Obedience is one of the most studied and most misunderstood behaviors in psychology. The popular account goes something like this: put an ordinary person under the thumb of an authority figure, and they will do terrible things without question. That story, built largely on a handful of famous mid-twentieth-century experiments, captures something real but gets a lot of the details wrong. Decades of research since then have reshaped the picture, revealing that obedience is less about mindless submission and more about identification, context, and a surprisingly flexible sense of personal responsibility.
What the Classic Experiments Actually Showed
The most famous obedience research involved participants being told by an experimenter to deliver what they believed were increasingly painful electric shocks to another person. In the best-known version of the study, a striking majority of participants went all the way to the maximum shock level. That finding alone became a cultural touchstone, shorthand for the idea that anyone can be made to hurt a stranger if someone in a lab coat tells them to.
But the original research program ran dozens of variations, and the results shifted dramatically depending on the setup. When the physical arrangement was manipulated so that participants were in the same room as the person being shocked, obedience dropped. When the experimenter was physically present, obedience increased. The takeaway was not that people always obey; it was that the spatial and social arrangement of authority and victim powerfully shaped whether they did.1PubMed. Obedience to authority as a function of the physical proximity of the student, teacher, and experimenter In other words, proximity to the suffering made it harder to comply, while proximity to the authority made it harder to resist.
A partial replication conducted in 2006 found that obedience rates were only slightly lower than those observed in the original studies decades earlier, suggesting that this effect is not a historical quirk tied to a more deferential era.2PubMed. Replicating Milgram: Would people still obey today? Whatever drives obedience in these paradigms, it has not faded with changing social norms.
Not Blind Submission but Active Followership
The traditional interpretation of these experiments was that obedience reflects a kind of psychological shutdown, where people enter an “agentic state” and hand over their moral reasoning to the authority. That framing has been substantially challenged. A more recent model, called “engaged followership,” argues that people do not obey because they stop thinking. They obey because they actively identify with the authority figure and the cause that person represents.3Philosophia Scientiae. Obedience as “Engaged Followership”: A Review and Research Agenda
Reanalysis of the original experimental data supports this. Participants’ willingness to deliver maximum shocks changed predictably depending on whether the experimental setup encouraged them to identify with the experimenter and the scientific mission or with the person receiving shocks.4PubMed. Working Toward the Experimenter: Reconceptualizing Obedience Within the Milgram Paradigm as Identification-Based Followership When participants saw themselves as advancing a worthy cause, they were far more likely to comply with harmful instructions.5Confronting Humanity at its Worst. Engaged Followership and Engaged Fellowship
This reframing matters because it changes what we think the danger actually is. The old story said: authority flips a switch in the brain and people become robots. The new story says: people who believe they are serving science, or the nation, or God, or any cause they care about, will endure significant moral discomfort to advance that cause. That is a different and arguably scarier kind of vulnerability, because it does not require a person to stop caring about right and wrong. It just requires them to believe the authority shares their values.
The Stanford Prison Experiment and Its Unraveling
The other experiment most people associate with obedience is the Stanford prison study, in which college students assigned to play guards in a mock prison quickly became abusive toward students assigned to play prisoners. For decades it was taught as proof that situational roles could override personal morality. That interpretation has largely collapsed under scrutiny.
Multiple lines of criticism have emerged. Researchers who examined archival materials found that the study’s investigators held an orientation session the day before the experiment in which they communicated expectations for hostile behavior from guards, essentially coaching them toward abuse.6PubMed. Revisiting the Stanford prison experiment, again: Examining demand characteristics in the guard orientation Participants exposed to that orientation, even in follow-up studies decades later, reported greater expectations for hostile behavior compared to those given neutral instructions.
A broader investigation uncovered additional problems: biased data collection, precise instructions given to the guards about how to treat prisoners, the guards not being told they were experimental subjects, and participants rarely being fully immersed in the scenario.7PubMed. Debunking the Stanford Prison Experiment The study, in short, was less a demonstration of obedience to situational roles and more a demonstration of what happens when an authority figure tells people what he wants them to do and they try to deliver. Which, ironically, loops back to the engaged followership idea: the guards were not overwhelmed by their roles. They were trying to help the experimenter get the results he seemed to want.
What Obedience Does to the Brain
One of the most interesting developments in obedience research has come from neuroscience. When you freely choose to perform an action, your brain registers a tight connection between your action and its outcome. You feel like the author of what happened. Under coercion, that sense of authorship weakens.
Researchers measured this using a well-established perceptual effect: when you voluntarily press a button and hear a tone afterward, you perceive the gap between the press and the tone as shorter than it actually was. Your brain bundles the two together because it treats them as your doing. When participants were ordered to press the button, the perceived gap between action and outcome grew longer, suggesting a weakened sense of personal agency. Brain imaging confirmed that the neural processing of the action’s outcome was also reduced under coercion.8PubMed Central. Coercion Changes the Sense of Agency in the Human Brain In plain terms, people who obeyed orders experienced their actions as closer to passive movements than to deliberate choices.
Follow-up work added nuance. A study that combined brain scanning with the same type of timing task found that the coercion effect on implicit agency was weaker inside the MRI scanner, possibly because the scanner environment disrupted normal time perception. But explicit feelings of responsibility still dropped sharply under coercion: participants rated their own responsibility as far lower when they had been ordered to act compared to when they chose freely.9PLoS ONE. The obedient mind and the volitional brain: A neural basis for preserved sense of agency and sense of responsibility under coercion Participants who showed a stronger reduction in responsibility ratings also showed a stronger reduction in the implicit timing measure, linking the two effects.
Research has also looked at the other side of the equation: what happens in the brain of the person giving orders. Interestingly, giving coercive commands also altered the sense of agency for the person issuing them, with the commander’s implicit measure of agency shifting compared to free-choice conditions.10PLoS ONE. Only giving orders? An experimental study of the sense of agency when giving or receiving commands Obedience does not just change the psychology of the person who obeys. It reshapes the experience of everyone involved.
How Obedience Develops in Children
Obedience does not appear out of nowhere when a child enters school. Its earliest forms show up in the second year of life, as toddlers begin responding to parental requests and prohibitions. Developmental researchers have drawn an important distinction between two kinds of compliance. One is “situational compliance,” where a child goes along with a parent’s request only while the parent is watching. The other is “committed compliance,” where the child embraces the rule as their own and follows it even when unsupervised.
Committed compliance, but not situational compliance, predicted a child’s later internalization of the parent’s rules. Children who simply went along when monitored did not develop a deeper sense of the rules as their own. Passive noncompliance, where the child simply did not engage, predicted worse outcomes for internalization.11Child Development. Children’s Emerging Regulation of Conduct: Restraint, Compliance, and Internalization from Infancy to the Second Year This echoes the adult finding about engaged followership: compliance that comes from genuine buy-in looks psychologically different from compliance that comes from external pressure, and leads to different downstream behavior.
Evolutionary Roots of Deference
Obedience-like behavior is not unique to humans. In many primate species, subordinate individuals signal their lower status through ritualized gestures, and these signals serve a clear social function. Among Verreaux’s sifakas, a type of lemur, subordination signals like chatter vocalizations increased tolerance and affiliation between dominant and subordinate individuals, reduced conflict, and lowered the chance that a dominant animal would steal a subordinate’s food.12PubMed. Subordination signals improve the quality of social relationships in Verreaux’s Sifaka: Implications for the evolution of power structures and social complexity
Across primates more broadly, ritualized submission appears to function as a social glue. After a conflict, the loser is essentially offered a deal: accept a lower-status position and gain a tolerant relationship, or face continued hostility. Without these mechanisms, aggression tends to scatter group members. With them, aggression becomes a manageable and even constructive part of group life.13PubMed. The integration of dominance and social bonding in primates The takeaway is not that human obedience is “just” an animal instinct. It is that the tendency to defer to higher-status individuals has deep evolutionary utility for maintaining social cohesion, and that utility did not vanish when human societies became complex. It got repurposed.
Cultural Variation and Power Distance
Not every culture treats authority the same way. The concept of “power distance” captures how much a society expects and accepts unequal distributions of power. In high-power-distance cultures, junior employees are more likely to defer to senior ones, students to teachers, children to parents, and so on. In low-power-distance cultures, those gaps are smaller and people are more comfortable challenging those above them.
This is not just an abstract cultural variable. Research on workplace communication found that employees with high power-distance beliefs communicated less with their superiors, and the mechanism was straightforward: fear of authority. People who internalized the idea that bosses should not be questioned were literally more afraid to talk to them, which suppressed upward communication.14PubMed Central. Power Distance Belief and Workplace Communication: The Mediating Role of Fear of Authority The practical consequence is that in high-power-distance organizations, problems are less likely to be reported to the people who can fix them.
What Makes People Disobey
If obedience is shaped by identification with authority and social context, then disobedience should be similarly contextual. And it is. One of the strongest predictors of whether someone refuses an unjust order is whether they see other people refusing first.
In an experiment where participants were asked by an authority figure to write a persuasive but unethical message, the presence of two peers who openly refused the instruction was enough to produce a dramatic shift. Participants who saw two people disobey were more than twelve times more likely to refuse the order themselves compared to those in a control condition with no visible dissenters.15Psychology Research and Behavior Management. On the dynamics of disobedience: experimental investigations of defying unjust authority This “social modeling” effect was the strongest variable in the study. It was not personality traits, not moral reasoning skills, not even the severity of the unethical act. It was simply seeing others say no.
This finding cuts against the notion that resistance to authority is an individual heroic act. It is much more commonly a social act, dependent on whether the group gives a person permission to dissent. If you want to make disobedience more likely in a setting where it matters, the most effective intervention is not to train individuals to be braver but to ensure that the first dissenter is visible.
Obedience Pressure in Professional Settings
Outside the laboratory, obedience pressure plays out in workplaces, hospitals, and institutions with concrete stakes. Research on whistleblowing provides a useful window. Studies examining the decision to report wrongdoing have found that people under high obedience pressure from a superior are less likely to blow the whistle than those under low obedience pressure.16Jurnal Ekonomi Dan Bisnis. Obedience pressure and group pressure: An experimental study on whistleblowing Group pressure from peers compounds the effect: when both authority pressure and peer pressure are high, whistleblowing intentions drop further.
Similar patterns show up across different cultural contexts. A study comparing individualist and collectivist orientations found that groups under lower obedience pressure and with more individualist norms had higher intentions to report wrongdoing.17Asia Pacific Fraud Journal. An Experimental Study of Whistle-Blowing: The Influence of Obedience Pressure and Ethical Dilemmas That is, the cultural value placed on group harmony versus individual judgment interacts with authority pressure to shape whether someone speaks up.
Healthcare offers a particularly vivid example. In operating rooms, anesthesiologists and junior staff sometimes notice errors or safety concerns that the lead surgeon may not see. Yet speaking up remains difficult. A study of anesthesiologists found no improvement in speaking-up behavior even after a simulation-based training intervention. The most frequently cited hurdles were uncertainty about whether the issue was real, stereotypes about other team members, respect for the senior clinician’s experience, and fear of repercussions.18Academic Medicine. Improving Anesthesiologists’ Ability to Speak Up in the Operating Room: A Randomized Controlled Experiment of a Simulation-Based Intervention and a Qualitative Analysis of Hurdles and Enablers The fact that even targeted training failed to move the needle underscores how deeply embedded these dynamics are. You can tell someone that speaking up saves lives, and they can agree completely, and still stay silent in the moment because the social cost feels too high.
Historical Lessons and Professional Ethics
The most extreme historical consequences of obedience have sometimes been leveraged as teaching tools for professionals who face authority pressures. In nursing education, for instance, curricula have incorporated study of the role that medical and nursing professionals played during the Holocaust. A seminar-based intervention on this history led nursing students to report increased awareness and to identify themes around being a patient advocate, the importance of “judicious obedience” rather than reflexive compliance, and the need to adhere to personal and professional ethics even under institutional pressure.19PubMed Central. Legacy of the Role of Medicine and Nursing in the Holocaust: An Educational Intervention to Support Nursing Student Professional Identity Formation and Ethical Conduct The phrase “judicious obedience” is worth sitting with: it captures the idea that the goal is not to eliminate obedience entirely, which would make organized institutions impossible, but to make it thoughtful rather than automatic.
The challenge in any professional setting is that obedience and hierarchy exist for good reasons. Surgeons need operating teams to follow instructions. Military units need soldiers to execute orders rapidly. Organizations need employees who can carry out directives without relitigating every decision from scratch. The problem is not obedience itself but the absence of recognized channels and cultural permission for pushing back when something is genuinely wrong. Every institution that depends on compliance also needs mechanisms for dissent, and most of them underinvest in the latter.