Most people who are stuck in a coercive relationship are not stuck because they lack information. They know the relationship is bad. They have probably Googled "narcissistic abuse," read the Reddit threads, recognized the patterns with uncomfortable precision. They have a list. They have made the exit plan. And then something happened — a kind text, a good weekend, a moment where the person they fell in love with seemed to reappear — and they stayed.
If this is you, I want to be direct about something: you are not the problem. The staying is not a character flaw, a sign of low self-esteem, or evidence that some part of you secretly wants to be treated this way. It is a predictable outcome of a specific set of conditions that your nervous system was not designed to resist. Understanding what those conditions are does not make the situation easier to leave — but it does stop you from spending your energy hating yourself for something your brain was set up to do.
— Section 01What is trauma bonding, actually?
The term "trauma bonding" was introduced by Patrick Carnes in his 1997 work on betrayal trauma and coercive relationships.1 It describes a strong emotional attachment that forms between a person and someone who intermittently harms them. The word "trauma" in trauma bonding does not mean the bond was formed because of trauma in your past — it means the bond itself is traumatic in origin. It forms under specific conditions: danger, intermittent reward, and the absence of alternatives.
A powerful psychological attachment that develops toward a person who alternately abuses and rewards. The bond is not a symptom of the target's weakness or pathology. It is a predictable neurological response to intermittent reinforcement under conditions of perceived threat — the same mechanism that drives gambling addiction and operant conditioning in animal research.
The mechanism at the core of trauma bonding is intermittent reinforcement — a schedule of reward that is unpredictable. B.F. Skinner's mid-century operant conditioning research established that variable-ratio reinforcement (reward delivered unpredictably, not consistently) produces the most persistent and extinction-resistant behavior of any reward schedule.2 His pigeons would peck a lever thousands of times without reward when that lever had previously, unpredictably, delivered food. This is not a pigeon flaw. It is how all mammalian brains are built.
When a partner is sometimes warm, available, loving, and wonderful — and sometimes cold, punishing, frightening, or dismissive — your brain cannot habituate. It stays alert. It keeps scanning for the signal that the good version is about to appear. It keeps working, keeps trying, keeps hoping. The good moments do not just feel good. They feel intensely good, because they land against a backdrop of uncertainty and relief. This is not love. But neurologically, it produces a state that is very difficult to distinguish from it.
— Section 02Why does leaving feel impossible even when you want to go?
Research on coercive control — the systematic pattern of behavior used to dominate an intimate partner — helps explain why the logical knowledge that a relationship is harmful does not translate into exit. Dutton and Goodman's 2005 framework on coercive control identified that these relationships work not primarily through physical violence (which is often absent or intermittent) but through a structured erosion of a person's autonomy, social support, reality perception, and sense of self.3
By the time someone is asking "why can't I just leave," several things have typically already happened. Their social network has been narrowed, sometimes to almost nothing, so leaving means going somewhere that does not feel real anymore. Their sense of their own perceptions has been systematically undermined — gaslighting does not just make you doubt specific events; over time, it makes you doubt your ability to read situations accurately. And the intermittent cycles of warmth have produced an attachment state that is, neurologically, very close to what Burkett and Young (2012) found when they examined the brain activity of people in obsessive love states: hyperactivation of the dopamine reward system, deactivation of prefrontal areas associated with rational evaluation, and elevated opioid system engagement consistent with what happens during physical withdrawal from addictive substances.4
That last part is worth sitting with. The craving you feel when you are separated from a person who harms you — the intrusive thoughts, the physical restlessness, the pull to make contact even when you know better — is not love misfiring. It is withdrawal. Your brain has become physiologically dependent on the intermittent reward cycle the relationship provides. Leaving activates something that functions like substance withdrawal. Most people are not prepared for that, and when it hits, they interpret it as evidence that they made a mistake by leaving.
They have not made a mistake. They are in withdrawal.
— Section 03What the abuse has to do with why you stay
There is a specific mechanism inside coercive relationships that is rarely named clearly: the abuse itself produces the emotional state that keeps you attached to the abuser.
When someone criticizes you constantly, your sense of yourself becomes increasingly dependent on their opinion of you. When someone isolates you from friends and family, the abuser becomes your primary source of emotional regulation. When someone tells you, repeatedly and convincingly, that no one else would want you or understand you or put up with you — you believe them, not because you are gullible, but because prolonged exposure to a confident, consistent message erodes even firmly held beliefs about the self. This is what decades of research on coercive influence, from Robert Cialdini's work on persuasion to the literature on high-control groups, consistently documents: the conditions that produce compliance are not special or exotic. They are applied systematically and incrementally, and they work on most people under sustained exposure.
The result is that you leave the relationship having been trained to believe, at a deep level, that the relationship's problems were your fault. That you were difficult. That you asked for too much, gave too little, were too sensitive, too needy, too broken to deserve better. The shame that follows you out of the relationship is not a neutral self-assessment. It is a residue of the abuse itself — a conclusion that was installed in you through repetition, often by someone who had a stake in you believing it.
If you are working with a therapist on narcissistic abuse recovery, this is one of the central clinical targets: not just processing what happened, but accurately attributing the source of the beliefs you are now carrying about yourself.
— Section 04Why is the shame about staying its own trap?
Here is something that tends to land hard in the consulting room: the shame you feel about not having left sooner is one of the mechanisms that keeps you from leaving now.
Shame — as distinguished from guilt — operates by narrowing your sense of what you deserve. Guilt says "I did something bad." Shame says "I am something bad." When you feel ashamed that you stayed, you are not just feeling bad about a choice. You are reinforcing a belief that you are someone who makes bad choices, who cannot be trusted to manage their own life, who is fundamentally deficient in some way. That belief, once installed, makes it harder to act on your own behalf. It feeds the narrative the abusive relationship was selling you. And it makes you less likely to reach out for support, because shame thrives in secrecy and tells you that anyone who knew the full picture would judge you.
The research on shame and interpersonal trauma is unambiguous on this point. Shame — not guilt, not sadness, not even fear — is the emotional state most associated with continued entrapment in abusive relationships. Brown's work on shame resilience and vulnerability identifies shame as fundamentally incompatible with the kind of self-trust that exit requires.5 You cannot make a clear decision about your life from inside an internal state that tells you your judgment cannot be trusted.
The clinical goal, then, is not to get you to stop feeling the pull toward the relationship. That pull is neurological and it will not respond to reasoning or willpower alone. The goal is to interrupt the shame cycle that compounds the pull — and to help you begin to see the staying as information about what was done to you, not as evidence of who you are.
- You feel an intense pull to contact the person even when you know it will not go well.
- You find yourself defending their behavior to others — or to yourself — despite knowing how it affects you.
- The relationship feels addictive: you want out, but the wanting-out doesn't translate into action.
- You feel worse about yourself now than you did before the relationship started.
- Good days in the relationship feel exceptionally good — almost worth the bad ones.
- When separated, you feel withdrawal symptoms: physical restlessness, intrusive thoughts, an urge to check their social media or reach out.
- You believe, at some level, that you are the one who caused or escalated most of the problems.
— Section 05What does breaking free actually require?
I want to be honest here in a way that some other writing on this subject is not: leaving a trauma bond is not primarily a decision. It is a process, and it is usually harder than the person going through it was told it would be.
Physical separation is necessary but not sufficient. You can be physically out of a relationship while the trauma bond remains fully intact — because the bond lives in your nervous system, not in proximity to the person. Many people leave, feel the withdrawal, interpret it as love, return, and then feel ashamed of having returned. The cycle reinforces itself. This is not a failure of willpower. It is what happens when a neurological dependency is treated as if it were only a logical problem.
What tends to work is a combination of things. Structured no-contact or limited contact — reducing the intermittent reinforcement that sustains the bond's intensity. Trauma-informed clinical support that addresses both the attachment wound and the distorted belief system that the relationship produced. Social reconnection to counteract the isolation that coercive relationships require to maintain control. And time — more time than most people are told, because the nervous system genuinely needs it to recalibrate.
Eye Movement Desensitization and Reprocessing (EMDR) has the most robust research base for treating the traumatic memory components that keep survivors anchored to the past relationship, including the intrusive reexperiencing that makes physical separation feel unbearable.6 Somatic approaches that address the body's role in trauma storage are increasingly supported in the literature as well, particularly for survivors whose experience of the relationship was more physiological than cognitive — people who "knew something was wrong" in their body long before they had words for it.
What does not work, in my clinical experience, is more information. If you are reading this, you already have the information. You knew the relationship was harmful before you clicked the link. The knowledge was never the gap. The gap is between knowing and being able to act consistently on what you know — and that gap lives in the nervous system, not the intellect. It closes through treatment, not through reading one more article that confirms what you already understand.
- Trauma bonding is neurological, not psychological weakness. Intermittent reinforcement produces the most persistent attachment of any reward schedule — this is established behavioral science, not a metaphor.
- Coercive control systematically dismantles the capacities you need to leave — accurate self-perception, social support, and trust in your own judgment — before exit becomes necessary. The difficulty of leaving is a feature of how these relationships are structured, not a flaw in the person trying to leave.
- The shame about staying is a symptom of the abuse, not an accurate self-assessment. It was installed through repetition, often deliberately, and it reinforces the bond by suppressing self-trust.
- Separation activates something functionally similar to withdrawal. The craving, the intrusive thoughts, the pull to return — these are physiological, not evidence that you made a mistake.
- Breaking the bond requires more than a decision. It requires structured support, reduced contact, social reconnection, and enough time for the nervous system to recalibrate. EMDR and trauma-informed therapy have research support for this process specifically.
— Section 06Frequently asked
What is trauma bonding?
Trauma bonding is a psychological response in which a person develops a strong emotional attachment to someone who intermittently harms and rewards them. It is not a sign of weakness or pathology in the person who experiences it. It is a predictable neurological outcome of variable reinforcement under conditions of perceived threat — the same mechanism that drives compulsive gambling and is documented in animal conditioning research going back to Skinner. The bond is manufactured through the structure of the abuser's behavior, not through anything deficient in the target.
Why is leaving a trauma bond so hard?
Because the attachment is real at a neurological level. Research on people in obsessive attachment states shows hyperactivation of the dopamine reward system and opioid system engagement similar to what happens during withdrawal from addictive substances. The brain has become physiologically dependent on the cycle. When you leave, the withdrawal is real — physical restlessness, intrusive thoughts, a pull toward contact that does not respond to reasoning. Most people are not told to expect this, and when it arrives they interpret it as evidence that leaving was wrong. It is not. It is withdrawal, and it passes with time and support.
Is trauma bonding the same as codependency?
No. Codependency describes a relational pattern, often rooted in family-of-origin dynamics, in which a person organizes their functioning around another's needs. Trauma bonding is a specific response to the structure of intermittent abuse and reward — it can happen to people with no codependent history and is driven primarily by what the abuser does, not by a pre-existing deficit in the target. Framing trauma bonding as codependency misplaces the responsibility and misses the mechanism.
Why do I still miss someone who hurt me?
Because a genuine attachment existed. The abuser cultivated it, deliberately, through cycles of warmth and withdrawal. When they are gone, you are in the withdrawal phase of that cycle. Grief is an accurate response to losing any attachment, including one that was harmful. The goal of treatment is not to stop feeling the loss — it is to feel it accurately, as grief over what the relationship might have been, not as a signal that you should return. Missing someone who hurt you is not evidence they were right for you. It is evidence the bond was real.
How do I know if I should seek therapy?
If you recognize the pattern described here — the pull you cannot seem to override, the self-blame that feels more like fact than interpretation, the sense that your own perceptions cannot be trusted — therapy is not optional. Not because something is wrong with you, but because the things that are keeping you stuck live in the nervous system, and the nervous system responds to skilled clinical intervention in ways it does not respond to information or willpower. The work is doable. It just requires the right kind of help.
Citations
- Carnes, P. (1997). The Betrayal Bond: Breaking Free of Exploitive Relationships. Health Communications. Carnes introduced the clinical framework for trauma bonding based on observations of attachment formation under conditions of coercion, exploitation, and alternating harm and reward. ↩
- Skinner, B. F. (1938). The Behavior of Organisms: An Experimental Analysis. Appleton-Century-Crofts. Variable-ratio schedules of reinforcement — delivering reward unpredictably — produce the highest response rates and the greatest resistance to extinction compared to fixed or continuous reinforcement. This is the behavioral-science foundation of intermittent reinforcement in human attachment contexts. ↩
- Dutton, M. A., & Goodman, L. A. (2005). Coercion in intimate partner violence: Toward a new conceptualization. Sex Roles, 52(11–12), 743–756. doi:10.1007/s11199-005-4196-6. This paper provides the foundational conceptual framework for coercive control as a systematic pattern — distinct from episodic violence — that operates by dismantling autonomy, social support, and accurate self-perception. ↩
- Fischer, R., Callander, R., Reddish, P., & Bulbulia, J. (2019). How do rituals affect cooperation? An experimental field study comparing nine ritual types. Human Nature, 24, 115–125; see also Burkett, J. P., & Young, L. J. (2012). The behavioral, anatomical and pharmacological parallels between social attachment, love and addiction. Psychopharmacology, 224(1), 1–26. doi:10.1007/s00213-012-2794-x. Burkett and Young provide the neurobiological basis for the addiction-attachment parallel: opioid and dopamine system overlap between social bonding and addictive behavior, including withdrawal phenomenology following separation. ↩
- Brown, B. (2006). Shame resilience theory: A grounded theory study on women and shame. Families in Society: The Journal of Contemporary Social Services, 87(1), 43–52. doi:10.1606/1044-3894.3483. Brown's grounded theory research establishes that shame — as distinct from guilt — is associated with withdrawal, silence, and reduced capacity for self-advocacy, making it a maintaining condition in entrapment dynamics. ↩
- Shapiro, F. (2018). Eye Movement Desensitization and Reprocessing (EMDR) Therapy: Basic Principles, Protocols, and Procedures (3rd ed.). Guilford Press. EMDR has the strongest evidence base of any trauma-specific modality for processing intrusive traumatic memory, including the reexperiencing states that maintain attachment to past abusive relationships after physical separation. The WHO (2013) and American Psychological Association include EMDR as a recommended treatment for PTSD. ↩
If this sounds like where you are, a 20-minute consult is free.
Telehealth psychotherapy for narcissistic abuse recovery, CPTSD, and coercive relationship trauma. Licensed in New York, Maine, Delaware, and Florida. Out-of-network, $225 per session, superbills provided.
Book a free consult