Psychology of Social Situations — AP Psychology Notes

Key Features:

• Changing behavior to fit in with the group
• Can be explicit (direct pressure) or implicit (unspoken norms)
• Not necessarily bad — helps societies function smoothly
• But can lead to harmful outcomes when group norms are destructive

The Experiment:

• Participants shown lines and asked which comparison line matched the standard line
• Answer was obvious (correct answer clear to everyone)
• Real participant seated with 6-8 confederates (actors posing as participants)
• Confederates deliberately gave wrong answer on 12 of 18 trials
• Researchers measured whether real participant conformed to incorrect majority

Results:

• 75% of participants conformed at least once (gave wrong answer to match group)
• 37% average conformity rate across all critical trials
• When alone, participants made errors less than 1% of the time
• Participants’ reports varied: some doubted their own judgment and others publicly agreed despite privately disagreeing. Do not assume that every conforming response reflected the same mechanism.

Factors Affecting Conformity:

• Group size: Conformity increases up to 3-4 people, then levels off
• Unanimity: One dissenter dramatically reduces conformity (breaks unanimity)
• Status: Higher status group members elicit more conformity
• Public response: More conformity when answering publicly vs. privately
• Culture: Collectivist cultures show higher conformity than individualist

The Setup:

• Participants told they were in a study on learning and punishment
• Assigned role of "teacher" (real participant) and "learner" (confederate)
• Teacher instructed to deliver electric shocks for wrong answers
• Generator settings ranged from 15 to 450 volts with increasingly severe labels; 450 was marked “XXX.” These were simulated shocks to the learner, not actual electrical injuries.
• Milgram used several variants. In the condition published in 1963, the learner pounded on the wall at the higher settings; later voice-feedback variants used spoken protests. Do not combine all procedures into one supposedly identical baseline.
• Experimenter (authority) insisted: "Please continue" or "The experiment requires that you continue"
• No actual shocks delivered — learner was acting

Results:

• In the reported 1963 condition, 26 of 40 participants continued to the maximum setting: $$\text{Rate}=\frac{26}{40}\times 100\%=65\%$$ This is a descriptive result for that sample and procedure, not a population constant.
• 100% went to at least 300 volts (labeled "Intense Shock")
• Participants showed extreme stress (sweating, trembling, nervous laughter)
• Yet most continued when authority insisted
• Predicted only 1-3% would go to maximum — actual result shocked researchers

Factors That Increase Obedience:

• Proximity of authority: Authority figure present in room (obedience higher)
• Legitimacy of authority: Prestigious institution (Yale University)
• Institutional support: Authority backed by respected organization
• Distance from victim: Can't see/hear victim → higher obedience
• Gradual escalation: Small shocks first, then increase incrementally

Factors That Decrease Obedience:

• Proximity to victim: Touching victim → obedience dropped to 30%
• Distance from authority: Authority gives orders by phone → 21% obedience
• Presence of dissenter: Another "teacher" refuses → dramatically reduces obedience
• Lower authority legitimacy: Conducted in rundown office building → 48% obedience

The Pattern:

• Easy/mastered tasks: Presence of others enhances performance
• Difficult/new tasks: Presence of others impairs performance
• Arousal strengthens dominant (most likely) response
• For easy tasks, dominant response is correct → better performance
• For hard tasks, dominant response is incorrect → worse performance

Examples:

• Expert athlete performs better in front of crowd
• Beginner athlete chokes under pressure of audience
• Skilled pianist plays better in recital; novice plays worse

Why It Happens:

• Diffusion of responsibility — "others will pick up the slack"
• Individual effort not identifiable/evaluated
• Feeling dispensable to group outcome
• Free-rider problem — benefit without contributing fully

Examples:

Group project where some members contribute less because they know others will do the work. Tug-of-war where individuals pull less hard in a team than they would alone. Classic study: people clapping/shouting louder when alone than in groups.

Conditions That Promote It:

• Anonymity (masks, uniforms, crowds)
• Large group size
• Arousal and sensory overload
• Diminished personal responsibility
• Alcohol or drugs

Examples:

• Mob violence and rioting
• Online trolling and cyberbullying (anonymity of internet)
• Vandalism by groups in masks
• Halloween mischief when costumed

Symptoms of Groupthink:

• Illusion of invulnerability: Overconfidence, excessive optimism
• Self-censorship: Members withhold dissenting views
• Pressure on dissenters: Members who question are pressured to conform
• Illusion of unanimity: Silence viewed as agreement
• Mindguards: Some members protect group from contradictory information

Consequences:

• Poor decision-making
• Failure to consider alternatives
• Ignoring expert warnings
• Catastrophic outcomes

Examples: Bay of Pigs invasion, Challenger disaster, corporate scandals where boards ignored warning signs

How It Works:

• Group starts with slight tendency in one direction
• Discussion exposes members to persuasive arguments supporting that direction
• Members compare themselves and shift toward more extreme position to stand out
• Result: Group ends up more extreme than average of individual starting positions

Examples:

• Jury slightly leaning guilty becomes more certain of guilt after deliberation
• Online echo chambers radicalize political views
• Business team cautious about risk becomes even more risk-averse after discussion

Kitty Genovese Case (1964):

Widely publicized murder in New York where woman was attacked for over 30 minutes. Initial reports (later disputed) claimed 38 witnesses heard/saw attack but didn't intervene or call police. Sparked research into why people don't help.

Darley & Latané Research:

• Staged emergencies with varying numbers of bystanders
• Some laboratory designs found reduced individual reporting or helping when other bystanders were believed to be present. The direction and size of effects depend on the situation and measured outcome.
• Helping rates differed across particular studies and conditions. Do not combine percentages from different emergency paradigms as if they came from one uniform Darley–Latané experiment.
• Delay in helping also increased with more bystanders

Two Main Explanations:

When Are People More Likely to Help?

• When alone or with few others (no diffusion of responsibility)
• When they've witnessed others helping (modeling prosocial behavior)
• When in a good mood
• When they feel empathy for the victim
• When victim is similar to them
• When not in a hurry (Good Samaritan study — seminarians rushing didn't help)
• When situation is clearly an emergency

The Setup:

• College students randomly assigned to be "guards" or "prisoners"
• Mock prison created in Stanford psychology building basement
• Planned for 2 weeks
• Prisoners arrested at home by real police, stripped, deloused, given numbers
• Guards given uniforms, mirrored sunglasses, and authority

What Happened:

• Some guards engaged in abusive conduct; behavior varied, and researcher instructions and expectations complicate its interpretation.
• Participants assigned prisoner roles responded in varied ways, including distress, resistance and attempts to leave; one uniform passive response should not be assumed.
• Role performance, demand characteristics, researcher involvement and individual choices cannot be cleanly separated in this design.
• One prisoner had to be released after 36 hours (breakdown)
• Study stopped after only 6 days due to guards' cruelty

Significance & Criticisms:

The study prompted discussion of power, institutions and research ethics, but archival research and methodological criticism undermine simple causal conclusions. Selection, instructions, researcher participation and incomplete observation limit inference; it is not decisive proof that a role alone overrides personality.