A message goes unanswered for several hours. One person notices the delay and continues with the afternoon. Another reads the last exchange again, searching for a change in tone. Someone else decides that needing a reply was the mistake and resolves to ask for less. These imagined responses show how uncertainty in a relationship can draw us toward another person or encourage us to retreat.

Adult attachment research describes these differences through attachment anxiety and attachment avoidance. These patterns help explain why uncertainty can lead us to seek more reassurance or protect ourselves through distance (Mikulincer & Shaver, 2007).

These patterns help explain what happens when connection feels uncertain. This article focuses on adult close relationships, exploring how expectations about care develop and how greater security can support both intimacy and independent life.

What Insecure Attachment Means

Attachment theory begins with a practical concern: access to someone who can provide comfort and protection when we are distressed. In Bowlby’s account, attachment behavior becomes especially noticeable when a person is frightened, tired, ill, or uncertain of an attachment figure’s availability. A familiar relationship can provide a safe haven in distress and a secure base from which to explore the world (Bowlby, 2008; Mikulincer & Shaver, 2007).

Dependence and independence can therefore support one another. Knowing that someone will respond may make it easier to turn our attention outward, take on a challenge, or spend time apart. The need for comfort is not, by itself, evidence of insecurity. Nor does security require constant closeness or freedom from every anxious feeling.

Insecurity concerns the expectations and protective responses surrounding that need. A person may worry that care will disappear, feel uncomfortable revealing vulnerability, or struggle to trust support even when it is offered. These tendencies become more intelligible when we ask what the person expects to happen if they rely on someone else (Mikulincer & Shaver, 2007).

Adult attachment insecurity is also distinct from the childhood diagnoses reactive attachment disorder and disinhibited social engagement disorder. Those diagnoses should not be used as alternative names for an adult’s fear of rejection or discomfort with intimacy. An attachment style is a research description, not a clinical diagnosis (Zeanah et al., 2016).

Attachment Anxiety and Reassurance

Attachment anxiety involves concern about rejection, abandonment, or another person’s uncertain commitment. When this concern is activated, attention may narrow around possible signs of distance. An ambiguous reply can feel significant, and reassurance may seem urgent. Researchers describe intensified efforts to secure attention and closeness as hyperactivating strategies (Mikulincer & Shaver, 2007).

In the opening example, repeated checking might briefly reduce uncertainty. Yet relief may depend on receiving another confirming signal, leaving the person vulnerable to the next delay. This is an illustration of how reassurance seeking can become repetitive; it does not mean that asking for reassurance is inherently unhealthy. Sometimes a clear request is an appropriate response to a confusing relationship.

Attachment Avoidance and Protective Distance

Attachment avoidance involves discomfort with dependence, closeness, or emotional disclosure. A person may minimize a need, turn away from an offer of support, or rely heavily on handling distress alone. Deactivating strategies reduce the expression or acknowledgment of attachment needs, even when a sense of security has not been achieved (Mikulincer & Shaver, 2007).

Protective distance can look composed from the outside, but it does not establish an absence of social needs. In two laboratory studies, people higher in dismissing avoidance showed positive emotional and self-esteem responses to acceptance or favorable interpersonal feedback. These responses suggest that belonging can still matter, even when someone reports preferring distance; they do not reveal what a particular person secretly wants (Carvallo & Gabriel, 2006).

Enjoying solitude, maintaining privacy, and setting a reasonable boundary do not by themselves establish attachment avoidance. The useful distinction concerns flexibility: whether a person can seek or accept appropriate help when it is needed.

When Anxiety and Avoidance Coexist

The two dimensions can coexist. A person may strongly desire connection while finding dependence threatening, moving toward closeness in one moment and retreating in another. Familiar labels such as anxious-preoccupied, dismissing-avoidant, and fearful-avoidant summarize combinations of anxiety and avoidance; they should not become complete descriptions of a person (Mikulincer & Shaver, 2007).

These terms vary across research traditions. Infant observation, adult attachment interviews, and questionnaires assess different aspects of attachment. Fearful avoidance on a questionnaire should not automatically be equated with infant disorganization or an unresolved interview classification (Mikulincer & Shaver, 2007; Fraley & Hudson, 2017).

How Insecure Attachment Develops

Bowlby’s concept of internal working models describes the expectations people construct about themselves and important others. In ordinary terms, these models concern whether help will be available, how another person is likely to respond, and whether expressing a need seems acceptable. They guide attention and interpretation, often without becoming a fully articulated belief (Mikulincer & Shaver, 2007).

Early caregiving is one important part of this account. Ainsworth’s research connected observations of care at home with infants’ responses during brief laboratory separations and reunions. Fraley and Hudson’s review describes how sensitive responsiveness, attending to a child’s needs and responding appropriately, relates to attachment security. The same review emphasizes that attachment develops within a wider context that includes family conflict, caregiver difficulties, and economic circumstances (Fraley & Hudson, 2017).

The developmental story needs room for uncertainty. Fraley and Hudson describe relatively weak continuity from infancy to adulthood and explain why later relationships can modify earlier expectations. Attachment is shaped by multiple influences. An adult who fears abandonment cannot infer a particular childhood history from that feeling alone, and a difficult beginning does not prescribe one relational future (Fraley & Hudson, 2017).

Expectations can also differ across relationships. Someone may trust a close friend while remaining guarded with a parent or uncertain with a romantic partner. General expectations and expectations about a particular person overlap, but they need not agree completely. This makes a question such as “When, and with whom, does dependence feel difficult?” more informative than a fixed judgment about who we are (Fraley & Hudson, 2017).

Insecure Attachment in Close Relationships

An attachment response unfolds between people. In an illustrative exchange, one partner seeks an immediate explanation for a change in plans. The other feels pressed and becomes less communicative. The first experiences that distance as confirmation that something is wrong and pushes harder. The cycle involves an event, the meaning assigned to it, a protective response, and the other person’s reaction. A change in plans becomes evidence of not mattering; urgent questioning feels like pressure; withdrawal appears to confirm the original fear. Each response can make the other person’s concern feel more convincing.

Johnson describes related pursuit and withdrawal patterns in her clinical account of distressed couples. Overall and Simpson review research showing how partners’ responses can influence the expression of attachment insecurity during difficult interactions. These accounts help explain a possible cycle; they do not establish that anxious people inevitably pair with avoidant partners or that every conflict has attachment at its center (Johnson, 2008; Overall & Simpson, 2015).

Murray, Holmes, and Collins describe a related tension in their risk regulation model: people balance the wish to connect with the wish to avoid rejection. Believing that a partner values us may make dependence feel more tolerable; doubting that regard may encourage self-protection. The model helps interpret relational choices, while remaining an integrative account rather than a single proven causal explanation (Murray et al., 2006).

The broader evidence concerns tendencies across many relationships. Li and Chan’s meta-analysis combined 73 studies and found that anxiety and avoidance were associated with poorer indicators of romantic relationship quality. The patterns differed: avoidance showed stronger negative associations with satisfaction, connectedness, and support, while anxiety was more strongly associated with conflict. These associations cannot predict the outcome of a particular relationship or establish a single causal direction (Li & Chan, 2012).

Insecurity may affect giving care as well as receiving it. In a chapter of the third edition of the Handbook of Attachment, Shaver and colleagues explain how preoccupation with one’s own safety can compete with attention to another person’s distress. Security may leave more room to respond sensitively. This is a theoretical and research-based account of caregiving processes, not a judgment that insecure people lack empathy or love (Shaver et al., 2016).

Understanding a protective response does not remove responsibility for its expression. Fear may help explain an impulse to monitor a partner, demand proof of loyalty, or punish distance. It does not make coercion acceptable. Likewise, withdrawing to regulate emotion differs from using silence to intimidate or control. The behavior and its consequences still matter.

Attachment Expectations and Present Relationships

Earlier expectations and present behavior both deserve attention. An ambiguous delay, a disagreement about contact, and a pattern of dishonesty are different situations. Attachment concepts can help us examine our interpretation without requiring us to dismiss the information a relationship actually provides.

Fraley and Hudson distinguish general expectations from expectations about particular people. A person may usually find support easy to accept yet become guarded with someone who repeatedly disappoints them. Conversely, a dependable relationship may still activate fears that are familiar across several relationships. One stressful exchange cannot settle which process is operating (Fraley & Hudson, 2017).

Useful questions therefore concern behavior over time. Can each person explain a need without being ridiculed? Are commitments followed through, or renegotiated honestly? Is a missed connection acknowledged and repaired? These are reflective questions about the relationship, not a checklist that can diagnose an attachment orientation or certify safety.

An attachment injury is a more specific concern: a perceived failure of care at a critical moment. Makinen and Johnson’s study examined couples working through such injuries, while noting that their attachment patterns before the events were largely unknown. A present wound should be taken seriously without assuming that it created every broader insecurity (Makinen & Johnson, 2006).

Where abuse, stalking, threats, or retaliation are present, greater disclosure may increase exposure to harm. Safety, appropriate support, and workable boundaries take priority over efforts at joint vulnerability or repair.

Closeness Without Losing Independence

Attachment security matters outside moments of conflict. A dependable relationship can provide comfort when life is difficult and support for turning toward activities beyond the relationship. This secure-base function helps explain why connection and independence can develop together (Bowlby, 2008; Feeney, 2007).

Brooke Feeney studied a possibility that can seem surprising: accepting a partner’s need for help may leave that person more able to act independently. Across two studies, people whose partners were more accepting of their needs tended to function more independently and make progress toward personal goals. The research combined what couples reported, observations of their interactions, a laboratory task involving assistance, and a six-month follow-up. The findings support the value of dependable care, without showing that unlimited reassurance creates independence or changes attachment style (Feeney, 2007).

Autonomy in a relationship also has a more precise meaning than separation. It concerns experiencing our actions as chosen and personally endorsed. Someone can freely choose to rely on a partner; someone else can insist on doing everything alone because needing help feels intolerable. The amount of contact or assistance does not, by itself, tell us how much choice a person experiences (Kluwer et al., 2020).

Kluwer and colleagues examined whether feeling close to a partner and having a sense of choice work together when problems arise. Across three studies, people who experienced both tended to report more constructive responses, especially talking through a problem. Findings were less consistent for other responses, and the researchers relied on what participants said about their own behavior. The results suggest that keeping a sense of self can support engagement with relationship difficulties (Kluwer et al., 2020).

Feeney and Collins extend this perspective by distinguishing support during adversity from support for engaging with opportunities. In their theoretical account, relationships may help us explore, learn, and pursue meaningful goals as well as manage distress. This broadens the question from whether a partner can reassure us to whether the relationship leaves room for a fuller life (Feeney & Collins, 2015).

For example, imagine someone preparing to return to school. A partner might listen to their uncertainty, help with an agreed responsibility, and respect their decision about the course. The support can make room for action without taking over the goal. Such everyday possibilities illustrate the secure-base idea; they are not a prescribed test of attachment security.

Moving Toward Greater Attachment Security

How Attachment Patterns Can Change

Attachment theory includes explanations for both continuity and change. Expectations can persist because people interpret experiences through familiar assumptions and participate in interactions that reinforce them. Yet later relationships also provide new information, and general tendencies can coexist with different expectations of particular people (Fraley & Hudson, 2017).

Fraley, Gillath, and Deboeck repeatedly assessed attachment before and after life events in a large longitudinal study. Their broader sample included 4,920 people, with smaller event-specific samples. Many immediate changes diminished over time; some remained at the final assessment, and people differed considerably in their trajectories. The results allow for change without implying that one event affects everyone in the same way (Fraley et al., 2021).

Here, enduring change meant change within the observation period, not permanence across a lifetime. The study was observational and its participants were predominantly White women. More positive interpretations of events were associated with some changes, but that does not establish that positive thinking creates attachment security. A reassuring evening or a difficult week also cannot, by itself, establish whether a broader pattern has changed (Fraley et al., 2021).

Recognizing the Protective Response

A useful starting point in working with emotion is to notice what happens when connection feels uncertain. What occurred? What meaning did we give it? What did we want to do next? Separating the event from its interpretation can make it easier to consider a response while retaining the possibility that the concern is justified.

For example, “I have not heard back” identifies an event, while “I no longer matter” identifies an interpretation. The next step may be a direct question, a period of waiting, or a boundary informed by an established pattern of behavior. These reflective questions apply the preceding ideas; they are not an attachment test or a validated program for changing attachment style.

Building More Responsive Interactions

Predictability helps us know what to expect, while responsiveness concerns the quality of the response. In their review, Itzchakov, Reis, and Weinstein explain perceived partner responsiveness as feeling understood, valued, and cared for. A reply can arrive promptly and still leave someone feeling dismissed; an honest request to talk later can acknowledge the need even when immediate availability is impossible (Itzchakov et al., 2022).

The authors describe listening with attention, an effort to understand, and a caring intention as one possible contributor to that experience. In practice, someone might ask what felt most difficult, check whether they understood, and respond to the concern actually expressed. Feeling understood need not require agreement with every interpretation. The review proposes how listening and responsiveness may connect; it does not establish listening as a treatment for insecure attachment (Itzchakov et al., 2022).

Where both people can participate safely, small agreements can make interactions easier to understand. A person who needs a pause might explain the need and agree on a time to return. A person seeking connection might describe the uncertainty and ask for something specific. Following through, or acknowledging when an agreement cannot be kept, supplies information that an abstract promise cannot provide.

Partner support has possibilities and limits. Overall and Simpson review research in which responses sensitive to attachment concerns were associated with more constructive interactions. A later review asked a narrower question: does support make the link between attachment insecurity and negative emotions weaker? That pattern appeared in 9 percent of 151 analyses across 31 studies. The question concerns whether support changes that particular link, rather than whether support helps people feel better generally (Overall & Simpson, 2015; Mosannenzadeh et al., 2025).

The same review reported broader associations between receiving more support and fewer negative emotions. The 9 percent figure therefore does not mean that only 9 percent of people benefited or that support rarely helps. The review counted patterns across analyses instead of combining them into one estimate of the size of an effect. Its search also ended in September 2020. These limits leave uncertainty about the best support for a particular person (Mosannenzadeh et al., 2025).

Responsiveness belongs within a relationship that permits limits and reciprocity. People need room for work, illness, other relationships, and responsibilities. One partner cannot be assigned the task of preventing every anxious feeling or making dependence comfortable. The aim is a workable pattern of care in which both people can express needs and retain meaningful choice.

Seeking Professional Support

Professional support may help when recurring patterns cause distress or attempts at repair repeatedly reach an impasse. Emotionally focused couple therapy is one attachment-informed approach. A meta-analysis of 20 studies involving 332 couples supports improvements in couple functioning, although differences in study designs and the small evidence base constrain generalization (Spengler et al., 2024).

Relationship improvement should be distinguished from demonstrated attachment change. In Makinen and Johnson’s attachment-injury study, resolved couples improved in satisfaction and forgiveness, but significant group differences in general attachment anxiety and avoidance were not established. Repair can be meaningful without proving that a broad orientation has changed. Accountability remains necessary, forgiveness and reconciliation remain separate choices, and safety and freedom to participate should guide whether joint work is appropriate (Makinen & Johnson, 2006).

A Few Words by Psychology Fanatic

We can return to the unanswered message with a wider set of questions. Perhaps an old expectation has made uncertainty especially painful. Perhaps the relationship has become unreliable. Perhaps both are contributing. Attachment offers a way to explore these possibilities without turning an anxious afternoon into a permanent identity.

Greater security can involve becoming more able to ask for care, receive it, and judge when it is available. It can also involve setting a limit when a relationship cannot provide reasonable safety. Being connected need not mean giving up a life of our own. Sometimes the value of dependable care becomes visible in what it helps us do next: return to a task, pursue a goal, or spend time apart without treating the distance as a loss of love.

Associated Concepts

• Attachment Theory: The broader account of proximity, protection, and expectations that gives attachment insecurity its meaning.
• Fear of Abandonment: A related concern that is particularly relevant to attachment anxiety, while not explaining every form of insecurity.
• Attachment Injury: A specific relational wound that can undermine trust and complicate repair.
• Emotional Regulation: Ways of working with distress that may affect how attachment concerns are expressed.
• Risk Regulation Model: A related perspective on balancing the benefits of connection with the risk of rejection.
• Relationship Security: The experience of trust and dependability within an ongoing relationship.
• Intimacy: Emotional closeness that makes both connection and vulnerability relevant.

References

Bowlby, John (2008). A Secure Base: Parent-Child Attachment and Healthy Human Development. Basic Books. Ebook edition. ISBN: 9780786722570.
(Return to Main Text)

Carvallo, Mauricio; Gabriel, Shira (2006). No Man Is an Island: The Need to Belong and Dismissing Avoidant Attachment Style. Personality and Social Psychology Bulletin, 32(5), 697–709. DOI: 10.1177/0146167205285451.
(Return to Main Text)

Feeney, Brooke C. (2007). The Dependency Paradox in Close Relationships: Accepting Dependence Promotes Independence. Journal of Personality and Social Psychology, 92(2), 268–285. DOI: 10.1037/0022-3514.92.2.268.
(Return to Main Text)

Feeney, Brooke C.; Collins, Nancy L. (2015). A New Look at Social Support: A Theoretical Perspective on Thriving through Relationships. Personality and Social Psychology Review, 19(2), 113–147. DOI: 10.1177/1088868314544222.
(Return to Main Text)

Fraley, R. Chris; Gillath, Omri; Deboeck, Pascal 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. DOI: 10.1037/pspi0000326.
(Return to Main Text)

Fraley, R. Chris; Hudson, Nathan W. (2017). The Development of Attachment Styles. In Jule Specht (Ed.), Personality Development Across the Lifespan (pp. 275–292). Academic Press. DOI: 10.1016/B978-0-12-804674-6.00017-X.
(Return to Main Text)

Itzchakov, Guy; Reis, Harry T.; Weinstein, Netta (2022). How to Foster Perceived Partner Responsiveness: High-Quality Listening Is Key. Social and Personality Psychology Compass, 16(1), e12648. DOI: 10.1111/spc3.12648.
(Return to Main Text)

Johnson, Sue (2008). Hold Me Tight: Seven Conversations for a Lifetime of Love. Little, Brown and Company. First ebook edition. ISBN: 9780316031998.
(Return to Main Text)

Kluwer, Esther S.; Karremans, Johan C.; Riedijk, Larisa; Knee, C. Raymond (2020). Autonomy in Relatedness: How Need Fulfillment Interacts in Close Relationships. Personality and Social Psychology Bulletin, 46(4), 603–616. DOI: 10.1177/0146167219867964.
(Return to Main Text)

Li, Tianyuan; Chan, Darius 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. DOI: 10.1002/ejsp.1842.
(Return to Main Text)

Makinen, Judy A.; Johnson, Susan M. (2006). Resolving Attachment Injuries in Couples Using Emotionally Focused Therapy: Steps Toward Forgiveness and Reconciliation. Journal of Consulting and Clinical Psychology, 74(6), 1055–1064. DOI: 10.1037/0022-006X.74.6.1055.
(Return to Main Text)

Mikulincer, Mario; Shaver, Phillip R. (2007). Attachment in Adulthood: Structure, Dynamics, and Change. The Guilford Press. ISBN: 9781593854577.
(Return to Main Text)

Mosannenzadeh, Farnaz; Kouri, Georgia; Luijten, Maartje; Schoebi, Dominik; Karremans, Johan C. (2025). Does Romantic Partner Support Buffer the Effect of Attachment Insecurities on Negative Emotions? A Systematic Review. Collabra: Psychology, 11(1), 130079. DOI: 10.1525/collabra.130079.
(Return to Main Text)

Murray, Sandra L.; Holmes, John G.; Collins, Nancy L. (2006). Optimizing Assurance: The Risk Regulation System in Relationships. Psychological Bulletin, 132(5), 641–666. DOI: 10.1037/0033-2909.132.5.641.
(Return to Main Text)

Overall, Nickola C.; Simpson, Jeffry A. (2015). Attachment and Dyadic Regulation Processes. Current Opinion in Psychology, 1, 61–66. DOI: 10.1016/j.copsyc.2014.11.008.
(Return to Main Text)

Shaver, Phillip R.; Mikulincer, Mario; Gross, Jacquelyn T.; Stern, Jessica A.; Cassidy, Jude (2016). A Lifespan Perspective on Attachment and Care for Others: Empathy, Altruism, and Prosocial Behavior. In Jude Cassidy and Phillip R. Shaver (Eds.), Handbook of Attachment: Theory, Research, and Clinical Applications (3rd ed., pp. 878–916). The Guilford Press. Publisher chapter excerpt.
(Return to Main Text)

Spengler, Paul M.; Lee, Nicholas A.; Wiebe, Stephanie A.; Wittenborn, Andrea K. (2024). A Comprehensive Meta-Analysis on the Efficacy of Emotionally Focused Couple Therapy. Couple and Family Psychology: Research and Practice, 13(2), 81–99. DOI: 10.1037/cfp0000233.
(Return to Main Text)

Zeanah, Charles H.; Chesher, Tessa; Boris, Neil W.; AACAP Committee on Quality Issues (2016). Practice Parameter for the Assessment and Treatment of Children and Adolescents with Reactive Attachment Disorder and Disinhibited Social Engagement Disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 55(11), 990–1003. DOI: 10.1016/j.jaac.2016.08.004.
(Return to Main Text)