You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
Milgram's study told us that ordinary people obey destructive orders; it did not, by itself, tell us why. The Edexcel Social Psychology topic names two theories that attempt to explain obedience: agency theory (Milgram, 1974) and social impact theory (Latané, 1981). Agency theory is a specifically obedience-focused account rooted in a shift in perceived responsibility; social impact theory is a broader, quantitative account of social influence that treats the "pull" of other people as a force whose strength can, in principle, be predicted from three variables. This lesson develops both, shows how each maps onto Milgram's findings and variations, and evaluates them against each other. Because social impact theory generalises beyond obedience to conformity, bystander behaviour and persuasion, it is a genuinely synoptic idea within the topic.
Key Definition: An explanation of obedience is a theory that accounts for the psychological processes underlying why an individual follows the orders of an authority figure.
By the end of this lesson you should be able to:
Edexcel 9PS0 — Paper 1, Topic 1: Social Psychology. This lesson maps to the specification's two named explanations of obedience: agency theory (the agentic state, the autonomous state, the agentic shift and moral strain) and social impact theory (the roles of strength, immediacy and number, social force, and the division of impact). Both are required content and both are examinable for description, application and evaluation.
Assessment Objectives exercised:
Connects to…
Milgram proposed that people can operate in one of two psychological states, and that obedience results from a shift between them.
Key Definition: The agentic state is a mental state in which a person acts as an agent for an authority figure and consequently feels little personal responsibility for their own actions, because responsibility has been displaced onto the authority.
The transition from the autonomous to the agentic state is called the agentic shift. Milgram argued that we are socialised from an early age into hierarchical structures (family, school, workplace) in which deferring to legitimate authority is adaptive and expected. When we enter a situation governed by a legitimate authority — a laboratory run by a scientist, a hospital run by doctors — we readily make the agentic shift, handing over responsibility to the person in charge.
Being in the agentic state is not comfortable. Milgram noted that his participants experienced moral strain: the acute discomfort of doing something that conflicts with their conscience while feeling unable to stop. The visible distress in his study — the sweating, trembling and, in three cases, seizures — is the observable signature of moral strain.
To manage this strain and continue obeying, people rely on binding factors: aspects of the situation that allow the individual to reduce or ignore the damaging effect of their behaviour. Examples include:
Binding factors explain how a person can remain in the agentic state despite the moral strain pulling them toward disobedience.
graph TD
A["Autonomous state<br/>(I am responsible)"] -->|"Legitimate authority present<br/>→ agentic shift"| B["Agentic state<br/>(I am the authority's instrument)"]
B --> C["Moral strain<br/>(conflict with conscience)"]
C -->|"Binding factors manage the strain"| D["Obedience continues"]
C -->|"Strain overwhelms binding factors"| E["Return to autonomous state → disobedience"]
Agency theory explains Milgram's baseline result directly: participants entered the agentic state under the legitimate authority of a Yale scientist, felt the experimenter (not themselves) was responsible, and used binding factors to manage the strain. It also explains several variations: reducing the legitimacy of the authority (the run-down office, the ordinary clothes) made the agentic shift harder to sustain, so obedience fell; and the disobedient-models variation gave participants "permission" to return to the autonomous state.
Bibb Latané (1981) offered a very different kind of explanation. Rather than describing an internal state, social impact theory treats social influence as a force — analogous to a physical force such as the light falling on a surface — whose strength on any individual (the target) can be predicted from features of the source(s) exerting it. Latané called the total effect of other people on an individual social impact, and argued it depends on three factors.
Key Definition: Social impact is the effect that other people have on an individual's thoughts, feelings or behaviour, resulting from the real, implied or imagined presence or actions of those others.
| Factor | Meaning | Effect on impact |
|---|---|---|
| Strength | How important, powerful or authoritative the source is to the target (status, expertise, legitimacy) | The stronger the source, the greater the impact |
| Immediacy | How recent and how close (in time and space) the source is to the target, and whether anything intervenes between them | The more immediate the source, the greater the impact |
| Number | How many sources there are | More sources generally increase impact, but with diminishing returns |
A helpful mnemonic is SIN — Strength, Immediacy, Number. Latané expressed the relationship as a multiplicative one: impact is a function of the strength, immediacy and number of sources combined. Because the three combine, weakness on one factor can be offset by strength on another — a single but highly authoritative, physically present source can exert as much impact as several distant, low-status ones.
Latané observed that the effect of number is not linear. Adding a second source to one already present produces a large increase in impact; adding a tenth source to nine already present produces a much smaller increase. Impact grows with number but at a decelerating rate — each additional person has less marginal effect than the last. Latané described this as a marginally decreasing relationship, often illustrated by the fact that the impact rises roughly with the square root of the number of sources rather than in direct proportion. This is precisely the pattern Asch found when conformity plateaued after about three confederates: beyond a small majority, extra members add little further pressure.
Latané used the metaphor of a social force field. When a source (or group of sources) targets a single individual, the full force is concentrated on that person. But when the same source targets many people at once, the impact is divided among the targets — this is the division of impact (sometimes called the divisional effect or diffusion of impact). Each individual in a large audience feels only a fraction of the pressure that a lone individual would feel from the same source.
The division of impact explains diffusion of responsibility in bystander situations: when many bystanders witness an emergency, the "pressure to act" that a single witness would feel is divided among them, so each feels less compelled to help. It also explains why performing in front of a large crowd can feel less nerve-wracking to each performer in a large troupe than to a soloist: the audience's evaluative pressure is shared.
graph LR
A["Sources of influence"] -->|"Strength × Immediacy × Number"| B["Social impact on target"]
C["One source → many targets"] -->|"Division of impact"| D["Each target feels a<br/>fraction of the pressure"]
Social impact theory accounts for Milgram's findings and variations at least as neatly as agency theory:
| Milgram finding/variation | Social impact theory explanation |
|---|---|
| High baseline obedience (65%) | The experimenter is high strength (a Yale scientist), high immediacy (present, in the room, giving direct orders), and the sole source of authority targeting one participant — maximal impact |
| Proximity variations (obedience falls) | Bringing the learner closer increases the learner's immediacy as a competing source pulling toward disobedience; the experimenter's telephoned absence reduces the experimenter's immediacy, cutting their impact |
| Location variation (Bridgeport) | The run-down office reduces the experimenter's strength (less institutional legitimacy) |
| Uniform variation (ordinary clothes) | Removing the lab coat reduces the source's strength |
| Two disobedient teachers (obedience → 10%) | The confederates add number on the side of disobedience and divide the impact of the experimenter's authority, so the pressure to obey on the genuine participant drops sharply |
This is the theory's great advantage: it explains the whole family of variations with one set of levers (strength, immediacy, number), and it does so for conformity too.
The clearest point of contact between social impact theory and conformity research is Asch's group-size finding. Recall that Asch found conformity rose as confederates were added up to about three, and then levelled off — adding a fourth, fifth or tenth confederate made little further difference. Social impact theory predicts exactly this shape. Because the effect of number is marginally decreasing (the psychosocial law), the jump from one source to two, and from two to three, is large, but each further source adds progressively less. In the language of the theory, a majority of three is already exerting most of the impact that a much larger majority could exert, so the curve flattens. This is a genuinely impressive convergence: a theory built to explain influence in general correctly predicts the precise, non-linear pattern found in a specific conformity experiment it was not designed around. It is the kind of cross-domain prediction that gives social impact theory much of its evidential weight, and it is well worth citing when evaluating the theory's explanatory reach.
Note, however, the mirror-image limitation this reveals. If impact rises with number, then a source that is low on number — a lone individual or a tiny group — should exert little influence. Yet minority influence shows that a numerically weak but consistent source can produce deep, lasting change. Social impact theory therefore predicts the group-size plateau beautifully but is poorly equipped to explain minority influence, because it measures the quantity of social force rather than the quality and persistence of an argument. This tension — strong on majorities, weak on committed minorities — is one of the most incisive evaluation points you can make about the theory.
It is worth seeing how the two "directions" of Latané's model — impact concentrating on a lone target versus impact dividing across many — work in a concrete case, because being able to reason with the model, not just recite the three factors, is what separates strong application from rote description.
Suppose an authority figure exerts a fixed quantity of social force. Latané's psychosocial law holds that the impact felt by a target rises with the square root of the number of sources, so impact i scales roughly as i∝N, where N is the number of sources bearing on the target. Adding a second confederate to a single source multiplies impact by about 2≈1.41 — a large jump — whereas moving from nine sources to ten multiplies it by only 10/9≈1.05, a negligible gain. This single relationship reproduces the whole shape of Asch's curve without any extra assumptions.
The division of impact runs in the opposite direction. When one strong source is directed not at a lone individual but at a crowd of targets, the same total force is shared out, so the pressure on each person falls roughly with the square root of the number of targets: a single witness to an emergency feels the full pull to act, but each of ten witnesses feels only about 1/10≈0.32 of that pull — Latané's account of diffusion of responsibility. The elegance of using one square-root relationship to generate both the group-size plateau (multiplication onto a target) and bystander apathy (division across targets) is a genuine strength; the same arithmetic that predicts why a fourth confederate barely matters also predicts why a large audience dilutes each individual's sense of obligation.
The most searching criticism of social impact theory is not any single failed prediction but a limitation of kind: the theory is descriptive rather than explanatory. It specifies, often with impressive precision, how much impact a configuration of sources will have, but it does not say why those factors produce impact at all. Why does "strength" — status, expertise, legitimacy — move us? Why does physical closeness amplify influence? The model treats these as brute inputs to a formula and remains silent on the underlying psychological mechanism, whereas agency theory at least offers a process (the shift of perceived responsibility). Predicting the magnitude of a phenomenon is not the same as explaining its cause, and by that standard social impact theory is better understood as a powerful mathematical description of social influence than as a full theory of it.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.