The Great Pretender

Susannah Cahalan

4 ideas

  1. Rosenhan's pseudopatient study and its missing data

    In 1973 psychologist David Rosenhan reported in Science that eight sane people, including himself, got admitted to twelve psychiatric hospitals by claiming to hear voices say words like 'empty,' 'hollow,' and 'thud.' Nearly all were diagnosed with schizophrenia, held for an average of 19 days, and released as 'in remission,' which he took as proof that psychiatry could not tell sane from insane. Decades later Cahalan could confirm only a few of the pseudopatients, and Rosenhan's own hospital records showed he had reported more severe symptoms than the paper described, including distress serious enough to justify admission.

  2. Excluding the pseudopatient whose experience contradicted the thesis

    Harry Lando, one of the few pseudopatients Cahalan found, had a largely positive stay and saw staff as caring. Rosenhan dropped Lando's data from the published study, reportedly on procedural grounds, while keeping cases that supported the conclusion that institutions were dehumanizing. The episode shows how quietly removing one data point can turn an ambiguous finding into a clean indictment.

  3. A shaky study reshaped real psychiatric policy

    Cahalan argues that 'On Being Sane in Insane Places' had outsized influence despite its thin evidence. It fed the push to make diagnosis more reliable through the checklist-style criteria of DSM-III, and it strengthened the case for closing psychiatric hospitals. Unverified findings that reach policy can shape care for millions long before anyone checks the underlying records.

  4. Stories that confirm suspicions escape scrutiny

    A study that dramatizes what an audience already suspects, such as 'psychiatrists can't really tell who's sick,' spreads because it feels true, so few people ask for the raw data. It then gets cited in textbooks and replication is rarely attempted. To judge a famous finding, ask who has actually seen the primary records, not how often the finding is repeated.

Save and mark ideas in the app