Social Psychology: Perception, Influence, Attitudes and Groups

This chapter answers Section XH5.8 of XH5 Psychology, one of the six optional sections of the Humanities and Social Sciences (XH) paper, chosen at application: “Social perception: Attribution, Impression formation, Social categorization, Implicit personality theory. Social influence: Conformity, Compliance and obedience, Attitudes, Beliefs and Values. Evaluating the social world, Attitude formation, Attitude change and persuasion, Cognitive dissonance, Prejudice, Discrimination, Aggression, Power and prosocial behaviour, Belief systems and value patterns. Group dynamics, Theories of intergroup relations and conflicts.” It teaches how we read other people and explain their behaviour, how others change what we think and do, how attitudes form and shift, and how groups shape behaviour and come into conflict.

1. Social perception and attribution

Attribution is the process of inferring the causes of behaviour. Fritz Heider, who founded the field, argued that ordinary people act as “naive psychologists” who explain acts by either dispositional (internal) causes — traits, abilities, effort — or situational (external) causes. Jones and Davis's correspondent inference theory asks when we infer that an act reflects a stable disposition: we do so most when the act is freely chosen, unexpected (low in social desirability) and has distinctive, non-common effects. Harold Kelley's covariation model applies when we have information across time and people, using three kinds of evidence: consensus (do others act this way toward the same stimulus?), distinctiveness (does the person act this way only toward this stimulus?) and consistency (does the person act this way every time?).

Attribution is systematically biased. The fundamental attribution error (Ross) is the tendency, when explaining others, to overweight disposition and underweight the situation. The actor-observer effect adds that we explain our own behaviour more situationally than we explain the same behaviour in others. The self-serving bias credits our successes to ourselves and blames our failures on circumstances, protecting self-esteem. High consensus, high distinctiveness and high consistency together point to an external cause; low consensus, low distinctiveness and high consistency point to an internal cause.

⚠️ Error versus self-serving bias
Do not confuse the two. The fundamental attribution error is about explaining other people (over-attributing to their character). The self-serving bias is about explaining oneself (taking credit for success, deflecting blame for failure). A stem describing someone judging a stranger harshly for one slip uses the error; a stem describing someone excusing their own slip uses the self-serving or actor-observer pattern.

2. Impression formation, categorization and implicit personality theory

Impression formation is how scattered bits of information are combined into a unified picture of a person. Solomon Asch showed the process is configural, not a simple sum: some traits are central and reorganise the meaning of the rest. Changing a single word from “warm” to “cold” in an otherwise identical trait list transformed the whole impression, whereas changing “polite” to “blunt” did little. Two further regularities: the primacy effect, in which information presented early weighs more than later information, and the halo effect, in which one salient good (or bad) quality colours judgments of unrelated qualities.

To manage a complex social world the mind categorises people into groups — by age, gender, occupation, nationality — which saves effort but underlies stereotyping and the sharpening of in-group versus out-group boundaries. An implicit personality theory is each person's private web of assumptions about which traits go together (for example, the untested belief that a quiet person is also unfriendly). Such theories let us “go beyond the information given”, filling gaps with expectations, which speeds judgment but can entrench error.

3. Social influence: conformity, compliance and obedience

Conformity is adjusting one's behaviour or belief to match a group. Muzafer Sherif used the autokinetic effect (an ambiguous task, where a stationary point of light seems to move) and found individual estimates converged on a shared group norm. Solomon Asch used an unambiguous task — matching a line to one of three comparison lines — and still found that participants went along with a unanimous but plainly wrong majority on a substantial share of critical trials, with most conforming at least once. Deutsch and Gerard separated two motives: informational influence (accepting the group as evidence about reality) and normative influence (going along to be liked or accepted).

Compliance is agreeing to a direct request, and several techniques raise it. The foot-in-the-door technique secures a small request first, so a later large request fits the person's new self-image as someone who helps. The door-in-the-face technique does the reverse: a large request is refused, and the smaller real request that follows gains from contrast and from the sense of a reciprocal concession. Low-balling wins commitment at an attractive price, then reveals added costs. Obedience is acting on a direct order from an authority. Stanley Milgram's experiments showed that a majority of ordinary people would, under an experimenter's prodding, deliver what they believed were dangerous electric shocks to a protesting learner; obedience fell when the victim was closer, the authority was more distant or less legitimate, or a peer refused.

🎯 Why Asch's result was so striking
Sherif's conformity is unsurprising: the task was ambiguous, so leaning on others is reasonable informational influence. Asch's task had a visibly correct answer, so conforming to a wrong majority is hard to put down to information alone — it largely reflects normative pressure, the pull of not wanting to stand alone. That is why the two studies are a standard contrast.

4. Attitudes, persuasion and cognitive dissonance

An attitude is an evaluation of an object, often analysed into three components (the ABC model): affective (feelings), behavioural (action tendencies) and cognitive (beliefs). A belief is a cognition about what is true; a value is an abstract, enduring guiding principle (Rokeach separated terminal end-states from instrumental modes of conduct). Attitudes form through classical and operant conditioning, observational learning, and mere exposure, by which repeated contact alone increases liking.

Persuasion is attitude change through a message. The Elaboration Likelihood Model (Petty and Cacioppo) proposes two routes. The central route operates when the audience is motivated and able to think carefully, so the quality of the arguments decides the outcome and the change is durable. The peripheral route operates when motivation or ability is low, so people rely on cues — the source's attractiveness or expertise, the sheer number of arguments — and the change is shallower.

ℹ️ When do attitudes predict behaviour?
Attitudes often predict behaviour only weakly, so the link has conditions. Attitudes predict behaviour better when they are specific to the act in question rather than general, when they are strong, personally important and easily brought to mind, and when situational pressures and social norms are weak. Measuring a broad attitude but a single narrow act is the classic reason the two seem not to match.

Cognitive dissonance (Leon Festinger) is the discomfort of holding inconsistent cognitions, which pushes a person to restore consistency, often by changing an attitude. Festinger and Carlsmith found that participants paid only a trivial amount to say a dull task was fun later believed it more than those paid well: with little external justification to lean on, they reduced dissonance by shifting their own attitude.

5. Prejudice, aggression, power and prosocial behaviour

Prejudice is a (usually negative) attitude toward a group and its members; discrimination is the unfair behaviour that expresses it; a stereotype is the cognitive belief about the group's attributes. The three can come apart: prejudice can exist without acted-on discrimination, and discrimination can be driven by norms without personal prejudice. Aggression — behaviour intended to harm — is explained by the frustration-aggression account, by social-learning theory (Bandura's studies of children imitating aggression they observed), and by biological and situational cues together.

Power is the capacity to influence others. French and Raven named five bases: reward power (giving benefits), coercive power (threatening punishment), legitimate power (an accepted right to direct), referent power (identification with and admiration of the influencer) and expert power (superior knowledge). Prosocial behaviour benefits others. Latane and Darley's work on the bystander effect showed that a person in need is often less likely to get help as the number of bystanders grows, through diffusion of responsibility (each witness assumes another will act) and pluralistic ignorance (each reads the others' inaction as a sign nothing is wrong).

6. Group dynamics and intergroup relations

  • Social facilitation: the presence of others improves performance on simple, well-learned tasks but impairs it on complex, novel ones, because others raise arousal and arousal strengthens the dominant response (Zajonc's account of Triplett's early finding).
  • Social loafing: people often exert less effort on a collective task when individual contributions cannot be identified (the Ringelmann effect), the reverse of facilitation.
  • Groupthink (Janis): a cohesive group under pressure for unanimity suppresses dissent and realistic appraisal, producing poor decisions; group polarization makes a group's average view more extreme than its members' initial private views.

Two theories dominate intergroup relations. Realistic conflict theory (Sherif, from the Robbers Cave study) holds that hostility grows out of real competition over scarce resources and is reduced by superordinate goals that both groups need each other to reach. Social identity theory (Tajfel and Turner) holds that merely categorising people into “us” and “them” — even on a trivial basis, as the minimal-group studies showed — is enough to produce in-group favouritism, because people draw self-esteem from their groups. Gordon Allport's contact hypothesis says that contact reduces prejudice only under conditions: equal status, common goals, intergroup cooperation and the support of authorities or norms.

Key takeaways

  • Attribution explains behaviour by disposition or situation; Kelley's covariation model reads consensus, distinctiveness and consistency (high-high-high points external, low-low-high points internal). The fundamental attribution error over-attributes others' acts to character; the self-serving bias protects one's own esteem.
  • Impression formation is configural: central traits (warm-cold) reorganise the whole impression, and the primacy and halo effects bias it. Categorization underlies stereotyping, and an implicit personality theory supplies assumptions about which traits co-occur.
  • Conformity: Sherif's ambiguous task shows informational influence, Asch's unambiguous task shows chiefly normative pressure. Compliance uses foot-in-the-door (small then large) and door-in-the-face (large then small); obedience, shown by Milgram, falls with victim proximity and weaker or more distant authority.
  • Attitudes have affective, behavioural and cognitive components. The Elaboration Likelihood Model's central route (strong arguments, careful thought) gives durable change, the peripheral route (cues) gives shallow change. Festinger's cognitive dissonance drives attitude change when external justification is low.
  • Prejudice is an attitude, discrimination the behaviour, a stereotype the belief. French and Raven's five bases of power are reward, coercive, legitimate, referent and expert. The bystander effect arises through diffusion of responsibility and pluralistic ignorance.
  • In groups, others' presence facilitates simple tasks and impairs complex ones; social loafing cuts effort when contributions are anonymous; groupthink harms decisions. Realistic conflict theory ties hostility to resource competition (eased by superordinate goals); social identity theory ties it to mere categorization; Allport's contact hypothesis sets the conditions under which contact reduces prejudice.

Practice questions (18)

Attempt each one before opening the answer. Every explanation names the tempting wrong option as well as the right one, because that is where marks are lost.

  1. The tendency, when explaining another person's behaviour, to overestimate the influence of their personal disposition and underestimate the influence of the situation is known as:

    1. The self-serving bias, which protects a person's own self-esteem after one of their successes or failures
    2. The false-consensus effect, in which people overestimate how widely their own views are shared by others
    3. The halo effect, in which a single good quality colours judgments of a person's other, unrelated qualities
    4. The fundamental attribution error, the tendency to over-attribute others' acts to their character
    Show answer

    Answer: D — The fundamental attribution error, the tendency to over-attribute others' acts to their character

    Over-weighting disposition and under-weighting the situation when judging others is the fundamental attribution error. The self-serving bias is about explaining one's own outcomes, not others' behaviour; the false-consensus effect is about overestimating agreement; the halo effect is about one trait spilling over onto others.

  2. A student who scores well on a test says it shows her real ability, but after a poor score on the next test she blames an unfair paper and a noisy exam hall. This pattern of explanation is an example of:

    1. The self-serving bias, attributing successes to oneself and failures to external circumstances
    2. The fundamental attribution error, over-attributing another person's behaviour to their character
    3. The just-world hypothesis, the belief that people generally get the outcomes that they deserve
    4. Pluralistic ignorance, misreading others' calm inaction as a sign that nothing is wrong
    Show answer

    Answer: A — The self-serving bias, attributing successes to oneself and failures to external circumstances

    Crediting success to one's ability while blaming failure on outside factors is the self-serving bias, which guards self-esteem. The fundamental attribution error concerns judging others, not oneself; the just-world hypothesis and pluralistic ignorance describe unrelated phenomena, so they do not fit a person explaining her own mixed results.

  3. In Asch's studies of impression formation, what happened when a single 'central' trait in an otherwise identical list of trait words was changed, for example from 'warm' to 'cold'?

    1. It changed the overall impression only when that trait word happened to appear last in the list
    2. It markedly changed the whole impression that people formed of the described person
    3. It had no measurable effect, since impressions are a simple average of all the separate traits
    4. It affected people's memory for the trait list but left the overall impression unchanged
    Show answer

    Answer: B — It markedly changed the whole impression that people formed of the described person

    Asch found impressions are configural: central traits such as warm-cold reorganise the meaning of the other traits, so swapping one transformed the whole impression. The simple-average option is exactly the additive view Asch's result refutes; the list-position and memory options misstate the finding.

  4. In Deutsch and Gerard's distinction, going along with a group because you take it to be a better source of information about an uncertain reality is best described as:

    1. Normative social influence, driven by the wish to be liked and accepted by the group
    2. Obedience, which is acting on the direct command of a recognised authority figure
    3. Informational social influence, driven by taking the group as evidence about reality
    4. Compliance, which is agreeing to a direct request in order to gain a tangible reward
    Show answer

    Answer: C — Informational social influence, driven by taking the group as evidence about reality

    Relying on the group as a source of information about an uncertain reality is informational influence, which is especially strong on ambiguous tasks. Normative influence is about acceptance, not accuracy; obedience is about authority's commands; compliance is about agreeing to requests, none of which captures the informational motive described.

  5. A charity worker first asks a passer-by to donate a very large sum, which is refused, and then immediately asks for a small donation, which is given. Which compliance technique is this?

    1. The foot-in-the-door technique, in which a small request is granted before a larger one is made
    2. The door-in-the-face technique, in which a large request is refused before a smaller real one is made
    3. The low-balling technique, in which an attractive offer is accepted before added costs are revealed
    4. The scarcity technique, in which a request is made more tempting by stressing limited availability
    Show answer

    Answer: B — The door-in-the-face technique, in which a large request is refused before a smaller real one is made

    A large request refused, immediately followed by a smaller request granted, is the door-in-the-face technique, which works through contrast and reciprocal concession. The foot-in-the-door runs the opposite way (small then large); low-balling and scarcity use different mechanisms entirely.

  6. In the three-component (ABC) model of attitudes, the 'cognitive' component refers to:

    1. The beliefs and thoughts that the person holds about the attitude object
    2. The feelings and emotions that the attitude object arouses in the person
    3. The actions and behavioural intentions that the person directs toward the object
    4. The social norms and group pressures that surround the attitude object
    Show answer

    Answer: A — The beliefs and thoughts that the person holds about the attitude object

    The cognitive component is the set of beliefs and thoughts about the object. Feelings are the affective component and action tendencies are the behavioural component; social norms are not one of the three ABC components at all, so that option is a distractor.

  7. According to the Elaboration Likelihood Model of persuasion, attitude change through the central route occurs mainly when:

    1. The audience relies on peripheral cues such as the speaker's attractiveness or the number of arguments
    2. The audience is distracted and processes only the surface features of the message
    3. The message is simply repeated so often that it is accepted without any real thought
    4. The audience is motivated and able to think carefully, so the strength of the arguments decides the outcome
    Show answer

    Answer: D — The audience is motivated and able to think carefully, so the strength of the arguments decides the outcome

    The central route operates when people are both motivated and able to scrutinise a message, so argument quality drives the change, which tends to be durable. Reliance on peripheral cues, distraction and mere repetition all describe the peripheral route, which produces shallower, less stable change.

  8. In social psychology, how are prejudice and discrimination most precisely distinguished?

    1. Prejudice is a negative attitude toward a group, while discrimination is the unfair behaviour of treating its members badly
    2. Prejudice is an observable behaviour that others can readily see, while discrimination is the hidden inner attitude lying behind it
    3. Prejudice is a belief that is always factually accurate, while discrimination is one that is always false
    4. Prejudice is an emotion felt only by groups, while discrimination is an emotion felt only by individuals
    Show answer

    Answer: A — Prejudice is a negative attitude toward a group, while discrimination is the unfair behaviour of treating its members badly

    Prejudice is the attitude (an evaluation of a group), and discrimination is the behaviour that acts it out unfairly. The choice that labels prejudice a behaviour and discrimination an inner attitude reverses the two; the others misdescribe both terms, since neither is a claim about factual accuracy or an emotion restricted to groups or individuals.

  9. The finding that a victim in need is often less likely to receive help as the number of onlookers grows (the bystander effect) is explained largely by:

    1. Social facilitation, whereby the presence of others raises arousal and improves task performance
    2. The mere-exposure effect, whereby repeated contact with a stimulus steadily increases a person's liking for it
    3. Deindividuation, whereby people in a crowd lose self-awareness and act on impulse more freely
    4. Diffusion of responsibility, whereby each witness assumes that someone else present will take action
    Show answer

    Answer: D — Diffusion of responsibility, whereby each witness assumes that someone else present will take action

    As the group of witnesses grows, responsibility to act is spread thinner across them, so each feels less personally obliged — diffusion of responsibility, along with pluralistic ignorance. Social facilitation and mere exposure concern performance and liking; deindividuation concerns loss of self-awareness in crowds, not failure to help.

  10. Using Kelley's covariation model, an observer concludes that a colleague laughed at a particular comedian because of something about the comedian (an external, stimulus attribution) rather than about the colleague. This conclusion is best supported when the observer learns that:

    1. Almost no one else laughs at this comedian, the colleague laughs at many different comedians, and the colleague laughs at this one every time
    2. Almost everyone laughs at this comedian, the colleague laughs at few other comedians, and the colleague laughs at this one every time
    3. Almost everyone laughs at this comedian, but the colleague has happened to laugh at him on this one occasion only
    4. Almost no one else laughs at this comedian, and the colleague laughs readily at nearly every comedian he ever sees
    Show answer

    Answer: B — Almost everyone laughs at this comedian, the colleague laughs at few other comedians, and the colleague laughs at this one every time

    An external (stimulus) attribution needs high consensus, high distinctiveness and high consistency: most people laugh at this comedian, the colleague laughs at few others, and he laughs every time. The option with low consensus and low distinctiveness points to the person (internal); low consistency points to the particular circumstance; and the choice where the colleague laughs readily at nearly every comedian again points to the person, not the comedian.

  11. The foot-in-the-door technique raises compliance mainly because:

    1. A large first request makes the smaller real request that follows seem quite modest to the person through a process of perceptual contrast
    2. People feel socially obliged to reciprocate a concession that the requester appears to have just made
    3. Agreeing to a small request shifts a person's self-image toward being a helper, so a later larger request fits that image
    4. A low opening price secures commitment before additional hidden costs are gradually revealed
    Show answer

    Answer: C — Agreeing to a small request shifts a person's self-image toward being a helper, so a later larger request fits that image

    Foot-in-the-door works through self-perception: granting a small request leads people to see themselves as the kind of person who helps, so they agree to a larger related request to stay consistent. Perceptual contrast and reciprocal concession are the mechanisms of door-in-the-face, and the low opening price is low-balling, so each of those belongs to a different technique.

  12. In Festinger and Carlsmith's classic study, participants who were paid only a trivial sum ($1) to tell a waiting person that a boring task was enjoyable later rated the task as more enjoyable than those paid a large sum ($20). Cognitive dissonance theory explains this because:

    1. The participants who were paid the large sum of money enjoyed the task more because the generous payment had simply lifted their general mood at the time
    2. The trivially-paid participants lacked external justification for lying, so they reduced dissonance by shifting their attitude to fit what they said
    3. The participants paid a trivial sum were simply bribed in a more effective way than the others were
    4. Paying people to do any task always makes them like that task less, whatever the size of the payment
    Show answer

    Answer: B — The trivially-paid participants lacked external justification for lying, so they reduced dissonance by shifting their attitude to fit what they said

    With only $1, participants had too little external reason to justify saying something they did not believe, leaving dissonance between 'the task was dull' and 'I said it was fun'; they resolved it by genuinely deciding the task was more enjoyable. The $20 participants had ample justification, so they felt little dissonance. The mood, bribe and blanket-payment options all miss the insufficient-justification mechanism.

  13. Volunteers are split into two groups by the toss of a coin, never meet, and have no resources to compete for, yet they still allot more reward to members of their own group than to the other group. Which theory most directly explains this result?

    1. Realistic conflict theory, because hostility arises from competition between groups over scarce resources
    2. Groupthink, because a cohesive group under pressure for unanimity suppresses dissent and realistic appraisal
    3. Social identity theory, because mere categorization into ‘us’ and ‘them’ is enough to produce in-group favouritism
    4. Social loafing, because individuals exert less effort when their personal contribution cannot be identified
    Show answer

    Answer: C — Social identity theory, because mere categorization into ‘us’ and ‘them’ is enough to produce in-group favouritism

    Favouring the in-group when the groups were created trivially and have no competition is the minimal-group result that social identity theory explains: categorization alone, with people drawing self-esteem from group membership, is enough. Realistic conflict theory is the tempting rival but requires real competition over resources, which is absent here; groupthink and social loafing concern decision quality and effort within a group, not bias between groups.

  14. Consider these statements about social perception, labelled P, Q, R and S. Which are correct?
    P: The primacy effect means that information encountered early about a person tends to weigh more heavily than information encountered later.
    Q: The halo effect is the tendency to let one salient good or bad quality colour judgments of a person's unrelated qualities.
    R: An implicit personality theory is a person's set of assumptions about which traits tend to go together.
    S: Social categorization eliminates stereotyping by forcing people to judge each individual entirely on their own merits.

    1. P
    2. Q
    3. R
    4. S
    Show answer

    Answer: A — P; B — Q; C — R

    P, Q and R are correct definitions. S is false and reverses the truth: categorization underlies stereotyping rather than eliminating it, because it leads people to judge individuals by their group. P states the primacy effect, Q the halo effect, and R the implicit personality theory accurately.

  15. Consider these statements about Milgram's obedience research, labelled P, Q, R and S. Which are correct?
    P: A clear majority of participants, under the experimenter's prodding, continued to the maximum shock level on the apparatus.
    Q: Obedience fell when the learner was brought physically closer to the participant.
    R: Obedience fell when the authority figure was made more distant or gave orders by telephone.
    S: Obedience was unaffected by whether another participant was seen to refuse to continue.

    1. P
    2. Q
    3. R
    4. S
    Show answer

    Answer: A — P; B — Q; C — R

    P, Q and R are correct. S is false: a disobedient peer who refused to continue sharply reduced obedience, one of Milgram's clearest situational effects. P reports the headline result (a majority reached the maximum), Q the proximity-of-victim effect and R the proximity-and-legitimacy-of-authority effect.

  16. Consider these statements about group processes and intergroup relations, labelled P, Q, R and S. Which are correct?
    P: Social facilitation predicts that the presence of others improves performance on simple, well-learned tasks but impairs it on complex, unfamiliar ones.
    Q: Realistic conflict theory holds that intergroup hostility grows from real competition over scarce resources and can be eased by superordinate goals.
    R: Social identity theory holds that merely categorizing people into in-group and out-group can produce in-group favouritism, even on a trivial basis.
    S: Allport's contact hypothesis holds that any contact between groups, under any conditions, reliably reduces prejudice.

    1. P
    2. Q
    3. R
    4. S
    Show answer

    Answer: A — P; B — Q; C — R

    P, Q and R are correct. S is false: the contact hypothesis holds that contact reduces prejudice only under specific conditions — equal status, common goals, cooperation and institutional support — not under just any conditions. P states social facilitation, Q realistic conflict theory and R social identity theory accurately.

  17. French and Raven identified several bases of social power. Consider the statements below, labelled P, Q, R and S. Which correctly match a base of power to its description?
    P: Reward power rests on the ability to give valued benefits for compliance.
    Q: Coercive power rests on the ability to threaten or impose punishment.
    R: Referent power rests on the target's identification with and admiration of the influencer.
    S: Expert power rests purely on holding a formally recognised position or office.

    1. P
    2. Q
    3. R
    4. S
    Show answer

    Answer: A — P; B — Q; C — R

    P, Q and R match correctly. S is false: resting on a recognised position or office is legitimate power; expert power rests on superior knowledge or skill. So the mismatch in S is the error, while reward, coercive and referent power are described accurately.

  18. In a replication of Asch's conformity paradigm, each participant faced 12 critical trials on which a unanimous majority gave a clearly wrong answer. One participant conformed to the wrong majority answer on 9 of these 12 critical trials. On what percentage of the critical trials did this participant conform? Give your answer as a whole number.

    Numerical answer — type the value.

    Show answer

    Answer: 75

    The conformity rate is the number of conforming trials divided by the number of critical trials, as a percentage: 9 / 12 = 0.75, which is 75%. The data needed are entirely in the stem, so the answer is 75.