Pandemic Separation Tested Polyamorous Attachment Bonds in Unprecedented Ways

When COVID-19 lockdowns split households apart in the spring of 2020, most research on romantic relationships focused on couples confined together under one roof. But a substantial number of people maintain more than one committed romantic relationship at the same time, and for them the pandemic produced a peculiar psychological experiment: some partners were suddenly locked in, while others were locked out. A new study published in the Archives of Sexual Behavior takes advantage of that natural experiment to ask how attachment bonds to both a cohabiting partner and a separated partner shaped relationship satisfaction, commitment, and psychological distress during the first wave of the pandemic.

The research team, led by Rose Bern of the University of California, Davis, together with colleagues at the University of Texas at Austin, the University of Toronto Mississauga, and Chapman University, surveyed 186 people practicing polyamory who were living with at least one romantic partner while being physically separated from at least one other. Participants completed their surveys between March and May of 2020, the narrow and anxious window when much of the world first went into lockdown. That timing matters. The respondents were describing their attachment bonds and their emotional states in real time, not reconstructing them from memory months or years later.

Attachment theory, first formulated by John Bowlby and later extended to adult romantic relationships, holds that intimate partners serve two psychological functions: a safe haven that provides comfort in times of threat, and a secure base from which people explore the world. Two dimensions of insecurity have been mapped across decades of research. Attachment anxiety reflects hypervigilance to rejection and an intense craving for reassurance, while attachment avoidance reflects discomfort with closeness and a tendency to suppress reliance on others. Classic theory assumed a single primary attachment figure, but people in polyamorous relationships maintain multiple concurrent romantic bonds, raising the question of whether each partner can independently function as an attachment figure and whether insecurity toward one partner spills over into another relationship.

To measure these dynamics, the researchers administered the short form of the Experiences in Close Relationship Scale separately for the cohabiting partner and the separated partner, allowing anxiety and avoidance to be assessed within each bond. Psychological outcomes were captured with the Patient Health Questionnaire-9 for depressive symptoms and the Generalized Anxiety Disorder-7 scale for anxiety symptoms, alongside a measure of perceived COVID-19 threat and distress. Relationship outcomes were assessed with the Couples Satisfaction Index and commitment items adapted for each specific relationship. The analytic strategy then regressed each outcome on attachment anxiety and avoidance toward both partners simultaneously, which is what allowed the team to detect cross-relationship effects that would be invisible in studies of monogamous couples.

The within-relationship findings were consistent with the broader attachment literature. Higher attachment anxiety toward a partner predicted lower relationship satisfaction with that partner, whether the partner was cohabiting or separated. Higher avoidance likewise predicted lower satisfaction within each corresponding relationship. These results reinforce the idea that each relationship in a polyamorous network functions as its own attachment system, with insecurity in a given bond degrading the quality of that bond rather than being diluted across multiple partnerships.

The pandemic-specific findings were more surprising and more finely differentiated. Attachment anxiety toward both the cohabiting partner and the separated partner predicted greater COVID-related distress, suggesting that anxious insecurity in any attachment bond left people more emotionally exposed to the global threat. Avoidance, however, worked in opposite directions depending on living arrangement. Avoidance toward a cohabiting partner predicted lower COVID distress, plausibly because avoidantly attached individuals who were confined with a partner experienced the enforced closeness as more stressful when their defensive distance was unavailable, or alternatively because distancing strategies buffered them. Avoidance toward a separated partner, by contrast, predicted higher COVID distress, consistent with the interpretation that avoidant individuals tolerate separation poorly when the pandemic forecloses the possibility of reconnection on their own terms.

Perhaps the most theoretically provocative result was the single cross-relationship effect that emerged: attachment anxiety toward a separated partner predicted increased commitment to the cohabiting partner. In other words, people who felt insecurely anxious about a partner they could not reach became more devoted to the partner they were locked in with. The authors interpret this pattern through the lens of interdependence and attachment dynamics: the physical presence of a safe haven during a period of acute threat may strengthen bonds to that haven, particularly when an alternative attachment figure is inaccessible and the anxiety associated with that inaccessibility heightens the value of the available one. For a literature that has often treated relationships in isolation, this finding demonstrates that attachment processes in one romantic bond can meaningfully shape outcomes in another.

The study also carries methodological and representational significance. Relationship science has been repeatedly criticized for relying overwhelmingly on young, White, heterosexual, monogamous samples, a limitation that constrains the generalizability of its theories. By recruiting polyamorous individuals during a global crisis, the researchers extended attachment theory into a relationship structure that Bowlby-era frameworks never anticipated, testing whether the safe haven concept scales across multiple concurrent figures. The answer appears to be yes, with important qualifications: each bond operates as an attachment system, but the systems interact, and the physical arrangement imposed by lockdowns modulated how insecurity translated into distress.

The findings also speak to the mental health consequences of pandemic separation more broadly. Prior work during COVID-19 consistently linked insecure attachment to greater depression, anxiety, and coronavirus-related worry, but nearly all of that work examined single dyads. The present results suggest that the geography of attachment, meaning which attachment figures are physically available when a threat strikes, is a critical variable. Being separated from an attachment figure during a mass threat event was not psychologically neutral; anxiety about that separated partner amplified pandemic distress, while the presence of a cohabiting partner partially compensated, at least in terms of commitment.

Caveats remain. The survey was cross-sectional, so the direction of the associations cannot be established with certainty, and the sample of 186 polyamorous individuals, while valuable, may not generalize to all consensually non-monogamous populations or to monogamous couples. Data were not released in an open repository, and the study was approved by the Institutional Review Board at Chapman University with informed consent from all participants. Still, as one of the first studies to measure dual attachment bonds to cohabiting and separated partners during lockdown, the work opens a genuinely new line of inquiry: how the architecture of a person’s entire attachment network, not just a single relationship, determines resilience when a crisis divides intimate life between those who are present and those who are kept apart.

Subject of Research: Attachment bonds to cohabiting and separated romantic partners among polyamorous individuals during COVID-19 lockdowns

Article Title: Together and Apart: Dual Attachment Bonds to Cohabiting and Separated Partners Predict Relationship and Psychological Outcomes During COVID-19 Lockdowns

Article References: Bern, R., Wooten, H., Moreno, E., Eastwick, P. W., & Moors, A. C. (2026). Together and Apart: Dual Attachment Bonds to Cohabiting and Separated Partners Predict Relationship and Psychological Outcomes During COVID-19 Lockdowns. Archives of Sexual Behavior. https://doi.org/10.1007/s10508-026-03538-7

Image Credits: AI Generated

DOI: 10.1007/s10508-026-03538-7

Keywords: attachment theory, polyamory, consensual nonmonogamy, COVID-19, lockdown, relationship satisfaction, commitment, mental health, psychological distress, romantic relationships, safe haven, Archives of Sexual Behavior