You are viewing a free preview of this lesson.
Subscribe to unlock all 11 lessons in this course and every other course on LearningBro.
Every Edexcel clinical-psychology topic is anchored by a prescribed classic study, and for Clinical Psychology that study is David Rosenhan's (1973) "On being sane in insane places", published in Science. It is the single most influential critique of psychiatric diagnosis ever conducted, and it is required knowledge: you must be able to describe its aims, method, results and conclusions precisely, and evaluate it rigorously. Rosenhan asked a question that sounds almost naive but turned out to be devastating — can the sane be distinguished from the insane? — and answered it by sending healthy volunteers into psychiatric hospitals to find out whether they would be detected. They were not. The study exposed how far diagnosis depended not on the patient but on the context and the label, and it became a catalyst for the operationalised, criteria-based diagnosis that followed. This lesson treats the study as an object of careful scientific and ethical analysis; the people at its centre — both the pseudo-patients and the genuine patients whose experiences Rosenhan reported — are discussed throughout with the seriousness the subject demands.
Key Definition: A pseudo-patient is a healthy participant who presents at a psychiatric facility feigning a symptom in order to test whether, and how, they are diagnosed and treated. Labelling is the process by which a diagnostic category, once applied, comes to dominate the interpretation of everything the labelled person subsequently does.
This lesson addresses the Edexcel 9PS0 — Paper 2, Topic 5: Clinical Psychology requirement to study the topic's classic study, Rosenhan (1973), "On being sane in insane places", in full: its aims, procedure (the pseudo-patient study and the follow-up study), results and conclusions about the (in)validity of psychiatric diagnosis and the effects of labelling and institutionalisation; and its detailed evaluation — methodological (design, controls, generalisability, replicability) and ethical (deception, consent, protection of staff and genuine patients) — together with its bearing on the reliability and validity of diagnosis. It builds directly on the reliability-and-validity lesson, where Rosenhan appears as evidence, and it supplies the classic half of the classic/contemporary study pairing that the next lesson completes. In assessment-objective terms, you should be able to describe the study accurately (AO1), apply its findings to scenarios or to debates about diagnosis (AO2), and evaluate it methodologically and ethically, reaching a reasoned judgement about what it does and does not show (AO3).
Connects to…
By the early 1970s, unease about psychiatric diagnosis had been building. Critics argued that diagnostic categories were unreliable (clinicians frequently disagreed) and that they might not correspond to anything real in the person — that "mental illness" could be, at least in part, a judgement about deviance rather than a medical fact. Rosenhan set out to test this empirically rather than merely argue it. His central aim was to investigate whether the sane can be distinguished from the insane in psychiatric hospitals — that is, whether the normality or abnormality of a person resides in the person themselves (in which case healthy people should be detected) or in the context and the labels that observers bring to bear (in which case healthy people might be admitted, held and interpreted as ill).
A second, connected aim emerged from the design: to examine the consequences of psychiatric labelling and hospitalisation — what it is like to be a patient in such an institution, how a diagnostic label shapes the interpretation of ordinary behaviour, and how staff and patients relate to one another. The study is therefore two things at once: a test of the validity of diagnosis and a piece of qualitative, observational field research into the experience of institutionalisation.
| Aim | Question it addresses |
|---|---|
| Can the sane be distinguished from the insane? | Does normality reside in the person, or in the diagnostic context and labels? |
| What are the consequences of labelling and hospitalisation? | How does a diagnostic label affect the interpretation of behaviour and the treatment of the patient? |
Rosenhan's design was, in effect, a field experiment combined with covert participant observation. The pseudo-patients gained admission to hospitals and then observed and recorded life on the wards from the inside.
Eight sane pseudo-patients took part — a mixed group including Rosenhan himself, and comprising (by his account) a psychology graduate student, three psychologists, a paediatrician, a psychiatrist, a painter and a housewife; three were women and five were men. Between them they sought admission to twelve hospitals across five US states, on both coasts, ranging from old and under-resourced to modern and well-staffed, including one private hospital. This spread was deliberate: it meant the findings could not be dismissed as a quirk of one poor institution.
Each pseudo-patient telephoned a hospital, arranged an appointment, and arrived at admissions complaining of a single, carefully limited symptom. They said they could hear an unfamiliar voice, of the same sex as themselves, which was often unclear but seemed to be saying the words "empty", "hollow" and "thud". These words were chosen because they suggested an existential crisis — a concern about the meaninglessness of one's life — and, crucially, because no study of such symptoms existed in the literature, so the "symptom" could not correspond to any recognised diagnosis.
Beyond this single fabricated symptom, everything the pseudo-patients reported was truthful. They gave false names and (where relevant) false occupations, to protect their careers and prevent special treatment, but their accounts of their life histories, relationships, frustrations and joys were entirely genuine, and described a life without significant pathology. The design point is important: apart from one auditory symptom and a false name, the pseudo-patients presented a normal history to the admitting clinician.
Once admitted, each pseudo-patient stopped simulating any symptom whatsoever. They reported that the voices had gone, behaved normally on the ward, and cooperated fully — with one striking exception: they did not swallow their medication (which they discarded), and they spent their time observing and making notes on ward life, staff and other patients. Their task was simply to behave sanely and to secure their own release by convincing staff they were well enough to be discharged.
flowchart TD
A["8 sane pseudo-patients<br/>telephone 12 US hospitals"] --> B["At admission: report ONE symptom<br/>unfamiliar voice — 'empty', 'hollow', 'thud'"]
B --> C["All other history reported TRUTHFULLY<br/>(only name/occupation falsified)"]
C --> D["Admitted to a psychiatric ward"]
D --> E["Symptom simulation STOPS immediately<br/>behave normally; observe and take notes"]
E --> F["Seek discharge by convincing staff they are well"]
style D fill:#e74c3c,color:#fff
style F fill:#27ae60,color:#fff
The findings were consistent across the very different hospitals and are worth stating precisely.
Alongside these headline findings, the pseudo-patients' observational notes documented the experience of institutionalisation: powerlessness and depersonalisation (patients were often treated as though they were not present — for example, staff discussing them within earshot, or attending to them with minimal eye contact or engagement), a profound imbalance of power between staff and patients, and remarkably little direct contact with psychiatric staff. Rosenhan also reported a small quantitative observation on staff avoidance: when pseudo-patients approached staff with a polite, neutral question, they were frequently ignored or answered while the staff member kept walking, without eye contact — a pattern he contrasted with how the same question would be answered in an ordinary setting.
The pseudo-patient study showed that healthy people were admitted and not detected — a failure to recognise the absence of illness (a false positive in diagnostic terms: calling a well person ill). Rosenhan anticipated the natural objection, that hospitals might sensibly err on the side of caution and admit doubtful cases, and designed a second study to test the reverse question: could staff be induced to over-detect and wrongly reject genuine patients as impostors?
A teaching and research hospital, whose staff had heard of the first study and insisted that such errors could not occur in their institution, provided the setting. Rosenhan informed the staff that, at some point during the following three months, one or more pseudo-patients would attempt to gain admission. Staff were asked to rate, on a confidence scale, how likely it was that each patient presenting for admission was in fact a pseudo-patient. In reality, Rosenhan sent no pseudo-patients at all — every person who presented during the period was a genuine patient seeking help.
Out of 193 patients admitted for treatment in that period:
Since none of them was a pseudo-patient, every one of these judgements was an error — this time a false negative for illness, or equivalently a false accusation of sanity: genuine patients wrongly suspected of faking. Taken with the first study, the pattern is decisive: the diagnostic process could be pushed either way. In the first study, expecting illness, staff saw illness where there was none; in the second, primed to expect deception, they saw deception where there was none. In neither case did the judgement track the actual state of the person.
| Study | What was tested | Key finding | Type of error exposed |
|---|---|---|---|
| Pseudo-patient study | Can staff detect the sane? | 8/8 admitted; 11/12 diagnosed schizophrenia; discharged "in remission" | False positive — the well judged ill |
| Follow-up study | Can staff be made to over-detect? | 41/193 genuine patients suspected of faking (no pseudo-patients were sent) | False negative for illness — the ill judged well/fakers |
Rosenhan drew several linked conclusions, and being able to state them clearly is essential.
The overarching message is that we cannot, from behaviour and diagnosis alone, be confident that we are identifying real, distinct illness accurately — a conclusion that fed directly into the reform of diagnostic practice.
graph TD
CTX["Context + expectation<br/>drive clinical judgement"] --> ADMIT["Sane admitted &<br/>diagnosed (study 1)"]
CTX --> REJECT["Genuine patients suspected<br/>of faking (study 2)"]
ADMIT --> LABEL["'Sticky' label reinterprets<br/>all subsequent behaviour"]
LABEL --> INST["Institutionalisation:<br/>depersonalisation, powerlessness"]
ADMIT --> VALID["Diagnosis lacks validity:<br/>sane ≠ reliably distinguished from insane"]
REJECT --> VALID
style VALID fill:#7c3aed,color:#fff
Because this study is examined chiefly as evidence about diagnosis, you should be able to state precisely what it shows about each property.
Subscribe to continue reading
Get full access to this lesson and all 11 lessons in this course.