If you have ever asked yourself why am I addicted to drama, I want to offer you a more useful answer than the one you have probably been given. You are not dramatic. You are not broken, and you are almost certainly not “choosing chaos” in any sense you would recognize as a choice. What is happening is that your midbrain is doing precisely what it evolved to do, and that thing it evolved to do turns out to be inconvenient for your relationships.
The short version: your brain’s reward system is most active not when something good happens, but when you do not know what is going to happen. Which means the not-knowing — the unread message, the four-word reply, the silence you cannot interpret — is not a problem your system is trying to solve. It is a state your system is being paid to maintain.
Let me show you the mechanism, because once you can see it, you can stop confusing it with intuition.
The question underneath the question
Most people arrive at “why am I addicted to drama” from a specific, repetitive experience. Someone said something, or didn’t. The tone was off. The reply was shorter than usual. And instead of walking over and asking what they meant, you went home and built them a motive.
As Peter Crone would say: Past hurt informs future fear. You assembled it out of your own material — old rooms, old people, old evidence — and then you responded to the thing you built as though they had handed it to you.
I have watched this in myself and in nearly every client I have worked with, and the question I could never answer was the simple one: why don’t we just ask?
“What did you mean when you said that? What were you trying to do?”
It costs nothing. It takes eleven seconds. It resolves almost every version of this. And virtually nobody does it.
For years I assumed the obstacle was courage, or vocabulary, or not having the right script. It is not. The people who do this most are frequently the people who have read the most about communication. The obstacle is chemical, and it is more interesting than a confidence problem.
Dopamine is not the pleasure chemical
Start here, because this misunderstanding is doing enormous damage in wellness content.
Dopamine does not reward you for good things happening. It is not a happiness molecule and it is not a motivation molecule in the way the internet uses the term. What dopamine encodes is reward prediction error — the gap between what you expected and what actually arrived [1].
It is a learning signal. Better than expected, dopamine rises. Exactly as expected, dopamine barely moves. Worse than expected, it dips. The system exists to update your model of the world, not to make you feel good.
This matters because it means dopamine is fundamentally about uncertainty, not about pleasure. And that changes everything about what follows.
The finding: uncertainty is the dose
In 2003, Fiorillo, Tobler, and Schultz did something elegant. They recorded directly from dopamine neurons while animals learned that different cues predicted reward with different probabilities — sometimes reward was certain, sometimes it never came, and sometimes it was a coin flip [2].
As expected, the neurons responded to the cue and to the reward. But they found a third signal nobody was looking for: a slow, sustained ramp of dopamine activity during the waiting period, between the cue and the outcome.
And that ramp was not largest when reward was guaranteed. It was not largest when reward was hopeless. It peaked when the outcome was maximally uncertain — right around fifty-fifty.
Not the reward. The not-knowing.
This has since been formalized for behavioral addiction. Variable, unpredictable reinforcement schedules produce larger and more persistent dopaminergic activation than predictable schedules of identical average value [3]. Same reward, same frequency, different certainty — and the uncertain one is dramatically more compelling.
It is the entire design principle of a slot machine. It is why nobody has ever developed a compulsive relationship with a vending machine.
Why “just ask them” is so much harder than it sounds
Now apply this to the four-word reply.
When you do not know what someone meant, you are sitting in a state of maximum reward uncertainty about a socially significant outcome. Your midbrain is not bored by this. Your midbrain is engaged in a way that ordinary, resolved, comfortable interactions simply do not produce.
And the clarifying question does something your system registers as a loss. It collapses the uncertainty. It ends the ramp.
So the reason you have not asked is not, primarily, cowardice.
You are not avoiding the conversation. You are protecting the dose.
I find this genuinely irritating, having spent years telling people — and myself — that clarity is the simpler, cheaper, kinder option. It is all three of those things. It is also, neurochemically, the boring one, and that was missing from my model for a long time.
If you want to catch it in real time, look for the ramp: the 3 am rehearsal, the message drafted and deleted four times, the ninth reconstruction of their tone. That is not processing. That is not intuition working. That is the uncertainty paying out.
The relief loop, and why the fights get bigger
There is a second mechanism sitting underneath the first, and it explains the escalation that people find most frightening about their own patterns.
Acute stress triggers the release of endogenous opioids — your own beta-endorphin and enkephalins, producing real, measurable analgesia [4]. This is why you can be strangely functional in the middle of a conflict, and why the pain often arrives afterward rather than during.
Then the conflict resolves. The apology, the reunion, the make-up. The opioid tone withdraws against a dopaminergically sensitized background, and what you experience is not neutral. It is a specific, glowing, better-than-baseline relief.
That relief is the reinforcer. Not the peace — the relief.
And like any opioid-mediated loop, it builds tolerance [5]. The intensity of conflict required to reach the same relief goes up over time.
This is the mechanism in couples who can no longer have a small disagreement. It is not that they have become worse people or worse communicators. It is that the dose required has escalated, and nobody in the room can see the prescription.
Why you bond to the person who caused the spike
Here is the part that tends to land hardest, and I want to be careful with it.
Oxytocin has a well-documented role in buffering the stress axis — it inhibits the CRH/ACTH cascade and helps the body return to baseline [6]. It is genuinely a calming, bonding molecule.
But oxytocin is released by touch, by eye contact, by emotionally intense reunion. Which means the reconciliation phase floods you with the molecule that deepens attachment — and it deepens it toward the source of the original threat.
Cycles of threat followed by intermittent relief from the same person produce a physiologically real attachment. This is what the literature calls traumatic bonding, and it is described as mechanistically convergent with substance dependence rather than as a metaphor for it [7].
I want to say clearly: this is a description of a mechanism, not a verdict on your relationship or a diagnosis of anyone in it. Understanding the chemistry of why something is hard to leave is not the same as being told to leave.
If you are in a relationship where you are being harmed, please talk to someone qualified and local. This article describes a neurobiological pattern. It is not advice about your specific situation, and it is not a substitute for professional or legal support.
Why calm starts to feel like “no spark”
One more consequence, and this one quietly ruins good relationships.
Chronic exposure to high-variability relational reinforcement blunts the dopaminergic response to ordinary, predictable, positive events [3]. Reward that arrives reliably stops producing much signal at all.
So the steady person — the one who replies when they say they will, who tells you what they mean, who is not a mystery — does not register as safe. They register as nothing.
You will experience this as an absence of chemistry. You will describe it as “no spark.” And you will be entirely sincere, and entirely wrong about what you are describing.
You are not reporting on them. You are reporting on your calibration.
Where the calibration came from
None of this is a character failure, and I want to close that door firmly, because this is exactly the territory where people take a mechanism and convert it into evidence against themselves.
Early relational chaos does measurable work on the stress system, including epigenetic modification of the glucocorticoid receptor gene NR3C1 through DNA methylation, altering lifetime HPA reactivity and autonomic flexibility [8].
If your first environment ran on unpredictability — on screaming, or on the thing that is worse than screaming, which is four days of silence — then your set point for arousal was calibrated there. In a room you do not remember. By people who were, almost certainly, also just trying to get through the week.
You did not choose it. You were tuned by it.
What actually interrupts this
Not “communicate better.” You know how to communicate. You have probably read more about communication than the person you are currently avoiding.
1. Name the ramp as a sensation, not as information. The 3 am rehearsal is not analysis, and it is not intuition. It is a chemical event with a texture. Once you can feel it as a chemical event — a tightening, a leaning-forward, a pull toward the phone — you get a choice you did not have while you believed it was thinking.
2. Regulate before you ask, not after. The clarifying question is close to unaskable from sympathetic arousal, because from that state everything you say arrives as an accusation regardless of your wording. State first, sentence second. This is also why this work is done with another nervous system present rather than alone with a journal — co-regulation is the condition under which patterns can actually update, not a nice extra.
3. Ask the boring question. “What did you mean when you said that? What were you trying to do?” Confrontation without the energetics of conflict.
4. Expect flatness, and do not misread it. When you finally ask, and they say something ordinary and non-sinister, you may feel deflated rather than relieved. That flatness is not disappointment, and it is not proof that they are lying. It is withdrawal. It passes. What is on the other side of it is a real person instead of the one you built.
5. Be kind and take no bullshit. Understanding why someone did something does not oblige you to keep standing there while they do it. This is not a permission slip for anyone’s behavior, including your own. Ask, believe the answer, and then decide — from a regulated system rather than a spinning one.
The harder version: stop needing to ask
Everything above is about asking better. That is worth real money, and it is still a skill built on top of a job you should not be doing.
Most people who read a room surgically learned to do it young, in an environment where what was said and what was done did not match, and where the gap between the two was where the danger lived. So they got very good at the gap. Microexpressions, tone shifts, the half-second before someone changes their mind.
People will call this intuition. It is closer to surveillance, and it was built because it was needed.
Here is the part that takes decades to see: the room-reading and the rumination are the same apparatus. The faculty that reads a room in four seconds is the faculty that spends four months modeling one person’s intentions. You cannot keep the gift and discard the cost, because they are not two things.
Which means the real intervention is not sharper questions.
It is choosing rooms — and people, and jobs — that do not require the machine to be running at all.
A person whose words and actions match is not a romantic preference or a nice-to-have. Structurally, they remove the gap the machine exists to monitor. No gap, nothing to model. Nothing to model, and the faculty comes back to you.
One honest caveat, because this is where people go wrong. If your system was calibrated on unpredictability, then a person who leaves no room for doubt will not register as relief. They will register as flat. No spark. Suspiciously easy. That is the blunted dopaminergic response to predictable reward described above [3] — operating exactly as the literature says it will, on the person who is actually good for you.
So this is not an aspiration. It is a recalibration target; it comes with a withdrawal phase, and you will want to leave during it.
Boring is not the absence of chemistry. Sometimes boring is the first thing safety feels like when you have never had it.
The part the science does not resolve
I would rather be useful than authoritative, so: the neuroscience above is the best answer I have found to a question I have been asking since I was small enough to be standing in a doorway watching it happen. It is a genuinely good answer. It does not fully close the question.
What I have arrived at instead is this: almost nobody is trying to hurt you. Most people are trying to survive, most of the time, inside a story you cannot see from where you are standing. Everyone here is trying to love and be loved.
And your dopamine system finds that unbearably boring.
The work is not becoming a person who never guesses. It is becoming a person who notices the guess, asks anyway — and then, slowly, stops choosing lives that require so much guessing.
Frequently asked questions
Is drama addiction a real addiction?
Not as a formal diagnosis — “drama addiction” does not appear in the DSM-5. But the underlying neurocircuitry overlaps substantially with substance addiction: the same mesolimbic dopamine pathway, the same uncertainty-driven reinforcement, the same opioid-mediated relief cycle and tolerance [3,5]. Researchers describe behavioral and relational patterns using the same allostatic framework developed for substance dependence. So: a real mechanism, not a formal disorder.
Why do I always assume people are trying to hurt me?
Because your brain must fill the gap between what someone did and what they meant, and it fills it with priors — patterns learned from previous relationships, usually early ones. If your early environment was unpredictable or unsafe, a threat-shaped guess is the cheaper, faster prior. It is not paranoia. It is an efficient system running on outdated training data.
How do I stop overthinking what someone meant?
Interrupt the ramp rather than trying to think your way to the right conclusion — there is no right conclusion available from inside your own head. Notice the rehearsal as a physical sensation, regulate your state, then ask the direct question. Expect the answer to feel anticlimactic.
Is this the same thing as trauma bonding?
They share a mechanism — intermittent reinforcement, opioid relief cycling, and oxytocin released at reconciliation deepening attachment to the source of the threat [7]. Trauma bonding specifically describes this in the context of an abusive relationship. The chemistry described here operates on a spectrum that includes many non-abusive but exhausting dynamics. If you are being harmed, please seek qualified local support.
Can this be changed?
Yes, though not primarily through insight. Interventions that work borrow from addiction medicine and from somatic practice: building tolerance for the flatness that follows leaving an unpredictable dynamic, and restoring autonomic flexibility through regulation practices and safe co-regulation [9]. Insight tells you what is happening. State change is what alters it.
You can keep analyzing the pattern. Or you can change the state it runs on.
The Breakthrough Session is 90 minutes to find the root pattern, interrupt it, and leave with a written 90-day roadmap and 7 days of integration support. €222, online, and 50% credits toward any program within 30 days.
References
- Schultz, W. Dopamine Reward Prediction Error Coding. Dialogues Clin. Neurosci. 2016, 18 (1), 23–32.
- Fiorillo, C. D.; Tobler, P. N.; Schultz, W. Discrete Coding of Reward Probability and Uncertainty by Dopamine Neurons. Science 2003, 299 (5614), 1898–1902. DOI: 10.1126/science.1077349.
- Clark, L.; Boileau, I.; Zack, M. Engineered Highs: Reward Variability and Frequency as Potential Contributors to Behavioral Addiction. Neurosci. Biobehav. Rev. 2023, 148, 105143.
- Valentino, R. J.; Van Bockstaele, E. Endogenous Opioids: The Downside of Opposing Stress. Neurobiol. Stress 2015, 1, 23–32.
- Bruchas, M. R.; Land, B. B.; Chavkin, C. The Dynorphin/Kappa Opioid System as a Modulator of Stress-Induced and Pro-Addictive Behaviors. Brain Res. 2010, 1314, 44–55.
- Neumann, I. D. Involvement of the Brain Oxytocin System in Stress Coping: Interactions with the Hypothalamo-Pituitary-Adrenal Axis. Prog. Brain Res. 2002, 139, 147–162.
- Dutton, D. G.; Painter, S. Emotional Attachments in Abusive Relationships: A Test of Traumatic Bonding Theory. Violence Vict. 1993, 8 (2), 105–120.
- Chubar, V.; Vaessen, T.; Van den Noortgate, W.; et al. Mild Daily Stress, in Interaction with NR3C1 DNA Methylation Levels, Affects Autonomic Stress Responses. Psychoneuroendocrinology 2023, 152, 106083.
- Lehrer, P. M.; Gevirtz, R. Heart Rate Variability Biofeedback: How and Why Does It Work? Front. Psychol. 2014, 5, 756.
Rosa F. Brissos, PhD, is a Medicinal Chemist and holistic therapist specializing in nervous system regulation, somatic trauma work, and preparation and integration of altered states. Based in Portugal. Working internationally.
