5.3.1 – Rosenhan (1973) – On being sane in insane places

Description (AO1)


To investigate whether psychiatrists could distinguish 8 sane pseudopatients from the insane.

To investigate the conditions in the mental hospitals.


8 pseudopatients (5 men, 3 women) went to 12 different hospitals across 5 states in the US and reported hearing the words “empty”, “hollow” and “thud”. They chose a variety of hospitals: some old, some new, some underfunded, some wealthy, some public, some private.

Once admitted, they started acting completely normally and reporting that their symptoms had gone. They wanted to see how long it would take for the psychiatrists to realise that they weren’t insane.


All of the pseudopatients were admitted with the diagnosis of schizophrenia, except one who was admitted with manic depression.

The pseudopatients spent an average of 19 days in the institution, with the shortest stay being 7 days and the longest being 52. None of the pseudopatients were declared sane, but all of those diagnosed with schizophrenia were released with a diagnosis of schizophrenia in remission.

Many of the other patients suspected that they were sane. However, the staff didn’t spot this, and many of them attributed their normal behaviour to their illness. For example, when the pseudopatients were taking notes, the psychiatrists recorded this as “writing behaviour” and saw it as a symptom of the illness.

The patients were depersonalised by the staff, meaning they were treated as being less than other humans. When the pseudopatients tried to talk to the staff, they were ignored 71% of the time. They were only given a verbal response 2% of the time.


After this study, he agreed with another hospital that he would send more pseudopatients and they would have to guess which ones they were. In reality he didn’t send any, and out of the 193 real patients who were admitted, 41 of them were suspected by at least 1 staff member as being a pseudopatient.

Rosenhan suggested that the label caused people to treat them differently, and to misinterpret their behaviour. He used the self-fulfilling prophecy to explain this.

Evaluation (AO3)


The sample was generalisable because they used a large variety of hospitals.

However, it was only conducted in the USA, meaning that the results may not be generalisable to the rest of the world, and also may not be generalisable to today because treatment has changed significantly since then.


They didnt all follow standardised procedures. For example, one of the pseudopatients revealed he was going to become a psychologist.

11 of the 12 diagnoses were consistent, schizophrenia, and therefore it could be argued that it is reliable.


The research was used to change the world of psychological treatment. It helped to bring light to the conditions in which they were kept.


High ecological validity because it was in a realisitic setting.


Deception –> The hospitals were not informed of the experiment.

They also could have taken the attention away from people who really were mentally ill, however, as they only had an average of 6.8 minutes per day per pseudopatient of interaction, it suggests that this was not true.

However, it could be argued that this experiment had a massive impact on psychological treatment and helped to improve the standards, meaning that it could be argued that the ethical issues were outweighed by the impact of the research.