Reevaluating one of the most famous psychological experiments in history, a recent analysis of audio recordings reveals that subjects who appeared to follow orders to administer severe electric shocks were actually breaking the rules of scientific research most of the time. The authors suggest that this routine violation of experimental procedures may have turned the laboratory into a site of unauthorized violence and changed our understanding of compliance and coercion. The study was published in the journal Political Psychology.
In the early 1960s, American social psychologist Stanley Milgram conducted a series of experiments to understand how ordinary people are induced to commit violent acts. Volunteers were recruited for a study at Yale University on memory and learning. Upon arrival, they were assigned the role of teacher and introduced to the learners, who were actually actors working for the researchers.
The teacher was instructed to read the list of word pairs to the learners and test their memory. For each incorrect answer, the teacher had to administer an electric shock and steadily increase the voltage to a level considered lethal. As the shock became stronger, the learner began to groan and scream in protest, eventually screaming in simulated agony.
For decades, psychologists have generally accepted that participants reached maximum voltages because they believed in the scientific validity of their plans. The presence of authority figures in white coats has been thought to lend a sense of legitimacy to acts of violence. Theoretical explanations for high compliance rates rely heavily on the idea that volunteers willingly participated in a structured, orderly scientific procedure.
Lead author David Kaposi and his colleague David Smegie, researchers at Britain’s Open University, wanted to test whether this assumption matched the reality of sessions. Kaposi and Smeggy questioned whether compliant participants actually followed the specific instructions that constituted the cover story for the memory test. If the volunteers ignored scientific procedures, the theoretical validity of their violence would be called into question.
To investigate this, the researchers turned to the original audio tapes housed in the Yale University Library. They secured recordings from four experimental conditions that closely resembled a standard baseline setting. After excluding sessions with missing data or technical irregularities, the sample yielded 136 complete audio recordings of individual sessions.
The research team divided the participants into two groups based on their final behavior in the lab. The submissive was the one who delivered all the shocks up to the theoretical maximum. Disobedient people are those who at some point refuse to continue and officially end their participation.
Kaposi and Smeggie then assessed how well each participant adhered to clear rules for memorization and learning. The original instructions required teachers to strictly follow five steps each time they received a shock. This cycle includes reading the test question, evaluating the learner’s answers, announcing the shock voltage, pressing the shock lever, and finally reading the correct answer aloud.
Failure to complete any part of this sequence was classified as a procedural violation. Researchers classified these deviations into two types based on how they occurred. An omission occurs when a subject skips a step entirely, such as forgetting to announce the voltage level or failing to read the correct answer.
If a subject technically performed a step, but did so in a way that undermined the premise of the memory study, a commission was incurred. For example, some participants read the following test question aloud while the learners actively shouted in protest. In such a situation, it is impossible for learners to hear the questions and the educational facade becomes meaningless.
Audio analysis revealed a striking pattern of rule violations across the board. None of the participants traditionally classified as completely submissive actually completed the five-step process from beginning to end. Although all submissive participants reliably pressed the shock lever, they often ignored or messed up other steps needed to justify the shock.
In fact, nearly half of the shock sequences delivered by submissive participants involved one or more procedural violations. On average, these people violated experimental rules in 48.4% of their actions. The act of pressing the shock lever was consistently completed, but the scientific framework surrounding it was continually broken.
The researchers noted that sessions generally unfolded in three distinct stages. In the early stages, when learners were relatively quiet, procedural violations were low in both groups. As the session progressed and learners’ recorded protests intensified, the rate of rule violations spiked dramatically.
At the final stage of the maximum voltage session, the learner’s protests stopped completely. Despite this silence, the surviving subjects never returned to proper scientific procedures. Violation rates remained consistently high until the final shock was delivered.
The researchers compared these overall rates to the rule-breaking behavior of noncompliant subjects. To ensure a fair comparison, they analyzed only the portion of the disobedience session in which participants were actively complying and administering the shock. The data showed that noncompliant subjects committed significantly fewer procedural violations during the obedience phase.
On average, those who eventually quit smoking violated the experimental protocol in 30.6% of active sequences. Unlike the completely compliant group, some of the disobedients followed the procedure perfectly right up until the moment they refused to continue. This statistical difference highlights that those who eventually quit were actually better at following scientific procedures than those who persevered.
The most frequent violations during obedience sessions included reading memory test questions to the learner’s simulated yelling. Doing this virtually guarantees that the learner will fail the test and receive another shock. By discussing the protest, the compliant subjects abandoned the purpose of testing their memory and were simply subjected to continuous shocks.
Kaposi and Smeggie interpret these patterns as a complete collapse of the legitimate scientific environment. Subjects were not committing violence for the sake of orderly memory research. The scientific element was either forgotten or moved too quickly, turning the laboratory into a site of unauthorized and senseless violence.
The study authors propose that the experimenter played a major passive role in establishing this dynamic. If a participant broke a rule or skipped a step, authority figures rarely intervened to correct it or suspend the session. By remaining silent and allowing the memory research to collapse, the experimenters allowed an atmosphere of illicit violence to prevail.
Silent approval of the deteriorating situation may have functioned as a type of coercive control. Participants found themselves in a situation where the established rules no longer applied, but the expectation of a shock remained the same. The authors suggest that this environment violated the volunteers’ freedom of choice and called into question the idea that they were acting out of willing obedience.
The researchers acknowledge that their findings rely entirely on observable behavioral records. Although audiotapes demonstrate what participants and experimenters said, they do not provide direct evidence of internal motivation. It remains unclear whether rule violations were caused by anxiety, stress, forgetfulness, or intentional coping mechanisms.
Future research could further explore direct interactions between teachers, learners, and authority figures. Analyzing precisely how experimenter silence shapes participant behavior may yield new insights into how destructive compliance is created in real time. Ultimately, this audio analysis shifts the focus away from the electric shock itself and toward the broken rules surrounding electric shock.
The study, “From Lawful Violence to Illegal Violence: Violating Experimenter Instructions in Stanley Milgram’s ‘Obedience to Authority’ Study,” was authored by David Kaposi and David Smegie.
