65% of Ordinary People Would Hurt a Stranger If Someone in a Lab Coat Said To
Sixty-five percent of ordinary people would electrocute a stranger if someone in a lab coat told them to.
Not sociopaths. Not the clinically disturbed. Ordinary people. Teachers. Accountants. Nurses. People who loved their families and helped their neighbors and would describe themselves as fundamentally decent.
They hesitated. They sweated. They asked to stop. And when the authority figure said "please continue" — most of them did.
What Stanley Milgram Actually Found
In 1961, Yale psychologist Stanley Milgram designed what became the most disturbing experiment in the history of psychology. Participants were recruited through newspaper ads for a study on "learning and memory." They arrived at a Yale laboratory and were introduced to another participant — who was, in fact, an actor.
The setup was specific: the real participant would be the "teacher," administering word-memory tests to the actor "learner" in the next room. Each wrong answer would be punished with an electric shock from a panel the teacher controlled, starting at 15 volts and escalating in 15-volt increments to a maximum of 450 volts — labeled, at the highest levels, "Danger: Severe Shock" and then simply "XXX."
The shocks were not real. The screams were recorded and scripted. But the participants did not know this. They heard anguished pleas from the next room. They heard the learner complain of a heart condition. They heard silence — and were told that silence meant an incorrect answer and they should continue.
When participants expressed reluctance, the researcher in the lab coat gave a scripted response: "Please continue." If they persisted in wanting to stop: "The experiment requires that you continue." If they still refused: "It is absolutely essential that you continue." And finally: "You have no other choice, you must go on."
The original result, published in Milgram's 1963 paper in the Journal of Abnormal and Social Psychology, was 65% compliance to the maximum 450-volt level. The finding was replicated across genders, nationalities, and decades. A 2009 replication by Jerry Burger at Santa Clara University found compliance rates statistically identical to the original.
This was not an artifact of the 1960s. This is human behavior under authority.
What the Participants Actually Felt
The word "compliance" makes it sound easy. It was not.
Milgram documented what happened to participants during the experiment. Many showed visible distress — trembling, sweating, nervous laughter, stuttering. Some wept. Some protested repeatedly before ultimately continuing. Post-experiment interviews revealed that nearly all participants found the experience genuinely disturbing.
And yet they continued.
This is the finding that most accounts of the Milgram experiment miss: the participants were not comfortable obeying. They were conflicted, anxious, and often distraught. But their discomfort was insufficient to produce defiance. They kept turning the dial.
The psychological mechanism, Milgram argued, was what he called the "agentic state" — a condition in which the individual shifts from seeing themselves as a self-directed agent to a component of a system directed by someone else. When you enter the agentic state, responsibility feels like it belongs to the authority, not to you. The participant stops asking "is this right?" and starts asking "am I following instructions correctly?"
The Same Conditions Exist Right Now
The Milgram findings are regularly cited in discussions of historical atrocities, but that framing lets the rest of us off the hook. The conditions that produced his results are not confined to authoritarian regimes or crisis moments. They are present in ordinary life.
Consider the workplace. Hierarchies create permanent authority structures. A manager can direct employees toward behavior they would privately find troubling — and the agentic state kicks in. "I was just following orders" is not an excuse unique to war criminals. It is the daily reasoning of people who compromise their own ethics to comply with institutional authority and avoid professional consequences.
Consider the family system. Parental authority — especially in households where questioning was punished — creates deeply embedded obedience patterns. Children who learned that defiance meant abandonment, rage, or punishment carry an automatic compliance response into adulthood. The way trauma rewrites the nervous system's threat-detection maps directly onto this dynamic.
Consider online dynamics. Viral content, high follower counts, and the visible consensus of a crowd all function as authority signals. The brain does not distinguish between a scientist in a lab coat and a thousand people agreeing on your screen. The compliance instinct activates either way.
The 35% Who Said No
Thirty-five percent of participants refused to continue to the maximum voltage. This gets far less attention than it deserves.
What distinguished them was not a fundamentally different character. What distinguished them was a specific cognitive move: they maintained their own perspective as authoritative alongside the researcher's. They said, explicitly or internally, "I don't care what you say. I'm the one pushing this button. This is my responsibility."
The agentic state is not a lock. It is a frame. And it can be broken.
The defiant participants were also more likely to have questioned the researcher's credibility early in the process — not hostilely, but with genuine inquiry. "Who is actually in charge here? Where is this going?" The questioning itself created distance from pure compliance.
The Dial Is Still in Your Hands
You have almost certainly done something you privately believed was wrong because an authority structure told you it was required. And you almost certainly told yourself some version of "I didn't have a choice."
You did have a choice. So did the 65%.
What they lacked was the belief that their judgment was legitimate — that it counted for anything in the presence of institutional authority.
The application isn't only about resisting harmful orders. It's about recognizing when you've entered the agentic state in everyday relationships. When you suppress your read on a situation because someone confident told you you were wrong. When you go along with something that troubles you because questioning feels like overstepping. When you apologize for having a perspective at all.
The lab coat isn't always literal. Sometimes it's the certainty in someone's voice. The size of their following. The fact that everyone else in the room is nodding.
Milgram's participants had a dial in front of them. Most turned it.
Yours is in your hands right now.
Follow The Shadow Self Media on TikTok · Instagram · Facebook
Cover photo by cottonbro studio via Pexels.