What CBT for psychosis was actually like for me

This is a composite story, drawn from common experiences shared in the schizophrenia community. It does not depict a real individual.

I am Hannah. I am thirty-one, I live in a third-floor walk-up in Brooklyn with my partner, and I have heard voices since I was nineteen. I was diagnosed with schizophrenia at twenty-two, after a hospitalisation. I am stable on aripiprazole and a small dose of an SSRI for anxiety. Two years ago I completed twenty-four sessions of Cognitive Behavioural Therapy for psychosis (CBTp). People in my support group ask me about it constantly, so I am writing this in the hope it answers some of those questions.

What I expected

I expected something soft. I had been to general talk therapy in college for stress and depression, and it had been pleasant and a bit aimless. I assumed CBTp would be like that, except about voices. I was wrong about the second part. CBTp is structured. There is homework. There is a session agenda. There is a focus on specific beliefs and behaviours. It feels less like a friendly conversation and more like learning a new framework for analysing your own experience.

What my voices were like before

For context: I had two main voices. One I called the Commentator — a male voice that narrated what I was doing in a critical tone ("she's failing again, she's going to be late"). The other I called the Director — a more frightening voice that gave commands, sometimes about self-harm. Both had been roughly the same volume and frequency for about eight years. Medication had reduced them somewhat but not eliminated them. I had built a fragile life around managing them: avoiding crowded spaces, never being alone in my apartment for long, keeping music on at night.

Finding the therapist

It took four months. My psychiatrist gave me a list of three therapists trained in CBTp. The first was not taking new patients. The second did not accept my insurance. The third had a six-week wait. I waited.

The therapist's name (in my head) was Dr R. She was about my age, blunt without being cold, and she had a CBTp manual on her desk that I noticed in the first session. She used the standard CBTp framework that descends from the work of Birchwood, Chadwick, and Kingdon (you can see overviews of the approach in the research literature).

Sessions 1-4: building the foundation

The early sessions were not about challenging my voices at all. They were about:

• My history (slowly, in pieces)
• Mapping my voices in detail — frequency, content, triggers, what made them better or worse
• Building a "formulation" — a shared written model of how my early experiences, current stresses, and beliefs about the voices fit together
• Establishing what I wanted to get out of the work

The formulation was the part that surprised me. Dr R drew it on a whiteboard during session four. There was a box for early life experiences (my mother's anxiety, a frightening event when I was sixteen), a box for the onset of voices in college, a box for my current beliefs about the voices ("they know everything about me, they are dangerous, I cannot disobey them"), and a box for what I did in response (avoid, comply, isolate). Seeing it on a whiteboard, in someone else's handwriting, made it feel like a thing rather than a doom.

Sessions 5-10: examining beliefs

This is where the real work began. The CBTp model holds that the distress caused by voices comes less from the voices themselves and more from what we believe about them. We started examining specific beliefs.

For example, the belief: "If I do not do what the Director says, something terrible will happen." Dr R did not tell me this belief was wrong. She asked: how do we know? What is the evidence for it? What is the evidence against it? Have you ever not done what the Director said? What happened?

I had, in fact, frequently not done what the Director said. Nothing terrible had happened. I had spent eight years, however, never noticing this, because the belief was so strong it dictated my interpretation of every event.

This was uncomfortable work. It was not a simple matter of "now I will believe something different." Old beliefs have grooves cut into them. We worked the same belief for weeks.

Sessions 11-16: behavioural experiments

The middle of CBTp is where you start testing things. Dr R and I designed small experiments. The first: spend fifteen minutes alone in my apartment without music. Predict what would happen (I predicted: voices would escalate, I would have to leave). Record what actually happened. Compare.

What actually happened the first time: the voices got slightly louder for about three minutes, then settled. I did not leave. I had a cup of tea. I felt strange afterwards — partly proud, partly destabilised, because the prediction had been so wrong.

We did dozens of experiments. Going to a coffee shop alone. Riding a crowded subway. Spending an evening in the apartment without my partner home. Each one chipped at a slightly different belief. None of them eliminated the voices. All of them weakened the rules I had been living by.

Sessions 17-22: building a new relationship

The last third of the work shifted from testing beliefs to building a different stance toward the voices. We worked on what's called a "voice diary" — daily notes on what the voices said and what I did, with a focus on noticing rather than reacting. We worked on responses I could give to the voices in my own head ("I hear you. I am not going to do that."). We worked on what to do when the voices spiked under stress.

I want to name something honest: I never confronted the Director directly the way some accounts of CBTp suggest. My version of the work was more sideways. I learned to coexist with him rather than to argue with him. Dr R said this was a perfectly valid path.

Sessions 23-24: relapse prevention

The last two sessions were about consolidation. We wrote a "blueprint" of what I had learned, what triggers to watch for, what to do if the voices got worse. We agreed I would come back for booster sessions every three months for the first year. We planned an end. I cried at the last session.

What changed

The voices did not go away. They are slightly less frequent but mostly the same volume. What changed:

• I no longer believe the Director can hurt me if I disobey
• I no longer organise my life around avoiding voice triggers
• I can be alone
• I can be on the subway
• I can go to a movie by myself
• The voices are background noise I can choose, mostly, not to listen to
• I have a written plan for what to do if they spike

What CBTp is not

• It is not a replacement for medication. I am still on aripiprazole. The two work together.
• It is not about telling you your voices are not real or that they are "just your imagination."
• It is not a quick fix. The full course is typically 16-30 sessions.
• It is not for everyone. Some people prefer other approaches like ACT or Avatar Therapy.

What I would tell someone considering it

Do it. If you can find a trained therapist and you can stick with the structure, it is worth the work. Do not expect the voices to disappear. Expect, instead, a different relationship with them. That is a smaller-sounding goal that turned out to be much bigger than I had imagined.

This article is for educational purposes only and is not medical advice, diagnosis, or treatment. Always consult a qualified mental health professional. If you or someone you know is in crisis, call or text 988 in the US, or your local emergency number.