Fearful-avoidant attachment, also termed disorganized attachment in the developmental literature, represents the most complex and internally contradictory attachment pattern. First identified by Main and Solomon (1986) as the "D" classification in the Strange Situation, disorganized attachment was observed in infants who displayed contradictory, disoriented, or inexplicable behaviors in the presence of their caregiver — simultaneously approaching and avoiding, freezing mid-movement, displaying apprehension toward the attachment figure, or engaging in stereotyped or dissociative behaviors. Unlike the organized insecure patterns (anxious and avoidant), which represent coherent if costly strategies for managing an imperfect caregiving environment, disorganized attachment reflects a fundamental breakdown in the attachment strategy itself. Main and Hesse (1990) described this as "fright without solution" — a paradox in which the attachment figure who should serve as the source of safety is simultaneously the source of alarm.
Bartholomew and Horowitz (1991) extended the concept to adult relationships under the label fearful-avoidant, placing it in the quadrant defined by negative models of both self and others. Unlike dismissive-avoidant individuals, who maintain a positive self-model through self-reliance, fearful-avoidant individuals hold the view that they are unworthy of love (negative self-model) while simultaneously believing that others are untrustworthy or dangerous (negative other-model). This dual negativity creates a characteristic oscillating approach-withdrawal pattern in close relationships: the individual craves intimacy to soothe their negative self-regard but fears intimacy because they expect rejection, exploitation, or harm. Brennan, Clark, and Shaver (1998) documented that fearful-avoidant attachment corresponds dimensionally to high scores on both attachment anxiety and attachment avoidance, a combination that produces the most internally conflicted relational experience of the four attachment orientations.
The developmental origins of disorganized attachment are among the most extensively researched in the field. Main and Hesse (1990) proposed that disorganized attachment develops primarily when the caregiver is frightening or frightened — engaging in behaviors that are intrinsically alarming to the infant (hostile, threatening, or dissociative behaviors) or displaying unresolved fear or trauma that manifests as sudden lapses in caregiving coherence. Subsequent research has consistently supported this link: meta-analyses by van IJzendoorn, Schuengel, and Bakermans-Kranenburg (1999) found that parental frightening behavior and parental unresolved loss or trauma are the strongest predictors of infant disorganized attachment. It is important to note, however, that disorganized attachment is not synonymous with abuse — it can develop in response to subtle forms of caregiver dysregulation, and many caregivers of disorganized infants are not abusive but are themselves struggling with unresolved traumatic experiences.
Prevalence estimates for fearful-avoidant or disorganized attachment vary considerably depending on the population studied and the assessment method used. In low-risk community samples, estimates typically range from 3 to 15 percent of adults (Bartholomew & Horowitz, 1991; Mickelson, Kessler, & Shaver, 1997), while in high-risk or clinical samples the prevalence may be substantially higher. This variability has generated debate about whether fearful-avoidant attachment is being over-pathologized in popular and clinical literature. Some researchers (Roisman et al., 2007) have cautioned that the disorganized classification may function as a catch-all category for attachment behaviors that do not fit neatly into the organized categories, and that its boundaries may be less clearly defined than those of secure, anxious, or avoidant patterns. Dimensional approaches (Fraley, Waller, & Brennan, 2000) would frame fearful-avoidant attachment not as a discrete type but as the high-anxiety, high-avoidance region of continuous attachment space, a framing that may reduce the tendency toward essentialist labeling.
Common Patterns
Researchers have identified these characteristic patterns in individuals with fearful-avoidant attachment:
- Oscillation between seeking closeness and withdrawing when closeness is achieved
- Simultaneous use of both anxious (hyperactivating) and avoidant (deactivating) strategies
- Difficulty maintaining a consistent emotional stance within relationships
- Heightened emotional reactivity combined with difficulty processing intense feelings
- Tendency to experience relationships as both deeply desired and fundamentally threatening
- Unpredictable responses to intimacy — warmth and engagement may shift quickly to withdrawal or distancing
- Difficulty trusting others while also struggling with self-worth in relational contexts
Research Foundation
The research foundation for fearful-avoidant or disorganized attachment is distinctive in its origins in developmental psychopathology and its close links to the literatures on trauma, dissociation, and emotional dysregulation. The disorganized infant classification (Main & Solomon, 1986) was developed specifically to capture the subset of Strange Situation behaviors that could not be classified within Ainsworth's original three-category system, and it has since generated a substantial body of research connecting early attachment disorganization to a range of developmental and relational outcomes. However, the translation from infant disorganized attachment to adult fearful-avoidant attachment involves conceptual and methodological complexities that warrant careful examination.
Main & Solomon (1986)
Identified the disorganized/disoriented ("D") classification in infant attachment, describing a pattern of contradictory, incomplete, or anomalous behaviors in the Strange Situation that could not be assigned to Ainsworth's A, B, or C categories. Characteristic behaviors included simultaneous approach and avoidance movements, freezing or stilling, apprehensive or dazed expressions, and disoriented wandering.
Methodology: Systematic reanalysis of videotaped Strange Situation episodes from multiple research samples, focusing on infants whose behavior had been difficult to classify. Main and Solomon developed a detailed coding system for identifying disorganized behaviors and assigned a D classification as an overlay on the "best-fitting" organized classification (A, B, or C).
Honest Assessment: The D classification is applied as an overlay rather than a standalone category, meaning every disorganized infant also receives a "forced" organized classification. This dual-classification approach raises questions about whether disorganization represents a qualitatively distinct phenomenon or a breakdown that can occur within any organized strategy. Inter-rater reliability for the D classification, while adequate, is generally lower than for the organized categories, and the threshold for "disorganization" can be difficult to apply consistently, particularly for mild or brief instances of disoriented behavior.
Main & Hesse (1990)
Proposed that infant disorganized attachment is primarily caused by caregiver behavior that is frightening or frightened. When the attachment figure — who is supposed to be the infant's source of safety — simultaneously becomes a source of alarm, the infant faces an irresolvable paradox: the biologically driven impulse to approach the caregiver for comfort conflicts with the equally powerful impulse to flee from a source of fear. This "fright without solution" produces the contradictory and disoriented behaviors characteristic of the D classification.
Methodology: Theoretical framework supported by analysis of parent-infant interaction data and Adult Attachment Interview classifications of parents. Main and Hesse linked infant disorganized attachment to parental "unresolved/disorganized" AAI classifications, which indicate unresolved mourning or trauma in the parent.
Honest Assessment: While the "fright without solution" model is theoretically compelling and has received substantial empirical support, it may not account for all instances of disorganized attachment. Some studies have identified disorganized attachment in samples where parental frightening behavior is minimal, suggesting that other pathways (such as extreme parental insensitivity, neurobiological vulnerability in the infant, or chronic unpredictability) may also contribute. Additionally, the intergenerational transmission mechanism — how parental unresolved trauma translates into specific frightening behaviors — is not fully understood, and the mediating variables remain an active area of research.
Bartholomew & Horowitz (1991)
Described the adult fearful-avoidant style as characterized by negative models of both self and others, producing a desire for closeness that is inhibited by fear of rejection and distrust. Fearful individuals reported the highest levels of interpersonal distress and the lowest self-confidence of the four attachment groups.
Methodology: Multi-method assessment including self-report, semi-structured interviews, and peer ratings in a university sample.
Honest Assessment: The conceptual link between infant disorganized attachment and adult fearful-avoidant attachment is assumed rather than empirically demonstrated in this study. Longitudinal research connecting infant D classification to adult fearful-avoidant self-report has yielded mixed results, raising questions about whether these constructs from different measurement traditions are truly capturing the same underlying phenomenon. Some researchers argue that adult fearful-avoidant attachment, as measured by self-report, may be a broader category than infant disorganized attachment, encompassing individuals who developed conflicted attachment orientations through various pathways, not all of which involve the "fright without solution" dynamic.
Mikulincer & Shaver (2007)
Documented that fearful-avoidant individuals exhibit the most disorganized and inconsistent pattern of emotion regulation among the four attachment orientations. They alternate between hyperactivating strategies (intensified proximity-seeking and emotional escalation, characteristic of anxious attachment) and deactivating strategies (emotional suppression and withdrawal, characteristic of avoidant attachment), often within the same relational episode. This oscillation produces a chaotic relational experience for both the individual and their partners.
Methodology: Integrative review and theoretical synthesis drawing on experimental, correlational, and clinical research.
Honest Assessment: The characterization of fearful-avoidant attachment as involving simultaneous hyperactivation and deactivation is theoretically coherent but empirically difficult to measure, as most self-report instruments assess anxiety and avoidance as relatively stable dimensions rather than capturing the moment-to-moment oscillation that characterizes this pattern. Experience sampling methods and physiological monitoring may be needed to fully capture the dynamic nature of fearful-avoidant regulation, and such studies remain relatively rare.
Lyons-Ruth & Jacobvitz (2008)
Provided a comprehensive review of the developmental research on disorganized attachment, documenting its associations with later behavioral problems, dissociative symptoms, difficulty with emotion regulation, and vulnerability to psychopathology. The review also documented protective factors that can buffer against these outcomes, including the availability of alternative secure attachment figures and therapeutic intervention.
Methodology: Review chapter synthesizing longitudinal and cross-sectional studies of disorganized attachment from infancy through adulthood.
Honest Assessment: While the association between infant disorganized attachment and later difficulties is well-documented, the effect sizes are typically moderate, meaning that many individuals classified as disorganized in infancy do not develop significant psychopathology. This raises concerns about over-pathologizing the disorganized classification and about the potential for confirmation bias in clinical assessments. The relationship between early disorganization and adult outcomes is mediated by numerous intervening factors — including subsequent caregiving quality, peer relationships, life events, and individual temperament — that complicate deterministic predictions based on infant classification alone.
In Relationships
In romantic partnerships, individuals with fearful-avoidant attachment present what is arguably the most complex and contradictory relational pattern. The defining characteristic is an oscillation between approach and withdrawal that reflects the simultaneous activation of the attachment system ("I need closeness") and the defensive system ("Closeness is dangerous"). Mikulincer and Shaver (2007) describe this as the deployment of both hyperactivating and deactivating strategies, often in rapid succession: the individual may intensely pursue closeness, become overwhelmed by the vulnerability that closeness entails, abruptly withdraw or become hostile, experience distress at the resulting distance, and then pursue closeness again. This cycle can be disorienting for both the individual and their partner, creating a relational atmosphere characterized by unpredictability and emotional intensity.
The "fright without solution" dynamic that Main and Hesse (1990) identified in infant-caregiver relationships often replays in adult romantic partnerships of fearful-avoidant individuals. The romantic partner, like the original caregiver, represents both the desired source of comfort and a perceived source of danger. Levine and Heller (2010) observe that fearful-avoidant individuals may experience rapid shifts between idealization and devaluation of their partners, driven by the alternating activation of positive and negative internal working models. In moments when the attachment system is ascendant, the partner is experienced as essential and longed-for; in moments when the defensive system dominates, the same partner is experienced as threatening, untrustworthy, or suffocating. Partners frequently describe the experience of relating to a fearful-avoidant individual as "walking on eggshells" or as feeling that they cannot find a stable equilibrium.
Importantly, the relational difficulties associated with fearful-avoidant attachment are neither fixed nor inevitable. Johnson (2008) documents that Emotionally Focused Therapy can be effective in helping fearful-avoidant individuals identify the competing attachment and defensive impulses driving their oscillating behavior, and in developing a more integrated relational stance. Tatkin (2012) emphasizes the importance of establishing safety and predictability as foundational therapeutic and relational goals for fearful-avoidant individuals, noting that their nervous systems are often highly sensitized to threat cues and require consistent experiences of non-threatening closeness before the defensive system can begin to relax. The pathway toward more secure functioning for fearful-avoidant individuals is typically longer and more complex than for individuals with organized insecure patterns, but research on earned security (Roisman et al., 2002) demonstrates that meaningful change is possible.
Communication Patterns
Communication patterns in fearful-avoidant attachment are characterized by inconsistency, rapid shifts in tone and content, and difficulty maintaining a coherent relational stance during emotionally charged conversations. Mikulincer and Shaver (2007) describe a pattern in which fearful-avoidant individuals may begin a relational conversation with openness and vulnerability, become triggered by the intimacy of the exchange, shift abruptly to hostility or withdrawal, and then feel guilty about the withdrawal and attempt to re-engage. This pattern produces a communication style that partners experience as unpredictable and difficult to follow, as the emotional ground shifts rapidly beneath the conversation. In the Adult Attachment Interview tradition, adults with unresolved/disorganized classifications produce narratives characterized by lapses in monitoring of reasoning or discourse during discussion of loss or trauma — moments where the narrative becomes momentarily incoherent, suggesting the intrusion of unintegrated traumatic material (Main & Hesse, 1990).
Levine and Heller (2010) note that fearful-avoidant individuals often lack a consistent communication template for relational conversations because their internal working models of both self and other are negative and unstable. Without a stable expectation of either responsiveness (as in secure attachment) or rejection (as in organized insecure attachment), the fearful-avoidant individual approaches each relational conversation without a clear strategy, improvising between competing impulses to connect and to protect. Therapeutically, Johnson (2008) and Tatkin (2012) both emphasize that developing consistent communication protocols — predictable, brief, and structured ways of checking in, expressing needs, and managing disagreements — can provide the external scaffolding that the fearful-avoidant individual's internal working models do not yet provide, gradually building the experience of safe communication that was missing in their developmental history.
Under Stress
The stress response in fearful-avoidant attachment is distinctively disorganized and dysregulated, reflecting the absence of a coherent strategy for managing attachment-related distress. Mikulincer and Shaver (2007) document that fearful-avoidant individuals under stress may simultaneously experience the anxious individual's desperate need for comfort and the avoidant individual's impulse to withdraw, without the capacity to commit fully to either strategy. This produces a chaotic internal experience in which the individual may seek help and then reject it, express vulnerability and then become hostile, or vacillate between clinging and fleeing. Physiologically, disorganized attachment has been associated with elevated and poorly regulated cortisol responses (Spangler & Grossmann, 1993; Hertsgaard, Gunnar, Erickson, & Nachmias, 1995), suggesting that the stress regulation system itself reflects the absence of a coherent attachment strategy.
Main and Hesse (1990) theorized that the disorganized stress response originates in the fundamental paradox of having an attachment figure who is simultaneously the source of comfort and the source of fear. When stressed, the biological imperative to seek proximity to an attachment figure is activated, but so is the learned expectation that proximity will lead to further distress. This creates what Lyons-Ruth and Jacobvitz (2008) describe as a "collapsed" behavioral strategy — the individual literally does not know whether to approach or flee, and the resulting behavior appears fragmented, contradictory, or frozen. In adult relationships, this pattern may manifest during conflicts as abrupt emotional shutdowns, dissociative episodes, or sudden shifts from engagement to rage. Tatkin (2012) emphasizes that partners of fearful-avoidant individuals can support stress recovery by providing calm, predictable, non-retaliatory presence during these episodes, allowing the individual's arousal to subside without the additional threat of partner withdrawal or escalation.
Path Toward Growth
Growth for individuals with fearful-avoidant attachment is often the most challenging and the most transformative of the four attachment orientations, as it requires the integration of deeply conflicting internal working models and the development of a coherent attachment strategy where none previously existed. Mikulincer and Shaver (2007) describe the core developmental task as building a stable sense of both self-worth and other-trust — addressing the negative self-model ("I am unworthy of love") and the negative other-model ("Others will hurt me") that together produce the fearful-avoidant stance. Because these negative models are often rooted in early experiences of frightening or traumatically inconsistent caregiving, growth frequently requires some form of therapeutic support, and the process may be longer and more intensive than for individuals with organized insecure attachment.
Johnson (2008) emphasizes that therapy with fearful-avoidant individuals often involves processing unresolved traumatic experiences that are driving the disorganized attachment pattern. In the Emotionally Focused Therapy framework, this means helping the individual access the fear, grief, and rage associated with early experiences of "fright without solution" and integrate these emotions into a coherent narrative rather than allowing them to intrude as fragmented, dysregulated reactions in current relationships. Main and Hesse (1990) identified the resolution of unresolved loss and trauma as a key mechanism through which caregivers can shift from disorganized to organized attachment representations, and this same process applies to the individual's own growth: developing a coherent account of one's attachment history, complete with acknowledged pain and integrated understanding, is associated with movement toward earned security.
Practically, growth for fearful-avoidant individuals involves developing tolerance for the "middle ground" of relational experience — learning to remain present during moments of intimacy without being overwhelmed by fear, and to tolerate temporary distance without interpreting it as abandonment. Tatkin (2012) recommends structured relational practices that build predictability and safety: consistent daily rituals of connection, clear agreements about how conflicts will be managed, and explicit reassurances that the relationship can survive disagreement. Levine and Heller (2010) note that fearful-avoidant individuals benefit from developing what they call "effective dependency" — the ability to identify what they need from a partner and ask for it directly, rather than oscillating between desperate pursuit and defensive withdrawal. The earned security research (Roisman et al., 2002) provides the most encouraging finding for fearful-avoidant individuals: that early disorganization does not determine lifelong relational capacity, and that coherent, reflective integration of difficult experiences is possible and associated with secure adult functioning.
Frequently Asked Questions
Is fearful-avoidant attachment the same as having a personality disorder?
No, though there is documented overlap between disorganized attachment and certain personality disorder presentations, particularly borderline personality disorder. Research has found elevated rates of disorganized attachment among individuals diagnosed with BPD (Levy, 2005), and both phenomena involve emotional dysregulation, unstable relational patterns, and oscillation between idealization and devaluation. However, attachment orientation and personality disorder are distinct constructs: many fearful-avoidant individuals do not meet criteria for any personality disorder, and personality disorders can develop in individuals with any attachment orientation. The association between disorganized attachment and BPD likely reflects shared developmental risk factors — particularly early relational trauma and inconsistent caregiving — rather than an identity between the two constructs. Mikulincer and Shaver (2007) caution against equating attachment classifications with diagnostic categories, noting that attachment patterns exist on a continuum and do not inherently constitute psychopathology.
How common is fearful-avoidant attachment, and are prevalence estimates reliable?
Prevalence estimates for fearful-avoidant or disorganized attachment vary considerably depending on the population and measurement method. In low-risk community samples using self-report measures, approximately 3 to 5 percent of adults identify with the fearful-avoidant description (Bartholomew & Horowitz, 1991). In infant samples using the Strange Situation, disorganized classifications range from approximately 15 percent in low-risk samples to over 80 percent in maltreated samples (van IJzendoorn et al., 1999). This wide variation reflects both genuine population differences and measurement challenges. Some researchers (Roisman et al., 2007) have noted that the disorganized classification may be applied inconsistently, as the threshold for "disorganization" involves judgment calls about the severity and duration of contradictory behaviors. Dimensional approaches (Fraley, Waller, & Brennan, 2000) may provide more reliable assessment by measuring the degree of co-occurring anxiety and avoidance rather than assigning categorical labels.
Can a person with fearful-avoidant attachment develop a secure attachment style?
Yes. The research on earned security demonstrates that individuals can develop secure attachment representations regardless of early attachment history, including disorganized beginnings. Roisman et al. (2002) found that earned-secure adults — those who reported insecure early experiences but produced coherent, reflective narratives about those experiences on the Adult Attachment Interview — functioned comparably to continuously secure adults in their romantic relationships and parenting. The pathway to security for fearful-avoidant individuals typically involves some combination of therapeutic support (particularly approaches that address unresolved trauma, such as EFT or trauma-focused therapies), sustained relationships with securely attached partners, and the gradual development of a coherent narrative about one's attachment history. Johnson (2008) emphasizes that the process is neither quick nor linear — fearful-avoidant individuals may cycle through periods of progress and regression — but that meaningful and lasting change in attachment orientation is well-documented in the research literature.