Can Attachment Style Change? Stability, Life Events, and Earned Security
Updated: 6 days ago
Author: Ukrainian Psychological Hub · Published: September 22, 2026 · Editorial Policy
Yes. Adult attachment can change. The strongest answer from research, however, is more precise than the popular idea of “changing your attachment style.” Adult attachment shows both continuity and change: people often carry relatively stable tendencies, yet attachment-related anxiety and avoidance can shift across time, relationships, and major life experiences. A large naturalistic longitudinal study found both immediate and enduring changes around life events, while also showing that many changes were temporary and that people often moved back toward their pre-event trajectories. Fraley, Gillath, and Deboeck (2021).
That combination—stability plus revisability—is the central point. Attachment is not best understood as a permanent personality label, and it is not a switch that can be flipped on command. In adult romantic attachment research, anxiety and avoidance are commonly measured as continuous dimensions. Terms such as secure, anxious or preoccupied, dismissive-avoidant, and fearful-avoidant are useful summaries of patterns on those dimensions, not diagnoses or immutable human types. Relationship-specific measures also show that the same person can hold different attachment expectations in different close relationships. Fraley, Heffernan, Vicary, and Brumbaugh (2011). For the broader scientific model, measurement traditions, and anxiety × avoidance framework, see Adult Attachment Theory: How Anxiety and Avoidance Shape Relationships. For the cluster-wide taxonomy and four familiar style summaries, see Attachment Styles in Adults: The Four Patterns, Two Dimensions, and What They Mean.
This article owns one question: how stable adult attachment is, what kinds of change researchers actually observe, how life events and close relationships relate to that change, what psychotherapy evidence can and cannot establish, and how the research construct of earned security fits into the picture. It does not treat a quiz result as a diagnosis, promise a timetable for “healing an attachment style,” or assume that a new relationship automatically rewrites a person’s attachment system.
Can Attachment Style Change? The Short Answer
Adult attachment can change, but several different kinds of change are often collapsed into one sentence. A person can show a temporary state shift, a gradual change in average anxiety or avoidance, a relationship-specific change with one partner, a change in a categorical label because a score crossed a cutoff, or a broader long-term change in expectations about close relationships. These are not interchangeable outcomes.
Intensive longitudinal research illustrates why “stable” and “fixed” are different ideas. In two studies that measured adults daily for 30 days and weekly for a year, attachment showed a stable underlying component alongside temporary variation. The pattern fit a prototype model better than a model in which earlier attachment simply faded away with time. Fraley, Vicary, Brumbaugh, and Roisman (2011). The practical implication is that a stable baseline can coexist with meaningful short-term movement and with longer-term updating.
Across much longer spans of life, mean levels can also change. In longitudinal data spanning ages 13 to 72, attachment anxiety declined on average, especially in middle and later adulthood, while avoidance declined more linearly across the life span. Being in a relationship was associated with lower anxiety and avoidance. Chopik, Edelstein, and Grimm (2019). Those findings describe group-level trajectories; they do not mean that aging by itself makes every individual more secure.
What Does “Attachment Change” Actually Mean?
The phrase attachment style makes change sound categorical: anxious today, secure tomorrow. Much adult self-report research is better represented by two dimensions. Attachment anxiety concerns fears of rejection, abandonment, or insufficient availability. Attachment avoidance concerns discomfort with dependence, intimacy, vulnerability, or relying on others. A relatively secure self-report pattern generally means lower anxiety and lower avoidance. Movement can therefore happen without a person ever crossing a category boundary. For the low-anxiety, low-avoidance pattern itself, see Secure Attachment Style: Signs, Relationships, and How Security Works.
This matters because categorical language can exaggerate both stability and change. Someone near a cutoff could receive a different “style” label after a small score shift. Someone else could make a substantial reduction in anxiety while remaining on the same side of a cutoff. The dimensional question—how much anxiety and avoidance are present, in what relationship, and over what time period—is usually more informative than asking whether a fixed type has been replaced.
Global attachment and relationship-specific attachment can move differently
Adults do not necessarily carry one identical attachment representation into every relationship. The ECR-RS was designed to measure attachment anxiety and avoidance across specific relationships such as mother, father, romantic partner, and best friend, as well as more general patterns. Relationship-specific attachment measures can predict relational outcomes differently from broad attachment measures. Fraley et al. (2011).
A later longitudinal study followed 4,904 adults across repeated assessments and found that global and relationship-specific working models changed together over the long run and also showed coordinated short-term fluctuations. People who became more secure in specific close relationships also tended to become more secure globally, but the authors explicitly cautioned that correlated change does not prove that change in one domain caused change in another. Dugan, Fraley, Gillath, and Deboeck (2022).
So a person can plausibly feel secure with one partner and less secure in another context. That is not evidence that the science is inconsistent. It is one of the reasons a single global label can miss important structure.
Stability Has More Than One Meaning
When researchers say attachment is stable, they may mean rank-order stability: people who are relatively more anxious or avoidant than others at one time tend, on average, to remain relatively higher later. They may mean mean-level stability: the group average changes little. Or they may mean individual-level stability: a particular person changes little. Those questions can yield different answers in the same dataset.
For example, a group could show a small average decline in anxiety while many individuals remain stable, some become substantially less anxious, and others become more anxious. A correlation across two time points can show continuity in relative ordering even while nearly everyone moves somewhat. This is why a slogan such as “attachment styles are 70% stable” is not a sufficient scientific summary unless the measure, population, interval, and statistical definition of stability are specified.
The same caution applies to popular claims that “30% of people change attachment style.” A classification-change percentage from one study is not a universal rate of adult attachment change. It may depend on categorical cutoffs, sample characteristics, age, relationship status, interval length, and whether attachment was measured by self-report, interview, or developmental classification.
What Longitudinal Research Says About Life Events
Life events can coincide with attachment change, but the evidence is more nuanced than “a breakup makes you anxious” or “a good relationship makes you secure.” In the naturalistic study of more than 4,000 people, researchers tracked attachment before and after events over periods ranging from 6 months to 40 months. About half of the studied event types were associated with immediate changes. On average, however, people often moved back toward levels expected from their pre-event trajectories. Enduring average change appeared for about a quarter of the event types, and individuals differed substantially in direction and durability of change. Fraley, Gillath, and Deboeck (2021).
How people construed an event also mattered: more positive versus more negative interpretations were related to the extent of attachment change. That finding fits an updating model in which experiences matter partly through the meaning they acquire within a person’s relational expectations. It still does not imply that interpretation alone causes durable change, or that a person can simply choose a more positive story and thereby become secure.
Change can move toward security or toward insecurity
Attachment change is bidirectional. New experiences can weaken expectations of rejection or unavailability, but betrayal, chronic unreliability, bereavement, relationship dissolution, frightening experiences, or repeated conflict can also coincide with increased insecurity. The same longitudinal literature therefore argues against treating security as a one-way developmental ladder.
A temporary increase in anxiety or avoidance during a difficult period also does not necessarily mean a person has acquired a permanent new attachment identity. Short-term deviations can settle as conditions change. The scientifically useful question is whether the shift persists across repeated assessments and comparable contexts.
Can a Relationship Make You More Secure?
A close relationship can provide experiences that are relevant to attachment expectations, but a partner is not an attachment-style treatment. Research supports the possibility of relationship-specific updating and, in some studies, broader change. It does not support the promise that dating a “secure person” will automatically transform an insecure partner.
One line of evidence comes from studies of positive relationship events. Across three dyadic studies, more frequent positive relationship experiences predicted short-term declines in romantic attachment avoidance, but not anxiety. Relationship-validating behavior also predicted later declines in avoidance over one month. Those effects did not continue as significant time interactions across longer follow-ups of two months to one year. Bayraktaroglu et al. (2023). This is a useful example of why immediate improvement should not automatically be called enduring attachment change.
Recent dyadic evidence adds another caution. A 2025 study followed 1,036 established couples for 20 months. Attachment anxiety decreased slightly over time in women, avoidance remained stable, there was no evidence that partners’ attachment trajectories codeveloped, and changes were largely unrelated to baseline relationship characteristics such as satisfaction, support, closeness, responsiveness, and disclosure. Traut et al. (2025). Relationship experiences matter, but the pathway from a “good relationship” to lower insecurity is neither automatic nor uniform.
One relationship process worth understanding on its own is perceived partner responsiveness—feeling understood, valued, and cared for by a partner. Responsiveness is highly relevant to intimacy and relational experience. It should be treated as a relationship process, not as proof that attachment anxiety or avoidance must decline.
Can Therapy Change Attachment Patterns?
Psychotherapy can be a context in which people examine expectations, emotion regulation, interpersonal behavior, and patterns of closeness or distance. The evidence is consistent with possible attachment-related change, but it does not support a single standardized treatment called “attachment-style healing,” and it does not justify promising that a client will become securely attached.
A meta-analysis of 36 psychotherapy studies involving 3,158 patients found that more secure attachment before treatment was associated with better psychotherapy outcomes. It also found that improvements in attachment security during therapy may coincide with better outcomes, while moderator findings about treatment focus were preliminary. Levy, Kivity, Johnson, and Gooch (2018). The wording “coincide with” matters: this literature does not establish that every effective therapy works by changing attachment style.
Couple-therapy studies are mixed in exactly the way a careful account should expect. In a sample of 32 couples receiving emotionally focused couple therapy, relationship-specific avoidance decreased, and anxiety decreased among participants who completed a particular therapeutic change event. Burgess Moser et al. (2016). In a much larger treatment-as-usual sample of 461 couples followed across eight sessions, anxiety and avoidance were relatively stable, with only a slight decline in anxiety among women. Johnson et al. (2016).
The conclusion is not that therapy cannot matter. It is that therapy research does not justify universal claims about speed, mechanism, or guaranteed conversion from one attachment category to another. Treatment should be chosen for the actual problem a person wants help with—relationship distress, trauma symptoms, depression, anxiety, interpersonal difficulties, or another clinically relevant concern—rather than for a promise to install a new label. For the separate question of recovery and evidence-based treatment when childhood adversity is linked to present trauma-related distress, see Healing From Childhood Trauma: What Recovery Can Mean and What Treatments Have Evidence.
Security Priming Is Not the Same as Lasting Attachment Change
Experimental attachment research can temporarily activate mental representations of safety, availability, and support. A meta-analysis of 120 security-priming studies found beneficial effects across affective, cognitive, and behavioral outcomes. Gillath, Karantzas, Romano, and Karantzas (2022). This is evidence that attachment-related states and processes can be influenced in the short term.
It is not evidence that a brief visualization, affirmation, meditation, or social-media exercise permanently changes a person’s adult attachment orientation. State activation, repeated priming, relationship-specific change, and durable global change are different outcomes. Popular programs often collapse them into one promise; research should keep them separate.
What Is Earned Secure Attachment?
Earned security is a research construct with a specific developmental and interview-based history. It should not be used as a casual synonym for “I used to be anxious and now I feel better.” In the Adult Attachment Interview tradition, earned-secure or earned-autonomous classifications have generally referred to adults who show secure or autonomous states of mind regarding attachment despite reported adverse or insecure childhood experiences. For a dedicated review of the construct, its AAI roots, and the current evidence, see Earned Secure Attachment: What It Means and What the Evidence Actually Shows.
That definition is more complicated than internet usage suggests. A 23-year longitudinal study found that retrospectively defined earned-secure adults were not necessarily the same people who had been observed as insecurely attached in infancy. Prospectively defined participants who did move from insecure infant classifications to secure adult states did show strong close-relationship functioning. Roisman, Padrón, Sroufe, and Egeland (2002). The study is important partly because it challenged an easy retrospective story.
A 2026 scoping review of 24 empirical studies found more formal than substantive agreement across the earned-security literature and described the evidence as heterogeneous and preliminary. It identified secondary attachment figures and reflective functioning as potentially important, while noting major unresolved questions. Filosa, Sharp, Gori, and Musetti (2026).
Therefore, “earned secure attachment” should not be certified by an online quiz or inferred from one lower score on attachment anxiety or avoidance. Adult self-report dimensions and Adult Attachment Interview classifications are related areas of attachment research, but they are not the same measurement system.
Childhood Attachment and Adult Romantic Attachment Are Not the Same Measurement Problem
Questions about adult change often pull in evidence from infancy. That literature is relevant to developmental continuity, but it must be labeled correctly. A classic 20-year study assessed infants with the Strange Situation and later assessed adults with the Adult Attachment Interview. Seventy-two percent of the recontacted participants had the same secure-versus-insecure classification across those assessments, and negative life events were associated with classification change. Waters, Merrick, Treboux, Crowell, and Albersheim (2000).
A later meta-analysis of 127 papers and more than 21,000 attachment observations found medium stability in attachment to parents from infancy to early adulthood, with stability generally lower across longer intervals. Pinquart, Feussner, and Ahnert (2013). These findings concern developmental attachment continuity and different measurement traditions. They should not be converted into a direct numerical estimate of how likely an adult’s romantic self-report attachment style is to change.
This distinction also prevents another common error: fearful-avoidant adult self-report patterns should not automatically be equated with infant disorganized attachment or with unresolved/disorganized classifications on the Adult Attachment Interview. They come from different measures and classification traditions. Likewise, insecure adult romantic attachment patterns are not Reactive Attachment Disorder or Disinhibited Social Engagement Disorder.
How Long Does It Take to Change an Attachment Style?
There is no scientifically established universal timetable. Claims such as “30 days,” “six months,” “one year,” or “three to five years” may sound concrete, but adult attachment research does not provide one conversion clock that applies across people, measures, relationships, and interventions.
The studies themselves operate on very different time scales. Intensive designs can detect fluctuations across days or weeks. Relationship-specific longitudinal studies can detect changes across months. Naturalistic life-event research has followed trajectories before and after events for many months. Life-span studies examine mean-level trends across decades. A change that appears after three weeks may be a meaningful state or relationship-specific shift without being evidence of a durable global change.
A stronger question is: does the change persist across repeated observations, does it generalize beyond one unusually calm or stressful period, and does it appear in the relationship context that matters? Durability and generalization are empirical features of change; they cannot be read from a calendar.
What Meaningful Movement Toward Security Can Look Like
There is no official checklist that proves someone has “become secure.” A practical way to translate dimensional attachment research into everyday observation is to look for repeated changes in the processes that anxiety and avoidance describe. For someone high in anxiety, that may mean less catastrophic interpretation of ordinary distance, more direct requests for reassurance or support, and greater ability to tolerate uncertainty without escalating checking or protest behavior. For someone high in avoidance, it may mean greater willingness to disclose needs, receive care, depend on a partner when appropriate, and remain engaged during manageable emotional closeness.
These examples are behavioral translations of the constructs, not diagnostic criteria. Meaningful change would be more convincing if it recurs across comparable situations and over time rather than appearing only on one unusually good week. It can also be relationship-specific: a person may become more secure with one close other before broader expectations change.
Conflict behavior is one place where change may become visible, but attachment labels should not be used to explain every interaction. A recurring demand-withdraw pattern can involve one partner pushing for engagement while the other pulls back, yet that interaction pattern is not itself an attachment diagnosis. Similarly, calling a partner “emotionally unavailable” can hide multiple possible mechanisms and should not be treated as proof of avoidant attachment.
What May Support Greater Attachment Security?
No single technique has been established as a universal attachment-change intervention. The evidence instead points toward several plausible routes that can support more secure functioning: repeated experiences of reliable and responsive connection; opportunities to revise relationship-specific expectations; improved emotion regulation and communication; reflective understanding of recurring interpersonal patterns; and psychotherapy when a person is dealing with distress, trauma-related symptoms, entrenched interpersonal difficulties, or other mental-health concerns. For the practical change intent specific to attachment anxiety, see How to Heal Anxious Attachment: Evidence-Based Ways to Build More Security.
The evidence strength differs across those routes. Positive relationship experiences have shown short-term associations with lower avoidance in some studies, but longer-term effects are less consistent. Psychotherapy studies suggest attachment security can change alongside treatment, but effect patterns vary and attachment change is not guaranteed. Earned-security research points to potentially important roles for secondary attachment figures and reflective functioning, while the current review literature still describes the evidence as heterogeneous and preliminary.
A useful goal is therefore not to force a permanent identity change. It is to build more flexible responses to closeness, support, separation, conflict, and dependence, then observe whether those changes become durable and generalize. For the practical change intent specific to attachment avoidance, see How to Heal Avoidant Attachment: Evidence-Based Ways to Build More Security. For combined high-anxiety/high-avoidance change work, see How to Heal Fearful-Avoidant Attachment: Working With Both Anxiety and Avoidance.
Can Attachment Become Less Secure Again?
Yes. A person who functions securely in one period can become more anxious or avoidant under sustained stress or in a relationship marked by unreliability, threat, repeated rejection, betrayal, or loss. Longitudinal life-event research shows that change can occur in both directions and that individuals vary greatly in their responses to the same broad type of event. Fraley, Gillath, and Deboeck (2021).
This does not make earlier growth unreal. It means attachment-related expectations remain responsive to experience. A flexible model of change is more accurate than imagining a final state in which old reactions can never reappear.
Why One Attachment Quiz Cannot Prove Change
A self-report questionnaire can be useful for research or structured reflection when the measure is appropriate, but one before-and-after score cannot by itself establish enduring psychological change. Scores can vary with relationship context, current circumstances, reference person, interpretation of items, and ordinary measurement error. Relationship-specific instruments were developed precisely because attachment representations can differ across close relationships. Fraley et al. (2011).
If a score changes, ask what else changed. Did the reference relationship change? Did the person move from acute conflict into a calmer period? Are the same patterns different across repeated situations? Does a lower anxiety score correspond to less hypervigilance to rejection, or did the person simply disengage? Does lower avoidance correspond to greater comfort with closeness, or only to a temporary honeymoon period? The score is one piece of evidence, not a certificate.
Attachment Change Is Not the Same as Treating a Disorder
Adult romantic attachment style is a research construct, not a psychiatric diagnosis. An insecure pattern is not an “attachment disorder.” Reactive Attachment Disorder and Disinhibited Social Engagement Disorder are separate clinical diagnoses with their own developmental and diagnostic criteria and should not be used as labels for ordinary adult relationship patterns. The World Health Organization’s ICD-11 diagnostic guidance lists Reactive Attachment Disorder and Disinhibited Social Engagement Disorder as distinct disorders specifically associated with stress. World Health Organization (2024).
Likewise, phrases such as attachment wounds, attachment trauma, and attachment issues are often used in therapy or popular discourse, but they do not by themselves name an official diagnosis. If someone has trauma symptoms, obsessive-compulsive symptoms, depression, anxiety, or another clinical concern, that concern deserves assessment in its own terms rather than being automatically explained by an attachment label.
This distinction is especially important when repetitive relationship doubts, checking, mental review, comparison, and reassurance seeking become the main problem. Those experiences can overlap superficially with attachment anxiety, but Relationship OCD involves obsessive-compulsive processes and requires a different conceptual lens.
Frequently Asked Questions
Can anxious attachment become secure?
Attachment anxiety can decrease, and some people show movement toward lower anxiety over time. That can happen gradually and may be relationship-specific or broader. Research does not provide a guaranteed route or timetable, and becoming “secure” should not be treated as passing a categorical threshold. Longitudinal work supports both continuity and change. For the definition, signs, triggers, and relationship pattern of anxious attachment, see Anxious Attachment Style: Signs, Triggers, Relationships, and What Research Shows.
Can avoidant attachment become secure?
Attachment avoidance can also decrease. Some studies have linked positive relationship experiences with short-term declines in romantic avoidance, and life-span data show lower average avoidance with age. Other longitudinal couple research has found avoidance to be quite stable over shorter periods. The evidence therefore supports possibility, not inevitability.
Can your attachment style change depending on the person you are with?
Yes, relationship-specific attachment is an established research idea. A person can report different levels of anxiety and avoidance with different close others, and relationship-specific working models can change over time. That is different from claiming that every partner creates a completely new personality.
Does a secure partner fix insecure attachment?
No partner can guarantee that outcome. A reliable, responsive relationship may provide experiences that challenge expectations of rejection or unavailability, but current longitudinal evidence does not show a simple partner-to-partner transmission of security. In a 20-month study of more than 1,000 couples, partners’ attachment trajectories did not codevelop, and relationship characteristics were largely unrelated to change. Traut et al. (2025).
Can trauma change attachment style?
Adverse and stressful life experiences can be associated with changes in attachment-related security, but attachment measures do not diagnose trauma and a change in attachment score does not prove a trauma history. Trauma-related symptoms and disorders require their own assessment. The broader life-event literature supports experience-related change without reducing every change to trauma.
Does attachment style change with age?
On average, some longitudinal studies find lower attachment anxiety and avoidance with increasing age. In the 59-year longitudinal analysis, anxiety declined especially in middle and later adulthood and avoidance declined more linearly. Chopik et al. (2019). Group averages, however, cannot predict one person’s trajectory.
Is earned secure attachment the same as scoring secure on a quiz?
No. Earned security comes from a developmental and Adult Attachment Interview research tradition. A self-report quiz that places someone in a “secure” category is measuring something different. The recent scoping review also shows substantial heterogeneity in how earned security has been defined and studied. Filosa et al. (2026).
Can attachment change without therapy?
Yes, attachment can change outside therapy. Naturalistic longitudinal studies document change around life events and across relationships. Therapy is one possible context for change, not a requirement for every person and not a guarantee. If distress, trauma symptoms, compulsive reassurance seeking, severe conflict, or other mental-health problems are present, professional support may be useful for those concerns.
How do I know whether a change is lasting?
Look for persistence across time and comparable situations rather than one test result. Durable change should be distinguishable from a short-term reaction to an unusually good or bad period. Repeated observations across several months, and attention to whether change is global or relationship-specific, are more informative than a single categorical label.
The Bottom Line
Adult attachment is neither destiny nor a freely editable profile setting. The best evidence supports a system with stable underlying tendencies, short-term fluctuations, relationship-specific representations, and genuine capacity for longer-term revision. Anxiety and avoidance can change; life events sometimes produce enduring shifts; close relationships can supply updating experiences; therapy can coincide with greater security; and earned security is a real but technically specific research construct.
The most scientifically defensible goal is not to chase a permanent label. It is to understand which attachment-related expectations and behaviors are changing, in which relationships, for how long, and with what evidence. That framing preserves both parts of the research record: continuity is real, and so is change.
Related Articles
References
Bayraktaroglu, D., Gunaydin, G., Selcuk, E., Besken, M., & Karakitapoglu-Aygun, Z. (2023). The role of positive relationship events in romantic attachment avoidance. Journal of Personality and Social Psychology, 124(5), 958–970. https://doi.org/10.1037/pspi0000406
Burgess Moser, M., Johnson, S. M., Dalgleish, T. L., Lafontaine, M.-F., Wiebe, S. A., & Tasca, G. A. (2016). Changes in relationship-specific attachment in emotionally focused couple therapy. Journal of Marital and Family Therapy, 42(2), 231–245. https://doi.org/10.1111/jmft.12139
Chopik, W. J., Edelstein, R. S., & Grimm, K. J. (2019). Longitudinal changes in attachment orientation over a 59-year period. Journal of Personality and Social Psychology, 116(4), 598–611. https://doi.org/10.1037/pspp0000167
Dugan, K. A., Fraley, R. C., Gillath, O., & Deboeck, P. R. (2022). Changes in global and relationship-specific attachment working models. Journal of Social and Personal Relationships, 39(10), 3015–3043. https://doi.org/10.1177/02654075211051408
Filosa, M., Sharp, C., Gori, A., & Musetti, A. (2026). A comprehensive scoping review of empirical studies on earned secure attachment. Psychological Reports, 129(3), 1807–1832. https://doi.org/10.1177/00332941241277495
Fraley, R. C., Gillath, O., & Deboeck, P. R. (2021). Do life events lead to enduring changes in adult attachment styles? A naturalistic longitudinal investigation. Journal of Personality and Social Psychology, 120(6), 1567–1606. https://doi.org/10.1037/pspi0000326
Fraley, R. C., Heffernan, M. E., Vicary, A. M., & Brumbaugh, C. C. (2011). The Experiences in Close Relationships—Relationship Structures questionnaire: A method for assessing attachment orientations across relationships. Psychological Assessment, 23(3), 615–625. https://doi.org/10.1037/a0022898
Fraley, R. C., Vicary, A. M., Brumbaugh, C. C., & Roisman, G. I. (2011). Patterns of stability in adult attachment: An empirical test of two models of continuity and change. Journal of Personality and Social Psychology, 101(5), 974–992. https://doi.org/10.1037/a0024150
Gillath, O., Karantzas, G. C., Romano, D., & Karantzas, K. M. (2022). Attachment security priming: A meta-analysis. Personality and Social Psychology Review, 26(3), 183–241. https://doi.org/10.1177/10888683211054592
Johnson, L. N., Tambling, R. B., Mennenga, K. D., Ketring, S. A., Oka, M., Anderson, S. R., Huff, S. C., & Miller, R. B. (2016). Examining attachment avoidance and attachment anxiety across eight sessions of couple therapy. Journal of Marital and Family Therapy, 42(2), 195–212. https://doi.org/10.1111/jmft.12136
Levy, K. N., Kivity, Y., Johnson, B. N., & Gooch, C. V. (2018). Adult attachment as a predictor and moderator of psychotherapy outcome: A meta-analysis. Journal of Clinical Psychology, 74(11), 1996–2013. https://doi.org/10.1002/jclp.22685
Pinquart, M., Feussner, C., & Ahnert, L. (2013). Meta-analytic evidence for stability in attachments from infancy to early adulthood. Attachment & Human Development, 15(2), 189–218. https://doi.org/10.1080/14616734.2013.746257
Roisman, G. I., Padrón, E., Sroufe, L. A., & Egeland, B. (2002). Earned-secure attachment status in retrospect and prospect. Child Development, 73(4), 1204–1219. https://doi.org/10.1111/1467-8624.00467
Traut, A. C., Gander, F., Uhlich, M., Weidmann, R., Chopik, W. J., & Grob, A. (2025). Examining change in attachment in romantic couples: The role of relationship characteristics and codevelopment between partners. Journal of Personality and Social Psychology, 129(3), 602–617. https://doi.org/10.1037/pspp0000558
Waters, E., Merrick, S., Treboux, D., Crowell, J., & Albersheim, L. (2000). Attachment security in infancy and early adulthood: A twenty-year longitudinal study. Child Development, 71(3), 684–689. https://doi.org/10.1111/1467-8624.00176
World Health Organization. (2024). Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders. https://www.who.int/publications/i/item/9789240077263