Haute Lumière · The reading

You have seen it enough times to know it is real, and the manual has no name for it

The emotional response that arrives in seconds, resolves in hours, and sits in almost every ADHD chart without a diagnostic code of its own.

She has been reading the same paragraph for four minutes, and nothing in the room shows it.

THE THING ITSELF

A client with ADHD is doing well. The schedule is holding, the routines are built, they have arrived at three sessions in a row and the work is moving. Then a friend takes six hours to answer a text, or a supervisor's tone shifts by a quarter turn, or a partner sighs at the wrong moment. By the next session they want to quit the job, end the relationship and disappear entirely. Not sulking, not frustrated, not disappointed. Something closer to grief, arriving at the speed of an electrical fault.

The word that fits the size of it is dysphoria, from the Greek dysphoros, meaning hard to bear. The choice of word is deliberate and it is worth keeping. Ordinary hurt feelings respond to the ordinary remedies: a walk, a night's sleep, a friend saying the obvious thing out loud. This does not. Clients describe it at a physiological intensity — a punch to the sternum, a floor giving way, a sudden and total conviction that everything is finished — and they describe it while knowing, in full, that the trigger was small.

That is the part clinicians most often miss, because it looks from the outside like a failure of insight. It is not. The insight is intact and arrives on schedule. What does not happen is any reduction in the pain as a result of it.

The understanding shows up on time and changes nothing. The distance between what the client knows and what the client feels is the most exhausting feature of the whole condition.

This phenomenon has a working name — Rejection Sensitive Dysphoria — and it describes an intense, frequently sudden emotional response to rejection, criticism, failure or teasing, whether that rejection is real or perceived. It is widely observed in people with ADHD. It has no diagnostic code in the DSM-5-TR. Those two sentences sit together uncomfortably, and the discomfort is the subject of everything that follows.

What it is not is worth stating plainly, because the alternatives are what fill the charts when the phenomenon goes unrecognised. It is not manipulation, though a client mid-episode can be very hard to be near. It is not immaturity, though it can look adolescent in a forty-year-old. It is not resistance to treatment, though it is often recorded as exactly that, in a note written by somebody who had no other language available.

NOT A SIDE EFFECT

For most of its diagnostic life, ADHD was understood as a disorder of attention and impulse control. The criteria say so: inattention, hyperactivity, impulsivity. That is a behavioural picture, and treatment followed the picture faithfully — behavioural interventions, stimulant medication, organisational scaffolding, timers and checklists and external structure. Emotion, where it appeared in the literature at all, was filed as secondary. Comorbid. Downstream of the real disorder.

Clinicians knew this was wrong, and so did their clients. There was something in the emotional dimension that the criteria did not reach, and it was frequently the thing doing the most damage to the person's life. The organisational scaffolding held; the friendships did not.

Russell Barkley's reframing of ADHD as a disorder of self-regulation is among the most clinically useful moves of the past three decades, and it is the one that makes RSD legible. In that model, ADHD does not impair attention alone. It impairs the executive function system as a whole, and the regulation of affect sits inside that system. Emotional impulses are generated, sometimes at considerable intensity, and the mechanisms that would ordinarily modulate, delay and contextualise them are inconsistently available. The pause that most people get between feeling something and acting on it is, in the ADHD nervous system, unreliable rather than absent.

Neuroimaging has since put substance under the model. Emotional processing regions — the amygdala and the anterior cingulate cortex among them — interact differently with prefrontal regulatory networks in people with ADHD than in neurotypical controls. This is not a metaphor about temperament. It is measurable, and it bears directly on what happens in a client's body when a text message reads as cold.

The quantitative picture has been catching up. A 2020 meta-analysis by Beheshti, Chavanon and Christiansen found emotional dysregulation significantly more prevalent in people with ADHD than in neurotypical controls across every age group examined, with effect sizes from moderate to large. Their argument was not that emotional dysregulation frequently accompanies ADHD. It was that emotional dysregulation should be considered a core feature of it. Shaw and colleagues, in a 2014 paper that has become a landmark, arrived at a compatible conclusion: the emotional dimension of ADHD is pervasive, impairing and clinically undertreated.

The consequence for practice is not subtle. If emotional dysregulation is a comorbidity, it is something to notice after the ADHD is handled. If it is a core feature, it is a treatment target from the first session, and a treatment plan that addresses only attention and organisation has left the most impairing part of the presentation untouched.

A comorbidity is something you get to later. A core feature is something you were supposed to be treating all along.

THE MISSING CODE

The absence of RSD from the DSM is not evidence of its absence from clinical reality. It is evidence about the classification system — which has historically lagged behind both neuroscience and lived experience — and about the particular populations that lag tends to fall on. Several distinct mechanisms produce the gap, and none of them is about whether the phenomenon exists.

The first is timescale. An RSD episode can begin within seconds of a trigger and resolve within hours. Standard assessment instruments are largely built to detect states that persist for two weeks, or patterns stable across years. Something that ignites during a phone call and has burned out before the next appointment is close to invisible to that apparatus, and a client who has recovered by Thursday will often not report Tuesday at all.

The second is mimicry. The emotional storms of RSD can look, on first presentation, like borderline personality disorder, like bipolar II, like social anxiety disorder, like major depressive disorder. A clinician working without the framework will reach for the nearest coded explanation, and the nearest coded explanation will be partially right — which is the worst possible outcome, because partially right treatments produce partial responses, and partial responses get read as the client's resistance rather than the formulation's error.

The third is that adult ADHD is itself underdiagnosed, particularly among women, people of colour, and high-IQ compensators who have built enough structure to pass. Quinn and Madhoo's review of ADHD in women documented how differences in symptom presentation, socialisation pressure and clinician bias combine to delay diagnosis — often into a woman's thirties or forties, by which point she has accumulated years of misdiagnosis, self-blame, and an entirely reasonable scepticism about a mental health system that has repeatedly failed to see her. Where the ADHD has been missed, the RSD is not merely undiagnosed. It has no category to be undiagnosed within.

The fourth is economic and it is the one that keeps the other three in place. No code means no reimbursement pathway. No reimbursement pathway means no research infrastructure. No research infrastructure means no evidence base, and no evidence base is precisely the argument offered for not creating a code. The cycle is self-sustaining and nobody in it has to act in bad faith for it to hold.

The most cited figure in the area belongs inside this frame rather than outside it. William Dodson estimated that as many as 99% of adults with ADHD experience RSD to some degree — a clinical estimate drawn from practice, not an epidemiological finding, and one that awaits replication in large-scale peer-reviewed work. It should be quoted that way every time, including by people who find it convincing. What can be said alongside it, without overreach, is that the informal recognition of RSD across patient forums, advocacy spaces and social media has been extraordinary in scale, and that a description resonating at that volume with the people it describes is itself clinically meaningful information.

What this leaves for the individual clinician is a specific job. You become the diagnostic bridge your client could not find anywhere else — not by diagnosing beyond your scope or your certainty, but by recognising in a way that lets the person in front of you finally feel seen. The act of naming something carefully, accurately and warmly is therapeutic in its own right. A great many clients have waited a very long time for one person to do it.

A classification system's silence is a fact about the classification system. The client in the room is not required to disappear into it.

The morning after an episode, most of the work is deciding whether to send an apology that is not owed.

FIVE FACES

Recognising RSD requires listening for texture rather than for content. Clients rarely arrive saying they are in unbearable pain about a perceived rejection; they arrive with the shape that pain has taken in them. The same underlying phenomenon wears several different faces in an office, and knowing which one is in front of you determines nearly everything about what to do next.

The shutdown presentation is emotional flatness arriving suddenly, mid-sentence, mid-session. The client goes quiet, or goes reasonable, or goes somewhere else behind their own eyes. It reads as withdrawal, and it is frequently recorded as avoidant or resistant, but functionally it is a protective response flooding the system with stillness before the pain is allowed to land. The stillness is not the absence of the episode. It is the episode, managed.

The explosion presentation is the one that gets remembered: rage, intense crying, agitated speech, accusations that land and then have to be lived with. What matters clinically is the shame that follows, which is typically larger than the outburst and which compounds the original wound rather than discharging it. A client who has detonated in your office is waiting to see what you do. Your steadiness in receiving it, without retaliation and without flinching, is itself corrective, and there is no substitute for it.

The avoidance presentation is a whole life quietly organised around rejection prevention. The application not submitted, the invitation declined, the relationship ended pre-emptively at the first ambiguity. From the outside it looks like self-sabotage, and gets treated as a motivation problem. It is closer to sophisticated anticipatory avoidance — a nervous system that has run the numbers on rejection and priced it accurately, then built a life that avoids paying.

The masking presentation is the hardest to catch and the most common in high-functioning clients. The surface is calm and the interior is drowning. What gives it away is peripheral: a flush rising at the collar, eyes tearing while the voice stays level, hands white-knuckled in a lap. These are clients who have almost never been given permission to stop performing competence, and who will keep performing it in your office unless you name what you are seeing gently enough that it does not feel like exposure.

The retroactive presentation happens after the session, and after every other interaction of the day. The client replays conversations for hours, hunting for evidence of a rejection they are certain they detected. They are exhausted by the hypervigilance and openly self-critical about its irrationality, which does nothing to reduce it. This one surfaces only if you ask directly what their mind does afterwards, because it is not what they came to talk about and they assume everybody does it.

Most clients inhabit a combination, and the combination is stable enough to formulate around. The same phenomenon in a shutdown presentation and in an explosion presentation calls for different opening moves, different pacing, and different language for the same truth. Working out which face you are looking at is where individualised treatment actually begins.

THE ALARM CIRCUIT

There are no RSD-specific neuroimaging studies. That is worth stating first and plainly, because it is another consequence of the diagnostic gap and because a framework built on overstated evidence collapses at the worst moment. What exists instead is substantial: the neuroscience of ADHD and of emotion dysregulation, which provides strong explanatory scaffolding for what clients describe. That scaffolding matters clinically, and not only for the clinician's satisfaction. It is what moves both people in the room out of the framework of blame.

Dopaminergic and noradrenergic dysregulation in ADHD is well established across genetic, neuroimaging and pharmacological research. Those same systems govern the salience network — the machinery that decides what matters, what deserves attention, what warrants an emotional response at all. When the salience network is dysregulated, social threat signals can be processed as disproportionately significant relative to their actual content. This is the mechanism underneath the complaint that clients find hardest to explain to the people around them.

Consider the unanswered message. For most people it enters awareness as ambiguity and stays there: they are busy, the phone is somewhere else, it will be answered later. For the client with RSD it is not processed as ambiguity. It is processed as evidence, and the emotional response scales to the perceived threat, not to the objective severity of six hours of silence. Nothing about that is a reasoning failure. The evaluation happened before reasoning was consulted.

The amygdala carries the next part. Research on emotion dysregulation more broadly indicates that people with poor prefrontal cortical control over amygdala reactivity experience more intense and longer-lasting emotional responses to social stressors. Prefrontal underactivation is a core, consistently documented feature of ADHD. Put those two findings beside each other and the RSD picture follows almost mechanically: the alarm goes off, and the circuit that would ordinarily say wait, gather more information, check this against what you know is slow to arrive or arrives without sufficient force. The alarm keeps sounding into a room where nothing is coming to switch it off.

Nigg and colleagues added a third component by implicating working memory in emotional dysregulation among ADHD populations — specifically the difficulty of holding contextual information online long enough to reappraise an experience. Reappraisal is not a mood. It is a cognitive operation with prerequisites, and one of them is keeping several pieces of context available at once while the emotion is running. When working memory is compromised, the sentence yes, but remember how they showed up for you last week is genuinely harder to execute, not merely harder to want.

This is the point at which a client usually goes quiet, because nobody has ever told them that the thing they have been convicted of failing at has a mechanism. The failure to reappraise is not a refusal to reappraise. The architecture that makes reappraisal possible is less reliably available, and no amount of being told to be reasonable installs it.

None of this is a character flaw, and none of it is manipulation. It is a nervous system meeting the world through a different filter, one calibrated over years by accumulated criticism, social failure, and the well-earned knowledge that one's own reactions are frequently unwelcome. The biology and the history are not competing explanations. They are the same explanation, told at two scales.

TWENTY THOUSAND

Children with undiagnosed ADHD are estimated to receive on the order of twenty thousand more negative or corrective messages by age twelve than their neurotypical peers. That figure comes from Barkley's clinical observation rather than from a controlled longitudinal dataset, and it should be carried with that qualification attached. What it points at, though, is something practitioners observe with striking consistency, and the consistency is the part that matters: by adulthood, most ADHD clients have a well-worn neural pathway running from social uncertainty directly to anticipated pain. The architecture of RSD is not born fully formed. It is built, incrementally, out of ordinary days.

Consider what the child actually learns. Told constantly to sit still, stop fidgeting, pay attention, slow down, try harder — and unable to comply, or able to comply only briefly and at enormous cost — they extract a lesson that is entirely logical given the evidence available to them. There is a gap between what is expected of them and what they can reliably produce. They did not choose the gap and cannot close it by wanting to. They are nonetheless held responsible for it, repeatedly, by people they love, and the message accumulates in a form that outlasts every individual instance of it: you are too much, or you are not enough, and which one you are right now matters a great deal to everyone around you.

The peer world runs the same lesson from a second direction. Meinzer and colleagues have documented that children with ADHD experience significantly more peer rejection than neurotypical children — not solely because of behavioural dysregulation, but because social processing differences make the subtle interpersonal cues governing childhood acceptance harder to read and harder to answer in time. They are, through nothing they did, more often left out. More often the one whose social moment went wrong for reasons nobody explained afterwards.

By adulthood the hypervigilance no longer requires a decision. The client does not choose to scan a room for signs of rejection; the scanning is running before they are through the door, a continuous low-level process checking the emotional weather of every interaction they enter. It has the quality of a reflex because functionally it has become one. A conclusion reached early enough, and confirmed often enough, stops being a belief and becomes equipment.

And when rejection does arrive — real or perceived, and the nervous system does not reliably distinguish — the pain is not abstract. It is somatic and it is immediate. Clients report a blow to the chest, a crumbling sensation, heat, nausea, a certainty that something irreversible has just happened. They may know perfectly well that the colleague's brisk email was brisk because the colleague had eleven minutes before a meeting. The knowing is real and it is present. It is simply not what the body is responding to.

The body is not waiting for the argument. By the time the argument arrives, the body has already paid.

A nervous system at rest, which is a different thing from one that was never on alert.

WHAT TO RULE OUT

Differential diagnosis here is not an administrative exercise, and the stakes are concrete. The difference between a formulation that fits and one that nearly fits is years — years of treatment that partially addresses the symptoms while the client concludes, reasonably, that this is simply what their life is going to be. Four differentials come up repeatedly, and each requires ruling in as carefully as ruling out.

Borderline personality disorder shares intense emotional reactivity and significant fear of abandonment with RSD, and the overlap is not incidental. Comorbidity between BPD and ADHD is substantial; impulsivity and emotional dysregulation belong to both, and the presence of one meaningfully raises the likelihood of the other. The differentiators worth attending to are chronic identity instability, idealisation and devaluation cycling within relationships, and self-harm used as a regulatory strategy, all of which are more characteristic of BPD. Episode duration helps too: RSD episodes tend to resolve within hours rather than days, and tend to be traceable to an identifiable cue. Where both are present, both need treating, and neither diagnosis excuses you from the other.

Bipolar II and cyclothymia can be suggested at first presentation because the emotional storms are intense and shift fast, sometimes with a euphoric flip side that Dodson has called rejection elation. The discriminator is trigger-specificity. Bipolar mood episodes arise relatively independently of environmental triggers and persist for days to weeks with their own momentum. RSD episodes are triggered, are frequently traceable to a specific interaction the client can name, and typically resolve within hours. A careful timeline, built around what happened immediately before each episode, usually separates these two faster than any symptom checklist.

Social anxiety disorder shares the fear of negative evaluation, but the mechanism sits in a different tense. Social anxiety is primarily anticipatory and cognitive, organised around the dread of future evaluation. RSD is the acute, affective, near-physiological pain response at the moment rejection registers. The clinical shorthand is worth keeping: the person with social anxiety dreads the party for a week beforehand, while the person with RSD may thoroughly enjoy the party and then lie awake for three hours convinced that one sentence they said was unforgivable.

Complex PTSD is the fourth, and here the framing matters more than the distinction. ADHD and trauma co-occur at significant rates. The presence of ADHD does not rule out trauma, and the presence of trauma does not rule out ADHD or RSD. These are frequently concurrent realities requiring concurrent attention, and a careful trauma history is not an optional extra to be taken if time permits.

The discipline underneath all four is the same. The question is never only what this is instead of, but what else is also true. A client can have ADHD with RSD, a trauma history, and a depressive episode layered on top, and a formulation that names one and quietly drops the others will fail slowly and expensively.

THREE FRAMES

Where RSD is suspected, three frames held simultaneously produce a fuller picture than any one of them can alone. They are not competing hypotheses waiting to be narrowed down to a winner. They are lenses, and the usefulness comes from keeping all three in hand at once.

The neurological frame says that this intensity is the predictable output of specific regulatory differences. Not moral weakness, not manipulation, not a deficit of effort. Holding this frame is what allows a clinician to receive an episode without editorialising about it, and it is what allows the client to stop litigating their own character in every session. It also protects against a subtle failure: treating the emotion as irrational and therefore as something to be argued with.

The developmental frame says that a long history gives change its particular shape and its particular pace. Twenty years of accumulated evidence does not revise in six sessions, and expecting it to is a setup for a rupture. This frame supplies patience with a reason attached. It is context, and it should never be allowed to become an argument for stasis — a history explains why change is slow, not why it is unavailable.

The parts frame treats the RSD response as protective rather than pathological. Something in the system learned, early and for good reasons, that detecting rejection first was the difference between being blindsided and being braced. That part has been doing an exhausting job for decades without relief and without thanks. The clinical aim is not to eliminate it but to relieve it of an assignment it took on when the client was eight and no longer needs performed at that intensity. Clients frequently soften at this reframe faster than at any other, because it is the first account of their worst trait that casts it as loyalty.

Together these three allow a clinician to be rigorous and warm at the same time, which is harder than it sounds and is the whole requirement of this work. The neurology keeps the formulation honest. The developmental history keeps the pace realistic. The parts frame keeps the client from being reduced to a mechanism in their own case conceptualisation.

Three lenses, none of them sufficient alone. A formulation that picks a favourite has already lost the client's actual shape.

WHAT IT COSTS TO MISS

When a clinician misses RSD — when the emotional intensity gets conceptualised as manipulation, immaturity or resistance — the consequences are specific and they are not confined to that course of treatment. The client learns, again, that the mental health system has no language for their experience. They may leave prematurely, and the note will record that they disengaged. What actually happened is that they responded rationally to an experience of not being seen, which is the same conclusion they have been reaching since childhood, now confirmed by a professional.

That is the part worth sitting with. A client who drops out after being misread has not failed treatment. They have been handed one more piece of evidence for the proposition that has organised their whole inner life, and it was handed to them by the person who was supposed to be the exception.

The literature on therapeutic alliance is consistent on this point: feeling understood is among the most robust predictors of positive treatment outcome across modalities. That finding is general, and it applies with particular force to clients whose presenting concerns carry a strong relational and identity dimension. RSD, rooted as it is in the fear of being unacceptable to other people, carries that dimension about as heavily as any presentation does. The alliance is not the soft precondition for the technical work here. It is a substantial part of the mechanism.

Accurate recognition therefore requires a particular kind of clinical humility — the willingness to sit with uncertainty, to update a formulation that has already been written down, and to ask a question that is uncomfortable in exactly the right way. Am I finding this client difficult because their presentation is genuinely complex, or because I have not yet developed the framework that would make their experience legible to me? Both answers happen. Only one of them is about the client.

And what the client needs understood first is not a technique. It is that the pain is real. That it is not weakness. That they are not too much. That the distance between how intensely they feel things and how intensely the world suggests they ought to feel them is not evidence of their brokenness, but of a nervous system shaped by history and by biology — one that carries real costs, and also carries a permeability to other people that a less porous nervous system never knows. Everything else in the treatment follows from that recognition landing.

That is the ground floor, and it is where this volume stops. The nine chapters run the argument in full — the epidemiology, the missing code, the five presentations, the neurobiology, the developmental arc, the differentials, the three frames, the ethics, and the sources behind every claim. It is the first article of a trilogy; the second and third go into the treatment room itself, into modalities, ruptures, language and the long arc of repair, including the application of IFS, DBT skills and appreciative coaching frameworks to RSD. The whole book is free to read at Haute Lumière — no account, nothing to sign, nothing held back behind a form. Buying a copy is for keeping it.

Free to read

Free to read, and free to hear. Every chapter of every book in this house, and every narration of it, is open to anybody. No account, no card, nothing to cancel.

What is in it

• What Is Rejection Sensitive Dysphoria?open this in the house search
• The Epidemiological Reality: ADHD and Emotional Dysregulationopen this in the house search
• Why It Doesn't Have a DSM Code — and Why That Matters for Youopen this in the house search
• How RSD Presents in the Clinical Houropen this in the house search
• The Neurobiological Signatureopen this in the house search
• The Developmental Arc: How RSD Builds Over a Lifetimeopen this in the house search
• Differential Diagnosis: What to Rule In and Rule Outopen this in the house search
• A Clinical Framework for Conceptualizing Your Clientopen this in the house search
• The Ethics of Accurate Recognitionopen this in the house search

A missing diagnostic code is a fact about the manual. It is never a fact about the patient.

Hypervigilance is not a choice made daily. It is a conclusion reached once, in childhood, and never revisited.

The alarm sounds before the argument arrives, and no argument has ever caught up with it.

Recognition is not the preliminary to the treatment. For this client it is the treatment's first act.

Keep looking

Every phrase on this page opens into the house search. The shelf holds Luminous Clinical and six other shelves, and the reading is free.