CBT for delusions: how therapists work with strongly held beliefs

Of all the things CBT for psychosis (CBTp) does, working with delusions is probably the most misunderstood — both by patients who imagine the therapist will spend forty-five minutes a week trying to argue them out of a belief, and by clinicians outside the field who assume nothing useful can be done. In practice, careful CBT work with delusions looks very different from either of those pictures. It is slow, collaborative, and rooted in a long tradition stretching back to Aaron Beck's earliest case reports.

A brief history

Aaron Beck published the first case study of cognitive therapy for a delusion in 1952 — a paranoid patient who, over months of structured conversation, came to consider alternative explanations for events he had previously interpreted as a coordinated conspiracy. The work largely sat dormant until the 1990s, when British clinicians including David Kingdon and Douglas Turkington, and separately Philippa Garety, Daniel Freeman, and Elizabeth Kuipers, built it into a manualised therapy. Their textbooks (Kingdon & Turkington's Cognitive Therapy of Schizophrenia, Morrison's A Casebook of Cognitive Therapy for Psychosis) remain core references today.

The NICE guideline on psychosis and schizophrenia (NICE CG178) — available at nice.org.uk/guidance/cg178 — recommends offering CBTp to all adults with psychosis or schizophrenia, in part because of consistent evidence for its effect on delusional distress.

What a delusion actually is

The textbook definition is "a fixed, false belief held with conviction despite contradictory evidence." That is a useful shorthand but a misleading one. Real delusions sit on a continuum:

• Conviction — how strongly the person holds the belief
• Preoccupation — how much of their attention it consumes
• Distress — how upsetting the belief is
• Behavioural impact — how much it shapes daily action

CBTp targets all four dimensions. A patient may end therapy still believing that a neighbour is monitoring them — but holding the belief with less certainty, thinking about it less often, feeling less afraid, and going about a normal day rather than barricading the windows. That is a clinically meaningful change.

The first job: engagement

Almost every CBTp manual emphasises that the first three or four sessions should not address the delusion at all in any direct way. The therapist asks about the patient's life, their concerns, their goals, what is working and what is not. They listen to the delusional content respectfully without endorsing or challenging it. For many patients this is the first time a clinician has shown that kind of sustained, non-judgmental interest. The relationship that is built in this phase is what makes the later work possible.

Formulation: how the belief came to make sense

Once trust is established, therapist and patient construct a shared formulation — a story of how the belief developed. This usually includes early experiences (sometimes trauma), recent triggers, the emotional state at the time the belief crystallised, and the way subsequent events have been interpreted in its light. The formulation is collaborative; it is not handed down from the therapist. Many patients find this the single most useful part of the work, because it places their experience in a recognisable human pattern.

Working with the belief itself

Several techniques are used, almost always in combination.

1. Peripheral questioning

Rather than challenging the central belief ("the neighbours are spying on me"), the therapist explores its edges. How would the spying actually be done? What evidence convinced you first? What would have to be true for it to be wrong? The aim is not to "win" but to understand the belief in detail and surface places where the patient may already feel some uncertainty.

2. Generating alternative explanations

The therapist may eventually offer — tentatively, and only with permission — other ways of understanding the same observations. "Some people who have been through what you've been through find that ordinary events start to feel meaningful in a way they didn't before. Could that fit any of what's happening?" The patient is not asked to accept the alternative; they are asked to hold it alongside their original belief as one possibility among others.

3. Examining evidence

Together they look at the evidence for and against the belief — not as a courtroom exercise but as a way of teaching the patient a habit of evaluation. Beck described this as moving from fixed conviction to tentative hypothesis.

4. Behavioural experiments

If the patient is willing, small behavioural experiments can directly test predictions the belief generates ("if I leave the curtains open the people watching will retaliate within a week"). This is done collaboratively and slowly, with safety planning.

5. Working on the meaning, not the content

Often the most important question is not "is this belief true?" but "if it were true, what would it mean about you, your safety, your worth?" Many delusions encode fears about being targeted, exposed, worthless, or special. Working on these underlying meanings can reduce distress even while the surface belief remains.

What therapists do not do

• Argue, contradict, or "reality-test" aggressively
• Pretend to agree with the belief
• Promise that the belief will go away
• Push behavioural experiments the patient is not ready for
• Treat the patient's belief as evidence of stupidity or weakness

Special situations

Paranoid delusions

Often respond well to a combination of normalising (see normalising psychotic experiences), examining safety behaviours, and graded behavioural experiments. Daniel Freeman's group at Oxford has built specific protocols around persecutory delusions with strong outcome data.

Grandiose delusions

Trickier, because the belief itself is often pleasurable. The therapist generally focuses on the costs and consequences, and on any underlying low self-esteem the grandiosity may be defending against.

Religious or culturally-rooted beliefs

Therapists are taught to distinguish between beliefs that are shared by the person's community (which are not delusions) and beliefs that are idiosyncratic and impairing. Cultural humility matters here.

What the evidence shows

Meta-analyses, including a frequently cited 2014 paper by Mehl, Werner, and Lincoln in Frontiers in Psychology, show small-to-moderate effect sizes for CBTp on delusional severity, with larger effects on distress and conviction than on the presence of the belief itself. A 2023 update by Sitko and colleagues, indexed at PubMed, broadly supports the same pattern. NICE concluded that the effect is large enough, and the harms small enough, that CBTp should be routinely offered.

For people considering this work

If you are thinking about CBTp for delusional beliefs, a few things help:

• Look for a clinician with explicit CBTp training, not generic CBT
• Expect the early sessions to feel slow — this is intentional
• You do not need to "give up" any belief to start the work
• Keep your prescriber in the loop; CBTp works best alongside medication for most people

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.