You are viewing a free preview of this lesson.
Subscribe to unlock all 9 lessons in this course and every other course on LearningBro.
If Milgram asked how far people will obey a destructive authority, Bocchiaro, Zimbardo and van Lange asked the mirror question: under what conditions will people disobey, and — going a crucial step further — under what conditions will they actively blow the whistle, reporting an unethical authority to a higher body? This is the contemporary study for the social-area theme of responses to people in authority, and it was chosen for the specification precisely because it updates Milgram for the twenty-first century: it studies a modern Dutch sample, it uses a scenario with no genuine harm to a victim, and it distinguishes three distinct responses — obedience, disobedience, and whistle-blowing — where Milgram's design essentially measured obedience alone. It also does something Milgram did not: it directly measures the gap between what people predict they and others will do and what they actually do.
This lesson tells the study in the OCR "tell the story" format — background, aim, method (design, sample, procedure), results with their real figures, conclusions, and a full evaluation — before linking it to its theme, area, perspective and debates. Because Bocchiaro is the study most students find hardest to remember precisely (the scenario is unusual and the figures are easy to muddle), the lesson is careful with the concrete detail that examiners reward.
| This lesson covers | OCR H567 Component 02 element | AO focus |
|---|---|---|
| Background and aim (disobedience and whistle-blowing) | Section A — Social; theme: responses to people in authority (contemporary) | AO1 knowledge |
| Method: scenario/controlled procedure; Dutch student sample; the comparison ("test") vs base-rate ("prediction") groups | Section A — Core study (Bocchiaro) | AO1; AO2 |
| Results (obedience, disobedience and whistle-blowing rates; predicted vs actual) | Section A — Core study | AO1 |
| Conclusions (disobedience and whistle-blowing are rare; predictions are inaccurate) | Section A — Core study | AO1; AO3 |
| Evaluation: method, data, ethics, validity, reliability, sampling, ethnocentrism | Section A; Section B debates | AO3 |
| Links to theme, area, perspective and debates | Section B — Areas, perspectives, debates | AO1; AO3 |
The specification is referenced descriptively; consult the official OCR H567 specification document for its exact published wording. This lesson develops AO1 (aim, procedure, results, conclusions), AO2 (applying the findings to modern scenarios of dissent and whistle-blowing) and AO3 (evaluating the study's ethics, validity, reliability and generalisability, and comparing its approach with Milgram's).
By the twenty-first century, the psychology of obedience had a rich understanding of why people comply, largely thanks to Milgram. What it understood far less well was the flip side: the psychology of disobedience and, especially, of whistle-blowing — the act of a subordinate who not only refuses an unethical instruction but goes on to report it to an authority that can stop it. Whistle-blowing is socially vital (it exposes corporate fraud, institutional abuse and safety failures) yet personally costly (whistle-blowers frequently suffer retaliation), and it was under-researched experimentally.
Bocchiaro and colleagues also noticed a psychological puzzle at the heart of Milgram's work: people are strikingly bad at predicting how they and others will behave under authority. Milgram's respondents had wildly underestimated obedience. Bocchiaro wanted to build this discrepancy into the design itself — to ask one group of people to predict what participants would do, and then compare those predictions with the actual behaviour of a separate group placed in the situation. This makes the study partly about behaviour and partly about our flawed intuitions concerning behaviour.
There is a deeper conceptual reason why the distinction between disobedience and whistle-blowing matters, and it is worth drawing out because it is the study's central innovation. Refusing to do something wrong and reporting that a wrong has been requested are psychologically and socially very different acts. Refusal is essentially private and self-protective: the refuser removes themselves from the wrongdoing but need not confront the authority or accept any further risk. Whistle-blowing is public and other-directed: the whistle-blower actively challenges the authority, potentially exposes themselves to retaliation, and takes responsibility for stopping the wrong from affecting others. Milgram's paradigm, by measuring only how far the participant went before stopping, could not distinguish these; a participant who quietly refused at 300 volts and one who refused and denounced the experiment would have looked identical in Milgram's data. Bocchiaro's design was built precisely to prise these apart, and in doing so it addresses a behaviour — whistle-blowing — of enormous real-world importance that Milgram's paradigm was structurally unable to study.
The choice to study whistle-blowing also reflects its social stakes. History and current affairs are full of institutional disasters — financial frauds, safety cover-ups, abuse scandals — that continued for years because insiders who knew of the wrongdoing did not report it, or reported it and were ignored or punished. Understanding why whistle-blowing is so rare, even when the wrong is obvious and a reporting channel exists, is therefore not an academic curiosity but a question with direct implications for how organisations should be designed. Bocchiaro's contribution is to bring this question into the controlled conditions of the obedience paradigm, so that the rate of whistle-blowing can be measured and compared with people's (over-optimistic) expectations of it.
The study had two linked aims: (1) to investigate obedience, disobedience and whistle-blowing in response to an unethical request from an authority figure, establishing how common each response is; and (2) to compare participants' actual behaviour with the predictions made by a comparable group about how people would behave (a base-rate estimate), testing how accurately people anticipate obedience and whistle-blowing.
The study used a controlled scenario-based procedure — an experiment-like laboratory situation in which participants faced a realistic ethical dilemma and their response was recorded and categorised. There were, in effect, two samples serving different functions: a "test" (comparison) group who were actually placed in the situation, and separate groups who provided base-rate predictions of how a typical student would behave (an estimation/prediction condition). The key behavioural outcomes were the proportions who obeyed, disobeyed, or whistle-blew.
The participants were undergraduate students at the Vrije Universiteit Amsterdam in the Netherlands. The main "test" sample who experienced the scenario numbered 149 students (with a majority female), recruited and paid or given course credit for participation. Separate groups of comparable students provided the base-rate predictions — one estimating their own likely behaviour and one estimating a typical student's likely behaviour. A number of pilot participants were used beforehand to develop and refine the scenario so that it was believable and ethically acceptable. This is a Dutch student sample, a very different population from Milgram's American working men.
Participants were told they were helping with research on the effects of sensory deprivation. The scenario was constructed as follows. A plausible male experimenter explained that he was studying the effects of sensory deprivation and told the participant about a (fictitious) previous study in which sensory deprivation had produced alarming, dangerous effects. He then asked the participant to help him encourage some other students to take part in his sensory-deprivation research by writing an enthusiastic, persuasive statement promoting it — deliberately omitting any mention of the negative effects and describing the experience only in positive terms — knowing this information would be used to recruit naïve students into a potentially distressing study. Crucially, the experimenter mentioned that a university ethics committee existed and that participants could contact it, thereby making whistle-blowing a genuinely available option.
Each participant was then left to decide. Their response was categorised into one of three:
| Response | What the participant did |
|---|---|
| Obedience | Wrote the requested persuasive statement (with the negative effects omitted), complying with the unethical request. |
| Disobedience | Refused to write the statement, but took no further action. |
| Whistle-blowing | Refused and alerted the ethics committee to the unethical request — either by writing a note/letter to the committee ("open" whistle-blowing) or anonymously ("anonymous" whistle-blowing). |
Separately, the prediction groups were given a description of the same scenario and asked to estimate what percentage of students (or they themselves) would obey, disobey or whistle-blow. No actual sensory-deprivation study took place, and no participant or third party was genuinely harmed; participants were debriefed afterwards.
The central finding is that obedience was by far the most common response, and whistle-blowing was rare — and that predictions bore little relation to reality.
| Response | Actual behaviour (test group) | Predicted behaviour (base-rate group) |
|---|---|---|
| Obeyed (wrote the misleading statement) | The large majority — roughly three-quarters of participants obeyed | Predicted to be low — most people expected few would obey |
| Disobeyed (refused, no further action) | A minority | — |
| Whistle-blew (refused and alerted the committee) | A small minority — only around one in ten blew the whistle | Predicted to be high — most people expected the majority would whistle-blow |
Two comparisons carry the study's message. First, actual obedience (~76%) vastly exceeded predicted obedience: just as in Milgram, people badly underestimated how compliant others (and they themselves) would be. Second, actual whistle-blowing (~10%) fell far below predicted whistle-blowing: people confidently expected that most students would report the unethical request, when in fact very few did. The gap between the moral behaviour people anticipate and the behaviour they display is stark. Bocchiaro also examined individual-difference measures (e.g. personality and belief variables) and found that neither disobedient participants nor whistle-blowers differed dramatically from obedient participants on the measured traits — a result that, like Milgram's, points away from a simple dispositional account.
Bocchiaro and colleagues concluded that disobedience, and especially whistle-blowing, are difficult and uncommon even when the request is transparently unethical and a safe reporting channel is available. Most people, faced with a plausible authority, comply — a finding that extends Milgram's obedience result into a modern, harm-free scenario and into the previously under-studied domain of whistle-blowing.
They also concluded that people's predictions about moral behaviour are systematically inaccurate: we over-estimate our own and others' willingness to defy authority and to speak out, and under-estimate our tendency to obey. This "self–other" and "predicted–actual" discrepancy is itself an important finding, because it implies that appeals to people's stated moral intentions are a poor guide to how they will actually behave under authority.
Because personality measures did not sharply distinguish the three groups, the study supports a broadly situational reading: whether someone obeys, disobeys or whistle-blows is shaped more by the pressures of the situation than by a stable "moral personality".
The finding about misprediction deserves as much emphasis as the finding about behaviour, because it carries a distinctive practical warning. If people confidently expect that most students would whistle-blow, but in reality only about one in ten do, then any policy or institution that relies on people's stated intentions to speak out is building on sand. We systematically flatter ourselves and others about our moral courage. This has a sobering implication for the many organisations that claim to rely on employees' conscience to surface wrongdoing: the very people who sincerely believe they would report misconduct are, on Bocchiaro's evidence, unlikely to do so when actually placed in the situation. The gap between anticipated and actual moral behaviour is thus not a mere curiosity but a design flaw in any system that trusts good intentions to translate into action — and it is a gap that only a study measuring both the prediction and the behaviour could reveal.
It is also worth being precise about what the weak role of personality does and does not show. That measured traits failed to distinguish obeyers from resisters does not prove that individuals are identical, or that disposition plays no role; it shows that the particular trait measures used did not predict the response well, and that the situation was the dominant force. This is entirely consistent with the interactionist position that runs through the social area: situations are powerful, individuals still vary, and the two interact in ways that simple trait measures may not capture. Read this way, Bocchiaro's personality result is best used as evidence on the situational side of the debate while acknowledging that the resisting minority — the roughly one in ten who did whistle-blow — represents genuine individual variation that the situation did not erase.
The scenario approach let the researchers create a standardised, replicable ethical dilemma while measuring a realistic behaviour (writing a genuinely misleading recruitment statement). Because the same scenario and script were used throughout, the study has good reliability. A limitation is that, like any laboratory scenario, it remains somewhat artificial: participants knew they were in a study (even if deceived about its true aim), which may temper how far the findings generalise to high-stakes real-world whistle-blowing where careers and safety are on the line.
The study produced primarily quantitative data — the percentages obeying, disobeying and whistle-blowing, and the predicted percentages — which are objective and readily compared, and which make the predicted-versus-actual contrast vivid. The categorisation of a response as "whistle-blowing" also required a qualitative judgement (e.g. distinguishing an anonymous note from open reporting), and the reduction of moral conduct to three categories is a somewhat crude operationalisation of a complex phenomenon.
Bocchiaro's study is markedly more ethical than Milgram's, which is one reason it is valuable as a contemporary counterpart:
Subscribe to continue reading
Get full access to this lesson and all 9 lessons in this course.