibilities with credible comparisons, rather than prescribe novelty as a universal response to low passion.

Coping Depends on Stress and the Feasibility of Responding

Falconier et al. (2015) synthesized 72 independent samples involving 17,856 participants and found an aggregate correlation of r = .45 between dyadic coping and relationship satisfaction. This is evidence of substantial covariation, not an estimate of the effect of coping instruction. Rusu et al. (2020), in five annual assessments of 240 couples, separated stable between-person differences from year-specific deviations and found associations at both levels, with some associations involving partner and common coping varying with stress. The within-person findings narrow the gap between individual change and between-couple comparison, but same-year covariation does not identify whether coping preceded satisfaction.

Roth et al. (2025) followed 300 mixed-gender couples across ten years and related communication and dyadic coping to satisfaction subgroups. The focus on stable couples makes the study informative about heterogeneity among continuing relationships, while limiting conclusions about relationships that dissolve. Annual and daily findings should consequently be integrated as evidence at different timescales, not counted as repeated demonstrations of the same causal pathway.

Material conditions affect what maintenance is possible. Karney (2021) describes how socioeconomic circumstances shape intimate relationships, including exposure to demands and opportunities to respond. A partner may understand the other’s needs yet lack time, money, or schedule control to act on them. A recommendation that requires additional childcare, travel, or planning can increase the burden it was intended to relieve. Distinguishing motivation from feasible opportunity prevents the model from treating resource constraints as a personal failure of caring.

Desire Requires Attention to Autonomy and Unequal Burdens

Birnbaum et al. (2016) linked perceived responsiveness with partner-directed sexual desire across complementary methods, with stronger associations among women in the studied samples. This evidence challenges a necessary opposition between closeness and desire, while not establishing that emotional responsiveness resolves every sexual difficulty. Desire, willingness to engage in sexual activity, and sexual satisfaction must remain distinct outcomes.

The content of sexual responsiveness matters. Vowels et al. (2022) used responses from 305 participants to develop a participant-informed definition that included listening and accommodation alongside respect for safety and boundaries. As qualitative definitional evidence, this study clarifies the construct rather than demonstrating an intervention effect. It also identifies a problem for maintenance research: a measure that rewards compliance without attending to either partner’s preferences can misclassify costly self-suppression as successful responsiveness.

Household labor provides an instructive test of contextual explanations. Harris et al. (2022), in two observational studies of women partnered with men who had children, associated a greater share of household labor with lower partner-directed desire; perceived unfairness and partner dependence were implicated in statistical mediation. A newer pair of preregistered studies did not find an overall association between doing more labor and lower desire. Instead, Liepmann et al. (2026) reported patterns that depended on gender-related beliefs, with the predicted association more evident among women lower in benevolent sexism and less consistent across designs. These findings support measuring meanings and expectations, but they do not establish a universal causal effect of chore redistribution on desire. Nor does acceptance of an unequal arrangement establish its fairness.

The broader implication is that low desire can have relational and contextual correlates that are obscured by locating the problem entirely within an individual’s biology. It does not follow that the higher-desire partner’s preferred frequency should determine successful maintenance. The relevant outcomes include mutual willingness, sexual satisfaction, and distress, rather than activity frequency alone.

Commitment Organizes Persistence But Does Not Guarantee Quality

The Investment Model links commitment with satisfaction, investments, and the quality of alternatives. Tran et al. (2019) provide meta-analytic support for these associations. Their magnitude should not be read as causal weights or as a universal ranking of what a couple should change. Investments can motivate constructive effort and can also increase the cost of leaving an unsatisfactory relationship.

Schoebi et al. (2012) distinguished wanting a marriage to continue from the behavioral inclination to maintain it. Among 172 couples, the latter predicted interaction, steps toward dissolution, and 11-year divorce outcomes beyond satisfaction. The lower of the two partners’ maintenance inclinations was consequential for divorce risk. This is a useful corrective to explanations that presume one highly committed partner can sustain a mutually valued relationship alone.

Rhoades et al. (2010) further distinguished dedication, material constraints, perceived constraints, and feeling trapped in 1,184 unmarried adults followed for eight months. Each uniquely predicted stability, but felt constraint differed from the other components in direction. The findings make it inappropriate to treat every source of persistence as evidence of flourishing. They also caution against transferring associations from dating relationships directly to decades-long marriages without further testing.

Beliefs about passion supply a bridge from experience to action. Carswell and Finkel (2018) found that low passion was more strongly associated with reduced commitment and interest in alternatives among people who regarded passion decline as irreversible. This supports examining interpretation alongside passion level. It does not show that teaching a belief in renewal prevents divorce, or that continued investment is always adaptive. Whether confidence encourages effective action or prolongs frustration should depend partly on what partners subsequently do and experience.

Intervention Evidence Supports Change But Not a Single Mechanism

Lavner et al. (2016) found that longitudinal links between observed communication and marital satisfaction were less consistent than a simple communication-to-satisfaction explanation would imply. This limits claims about a single mechanism; it does not imply that interventions involving communication cannot help. Roddy et al. (2020) synthesized 58 couple-therapy studies representing 40 independent samples and 2,092 couples and found improvements across major relationship outcomes. Because the synthesis included different designs, its large within-treatment pre–post satisfaction effect should not be interpreted as an equivalent randomized between-group effect.

A trial supplies a more specific comparison. Doss et al. (2020) randomized 742 low-income couples to OurRelationship, ePREP, or a waitlist. Both programs improved several dimensions of functioning relative to the waitlist, with benefits maintained over four months after treatment; neither significantly reduced breakups at the six-month follow-up. Differences between programs were generally small. Thus, distress and functioning can improve through interventions with different emphases, while improvement and persistence remain empirically distinct. This finding challenges a strong claim that only a precisely matched response can work. Any proposed advantage of matching must be demonstrated over interventions that already provide attention, structure, and opportunities for constructive engagement.

Maintenance as Selection and Revision of Responses

The Proposed Sequence

The proposed account concerns an episode of change embedded in an ongoing relationship. A partner detects a difference between desired and experienced connection. The partners interpret that change, identify relevant demands, and select a response from what they can feasibly do. The enacted response may alter the demand directly, change how cared for the recipient feels, or generate a rewarding experience. Each partner then updates expectations and either sustains, revises, or abandons the response. Repeated episodes may accumulate into security and valued connection or into frustration and withdrawal.

This sequence is a theoretical synthesis, not a causal chain established by the reviewed studies. Vulnerability–stress–adaptation theory explains the conditions under which it occurs; dynamic goal theory specifies why priorities can change; responsiveness describes an interpersonal experience through which actions acquire meaning; self-expansion and coping specify different candidate functions. Commitment influences willingness to continue investing, and constraints influence whether continued participation is freely chosen. The framework does not require these processes to occur in a rigid order or imply that all demands can be resolved.

Three distinctions make the account testable. Demand–response correspondence concerns whether an action addresses a previously measured difficulty. Received responsiveness concerns the recipient’s experience of understanding and care. Feasibility concerns whether the action can be enacted without unacceptable cost or burden. A response can score differently on these dimensions: listening may convey care without reducing workload, while taking over a task may reduce workload without repairing distrust. “Fit” cannot simply be assigned after satisfaction improves, because that would make the theory circular.

Specificity and Sequencing

The framework predicts relative, not exclusive, advantages. An enjoyable activity could benefit an overloaded couple through shared positive affect even if it leaves workload unchanged. The more discriminating claim is that, holding attention and credibility reasonably constant, addressing overload should produce larger or more durable benefits when overload is the principal impediment. If outcomes improve equally across initially different demands and responses, a general positive-engagement account is sufficient.

Sequencing offers an additional prediction. Where an activity requires resources currently depleted by strain, practical relief may improve its uptake and subsequent benefit. Where the dominant difficulty is loss of shared reward despite adequate resources and preserved trust, requiring extensive problem discussion before engaging in a rewarding activity may add little. These are hypotheses about order, not a recommendation to classify couples into fixed types. Partners often face multiple demands, and their priorities may conflict. The relevant test compares sequences selected from measurements made before treatment, rather than retrospectively explaining any successful sequence as well matched.

Feedback from Both Partners

Effective revision requires information from both members of the dyad. One partner’s satisfaction with an attempted solution cannot establish that it reduced the other’s burden. Disagreement is itself information about the target of maintenance, not measurement error to be averaged away. The model therefore predicts that an intervention can produce apparent couple-level improvement while creating unequal benefits, especially when the person selecting the response is not the person doing most of the work.

Feedback can also be misleading. If dissatisfaction reduces favorable interpretations, even constructive behavior may initially be discounted; if commitment encourages favorable interpretation, persistent problems may be underestimated. Joel et al. (2020), analyzing 43 longitudinal couple datasets, found strong prediction of reported relationship quality from relationship-specific variables but much weaker prediction of subsequent change. This warns against treating concurrent self-report agreement as an explanation of development. Independent observation, repeated measurement, and out-of-sample prediction are necessary to establish whether the proposed sequence adds explanatory value.

The framework’s boundary is mutual participation under conditions that permit choice. It cannot turn unilateral effort into a sufficient condition for a mutually valued partnership, nor can it establish that persistent disagreement is repairable. Successful assessment may reveal incompatible priorities or an unwillingness to invest. Following individuals after dissolution is therefore essential to distinguish preserving a relationship from improving the lives of its members.

Predictions That Could Distinguish the Account

Prediction 1 Changes in Interpretation Precede Response Selection

Among partners reporting comparable deviations from their usual passion, beliefs about reversibility should predict subsequent maintenance attempts, particularly when past interactions support expectations of responsiveness. Repeated assessments should separate passion, interpretation, and action in time and distinguish stable differences between people from deviations within people. The prediction is weakened if interpretation predicts only contemporaneous satisfaction, if action reliably precedes interpretation, or if effects disappear when prior action and contextual demands are accounted for. This extends passion-belief research to a specified behavioral sequence; it remains a hypothesis.

Prediction 2 Initial Demands Moderate Comparative Benefits

A randomized comparison should test at least two credible response conditions across independently measured demands. For example, a practical-coordination intervention and a shared-growth intervention could be compared among couples varying continuously in overload and restricted shared reward. The primary test is the prespecified demand-by-condition interaction, with attention to each partner’s outcomes, rather than separate significance tests within subgroups. Comparable improvement across demands would challenge specificity while remaining compatible with common benefits of engagement. A null result with poor uptake would be inconclusive about the response’s psychological mechanism but informative about feasibility.

Prediction 3 Removing Constraints Can Improve Later Uptake

When lack of opportunity is measured before intervention, reducing a practical constraint should increase uptake of a subsequent shared activity and may increase its benefit. A factorial or sequentially randomized design could distinguish the effects of relief alone, activity alone, and their combination or order. The model predicts an advantage of relief before activity specifically where that constraint is present. An equal sequencing benefit regardless of initial constraints would favor a nonspecific explanation, and no benefit despite verified constraint reduction would weaken the proposed pathway. Uptake, experienced responsiveness, and satisfaction should be measured separately so that a participation effect is not mistaken for an effect on experience once participating.

Prediction 4 Revision Improves Outcomes Beyond Repetition

After an initial response produces little improvement in its intended target, a strategy that incorporates both partners’ feedback and permits a prespecified change of response should outperform repetition of the initial strategy under equivalent contact and burden. A sequential trial could randomize eligible nonresponders to continuation or revision, while keeping all randomized participants in the analysis. The definition of nonresponse, the timing of reassessment, and the allowable revisions must be set prospectively. The prediction is challenged if revision provides no advantage over continued credible engagement, or if benefits reflect additional attention rather than change in response. This is the most direct test of adaptive maintenance as an ongoing process rather than a one-time matching decision.

Design Requirements Across the Predictions

The model should be evaluated against established responsiveness and support-adequacy measures, not only against a weak no-treatment comparison. Measures of initial demand must avoid incorporating the later outcome. Behavioral coding, time-use records, and partner reports can complement self-reports while introducing their own measurement limitations. Analyses should represent dyadic interdependence and distinguish within-person fluctuations, differences between partners, and differences between couples. Statistical mediation alone cannot establish the proposed mechanism, especially when mediator and outcome are jointly influenced by unmeasured daily events.

Timescales should follow the proposed process. Interaction-level measures can capture receipt of care; daily measures can capture uptake and immediate experience; repeated follow-ups over months and years are required for durable change. Prespecified primary outcomes, simulation-based power planning for interactions and dyadic dependence, and reporting of effect uncertainty are more informative than an accumulation of exploratory moderator tests. Relationship dissolution should be modeled as an outcome, with continued individual follow-up where possible, rather than treated solely as missing satisfaction data. Sensitivity analyses should examine informative attrition and selective participation by couples willing to complete research together.

Implications and Limits

The evidence supports asking what has changed before selecting a response. Low excitement with preserved affection, persistent overload, loss of trust, and unwanted sexual pressure are not equivalent presentations. The review does not supply a validated classification algorithm or justify replacing established clinical assessment with the proposed sequence. It does justify evaluating the target of an intervention, its feasibility, and both partners’ experience, rather than assuming that increasing the frequency of a positively labeled behavior is sufficient.

Midlife provides a particularly useful setting for these tests because changes in work, family responsibilities, health, and opportunities can be measured alongside longstanding relationship histories. Age should not determine the presumed need. Empty-nest status does not imply more enjoyable time together for every couple, and a long marriage does not establish either security or entrapment. Recruiting across resources, relationship durations, and family arrangements would help distinguish a developmental mechanism from characteristics of the samples in which it was first observed.

Several limits remain. This is a selective synthesis, and its proposed integration may underrepresent competing explanations. Many studies involve Western, mixed-gender, married or cohabiting samples, with uneven representation of sexual and gender diversity and nonmonogamous arrangements. A pathway can be plausible without functioning equivalently across cultural meanings of obligation, autonomy, and partnership. Measurement comparability and contextual validity require investigation rather than assumption.

The framework also risks relabeling established processes. It earns explanatory value only if prospective measures of demand, response sequence, and revision improve prediction or intervention outcomes beyond existing accounts. The propositions specify where that added value could be demonstrated and where evidence could favor a simpler explanation. Current findings warrant neither a universal hierarchy of protective factors nor a claim that ordinary relationship practices alone overcome structural deprivation, coercion, or persistent incompatibility.

Conclusion

Long-term relationship maintenance is not adequately explained by what replaces a depleted romantic chemistry. The evidence points instead to changing experiences of reward, security, cooperation, and commitment, operating under unequal opportunities and evolving priorities. Responsiveness, appreciation, shared growth, and coping are credible processes associated with relationship quality; experimental evidence shows that functioning can improve, while leaving the mechanisms and durability of particular components less certain. The proposed account organizes the next scientific question around how partners select, sequence, and revise responses to independently assessed demands. Its success should be judged by improvements in both partners’ valued experience and voluntary participation, with relationship persistence retained as a distinct outcome.

References

Acevedo, B. P., & Aron, A. (2009). Does a long-term relationship kill romantic love? Review of General Psychology, 13(1), 59–65. https://doi.org/10.1037/a0014226

Acevedo, B. P., Aron, A., Fisher, H. E., & Brown, L. L. (2012). Neural correlates of long-term intense romantic love. Social Cognitive and Affective Neuroscience, 7(2), 145–159. https://doi.org/10.1093/scan/nsq092

Algoe, S. B., & Zhaoyang, R. (2016). Positive psychology in context: Effects of expressing gratitude in ongoing relationships depend on perceptions of enactor responsiveness. The Journal of Positive Psychology, 11(4), 399–415. https://doi.org/10.1080/17439760.2015.1117131

Aron, A., Norman, C. C., Aron, E. N., McKenna, C., & Heyman, R. E. (2000). Couples' shared participation in novel and arousing activities and experienced relationship quality. Journal of Personality and Social Psychology, 78(2), 273–284. https://doi.org/10.1037/0022-3514.78.2.273

Aron, A., & Tomlinson, J. M. (2026). Self-expansion activities with a partner as one path to well-being. Behavioral Sciences, 16(5), Article 700. https://doi.org/10.3390/bs16050700

Berendzen, K. M., Sharma, R., Alvarado Mandujano, M., Wei, Y., Rogers, F. D., Simmons, T. C., Seelke, A. M. H., Bond, J. M., Larios, R., Goodwin, N. L., Sherman, M., Parthasarthy, S., Espineda, I., Knoedler, J. R., Beery, A. K., Bales, K. L., Shah, N. M., & Manoli, D. S. (2023). Oxytocin receptor is not required for social attachment in prairie voles. Neuron, 111(6), 787–796.e4. https://doi.org/10.1016/j.neuron.2022.12.011

Birnbaum, G. E., Reis, H. T., Mizrahi, M., Kanat-Maymon, Y., Sass, O., & Granovski-Milner, C. (2016). Intimately connected: The importance of partner responsiveness for experiencing sexual desire. Journal of Personality and Social Psychology, 111(4), 530–546. https://doi.org/10.1037/pspi0000069

Brock, R. L., & Lawrence, E. (2009). Too much of a good thing: Underprovision versus overprovision of partner support. Journal of Family Psychology, 23(2), 181–192. https://doi.org/10.1037/a0015402

Bühler, J. L., Krauss, S., & Orth, U. (2021). Development of relationship satisfaction across the life span: A systematic review and meta-analysis. Psychological Bulletin, 147(10), 1012–1053. https://doi.org/10.1037/bul0000342

Bühler, J. L., & Orth, U. (2025). Terminal decline of satisfaction in romantic relationships: Evidence from four longitudinal studies. Journal of Personality and Social Psychology. Advance online publication. https://doi.org/10.1037/pspp0000551

Carswell, K. L., & Finkel, E. J. (2018). Can you get the magic back? The moderating effect of passion decay beliefs on relationship commitment. Journal of Personality and Social Psychology, 115(6), 1002–1033. https://doi.org/10.1037/pspi0000147

Carswell, K. L., & Impett, E. A. (2021). What fuels passion? An integrative review of competing theories of romantic passion. Social and Personality Psychology Compass, 15(8), Article e12629. https://doi.org/10.1111/spc3.12629

Dobson, K., Veeravalli, A., Gazder, T., & Stanton, S. C. E. (2024). Self-expansion perceptions and behaviors uniquely contribute to relationship quality over time. Journal of Family Psychology, 38(3), 484–494. https://doi.org/10.1037/fam0001188

Doss, B. D., Knopp, K., Roddy, M. K., Rothman, K., Hatch, S. G., & Rhoades, G. K. (2020). Online programs improve relationship functioning for distressed low-income couples: Results from a nationwide randomized controlled trial. Journal of Consulting and Clinical Psychology, 88(4), 283–294. https://doi.org/10.1037/ccp0000479

Emery, L. F., Hughes, E. K., & Muise, A. (2025). Self-expansion theory: Origins, current evidence, and future horizons. Social and Personality Psychology Compass, 19(9), Article e70082. https://doi.org/10.1111/spc3.70082

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

Feeney, B. C., & Collins, N. L. (2015). A new look at social support: A theoretical perspective on thriving through relationships. Personality and Social Psychology Review, 19(2), 113–147. https://doi.org/10.1177/1088868314544222

Gorchoff, S. M., John, O. P., & Helson, R. (2008). Contextualizing change in marital satisfaction during middle age: An 18-year longitudinal study. Psychological Science, 19(11), 1194–1200. https://doi.org/10.1111/j.1467-9280.2008.02222.x

Gordon, A. M., Impett, E. A., Kogan, A., Oveis, C., & Keltner, D. (2012). To have and to hold: Gratitude promotes relationship maintenance in intimate bonds. Journal of Personality and Social Psychology, 103(2), 257–274. https://doi.org/10.1037/a0028723

Graham, J. M. (2008). Self-expansion and flow in couples' momentary experiences: An experience sampling study. Journal of Personality and Social Psychology, 95(3), 679–694. https://doi.org/10.1037/0022-3514.95.3.679

Harris, E. A., Gormezano, A. M., & van Anders, S. M. (2022). Gender inequities in household labor predict lower sexual desire in women partnered with men. Archives of Sexual Behavior, 51(8), 3847–3870. https://doi.org/10.1007/s10508-022-02397-2

Jacobs Bao, K., & Lyubomirsky, S. (2013). Making it last: Combating hedonic adaptation in romantic relationships. The Journal of Positive Psychology, 8(3), 196–206. https://doi.org/10.1080/17439760.2013.777765

Joel, S., Eastwick, P. W., Allison, C. J., Arriaga, X. B., Baker, Z. G., Bar-Kalifa, E., Bergeron, S., Birnbaum, G. E., Brock, R. L., Brumbaugh, C. C., Carmichael, C. L., Chen, S., Clarke, J., Cobb, R. J., Coolsen, M. K., Davis, J., de Jong, D. C., Debrot, A., DeHaas, E. C., . . . Wolf, S. (2020). Machine learning uncovers the most robust self-report predictors of relationship quality across 43 longitudinal couples studies. Proceedings of the National Academy of Sciences, 117(32), 19061–19071. https://doi.org/10.1073/pnas.1917036117

Karney, B. R. (2021). Socioeconomic status and intimate relationships. Annual Review of Psychology, 72, 391–414. https://doi.org/10.1146/annurev-psych-051920-013658

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

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

Lavner, J. A., Karney, B. R., & Bradbury, T. N. (2016). Does couples' communication predict marital satisfaction, or does marital satisfaction predict communication? Journal of Marriage and Family, 78(3), 680–694. https://doi.org/10.1111/jomf.12301

Li, T., & Chan, D. K.-S. (2012). How anxious and avoidant attachment affect romantic relationship quality differently: A meta-analytic review. European Journal of Social Psychology, 42(4), 406–419. https://doi.org/10.1002/ejsp.1842

Li, T., & Fung, H. H. (2011). The dynamic goal theory of marital satisfaction. Review of General Psychology, 15(3), 246–254. https://doi.org/10.1037/a0024694

Liepmann, A., Cross, E. J., & Muise, A. (2026). Division of household labor and sexual desire: The role of gender and benevolent sexism. The Journal of Sex Research, 63(8), 1465–1481. https://doi.org/10.1080/00224499.2026.2656775

Lin, I.-F., Brown, S. L., Wright, M. R., & Hammersmith, A. M. (2018). Antecedents of gray divorce: A life course perspective. The Journals of Gerontology: Series B, 73(6), 1022–1031. https://doi.org/10.1093/geronb/gbw164

McNulty, J. K., Meltzer, A. L., Neff, L. A., & Karney, B. R. (2021). How both partners' individual differences, stress, and behavior predict change in relationship satisfaction: Extending the VSA model. Proceedings of the National Academy of Sciences, 118(27), Article e2101402118. https://doi.org/10.1073/pnas.2101402118

O'Meara, M. S., & South, S. C. (2019). Big Five personality domains and relationship satisfaction: Direct effects and correlated change over time. Journal of Personality, 87(6), 1206–1220. https://doi.org/10.1111/jopy.12468

Ogolsky, B. G., & Bowers, J. R. (2013). A meta-analytic review of relationship maintenance and its correlates. Journal of Social and Personal Relationships, 30(3), 343–367. https://doi.org/10.1177/0265407512463338

Ogolsky, B. G., Monk, J. K., Rice, T. M., Theisen, J. C., & Maniotes, C. R. (2017). Relationship maintenance: A review of research on romantic relationships. Journal of Family Theory & Review, 9(3), 275–306. https://doi.org/10.1111/jftr.12205

Ogolsky, B. G., & Stafford, L. (2023). A systematic review of relationship maintenance: Reflecting back and looking to the future. Personal Relationships, 30(1), 19–43. https://doi.org/10.1111/pere.12429

Rhoades, G. K., Stanley, S. M., & Markman, H. J. (2010). Should I stay or should I go? Predicting dating relationship stability from four aspects of commitment. Journal of Family Psychology, 24(5), 543–550. https://doi.org/10.1037/a0021008

Roddy, M. K., Walsh, L. M., Rothman, K., Hatch, S. G., & Doss, B. D. (2020). Meta-analysis of couple therapy: Effects across outcomes, designs, timeframes, and other moderators. Journal of Consulting and Clinical Psychology, 88(7), 583–596. https://doi.org/10.1037/ccp0000514

Roth, M., Nussbeck, F. W., Landolt, S. A., Senn, M., Bradbury, T. N., Weitkamp, K., & Bodenmann, G. (2025). Dyadic coping and communication as predictors of 10-year relationship satisfaction subgroup trajectories in stable romantic couples. Behavioral Sciences, 15(10), Article 1361. https://doi.org/10.3390/bs15101361

Rusu, P. P., Nussbeck, F. W., Leuchtmann, L., & Bodenmann, G. (2020). Stress, dyadic coping, and relationship satisfaction: A longitudinal study disentangling timely stable from yearly fluctuations. PLOS ONE, 15(4), Article e0231133. https://doi.org/10.1371/journal.pone.0231133

Schoebi, D., Karney, B. R., & Bradbury, T. N. (2012). Stability and change in the first 10 years of marriage: Does commitment confer benefits beyond the effects of satisfaction? Journal of Personality and Social Psychology, 102(4), 729–742. https://doi.org/10.1037/a0026290

Shiota, M. N., & Levenson, R. W. (2007). Birds of a feather don't always fly farthest: Similarity in Big Five personality predicts more negative marital satisfaction trajectories in long-term marriages. Psychology and Aging, 22(4), 666–675. https://doi.org/10.1037/0882-7974.22.4.666

Tran, P., Judge, M., & Kashima, Y. (2019). Commitment in relationships: An updated meta-analysis of the Investment Model. Personal Relationships, 26(1), 158–180. https://doi.org/10.1111/pere.12268

Vowels, L. M., Roos, C. A., Mehulić, J., O’Dean, S. M., & Sánchez-Hernández, M. D. (2022). What does it mean to be responsive to a partner’s sexual needs? Toward a definition of sexual need responsiveness. Archives of Sexual Behavior, 51(8), 3735–3747. https://doi.org/10.1007/s10508-022-02432-2

Wright, M. R., Brown, S. L., & Manning, W. D. (2023). A cohort comparison of midlife marital quality: A quarter century of change. Journal of Family Issues, 44(2), 538–559. https://doi.org/10.1177/0192513X211054466

Table 1

Selected evidence and limits of inference

Note. This is a purposive evidence matrix, not an exhaustive review or formal grading of certainty. Each source is discussed in the text and listed in References. The proposed response-selection and revision sequence has not been tested as a complete model.