The most famous psychology experiment of the 20th century may need a footnote. Newly analyzed audio recordings from Stanley Milgram's obedience studies reveal that participants routinely violated experimental procedures—and that the violations were more common among those who appeared most compliant.

If you took Psychology 101, you know the setup: participants were told to administer increasingly painful electric shocks to a "learner" (actually an actor) whenever they answered questions incorrectly. The shocking finding—pun intended—was that 65% of participants continued to the maximum voltage, apparently willing to hurt or even kill a stranger when instructed by an authority figure.

But here's what the textbooks didn't mention: not a single participant actually followed the required five-step procedure from start to finish. That procedure was simple enough—read the test question, evaluate the answer, announce the voltage level, press the shock lever, then read the correct answer. Yet participants deviated from this protocol in nearly half of all shock sequences.

The analysis, based on original audio tapes from the 1960s, distinguishes between two types of rule-breaking. Some participants simply skipped steps—omitting the voltage announcement, for instance, or pressing the lever without reading the question. Others performed steps incorrectly, like trying to read test questions while the learner was screaming in (apparent) agony.

Here's the counterintuitive part: participants who appeared most obedient actually broke the rules more often. Those who went all the way to 450 volts violated protocols in 48.4% of shock sequences. Those who eventually refused to continue—the ones who disobeyed—violated rules in only 30.6% of sequences before they quit.

The researchers behind this analysis argue that these findings fundamentally alter our understanding of what was happening in that laboratory. The standard interpretation is that participants were caught in a psychological bind—torn between their conscience and their deference to authority. But if they were routinely breaking the experimental rules, they clearly weren't in a state of blind obedience.

Instead, the lab may have transformed into something messier and more disturbing: "a setting for unauthorized and senseless violence," as the authors put it. The experimenter's failure to correct rule violations—his passive acceptance of participants doing whatever they wanted—might have functioned as tacit approval for escalating aggression.