Episode guides

Episode guide What Makes a Belief a Delusion? Delusions, Thought Disorder, the TALD Scale

Jan 2026 · 22 key moments

Key moments, in order

  1. When you can reasonably identify a classical primary psychotic disorder or a serious bipolar disorder, tapering is quite dangerous; those conditions generally need lifetime medication.

  2. A syndrome is not an arbitrary collection of reported symptoms. Look at how the patient relates to and describes the symptoms themselves.

  3. Reserve delusions, hallucinations, negative symptoms and disorganization for primary psychotic disorders, and use neutral descriptive language until you have a reasonable degree of medical certainty.

  4. Defining psychosis as loss of contact with reality is too broad and nonspecific. The field needs more precise language over time.

  5. Mania can mimic psychotic features: distractible, rushing thoughts and speech can look like thought disorder.

  6. Many residents are taught only three or four levels of disorganization: linear, circumstantial, tangential. That is not sufficient.

  7. The Thought and Language Disorder scale (TALD) walks through the objective positive phenomena one by one: derailment, tangentiality, circumstantiality, crosstalk, logorrhea, pressured speech, paraphasias, neologisms, perseveration, poverty of content and more.

  8. Read the TALD even if you disagree with some of its definitions. If you believe at all in predictive coding, you cannot recognize something until you know it is there to be recognized.

  9. Challenge, but not as a true challenge: with the air of disbelief of someone who really wants to understand more and believe.

  10. 'Fixed false belief' is not enough, and the DSM-5 wording (fixed beliefs not amenable to change in light of conflicting evidence) is not a particularly good definition either. A delusion does not have to be false or fixed. It has to be relatively impervious to reasoning.

  11. Substance-induced and mood-induced delusions are still in the psychotic realm. They are just non-chronic.

  12. Do not rate a belief by how bizarre it is, because truly psychotic people can have ordinary, non-bizarre delusions. What is left is how the person came to the belief.

  13. Religious and cult beliefs need care: people who join new religions and cults tend to have more psychosis, more mental illness and more cluster A traits.

  14. Do not take statements at face value. Ask how they came to understand this to be true, what experiences led to it and when they started to believe or suspect it, and how they respond to alternatives. How did you come to understand that you are Jesus Christ?

  15. Do not take what patients say as an accurate picture of their inner state. Someone with schizophrenia may start denying the delusion mid-interview because they can still read your social judgment, and people with chronic psychosis learn not to talk about certain things. That is social learning, not remission.

  16. The more it arrives as a sudden, quasi-religious insight into a strange or highly valued idea, the more likely it comes from a psychotic disorder.

  17. When a patient holds a rigid theory about their own illness, overvalued idea goes at the top of the differential. People develop all kinds of wild theories about their illnesses.

  18. Have the patient walk through the exact last time, starting from the moment before, step by step and moment to moment, in hallucinatory detail. There is no time for this in medication management, but there is in therapy and the forensic interview.

  19. With the strongest delusions there is a short circuit of reasoning: outside information does not get absorbed and processed. A patient who can take in what you say and consider alternatives is not showing that.

  20. Do not argue with overvalued ideas or delusions, or challenge them head on. Some ways of framing a belief keep you stuck and others give you a chance to progress, so choose the frame that lets you keep working. This applies broadly, including to how patients understand themselves.

  21. Unless it directly interferes with a key or emergent part of the treatment plan, do not work on it much. Acknowledge that it is there, and if it distresses them, talk a little about that.

  22. Unless you detect an emergency, psychosis is technically not an emergency. If the patient is unsure about treatment or you are unsure of the diagnosis, forge the alliance and gather information before starting a medication; random medications cause side effects, and without an alliance they will not be taken.

Each moment is a point made in the episode, linked to the second it was said. The topic links open everything the podcast has said on that topic.

Educational content only. Everything here is what was said on the Psychofarm podcast, for clinicians. It is not medical advice and not a substitute for clinical judgment, current labeling or guidelines. An episode reflects what we thought when we recorded it, so check anything before you rely on it.