Notes on 'On being sane in insane places' by D.L.Rosenhan



What is sanity and insanity? Diagnostic categories are questioned. For example, psychiatrists for the defence and psychiatrists for the prosecution disagree in court over an accused person's sanity.

If sane people were not detected as such in a mental hospital, then this would suggest that the environment of the hospital, rather than the individual, is influencing the judgements of hospital staff. Eight sane people were admitted to 12 different hospitals. Once inside they acted normally.


The pseudo-patients came from varied backgrounds. There were 3 women and 5 men. All used pseudonyms. Some had to lie about their occupation, because if the hospital staff knew that they were a psychiatrist, for example, then they might be afforded special attention or caution. A variety of hospitals were used.

Pseudo-patients made an appointment, prior to presenting themselves at the admissions office. During the interview the pseudo-patients said that they heard an unfamiliar voice of the same sex as themselves. The voice said "empty", "hollow", and "thud". These symptoms were consistent with the symptoms of existential psychoses. This is where you find life meaningless, and you become obsessed with this thought. All were admitted. The pseudo-patients were surprised to find themselves admitted so easily, and initially felt ill at ease on the ward. Soon, however, they were able to act as normal. The quicker they were to be recognised as sane, the quicker they would be released.


The pseudo-patient's sanity went undetected. They spent an average of 19 days (range of 7 to 52 days) on the ward, before being released. When released, they were diagnosed as being `schizophrenic in remission' not as being sane. Some visitors and patients detected the pseudo-patients' sanity (35 out of 118 patients).

Doctors are most likely biased towards committing type-one errors. This is when one plays safe; causing a doctor to diagnose healthy people as sick more often than sick people as healthy. This is also called a false positive. A type-two error is the other way round; diagnosing sick people as healthy, and is also known as a false negative.

A follow up experiment was arranged at another hospital. They doubted that they would commit type-two errors. They were warned that over the next three months one or more pseudo-patients would present themselves for admission at their hospital. The staff was asked to make judgements on a 10-point scale. Forty-one patients were detected by at least one staff member. Twenty-three were suspected by at least one psychiatrist. A psychiatrist and one other staff member suspected nineteen. In fact, no pseudo-patients had been presented! So it seems that type-one errors can be made when the stakes are high (prestige or diagnostic ability).

The Stickiness of psychodiagnostic labels.

Once labelled as mentally ill, people will view a person as if they are still ill, even if all symptoms have disappeared. This fits in with the theories of Gestalt psychology, whereby elements are viewed in the context of the whole. This also accords with Asch's (1946) findings that there are powerful central traits. For example, whether a person is described as warm or cold has more effect on how others view them, than descriptions such as 'hard working' or 'laid back' would.

Pseudo patients could have fairly ordinary life histories; such as being close to mother than father in early childhood, and vice versa during adolescence. The medical staff would distort this information in their case notes; For example 'the patient manifests a long history of considerable ambivalence in close relationships, which begins in early childhood'. The truth becomes distorted in order to fit in with theories about the dynamics of schizophrenia.

The pseudo-patients took extensive notes recording the daily events on the ward. Fearing that the note taking would cause the staff to realise the pseudo-patient was not really a psychiatric patient, an effort was made to take notes covertly. Additionally, notes were removed from the ward each day. It was soon realised that precautions were not necessary, as the note-taking was seen by the staff as part of the pseudo-patients' symptoms.

Normal behaviour was interpreted as abnormal. A pseudo-patient pacing up and down was asked if he was nervous, when really he was bored. A patient might go berserk, because of being mishandled by the staff, but the medical staff would blame the behaviour on a recent visit from a relative or friend!

Patients would have very little to look forward to, so might queue outside the refectory half an hour before food was to be served. One psychiatrist described this behaviour as demonstrating the `oral-acquisitive nature of their syndromes'.

The Experience of Psychiatric Hospitalization.

Before patients were called `Mentally ill', over 200 years ago, they were thought to be possessed by evil spirits. The classification of being mentally ill should have put mental illnesses on a par with physical illnesses. Unfortunately, mental illness, unlike physical illness is still treated with little sympathy, and believed to be a chronic (long lasting) affliction. The general public seem to avoid mentally ill people. It is surprising to find, however, that medical staff who work with mentally ill people also seem to avoid them. Medical staff spent much of their time in their own areas, where patients were not allowed. Psychiatrists spending even less time than nurses, and nurses spent less time with patients than attendants. Attendants, who are supposed to be in contact with the patients much of the time, spent nearly all of their time in the staff area!

An experiment was conducted to test for the level of student and staff interaction. The Pseudo-patients asked staff "Pardon me, Mr [or Dr or Mrs] X, could you tell me when I will be presented at the staff meeting?" or "...when am I likely to be discharged?". As can be seen from the table, many psychiatrists ignored the question, or failed to respond to it sympathetically. Even more worrying, is the fact that even more nurses and attendants were as unhelpful.

As a control, a young person approached faculty members at a university asking for directions. She was helped on every occasion. This means we can discount the explanation that psychiatrists are too important to talk to the patients. The young person was helped on every occasion when she asked for a doctor whilst in the university medical centre. However, when she asked for a psychiatrist, 22% gave a minimal amount of interaction.

Percentage of people who interact with patients or control



Nurses & Attendants


Looking for a psychiatrist

Looking for a doctor







Moves on head averted %






Makes eye contact (%)






Pauses and chats(%)






Stops and talks (%)






Powerlessness and depersonalisation

The patients were powerless. Some patients were physically beaten for initiating a conversation with an attendant. Freedom of movement is restricted. Privacy is not respected, even for toilet visits! Patients would sometimes be beaten in the presence of other patients, but never in the presence of other staff members. It would seem that patients would not make credible witnesses!


The above evidence suggests that we are not that sure as to what a mental illness really is, and the categorisation of an illness is not easy as a result. It would seem that labelling plays a part in how `patients' are treated. They are seen as incurable, and treated as if sub-human. A broader concern is if we are so uncertain as to what constitutes a mental illness, this could lead to miscarriages of justice and the abuse of human rights. A plea of insanity might help a criminal escape a harsher punishment than treatment in a psychiatric hospital would afford. Labelling a political opponent as insane, might be a convenient way of suppressing him or her.

This experiment has recently been partially replicated Click here for Guardian article

Return to Gary Sturt's Home page _