If you are not alone right now, this page will be in your browser history. You can read it in a private window, or clear it when you are done.
Abuse and a psychiatric or neurodevelopmental condition are two different kinds of thing. Abuse is a pattern of behavior that produces and maintains control over another person. A condition is an impairment in how a person functions. Neither one rules out the other, and the honest answer to is it abuse or is it his condition is very often that both are true at the same time.
This page diagnoses nobody. A diagnosis of any person requires an assessment of that person, by a clinician who has met him, and nothing written here is that. What this page offers is a way of thinking about the difference — clinical reasoning, explicitly labeled as clinical reasoning — and a plain statement of where the research runs out, because on this particular question it runs out early.
Why does everything I find about this start from couples work?
Most of the material answering this question is published by sites organized around couples work, and a couples frame rests on two assumptions: that the relationship is the unit of treatment, and that each partner holds a share of the problem. Those assumptions are reasonable for a marriage strained by a real condition. They are unsafe where there is fear — and a site built on couples work is not in a position to say so about its own framework.
The mechanism is not anyone’s bad faith. Conjoint work runs on mutual disclosure and shared responsibility: both people say true, unflattering things in front of each other. Where one person can make the other pay for what was said, the session becomes an inventory of admissions and the drive home becomes the consequence. Inside the frame there is also the identified patient problem — a systems model looks for what each party contributes, so whoever is visibly dysregulated in the room becomes the visible problem, and that is very often the person who has been frightened for years.
The couples question and the differential question are therefore tangled together: to work out whether a pattern of behavior is impairment or control, a clinician needs an account that could not have been given in front of the person it is about. If you came out of six months of couples work feeling like the one being treated, there is a mechanism behind that, and it is mechanical rather than moral.
What does the research actually say about telling these apart?
The research does not answer this question, and naming the gap is more useful than dressing it. A search for studies comparing an abuse presentation against an ADHD, autism, PTSD or depression presentation, for the purpose of telling them apart, returns nothing. No differential instrument exists, and no diagnostic algorithm exists. Anyone offering you a checklist that sorts the two is offering you something the field has not built.
What does exist is association literature, and it is a different animal. In a questionnaire study of 316 German adults — 131 meeting ADHD criteria and 185 controls, recruited online — those meeting criteria reported higher rates of both perpetrating and experiencing psychological and physical partner violence. The study is a cross-sectional online survey, so it can show an association and cannot show a direction; the authors themselves note that comorbid conditions were uncontrolled and that online recruitment lowers internal validity (Merscher et al., 2025).
A group-level association like that cannot do the job you need done, and the reason has a name. Inferring something about one person from a group-level association is the ecological fallacy. A raised rate in a population tells you nothing about which mechanism is operating in one house, and it cuts both ways: it does not mean his ADHD explains what happens on Sunday nights, and it does not mean anything about anyone else with ADHD. The overwhelming majority of people with any of these diagnoses are not controlling their partners.
Everything from here down is clinical reasoning, not a finding. The questions below are useful because they ask about structure rather than about severity — and severity is the thing you have probably been arguing about with yourself for years.
What do these conditions actually look like?
Each of these four conditions has a shape, and the shapes are genuinely different from one another. Knowing what each looks like is what stops the question collapsing into “he’s ill” or “he’s cruel,” which are the only two answers most people will offer you.
ADHD
ADHD is a neurodevelopmental condition defined by inattention, hyperactivity and impulsivity, and its most disabling feature in adult life is usually executive dysfunction — trouble with planning, sequencing, holding intentions across time and stopping a response once it has started. It is common: an estimated 6.0% of US adults, about 15.5 million people, reported a current ADHD diagnosis in a national survey of 7,046 adults conducted in October and November 2023 (Staley et al., 2024, MMWR).
Emotional volatility is part of the adult picture even though it is not part of the diagnostic definition. A systematic review of 22 studies found consistently poorer emotion regulation in adults with ADHD than in controls, and noted that emotion dysregulation is still not in the DSM-5 criteria, which remain built on inattention, impulsivity and hyperactivity (Soler-Gutiérrez et al., 2023). In ordinary life it looks like this: things are forgotten rather than withheld, the flare is fast and recedes fast, and the same failures show up at work, with his mother, and with the car registration.
Autism
Autism is a neurodevelopmental condition involving differences in social communication and sensory processing, and much of what reads as coldness or contempt is a difference in how meaning is exchanged rather than a message being sent. Damian Milton’s double empathy problem describes this as “a breakdown in mutual understanding (that can happen between any two people) and hence a problem for both parties to contend with, yet more likely to occur when people of very differing dispositions attempt to interact” (Milton, Gurbuz & López, 2022). It is mutual, and it is not somebody being ignored on purpose.
A meltdown is a different event from targeted rage. In a descriptive phenomenological study of 32 autistic adults across seven countries, participants described being overwhelmed by sensory, social, informational or emotional load, losing the ability to think logically, and being unable to stop — “I lose control of my mind. If I try to be calm, I simply can’t” — followed by exhaustion, patchy memory and shame (Lewis & Stevens, 2023). A meltdown has a build-up, it is not aimed at a person, and it costs the person having it enormously.
PTSD
PTSD is a trauma- and stressor-related disorder, and the part of it that lands in a relationship is Criterion E, the alterations in arousal and reactivity. The DSM-5-TR wording repays close reading: “irritable behavior and angry outbursts (with little or no provocation) typically expressed as verbal or physical aggression toward people or objects.” The operative phrase is little or no provocation — the outburst is decoupled from what is happening rather than tuned to it. Around it sit hypervigilance, exaggerated startle, sleep disturbance, concentration problems and avoidance of reminders.
PTSD is also the condition most likely to be on both sides of a marriage: whether emotional abuse can cause PTSD is a live question for the person asking this one.
Depression
Depression in adults is characterized by depressed mood or loss of interest and pleasure — anhedonia — for at least two weeks, with changes in sleep, appetite, energy, concentration and self-worth. Irritability is common in depressed adults, but the DSM-5-TR allows irritable mood to substitute for depressed mood only in children and adolescents. In an adult, irritability travels with depression rather than defining it.
Depression’s characteristic movement is withdrawal, and the withdrawal is indiscriminate. It takes away the things he likes along with the things he owes you. He stops going to the game, stops calling his brother, stops caring about the yard. A man who has withdrawn from you and from nothing else has not withdrawn.
None of these four descriptions is a test. They are shapes, they overlap, and a person can have two at once. They are here to give the four questions that follow something to work on.
Does it happen with everyone, or only with you?
Selectivity is the first question a clinician asks, because impairment does not check who is in the room before it happens. Executive dysfunction does not remember to lose the car keys only in front of one person. A meltdown is not scheduled. Hyperarousal does not consult the guest list. Where a difficulty is genuinely a deficit it generalizes — at work, with his family, in traffic — and other people have noticed it too.
The honest complication is that everybody performs better in public than at home, and that is not evidence of anything. Autistic people and people with ADHD often hold it together all day at real cost and come apart in the one place it is safe to. Masking is well described and exhausting, so a man calmer at work than at home may simply be spending his whole reserve before he gets there.
The question is therefore not whether there is any difference between home and elsewhere, but whether the difference is total, whether it runs consistently in one direction, and whether other people describe a different man from the one you live with. A deficit that appears in one relationship and nowhere else in his life, over years, is doing something a deficit does not do. Clinically, the sentence that arrives most often at this point is not about him at all. It is “everyone thinks he’s lovely,” said as though it were evidence against herself.
Who ends up better off afterwards?
Asking who ends up better off is a way of asking whether there is a pattern of benefit, and it is one of the clearest lines available between impairment and control. Impairment costs everybody, including the person who has it: a man whose executive function fails misses his own deadlines and loses his own money along with yours. If every failure lands in a way that suits him — the plans that get forgotten are yours, the money that vanishes was earmarked for your thing, the argument erupts the night before your sister visits — that is a distribution, not a deficit.
The aggression literature has long distinguished reactive aggression, which is affect-driven and follows provocation, from proactive or instrumental aggression, which is goal-directed and produces something for the person doing it (Dodge & Coie, 1987) — though that work was done in children’s peer groups, so applying it to a marriage is an extrapolation. Better established, in the coercive control literature itself, is that control has a structure: Mary Ann Dutton and Lisa Goodman describe coercion as a demand, a credible threat of a consequence, surveillance to check compliance, and occasional delivery of the consequence (Dutton & Goodman, 2005). A condition does not build a contingency. Nothing about depression checks the mileage on the car.
Does it stop when somebody else walks in?
Behavior that stops when the doorbell rings and resumes when the door closes is being modulated by an audience, and in behavioral terms that is discriminative stimulus control — the presence or absence of a witness functioning as a signal about consequences. Something capable of being suspended in a second, on cue, and picked up again later, is not something that could not be helped when it happened.
The honest complication, again, is that suppression is possible for people with these conditions too, and the autistic adults in the meltdown research described exactly that: internalizing a meltdown in public “for survival,” and paying for it afterwards. So suppression alone settles nothing. What separates the two is what happens next. A meltdown held in until the visitors leave is followed by collapse — exhaustion, shutdown, sometimes hours of it, often shame the next day. Anger set down for company and picked back up afterwards is followed by nothing. There is no bill.
The second thing to notice is direction. A man who has been visibly struggling all evening and then snaps at whoever is nearest has snapped at the nearest person. A man who is patient with a colleague at six and precise with you at eight has aimed. Aim is not a symptom of anything.
If the person you are worried about here is yourself — if you have shouted, thrown something, or said the unforgivable thing and cannot now tell what you are — that is a different question, and it is not the one this page answers.
Does the apology change the behavior, or restore access?
Apologies are where the two explanations look most alike and behave most differently, and what separates them is what the apology is followed by. Remorse arising from a condition tends to produce machinery: alarms, lists, a medication review, a therapist, a system for the thing that keeps going wrong. The machinery is uneven and frequently fails. What matters is that it exists, that it is aimed at the behavior, and that it turns up in other parts of his life too.
An apology whose function is to restore access produces temperature change and no machinery. The atmosphere lifts, the affection returns, the subject is closed — and nothing in the structure is different. In behavioral terms that apology is being maintained by negative reinforcement: it reliably removes something aversive to him, which is your withdrawal, and behavior that reliably removes something aversive is extremely durable.
There is a second signal in the content, easy to miss because it sounds like accountability. Listen to what the apology is for. “I’m sorry you took it that way,” “I’m sorry I lost it, you know what I’m like,” and “I’m sorry, but you knew that would set me off” are apologies whose subject quietly relocates to you. An apology that names the behavior, with no clause attached, is doing something different.
Lundy Bancroft, who spent years running programs for men who had been abusive, put the underlying point in a sentence: “Abuse grows from attitudes and values, not feelings. The roots are ownership, the trunk is entitlement, and the branches are control.” A condition can supply the feelings. It cannot supply the entitlement.
If you are in immediate danger, call 911.
National Domestic Violence Hotline — 1-800-799-7233, 24 hours · text START to 88788 · thehotline.org
988 Suicide and Crisis Lifeline — call or text 988
Deaf, DeafBlind and hard-of-hearing callers: National Deaf Domestic Violence Hotline videophone 855-812-1001, 24 hours (a partnership between the National Domestic Violence Hotline and Abused Deaf Women’s Advocacy Services).
Calling leaves the number in your call log and texting leaves the thread in your messages. Both can be deleted afterwards. If that is not something you can do safely, a friend’s phone or a payphone is the safer route.
What if he has a condition and it is also abuse?
A condition and abuse being true at once is not a compromise answer, and for a lot of women it is the accurate one. A man can have genuine ADHD, be genuinely impaired by it, be doing genuine work on it — and also run a pattern of control that has nothing to do with it. Comorbidity is ordinary in every part of clinical practice. This particular combination feels impossible only because you have been offered two boxes and told to pick, usually by people who love you and want the question settled.
The clinical error that comes from picking has a name: diagnostic overshadowing, attributing everything a person does to a known diagnosis so that anything else stops being visible. It runs both ways here. A clinician who knows about the ADHD may file the control under the ADHD and never look again. Someone who has decided this is abuse may file the real impairment under the abuse and be surprised when medication changes something.
The consequence worth holding on to is this: treating the condition does not treat the control, and the two are not on the same schedule. Stimulant medication improves attention. It does not produce a man who stops reading your messages. Trauma therapy can reduce hyperarousal without touching an assumption about who decides how money is spent. If the change you are waiting for is a change in the pattern, a diagnosis and a prescription are not evidence that it is coming.
None of that makes the condition less real. It is real, it is his, and treating it is worth doing on its own terms. It is just not the answer to the question you came here with.
What this page can tell you, and what it cannot
This page can tell you what ADHD, autism, PTSD and depression actually look like, why the research does not settle the differential, and what a clinician listens for: selectivity, benefit, audience, and what an apology is followed by. This page cannot tell you what is happening in your marriage, and it cannot diagnose your husband, your partner or you. A diagnosis of anybody requires an assessment of that person, not an assessment of your account of him.
Two things before you close the tab. A real condition and a pattern of control can coexist, and if that is your answer, it is an answer and not a failure to decide. And the strength of an explanation is not the same as the strength of an obligation — being good at explaining him has never been a reason to keep doing it.
If what you want to know is why you are still here when you can see all of this clearly, that is a mechanism rather than a character flaw. If you would rather talk to somebody today, the National Domestic Violence Hotline is 1-800-799-7233, 24 hours, or text START to 88788.
The differential this page describes — abuse, or a condition, or both — is one of the more common reasons women look for a therapist at all, and it is worth asking enough to tell, before you book, whether somebody has actually worked with both sides of it.
You do not have to have decided anything to start. Not leaving. Not staying. Nothing at all.
SOURCES
- Staley, B.S., Robinson, L.R., Claussen, A.H., et al. (2024). Attention-deficit/hyperactivity disorder diagnosis, treatment, and telehealth use in adults — National Center for Health Statistics Rapid Surveys System, United States, October–November 2023. MMWR Morbidity and Mortality Weekly Report, 73(40), 890–895. PDF
- Soler-Gutiérrez, A.-M., Pérez-González, J.-C., & Mayas, J. (2023). Evidence of emotion dysregulation as a core symptom of adult ADHD: A systematic review. PLOS ONE, 18(1), e0280131. doi:10.1371/journal.pone.0280131
- Lewis, L.F., & Stevens, K. (2023). The lived experience of meltdowns for autistic adults. Autism, 27(6), 1817–1825. doi:10.1177/13623613221145783
- Milton, D. (2012). On the ontological status of autism: The “double empathy problem.” Disability & Society, 27(6), 883–887.
- Milton, D., Gurbuz, E., & López, B. (2022). The “double empathy problem”: Ten years on. Autism, 26(8), 1901–1903. doi:10.1177/13623613221129123
- Merscher, J., Barra, S., Xander, A.M., Patrasc, P.G., Retz-Junginger, P., & Retz, W. (2025). Adult ADHD predicts intimate partner violence perpetration and victimization irrespective of gender and age. Scientific Reports, 15, 4914. doi:10.1038/s41598-024-74222-w
- Dodge, K.A., & Coie, J.D. (1987). Social-information-processing factors in reactive and proactive aggression in children’s peer groups. Journal of Personality and Social Psychology, 53(6), 1146–1158. PubMed 3694454
- 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
- Stark, E. (2007). Coercive Control: How Men Entrap Women in Personal Life. New York: Oxford University Press.
- Bancroft, L. (2002). Why Does He Do That? Inside the Minds of Angry and Controlling Men. New York: G.P. Putnam’s Sons.
- American Psychiatric Association (2022). Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR). Washington, DC: American Psychiatric Association Publishing.
Educational content. Not therapy, legal, or financial advice. Nothing on this page is an assessment or a diagnosis of any person.