Rosenhan experiment
Editor-In-Chief: C. Michael Gibson, M.S., M.D. [1]
Background
The Rosenhan experiment was a famous experiment into the validity of psychiatric diagnosis conducted by David Rosenhan in 1972. It was published in the journal Science under the title "On being sane in insane places".
Rosenhan's study consisted of two parts. The first involved the use of healthy associates or 'pseudopatients', who briefly simulated auditory hallucinations in an attempt to gain admission to 12 different psychiatric hospitals in 5 different states in various locations in the United States. The second involved asking staff at a psychiatric hospital to detect non-existent 'fake' patients. In the first case hospital staff failed to detect a single pseudopatient, in the second the staff falsely detected large numbers of genuine patients as impostors. The study is considered an important and influential criticism of psychiatric diagnosis.
The study concluded "It is clear that we cannot distinguish the sane from the insane in psychiatric hospitals" and also illustrated the dangers of depersonalization and labelling in psychiatric institutions. It suggested that the use of community mental health facilities which concentrated on specific problems and behaviors rather than psychiatric labels might be a solution and recommended education to make psychiatric workers more aware of the social psychology of their facilities.
The Pseudopatient Experiment
For the purposes of the study, eight 'pseudopatients' (associates of Rosenhan selected to be a group of varied and healthy individuals) attempted to gain admission into psychiatric hospitals. During psychiatric assessment they claimed to be hearing voices that were often unclear, but noticeably said the words "empty", "hollow" and "thud". No other psychiatric symptoms were claimed, and apart from giving false names and employment details, further biographical details were truthfully reported. If admitted, the pseudopatients were asked to 'act normally', report that they felt fine and no longer heard voices.
The pseudopatients were: a psychology graduate student in his twenties, three psychologists, a pediatrician, a psychiatrist, a painter and a housewife. None had a history of mental illness. After being admitted, the experimental subjects acted normally and did not display any obvious psychopathology. Subjects were to remain as inpatients until they were discharged by the staff at their hospitals, who were not privy to the experiment and believed the subjects to be real psychiatric patients.
All eight were admitted, seven with a diagnosis of schizophrenia, the last with bipolar disorder. None of the pseudopatients were detected during their admission by hospital staff, although other psychiatric patients seemed to be able to correctly identify them as impostors. While the staff failed to identify sanity, in the first three hospitalisations notes of patient responses were kept and 35 of the total of 118 patients did express a suspicion that the pseudopatients were sane. All were discharged with a diagnosis of schizophrenia "in remission". Their stays ranged from 7 to 52 days and the average was 19 days.
During their stay, hospital notes indicated that staff interpreted much of the pseudopatient's behaviour in terms of mental illness. For example, one observer, apparently oblivious to the irony, labeled the note-taking of one pseudopatient as "writing behaviour" and considered it pathological.
The non-existent impostor experiment
For this experiment, Rosenhan used a well-known research and teaching hospital, whose staff had heard of the results of the initial study but claimed that similar errors could not be made at their institution. Rosenhan arranged with them that during a three month period, one or more pseudopatients would attempt to gain admission and the staff would rate every incoming patient as to the likelihood they were an impostor. Out of 193 patients, 41 were considered to be impostors and a further 42 were considered suspect. In reality, Rosenhan had sent no pseudopatients and all patients suspected as impostors by the hospital staff were genuine patients (unless they were other impostors unknown to the study, which seems unlikely). This led to a conclusion that "any diagnostic process that lends itself too readily to massive errors of this sort cannot be a very reliable one". Studies by others found similarly problematic diagnostic results.
Related experiments
Maurice K. Temerlin split 25 psychiatrists into two groups and had them listen to an actor portraying a character of normal mental health. One group was told that the actor "was a very interesting man because he looked neurotic, but actually was quite psychotic" while the other was told nothing. Sixty percent of the former group diagnosed psychoses, most often schizophrenia, while none of the control group did so.
Loring and Powell gave 290 psychiatrists a transcript of a patient interview and told half of them that the patients were black and the other half white; they concluded of the results that "Clinicians appear to ascribe violence, suspiciousness, and dangerousness to black clients even though the case studies are the same as the case studies for the white clients".
Impact
Rosenhan published his findings in Science, criticising the validity of psychiatric diagnosis and the disempowering and demeaning nature of patient care experienced by the associates in the study. His article generated an explosion of controversy.
Many defended psychiatry, arguing that psychiatric diagnosis must rely heavily on the patient's own report of their experiences. Hence, mis-diagnosis in the presence of fake symptoms no more demonstrates problems with psychiatric diagnosis than would lying about other medical symptoms. As an example, psychiatrist Robert Spitzer wrote in a 1975 criticism of Rosenhan's study:
- If I were to drink a quart of blood and, concealing what I had done, come to the emergency room of any hospital vomiting blood, the behaviour of the staff would be quite predictable. If they labelled and treated me as having a peptic ulcer, I doubt I could argue convincingly that medical science does not know how to diagnose that condition.
However, despite the perceived shortcomings of Rosenhan's study, Spitzer felt that there was still a laxness in the field. He played an important role in updating psychiatric diagnosis, eventually resulting in the Diagnostic and Statistical Manual of Mental Disorders, in an attempt to introduce more rigor and reliability.
Lauren Slater says in her 2004 book Opening Skinner's Box that she repeated Rosenhan's study, by presenting at the emergency rooms of different hospitals with a single auditory hallucination. She writes that she was not admitted to any of them but was instead given prescriptions for antipsychotics and antidepressants. Her claims were questioned by Robert Spitzer and others[2]; she replied (through her attorney) that she considered her work to be "anecdote, not systematic research, and certainly not a 'replication' of Rosenhan's study."[3]
References
- Rosenhan, D. (1973) On being sane in insane places. Science, 179, 250-8. Full text as PDF
- Slater, L. (2004) Opening Skinner's Box: Great Psychological Experiments of the Twentieth Century. ISBN 0-393-05095-5.
- Spitzer, R.L. (1975) On pseudoscience in science, logic in remission, and psychiatric diagnosis: a critique of Rosenhan's "On being sane in insane places". Journal of Abnormal Psychology, 84 (5), 442-52.
See also
- Confirmation bias
- Diagnosis
- Involuntary commitment
- Psychiatry
- Schizophrenia
- The Rosenhan experiment features in the recent BBC TV series The Trap.
- David Rosenhan on the trap Rosenhan on youtube