Social Psychology: What Other People Do to Your Mind

The core discovery of social psychology fits in one line: the situation is far more powerful than we believe, and character far less. We move through life explaining behaviour — ours and others’ — by personality (“he’s rude,” “she’s a natural leader”), while decades of experiments show that ordinary, decent people will deny the evidence of their own eyes, obey terrible instructions, or walk past someone in trouble, given the right situation. Learn the situations.

Conformity: Asch’s lines

Solomon Asch (1951) sat participants in a group asked to match line lengths — a task so easy that alone, people got it right >99% of the time. But the other “participants” were confederates who unanimously gave an obviously wrong answer. Result: ~75% of people conformed at least once, and about a third of all answers bent to the group.

The two engines, worth separating because the fix differs:

• Normative influence — you know they’re wrong but conforming avoids the social cost. (Fix: private voting, anonymous feedback.)
• Informational influence — you genuinely doubt yourself: “can eight people all be wrong?” (Fix: independent judgements before discussion.)

The most practical finding: a single dissenter cracks it. One confederate giving the right answer cut conformity by ~75% — the dissenter doesn’t even need to agree with you, they just need to break unanimity. Being the first voice to say “I’m not sure about this plan” is a structural gift to every silent doubter in the room. This is also the mechanism behind groupthink — cohesive teams converge on bad decisions because dissent has social cost — and behind the fixes: assigned devil’s advocates, leaders speaking last, estimates collected in writing first (which also blocks anchoring).

Obedience: Milgram, with the modern caveats

Stanley Milgram (1961) had participants deliver escalating “shocks” to a learner (an actor) at an experimenter’s calm insistence — “the experiment requires that you continue.” ~65% went to the maximum 450 volts, versus the ~1% that psychiatrists predicted. The finding replicated across countries and decades (a 2009 partial replication found similar rates).

Modern re-examination (archival work on Milgram’s own tapes) adds nuance without overturning the core: many participants disobeyed more than the famous number suggests, the least “order-like” prompt was the one most refused, and compliance looked less like blind obedience and more like trusting the system — people obeyed while believing the institution wouldn’t allow real harm. The transferable lessons:

• Authority is situational: the lab coat, the setting, the institution — not the person.
• Gradual escalation is the trap. Nobody starts at 450 volts; every step is only slightly worse than the last one you already accepted (the sunk-cost / commitment machinery from biases, weaponized).
• Distance matters: obedience collapsed when the learner was in the same room, or when the experimenter instructed by phone.

(The Stanford prison experiment, the other famous “situations corrupt” study, has fared much worse under scrutiny — guards were coached toward cruelty, and the famous breakdown was partly performed. Cite Milgram carefully; cite Zimbardo barely.)

Attribution: the errors that run your social life

The fundamental attribution error: explaining others’ behaviour by character and our own by circumstance. The driver who cuts you off is a psychopath; when you cut someone off, you were avoiding a merge. He’s late because he’s disorganized; you’re late because traffic. It’s the default setting — other people arrive in our lives context-free, so character is the only explanation available to System 1.

Its team-scale twin is the self-serving bias (successes are mine, failures are the situation’s) — run both in a group and you get each member honestly believing they do most of the work. In one classic study, spouses’ estimates of their own share of housework summed to well over 100%.

The repair is a habit, not an insight: for any annoying behaviour, generate one situational explanation before settling on a character one. You’ll be wrong sometimes — but the default is wrong more.

Persuasion: Cialdini’s six levers

Robert Cialdini spent years inside sales trainings and compliance professions, then organized what he saw into six principles — each a genuine social heuristic that works because it’s usually right:

Lever	The reflex	In the wild
Reciprocity	Return favours — even unrequested ones	Free samples; “no-strings” gifts before the ask
Commitment/consistency	Act like your past self	Foot-in-the-door: small yes → large yes
Social proof	When unsure, copy the crowd	Reviews, “bestseller”, ratings, queues
Liking	Say yes to people you like	Rapport-building, similarity, compliments
Authority	Defer to expertise signals	White coats, titles, “9/10 dentists”
Scarcity	Want what’s running out	”Only 2 left”, countdown timers (loss aversion again)

Notes for both directions of use: these are strongest under uncertainty — when people don’t know what to choose, social proof and authority carry the decision. Fake versions (invented scarcity, purchased reviews) are effective, corrosive, and increasingly illegal — the ethical line is whether the signal is true. The dark patterns discussion in user behaviour is this table wearing a UI.

Defence-wise, name the lever and it loses power: “I feel obligated because they gave me something” converts an automatic reflex into a System 2 decision.

Bystanders, briefly

Kitty Genovese’s 1964 murder launched the bystander effect literature (the “38 silent witnesses” story itself was largely a journalistic fabrication — but the effect it inspired is real): the more people present, the less likely any individual helps, through diffusion of responsibility (“someone will call”) and pluralistic ignorance (everyone scans everyone else’s calm and concludes it’s not an emergency). Modern meta-analyses confirm the effect while noting it shrinks or reverses in clear, high-danger emergencies — ambiguity is the real villain. The practical protocol, taught in every first-aid course: single out one person — “you, in the red shirt, call an ambulance.” Responsibility diffuses across a crowd; it sticks to a name.

Takeaways

• Situations beat dispositions — arrange the situation, don’t rely on character (yours included).
• Unanimity is fragile: one dissenter frees a room. Collect independent opinions before group discussion.
• Beware gradual escalation under authority — audit the staircase, not the step.
• Give others one situational excuse before a character verdict.
• Know the six levers — to persuade honestly, and to notice when they’re aimed at you.

The last piece of the foundation: the machinery underneath all of this — emotions, stress, and what regulation actually works.