Social Psychology: How Others Shape Us

Psychology Fundamentals

Chapter 6 · Social Psychology: How Others Shape Us

Social psychology studies how the presence, instructions, and expectations of other people change what we do — often more dramatically than we'd predict about ourselves. Three of psychology's most famous studies sit at the center of this chapter, and one of them holds up far worse under real, modern scrutiny than its textbook reputation suggests.

Milgram's Obedience Experiments: Real Numbers, Real Controversy

Stanley Milgram began his real obedience experiments in August 1961 at Yale University, first publishing the results in a 1963 journal article. Participants believed they were assisting in a memory study, taking on the role of "teacher" while a confederate actor played the "learner" in another room. For every wrong answer, the teacher was instructed to deliver an electric shock via a generator, increasing by 15 volts each time, up to a labeled maximum of 450 volts — while the learner's pre-recorded cries of pain grew steadily more distressed.

The Real Result

Every participant continued at least to 300 volts, and 65% went all the way to the full 450 volts — despite audible distress from the "learner" and, in many cases, visible discomfort in the participant themselves. A later meta-analysis of replications found obedience rates ranging from 28% to 91%, with U.S. studies averaging around 61%.

The study's real ethical fallout was substantial. Psychologist Diana Baumrind argued that even with informed consent to participate, Milgram bore real responsibility for the psychological distress his subjects visibly experienced, and that the study should have been halted once that distress appeared. Her critique is widely credited with driving a genuine, lasting revision of research ethics standards. Milgram himself pointed to a real follow-up finding in his own defense: 84% of former participants later reported being "glad" or "very glad" they had taken part.

Asch's Conformity Experiments: Bending to a Group You Can See Is Wrong

Solomon Asch ran his real conformity experiments in 1951 at Swarthmore College, using a simple perceptual task: a group of eight young men were shown a reference line, then asked which of three comparison lines matched it. Only one person in the room was a genuine participant — the other seven were actors instructed to unanimously give an obviously wrong answer on 12 of 18 total trials, with the real participant always answering last.

Alone, No Pressure

Real error rate below 0.7% — the task itself was genuinely easy.

Under Group Pressure

35.7% of individual responses conformed to the group's wrong answer.

At Least Once

74% of participants conformed to the incorrect majority on at least one of the 12 critical trials.

A Real, Easy-to-Miss Nuance
Even under direct group pressure, the majority of individual responses — 64.3% — still stayed correct. Conformity was real and measurable, but it wasn't total: most people resisted the incorrect majority most of the time, even while a real, substantial share caved at least once.

The Stanford Prison Experiment: A Real Story of Its Own Undoing

In 1971, Philip Zimbardo randomly assigned 24 screened male participants to play "guards" or "prisoners" in a mock prison built in Stanford's own basement, intending to run the simulation for two weeks. It was shut down after just six days, once guards' behavior toward prisoners had turned genuinely abusive. For decades, textbooks — including, honestly, many earlier drafts of exactly this kind of course — presented it as clean evidence that ordinary people will spontaneously turn cruel purely because of the situational power a role hands them.

The Real 2018-19 Reappraisal
French sociologist Thibault Le Texier gained access to newly released archival recordings and participant testimonies, and his real investigation found the study was far less spontaneous than its popular reputation claims:
  • Guards were directly coached by the research team to behave more aggressively and "tough," with the study's own conclusions substantially written in advance.
  • The famous "breakdown" of prisoner Douglas Korpi (Prisoner 8612), long treated as spontaneous proof of the situation's real psychological power, was later admitted by Korpi himself, in a real 2017 interview, to have been performed — he needed to leave to study for exams, not because he was genuinely breaking down.
  • Participants had answered an advertisement that specifically mentioned "prison life," raising a real, documented self-selection concern: those who volunteered may have already carried above-average aggressive tendencies.

A further real complication: the BBC Prison Study (2002-2006), which recreated a similar design but deliberately gave guards no behavioral coaching, produced genuinely different results. Zimbardo and colleagues have disputed parts of Le Texier's account, but multiple original participants have corroborated the coaching and staging he documented.

Reading These Three Studies Together

StudyReal Core FindingHow Well It Holds Up Today
Milgram (1961-63)65% obeyed to the full 450 voltsWidely replicated across real follow-up studies (28-91% range) — the core finding is real and robust
Asch (1951)74% conformed at least once to an obviously wrong group answerAlso widely replicated — considered a genuine, stable finding
Zimbardo (1971)Ordinary people spontaneously became abusive once given a "guard" roleSubstantially undermined — real 2018-19 evidence shows the abuse was significantly coached and staged, not spontaneous

This isn't a reason to distrust social psychology as a field — Milgram's and Asch's own real, independently replicated findings show genuine, powerful social influence effects do exist. It's a reason to apply this course's own Chapter 5 lesson one more time: a study's fame and its scientific soundness aren't the same thing, and checking whether a famous result has actually been re-examined is worth doing before repeating it as settled fact.

Hands-On Exercises

Exercise 1

Milgram's 84%-were-glad-they-participated finding is sometimes used to argue the study was ethically fine after all. Explain, in your own words, why this doesn't fully answer Baumrind's real ethical objection.

📄 View solution
Exercise 2

Explain, in your own words, why the fact that 64.3% of individual responses in Asch's study stayed correct under pressure doesn't contradict the headline "74% conformed at least once" finding.

📄 View solution
Exercise 3

Explain, in your own words, why the BBC Prison Study's different result under a no-coaching design is real evidence relevant to evaluating Le Texier's own critique of Zimbardo's study — not just an unrelated fact.

📄 View solution

Chapter 6 Quick Reference

  • Milgram (1961-63, Yale): 65% of participants obeyed instructions to the full 450 volts; the study is widely replicated (28-91% range) but sparked lasting ethical reform
  • Asch (1951, Swarthmore): 74% conformed to an obviously wrong group answer at least once across 12 critical trials, while 64.3% of individual responses still stayed correct
  • Zimbardo's 1971 Stanford Prison Experiment was substantially undermined by Thibault Le Texier's real 2018-19 archival investigation, which found guards were directly coached and the famous "prisoner breakdown" was later admitted to have been staged
  • Fame and scientific soundness aren't the same thing — checking whether a famous study has been re-examined is worth doing before treating it as settled