Sam de Selby (pseudonym) is a writer living in Germany with a background in nonfiction and research assistance. She is currently working on Fractal Procrastination, a project detailing the experiential nature of psychosis.
I think it is a mistake to frame treatment adherence as a one-time decision. Taking oral medication means the concession that you are in need of correction is made anew daily, or multiple times per day.
Even at the point when I voluntarily admitted myself to the hospital in the midst of psychosis, I was vehemently opposed to taking antipsychotic medication. If a lobotomy had been on offer, I would have signed the consent forms without hesitation if it seemed likely to stop the pain, yet I baulked at the idea of long-term medication. It wasn’t just fear of side effects or lack of awareness that it would probably help. I didn’t want to make that decision over again every day. I wanted a solution. It was painful to consider what I hoped was a temporary blip could be something deeper.
After about a week in the hospital, another patient persuaded me to start taking my prescribed medication. (I suspect that, where psychiatric hospitalization is beneficial, much of what is therapeutic happens between patients.) He had been in and out of treatment for almost as long as I’d been alive and painted for me a horrifying picture of how my life would go if I remained unmedicated, based on others he had met. I don’t think I would have made that decision without someone telling me I was an idiot not to do so. No amount of education makes it easy to communicate the possibility of escape to someone who is on a different plane of reality and doesn’t know it. I had no idea how much it was possible for things to change as a result.
When I found an antipsychotic that works for me, it was like drinking from a firehose of lucidity. It was too much, too fast, too soon. After years of unrecognized on-and-off psychosis, normal sanity felt alien. Everything I thought I knew about myself was wrong. Suddenly, I re-evaluated my entire life, seeing every single occurrence in a new light. Memories lost all emotional resonance. The nature of state-dependent memory makes it difficult to connect with the past self when your mindset is so different from then. At the time of writing, 5 months after exiting psychosis, I continue to reel at the change in the most basic aspects of experience, like the sense of time and space. I used to feel much older than my age. Now I feel younger, because I’m new to seeing this world without a coating of paranoia.
Every 24 hours still, there is a moment on which everything to follow pivots. Every 24 hours, I must decide which world I want to live in and which person to be. Every 24 hours, I must admit to myself anew that there is something terribly wrong with me. I can only paper over it for a day at a time. It buys me freedom, for one day at a time. The improvement that medication brings in the eyes of others feels like a condemnation of my undoctored self. The unmedicated me becomes the thing to fear. Insight is something that has to be re-enacted over and over, to this day. I take it, and resent it.
It sounds so simple to other people: just take your meds. But that’s a fridge magnet, not a solution to the complicated equation of choosing between living in a world that can feel meaningless and risking incomprehensibility again.
Yet medication also gives you at least a foothold in normality, a little distance to see yourself in cold, clear light. For that, I will always be grateful. The choice to stay in this world feels like a leap of faith anyway. It doesn’t get easier with time, at least not so far. Every pill is a reminder that who I am is not who I want to be. It’s an intervention in basic identity. If it were just about treating a temporary state, it would be easy. The past can be buried. Who I am follows me around. I have to reject that person over again every day.
As far back as I remember, I felt I had access to an extra plane of meaning. I didn’t think I was smarter than other people, but I was sure I knew something they didn’t. The fear setting was always turned up a bit too high. I never quite outgrew having imaginary friends or being scared of the dark. Around age 15, something shifted. That is the point in time where I can clearly see in hindsight how the ambient paranoia solidified, and the voices became problematic.
In the 13 years that followed, the paranoia, social withdrawal, and hallucinations waxed and waned, becoming a nagging background hum after about 8 years. Whether those early features are best categorized as prodrome, schizotypy, or neurodevelopmental difference will always be unclear to me. Certainly, there were previous times before the big episode when I fully detached from reality for days, weeks, or even a couple of months. The difference was that I always maintained some connection to the world, some obligations to keep me grounded, and managed to pull myself back. Looking back, it’s unimaginable that I coped with such intense fear for so long. But you have to get through the day one way or another. There is a widespread misconception that people in psychosis are too disorganized to manage basic functioning and that it should be obvious to other people. Psychosis is not always dramatic. When it starts slowly, you are forced to adapt.
It was only when things worsened to the point of complete functional collapse and suicidality that I was willing to accept help and received a diagnosis. But none of the symptoms were new or surprising to me—some had been worse at prior times. It just took feeling completely defeated by my own mind to break through my innate stubbornness. I no longer cared enough to be in denial. If there was an explanation someone could offer for what was wrong with me, then I wanted to hear it. On the one hand, I regret that I struggled unnecessarily for so long; on the other hand, I know that I would never have been willing to talk about it before. I just wasn’t ready to engage in what ended up feeling like the process of leaving a cult. Involuntary treatment would have only deepened the denial and made me less likely to ever engage with help.
If insight in psychosis is understood to be the awareness of a morbid change in oneself and understanding of how that differs from other people, what happens when there is no clear change, only an alteration in degree or distress? How can the impact on functioning be understood when there is no comparison? When you grow up with unrecognized psychosis, there is no obvious line between disorder and personality. The traits to be relabeled as symptoms may be unwanted, but they are still part of you. In that case, possessing insight becomes about acceptance of defect rather than acknowledgement of change.
Another aspect of how insight in psychosis is defined is recognition of how one’s experience of reality differs from others. This is a difficult ask for a group of people who tend to be socially isolated. If you don’t have much contact with other people, it’s tricky to reality test. If your psychosis is a bolt out of the blue, you’re more likely to have people around you to tell you you’re thinking abnormally. When onset is insidious, relationships erode or never develop in the first place. You need people most at the point where they’re least likely to be around.
Most people in the world are not particularly versed in psychopathology. Most people think everyone thinks like them. When those who’ve studied the mind assume everyone has that knowledge, they’re committing the same fallacy that psychotic people commit when they think they think normally. Until recently, I can’t recall ever having a conversation about the voices because they didn’t register as strange. Most people don’t just talk about how their consciousness is ticking along. There is a built-in assumption that psychotic people should know they’re abnormal. But every time I read this, I think, how should I know what normal looks like?
To accept treatment without ambivalence is then to let a professional carve out the parts of you considered detrimental, to remodel the internal landscape and ways of relating. The blow this strikes to the ego is considerable. Ambivalence in this case can come from the unfamiliarity of inhabiting consensus reality. The change being responded to is the arrival of health, not the arrival of sickness. After all, the gap between normal and psychotic experience is the same size as the one in the opposite direction. This is not to judge it as a bad change, but a drastic and unnerving one. Antipsychotics, when they work, tend to start taking effect relatively quickly compared to, for example, antidepressants. The change they bring is often described as a sudden epiphany, not a dawning awareness. Perhaps sometimes the problem can be that they work too well. They drag you, kicking and screaming, into a different world.
In late 2025, in the middle of a six-month psychotic episode, I found and printed out Aubrey Lewis’ classic paper The Psychopathology of Insight, while obsessively researching what might be going on and how to understand it. Next to Lewis’ definition of insight as ‘the correct attitude to a morbid change in oneself,’ I annotated: ‘but what if it is instead a failure to change?’
I knew, at least some of the time, that what I was experiencing was probably psychosis and was desperate for an explanation to give structure to the suffering. At the same time, I was terrified of being forced to reappraise myself as disordered. The idea that it had all been in my head, all along, was too much to bear. Whatever was wrong with me at that point had never been quite right. When I tried to trace the first appearance of various symptoms, I inevitably found myself going back to my childhood or early teens. In my diary, I wrote repeatedly that I was ‘scared of losing the people in my head’ because I knew they functioned as a shield against loneliness and boredom. I feared what it would do to me to turn something so intimate into a symptom.
Ambivalence about antipsychotic treatment is typically chalked up to lack of insight or intolerance of side effects. While the significant role of both is obvious, the picture is more complicated in the case of insidious, early onset. Where there is a significant duration of untreated psychosis during periods of identity formation, insight cannot mean awareness of a morbid change in oneself. There is no sense of a sudden rupture, no clear division between symptoms and personality—or adaptation to symptoms. In such cases, possessing insight requires acceptance of the self as defective, which is a different demand from recognising a change in the self.
The widely used Positive and Negative Symptom Scale (PANSS) assesses insight by asking if there is a perceived need for prescribed medication. What strikes me about that question is that it is about need — not whether medication is seen as necessary for certain aims, effective, helpful, or simply understood to be the clinical recommendation. Need is assumed irrespective of how well it actually works for a given patient and of whether negatives outweigh positives. Presumably it is insufficient to agree it was needed in the past or to aspire not to need it in the future. It’s unclear if it refers to an internal need, or a need in the sense of being told by others or even coerced.
To me, the question does connote a profound flaw in need of correction. No allowance is made for the possibility of recovery in the sense of functioning unmedicated. Nor is any allowance made for those who don’t respond to medication, would like to change the current one, or who experience intolerable, even life-threatening side effects. I recognise that rating scales are by definition devoid of nuance and are not intended as a full picture of illness. But I also know that no detail of the questions included is left to chance. My argument is about what the operationalisation actually encodes. Treatment ambivalence is cast as a component of sickness, even where there is continued adherence. Insight is not even equated with actual compliance, only with acceptance of fundamental brokenness. The question demands acceptance of its premises, acquiescence to needing fixing regardless of the experience of that repair.
It is hardly surprising to me that higher insight in schizophrenia, in what is known as the insight paradox, is associated with higher risk of depression and lower quality of life. Insight is approached practically, if not defined, in a way that is synonymous with complete pessimism. Schizophrenia has predominantly been seen as a chronic, degenerative disorder of basic brain functioning. While that view is increasingly challenged both by neuroscience and advocates for the recovery model, it remains the assumption underlying assessment of insight in many cases. Newer multidimensional measures of insight still arguably encode the same assumption.
The only way to have hope for the future while living with schizophrenia is to, at least partially, reject or question that one is fundamentally defective. To accept the need for lifelong treatment for something innate in you, with no hope of negotiating a recovered state on your own terms, is to give in to despair.
Ambivalence about treatment in early onset, prolonged, untreated psychosis can come from the sheer abrupt shock of consensus reality. Awareness of symptoms and the suffering they bring does not negate the challenges of this transition. Nor does a good treatment response make adherence easy. Disputing the premise can be a claim about identity, not failure to recognise suffering.
It makes a tremendous difference if someone has grown and adapted to symptoms, like a tree trunk warping around a fence. The basic disturbances underlying schizophrenia can, at least sometimes, be present many years before obvious illness. The personality traits comprising schizotypy are potentially lifelong. People may learn to balance inhabiting consensus reality with their own private framework before florid psychosis emerges, carrying a long-term sense of being different. It is a mistake to assume psychotic complaints are always mere states, unwanted and unwelcome.
Where there is no clear premorbid baseline, recovery must be about constructing a new self, not restoring an old one. The remittance, partial or full, of symptoms is only the beginning of this process. Delusional beliefs can be a self-protective mechanism for coping with traumatic life experiences, so simply removing them through medication without replacement complicates coping. Constructing a vision of the future is difficult work, especially for those, like me, who do not have access to therapy or any treatment other than medication and sporadic brief outpatient checkups.
Insidious trajectories in schizophrenia are associated with worse outcomes and only partial recovery, perhaps attributable to slowness to seek help once florid symptoms emerge. Longer durations of untreated psychosis are associated with a reduced likelihood of remission and higher risk of suicide. I wonder whether that has something to do with the difficulty in accepting treatment after you’ve lived with the slow disintegration of reality for years. It’s hard to have a sensible attitude to psychosis when it has crept up on you.
Perhaps rather than asking if medication is needed to assess insight, it makes sense to ask about the ways in which someone would like their experience of reality to change, and whether they understand that this is possible. To me, that sounds like hope. Wanting to change how you live and how you relate to the world does not require immediate acknowledgement that your entire way of thinking is wrong and different from other people. That’s a big ask for anyone. But it does open the door to understanding the ways in which treatment could be beneficial and desirable. Approaching insight this way may not always be possible — after all, assessing insight is pointless if everyone passes. For insight to do discriminatory work, someone has to lack it. But what I am suggesting can be a gentler starting point for those who are distress or impaired by psychosis and aware of it.
See also:
First-Person Psychiatric Accounts
First-person accounts of mental health challenges and mental healthcare have been a defining part of Psychiatry at the Margins from the start. Many of these guest contributions are among the most widely read and appreciated posts in this newsletter.
I did not more than scan this narrative. But the theme is the nature of the author's insight and the way he experienced psychiatric treatment on a psychiatric unit.
We are in the midst of a "backlash", the public is questioning concessions to the LGTBQ+ community, affirmative action and assorted concessions to people of color, and the rights and privileges of immigrants both documented and undocumented. This suggests a process by which we "liberalized" and now we "retrench".
I would argue that in-patient psychiatry treatment never went through that first phase of liberalization. Patients who have been hospitalized are remarkably consistent about their experience: they hate it. In comparison, patients who have received antidepressants (for example) might be 75% positive about their experience and 25% negative. Rare is the patient who absolutely refuses to try any antidepressant because of his past experience, and many of these patients have secondary gain to obtain other drugs to which they are addicted, like pain patients who refuse NSAIDs.
The negativity I have observed starts at the emergency setting and continues until discharge. There is no part of the hospitalization about which patients speak positively. These patients are not exclusively psychotic or have psychotic range illnesses. Their behaviors frequently seem to suggest a preference for in-patient care to out-patient care, but that doesn't make hospitalization seem more rosy in the recollection.
There is no discussion of this in psychiatry. There are discussions about insight, adherence, cooperation, and decreasing restraint use. But zippo about patient experience or recollection. There are no user surveys like there are for ER patients, for example.
We are in the dark ages of in-patient psychiatry services. I am not trying to say we don't have modern medications, limits to confinement, or modern conceptions of what requires in-patient treatment. We have long ago abandoned confinement as a sort of social control (arguably), for example.
Nevertheless in a world with MRI's catering to those with claustrophobia, Press-Ganey surveys of ER patients and Google reviews and USNews ratings, we have managed to maintain psychiatric in-patient treatment as a sort of backwater. I would say that this is definitely a desirable frontier to explore.
"In my diary, I wrote repeatedly that I was ‘scared of losing the people in my head’ because I knew they functioned as a shield against loneliness and boredom. I feared what it would do to me to turn something so intimate into a symptom."
An aspect of hearing voices that had never occurred to me. I also imagine that these relationships provided a sense of connection and orientation in the midst of unrelenting, unresolvable, global, fear.
Thank you for sharing this glimpse into your realities past and present. Until now, psychotic states seemed totally “other” to me.