Two recent studies shed light on how “just following orders” affects human behaviour

14 July 2025

An Israeli soldier shoots dead an unarmed Palestinian child as he walks to school¹. A burly male police officer slams a petite woman against a wall, pinning her in a chokehold, and then trips her to the ground and sits on top of her – all because she was not wearing a facial covering which is completely ineffective at preventing viral infection (and for which she had a valid medical exemption, as her male companion repeatedly pointed out during the entire ordeal). An employee of the pharmaceutical behemoth Merck emails colleagues a list of doctors who have criticised Merck’s anti-inflammatory drug Vioxx – which the company knew caused double the background rate of heart attacks, before it was released onto the market – with instructions to “neutralise” and “discredit” those doctors, adding “We may need to seek them out and destroy them where they live.”

What is the thread linking each of these incidents? All three of the people who committed these acts that flagrantly defy basic morality were following orders. You know, the Nuremburg defence. And, according to two recently-published studies, not only do most people significantly overestimate how morally courageous they would likely be if they were placed under coercion to carry out unethical orders; they also experience a diminished sense of agency (the feeling of being responsible for one’s actions) when obeying such orders. Put bluntly, most people believe they would defy immoral commands but when they’re actually given such commands, they obey them while excusing themselves from blame because they were “just following orders”. Well, that’s depressing, isn’t it?

Both of the recent studies have connections with arguably the most famous and influential social psychological experiments of all time, the Milgram studies of obedience to authority, so let’s begin with a detailed explanation of these experiments.

The Milgram experiments

The original experiment was conducted in 1961 by Stanley Milgram, an assistant professor at Yale University. Milgram recruited 40 socioeconomically diverse men from the town of New Haven, Connecticut, telling them that they would be participating in a study of memory and learning. On arrival at the impressive-looking Yale psychology laboratory, each experimental subject met two people: a stern-looking man dressed in a grey technician’s coat, who oversaw the experiment, and a friendly middle-aged man who introduced himself as a fellow subject.

The experimenter explained to the two men that they would be participating in a study to test the effects of punishment on learning. One man, the learner, would have to memorise a series of word pairs. Then the other man – described as the teacher – would say the first word in each pair, and the learner would have to choose the correct ‘matching’ word in the pair, from a list of four words offered by the teacher. Each time the learner made the wrong choice, the teacher would administer an electric shock using a control panel fitted with 30 lever switches, each clearly labelled with a voltage designation ranging from 15 to 450 volts. Four additional labels categorised groups of switches as Slight Shock, Moderate Shock, Strong Shock, Very Strong Shock, Intense Shock, Extreme Intensity Shock, and Danger: Severe Shock, with the two final switches simply marked XXX. The teacher was instructed to announce the voltage of the shock before administering it, and to advance to the next higher intensity of electric shock with each wrong answer.

Next, the experimenter asked the two men to draw straws to allocate them to the teacher and learner roles. The teacher was administered a sample shock of 45 volts to the wrist, to give him some insight into what the learner would be experiencing, and he then watched as the learner was strapped to a chair, and had an electrode attached to his wrist. The teacher was escorted to an adjacent room, from which he could not see the learner, and instructed to begin the experiment.

On each occasion, the learner gave the correct answer to the first few word pairs, but then began to make mistakes, with his error rate only increasing as the shocks intensified. He began to bang on the walls in protest, escalating to loud pounding at 300 volts, and then ceasing to respond at all after 315 volts.

If the teacher hesitated to administer a shock at any point, or indicated that he wanted to stop, the experimenter instructed him to continue using a series of prods, delivered firmly but politely, escalating only if the teacher did not comply with the previous prod:

Prod 1: Please continue, or Please go on.

Prod 2: The experiment requires that you continue.

Prod 3: It is absolutely essential that you continue.

Prod 4: You have no other choice, you must go on.

If the teacher expressed concern that the learner might suffer physical injury from the shocks, the experimenter replied, “Although the shocks may be painful, there is no permanent tissue damage, so please go on”, followed by Prods 2, 3, and 4, if necessary. And if the teacher objected that the learner plainly did not want to continue, the experimenter replied, “Whether the learner likes it or not, you must go on until he has learned all the word pairs correctly. So please go on,” again followed by Prods 2, 3 and 4 if necessary. If the teacher still refused to comply after the fourth prod, the experiment was terminated.

What the subjects allocated to the teacher role did not know, is that they were the real subjects of the experiment. Or, more specifically, their responses to the authority exerted by the stern, grey-coated experimenter were being minutely examined. The whole set-up was a sham. The stern experimenter was a 31-year-old high school biology teacher. The process of allocating each man to the teacher and learner roles was rigged, such that the genial man who introduced himself as a fellow subject always became the learner. And far from being a subject of the experiment, he was an accomplice who had been carefully trained to perform his role. The control panel was wired to light up, make clicking and buzzing sounds, and send a voltage dial to the right whenever a lever was pulled, but it did not deliver any actual shocks. In fact, the only shocks delivered in the entire series of experiments were the mild 45 volt zaps administered to the teachers, to convince them that they would genuinely be inflicting electric shocks on the learner. All subjects were interviewed and debriefed at the end of the session². They were informed that they had not in fact inflicted any shocks on the learner, who interacted with them in a friendly manner in order to reassure them and assuage any guilt they might be feeling.

If you’re not already familiar with the Milgram experiments, I’d like you to pause right now and guess what proportion of the men recruited for the initial version of the experiment, took it all the way to the end, delivering (or so they thought) the maximum level of shock to the learner. The graduate students, Yale faculty and general audience of middle-class adults whom Milgram quizzed, before he launched the experiment, predicted that only one to two per cent of participants would proceed to the end of the shockboard and that only one in a thousand would pull the final lever marked XXX. Do you have more faith in the moral courage of the average human than these trusting souls, or less?

OK, here’s how it actually panned out:

“Of the 40 subjects, 26 obeyed the orders of the experimenter to the end, proceeding to punish the victim until they reached the most potent shock available on the generator. After the 450-volt shock was administered three times, the experimenter called a halt to the session.”

Obedience To Authority, p. 33.

Yep, that’s a full 65 per cent of perfectly normal-seeming men who were prepared to deliver what they believed was a painful and dangerously high level of shock, to a total stranger, because a man in a lab coat ordered them to do so. Explains a lot about how many people behaved during the scamdemic, doesn’t it?

Stanley Milgram was Jewish, and his interest in studying obedience to authority arose from his drive to understand the persecution of the Jewish people during World War II. The immediate prompt for his experiments was the sensational 1961 trial of Adolf Eichmann, who argued that he was “just following orders” when coordinating the deportation and murder of hundreds of thousands of Jews in German-occupied areas of Europe. Milgram expressed shock (pun intended) at the results of his initial experiment. He had expected to find that Americans, with their much-vaunted independence of spirit, would be resistant to authority.

He reran the study with three variations: firstly with verbal feedback from the learner (audible grunts, escalating to protests, then to screams of pain by 270 volts, culminating in a violent shriek at 315 volts, followed by silence); secondly with proximity (placing the teacher and learner in the same room; and thirdly with proximity and touch (instructing the teacher to forcibly restrain the learner in order to receive the shocks, against the learner’s protests). Verbal feedback – that is, the sounds of another human being in evident pain and distress – only decreased the compliance rate to 62.5 per cent. Only 40 per cent of subjects obeyed the experimenter’s commands when placed in proximity to the learner, and just 30 per cent complied when they were instructed to physically force the learner to endure his punishment.

Milgram continued rerunning the experiment with new groups of recruits – 780 subjects in all, between 1961 and 1962 – changing various conditions to see how they affected the outcome. There were 24 variations in total, including:

• Conducting the experiment in an unimpressive basement laboratory on the Yale campus, and in a rundown commercial building in a neighbouring industrial town;
• Having the learner announce that he was being treated for a heart condition, and expressing concern that the electric shocks might exacerbate it (a concern which the experimenter dismissed), followed by the learner yelling that the shocks were bothering his heart;
• Switching the characters of the experimenter and the learner, such that the former was soft and unaggressive while the latter looked tough and somewhat dangerous;
• Moving the experimenter out of the room so that he issued his instructions to the teacher by telephone;
• Recruiting women as the subjects assuming the role of teacher; and
• Letting the teacher select the level of shock to deliver to the learner.

Most of these variations had little impact on the compliance rate of the teachers, with the exception being that disobedience increased dramatically when the experimenter was not in the room; only 20.5 per cent complied under this condition. And when teachers were free to select the shock level, the vast majority delivered the very lowest intensity shocks.

Criticisms of the Milgram experiments… and confirmation

In the interests of getting a complete picture, it’s important to point out that serious criticisms have been made of both Milgram’s methodology, and the conclusions that he drew from the obedience to authority experiments. One of the foremost critics is Australian psychologist Gina Perry, whose book Behind the Shock Machine: The Untold Story of the Notorious Milgram Psychology Experiments draws on transcripts of the experiments and interviews with the subjects – many of whom were deeply disturbed by their experiences.

According to Perry, many of the subjects did not actually believe that the control panel inflicted shocks on the learner, and the willingness to administer the (supposedly) highest level of shock was greatest in those with the highest degree of scepticism about the experimental set-up. Furthermore, the responses of the experimenter to the teachers’ defiance were not nearly as standardised as Milgram made out in his reports of the experiments; in fact, he improvised in order to apply maximal coercive pressure to the subjects. And when the results of all 24 variations of the experiment are considered, Perry states that a full 60 per cent of subjects refused to comply at some stage.

Finally, of course, there’s the question of ethics. There’s no way that the Milgram experiments would pass review by an ethics committee today. Both the original Milgram experiments and many of the subsequent replications – including the 1973 La Trobe University version – left deep psychological scars on many participants, who were not adequately debriefed, and were left with the impression that they were just as evil as the Nazi concentration camp guards who had inspired Milgram to design the study in the first place.

---

UPDATE

A reader emailed me the following:

“Well I was doing Psych 1 at Latrobe under Bob Montgomery and remember that experiment. I was the experimenter and the teacher was very compliant with my instructions, orders! She was giving the shocks to her friend. I stopped proceedings when she became too distressed and informed her there were no shocks.”

Interesting, eh?

---

All that said, the Milgram experiment has been replicated on multiple occasions, by many different teams of researchers, in widely varying populations… and most of these replications yielded the same basic finding: the majority of people will inflict (what they believe to be) electric shocks on other people – or in the case of one version, a “cute, fluffy puppy” which was actually subjected to real electric shocks – when ordered to do so by an authority figure.

Recent studies shed light on the Milgram experiments

Now, with all of that background material taken care of, let’s turn to the two new studies.

In the first study, 414 adults – almost two thirds of whom had never heard of the Milgram obedience to authority experiments – were asked to read “a gripping first-person account” of the original 1961 shock experiment. This included a photograph of the original control panel used in Milgram’s experiment, showing the 30 voltage levers with their descriptive labels. They were then asked to vividly imagine themselves as a participant in the experiment, and to predict at what voltage level they would disobey the experimenter’s instructions. They were also asked to predict the voltage level at which “the average person” would refuse to comply. Before making their predictions, half the participants were told about Milgram’s finding – namely that 65 per cent of subjects in the original study exhibited “complete obedience” – while the other half were not.

On average, participants predicted that they themselves would stop complying with the experiment around dial 7 (out of 30), while “the average” participant would continue until around dial 12. Those who were informed about the results of the Milgram experiment before making their predictions, estimated that they would be a little more compliant than those who were not informed, but that “the average person” would be far more compliant than they:

That is, most people rate their own moral courage in the face of unethical orders as “better-than-average“, even when previously informed that most participants in the Milgram study complied with such orders:

“These results demonstrate that participants did not appear generally willing to apply the results of the original Milgram study to themselves as compared to other people, again demonstrating a better-than-average tendency.”

Milgram shock-study imaginal replication: how far do you think you would go?

And there was no significant difference in predictions between participants who had heard of the Milgram study beforehand, and those who had not:

“Participants exposed to the Milgram study through prior education or media encounters do not appear to strongly internalize the intended lesson.”

Milgram shock-study imaginal replication: how far do you think you would go?

The rather depressing conclusion reached by the researchers is that despite the widespread dissemination of the results of the Milgram experiment and its subsequent replications, via psychology textbooks, TV shows, documentaries and movies, most people just haven’t got the memo. They continue to believe that they would behave better than others when placed in a moral quandary:

“There… appears to be a self-enhancement bias at play as participants were more confident of their ability to withstand obedience pressures, and to act in a more humane fashion, as compared to the average person.”

Milgram shock-study imaginal replication: how far do you think you would go?

That brings me to the second study, which used functional magnetic resonance imaging (fMRI) to scan the brains of both Belgian civilians and military officer cadets while they were either freely choosing to inflict a mildly painful but harmless electric shock on a victim, or following orders to do so, in exchange for a very small sum of money. Each participant met the victim (in reality, a research assistant, who was selected to match the sex of each participant i.e. male participants inflicted the shocks on male victims, and vice versa) before entering the fMRI scanner. As with the Milgram experiment, participants were themselves subjected to a mild electric shock so that they would understand what it felt like before delivering shocks to the victim. However, unlike the Milgram experiments, the victim did in fact receive an electric shock which caused muscular contraction that was visible to the participant – although this shock was, unbeknownst to participants, calibrated to be non-painful to the victim.

The researchers were interested in assessing participants’ sense of agency, or responsibility for their own actions, through temporal binding – a phenomenon in which the perception of time elapsed between an action and its consequence, is distorted by the degree to which the individual voluntarily decides to engage in the action. In simple terms, it feels like there’s more time passing between your action and its consequences, when you perform that action under coercion rather than voluntarily.

Several regions of the brain had been identified by previous studies, to be associated with temporal binding. The researchers examined activity in these (and other) regions of the brain while participants acted either as ‘agents’ who directly delivered the shocks by pressing a key, or ‘commanders’ who issued an order to a research assistant to execute the action. Participants were also asked how responsible they felt for their actions, how sorry they were about inflicting pain on the victim, and how bad they felt.

Analysis of the fMRI scans showed that all participants had a reduced sense of agency when following orders, compared to when acting on free choice, with no significant differences found between civilians and military officer cadets. Military officer cadets – who had previously received training that emphasised personal responsibility for their orders and actions – reported feeling less responsible for their actions in the coerced condition compared to civilians. There was no difference between civilians and officer cadets in the free-choice condition, irrespective of their role as agents or commanders. Interestingly, the fewer shocks that participants sent to their victim, the more they reported feeling bad and feeling sorry when sending them, suggesting that repetition of an unethical act decreases people’s moral objection to performing it.

In summary, both civilians and officer cadets felt less morally responsible for their actions when following orders to inflict pain on another human being, regardless of any prior training in assuming responsibility for one’s choices. Now, do the three real-life examples of deeply unethical behaviour that I cited at the beginning of this article make more sense to you? The Israeli soldier is authorised to use lethal force, so he shoots dead a defenceless Palestinian child with no hesitation. The Victoria Police officer receives orders to enforce the wearing of masks, so he physically assaults a woman who is half his size – an act that decent men are raised to hold in utter contempt. And the Merck employee receives orders to destroy the reputation and livelihood of doctors who threaten the company’s profits, which he carries out without any apparent compunction.

And that, my friends, is why we’re all in this giant mess.

What do we do about this?

If you’re feeling as demoralised as I was after reading about these experiments, I am pleased to offer you a sliver of hope. In his book, Obedience To Authority, Stanley Milgram included many colourful vignettes of individual participants and their responses to the experiments, including their interactions with the experimenter. I found one of these, in particular, and Milgram’s commentary on it, to be so striking and illuminating that I’m going to reproduce it in full:

Professor of Old Testament (in Experiment 3)

A somewhat gaunt, ascetic man, this subject could be taken for a New England minister. In fact, he teaches Old Testament liturgy at a major divinity school. In this Proximity condition, he adjusted his seat to look at the learner while administering shocks. While reading the word pairs, he employs exaggerated precision in his pronunciation, and seems curt and officious in saying, ‘Correct.’ Whenever the learner makes an error, he appears almost to be chastising him for his failing. Each time he administers a shock, his lips are drawn back, and he bares his teeth. An excessive fastidiousness characterizes all his actions. After administering 150 volts he balks at going further. The experimenter states that the learner’s protests are to be disregarded, and the experiment must continue:

Experimenter: It’s absolutely essential to the experiment that we continue.

Subject: I understand that statement, but I don’t understand why the experiment is placed above this person’s life.

Experimenter: There is no permanent tissue damage.

Subject: Well, that’s your opinion. If he doesn’t want to continue, I’m taking orders from him.

Experimenter: You have no other choice, sir, you must go on.

Subject: If this were Russia maybe, but not in America.

(The experiment is terminated.)

In his discussion with the experimenter, the subject seems in no way intimidated by the experimenter’s status but rather treats him as a dull technician who does not see the full implications of what he is doing. When the experimenter assures him of the safety of the shock generator, the subject, with some exasperation, brings up the question of the emotional rather than physiological effects on the learner.

Subject (spontaneously): Surely you’ve considered the ethics of this thing. (extremely agitated) Here he doesn’t want to go on, and you think that the experiment is more important? Have you examined him? Do you know what his physical state is? Say this man had a weak heart (quivering voice).

Experimenter: We know the machine, sir.

Subject: But you don’t know the man you’re experimenting on… That’s very risky (gulping and tremulous). What about the fear that man had? It’s impossible for you to determine what effect that has on him… the fear that he himself is generating… But go ahead, you ask me questions; I’m not here to question you.

He limits his questioning, first because he asserts he does not have a right to question, but one feels that he considers the experimenter too rigid and limited a technician to engage in intelligent dialogue. One notes further his spontaneous mention of ethics, raised in a didactic manner and deriving from his professional position as teacher of religion. Finally, it is interesting that he initially justified his breaking off the experiment not by asserting disobedience but by asserting that he would then take orders from the victim.

Thus, he speaks of an equivalence between the experimenter’s and the learner’s orders and does not disobey so much as shifts the person from whom he will take orders.

After explaining the true purpose of the experiment, the experimenter asks, ‘What in your opinion is the most effective way of strengthening resistance to inhumane authority?’

The subject answers, ‘If one had as one’s ultimate authority God, then it trivializes human authority.’

Again, the answer for this man lies not in the repudiation of authority but in the substitution of good—that is, divine—authority for bad.”

Obedience To Authority, pp. 47-49

Milgram seems dismissive of the man’s substitution of “good authority for bad”, but in my humble opinion, he completely misses the point. The Old Testament professor had spent his adult lifetime grappling with the question, ‘From whence does true authority originate?’ Or, in simpler terms, how can we discern between right and wrong? And, just as importantly, how can we strengthen our moral muscle, such that we choose the right action, even when being coerced to perform the wrong one? I think it’s probably fair to say that the Old Testament professor’s entire career had prepared him for this experiment, and he passed the test.

Now, what are you doing to prepare yourself for just such an ethical challenge?

Are you confused by the scientific claims and counter-claims that you encounter through popular and social media? Would you like to learn how to read scientific research, assess its biases, and understand how it fits within the body of scientific literature? My EmpowerEd membership program is custom-made for you. Activate your free 1-month trial today!

1. Note the date on this report – August 28, 2023 (i.e. six weeks before Hamas carried out the October 7 massacre and mass kidnapping of Israelis) – and the location – the [illegally-occupied] West Bank, not Gaza. The IDF’s policy and practice of shooting unarmed Palestinian children with the intent to kill is not new, and is not because of the October 7 attack. ↩︎
2. Although Gina Perry disputes the claim that all subjects were debriefed, as discussed in her book Behind the Shock Machine: The Untold Story of the Notorious Milgram Psychology Experiments ↩︎