Take the Free Diagnostic

An audience, a group, and individual performance

MCAT content library / Psychology & sociology / An audience, a group, and individual performance

Use the comparison conditions to separate being watched, having effort pooled, and sharing responsibility.

About 17 minutes, including the worked examples and checks.

Useful foundations
On this page

Concept: “other people are present” is not a mechanism

A musician makes fewer errors in a familiar piece when listeners arrive, yet struggles more with a newly learned sequence. In a different setting, team members submit fewer entries when only the total is evaluated. Both involve other people, but the comparisons target different processes. One concerns performing while observed; the other concerns how individual effort contributes to an evaluated collective outcome.

AAMC groups audience, loafing, deindividuation, and bystander effects together, but their evidence should remain distinct.5 Identify the outcome first: accuracy, effort, rule-following, or initiation of help. Then identify what changed about the social setting.

Pattern: an audience can help one response and hinder another

In the dominant-response account of social facilitation, an audience increases the likelihood of a response that is already dominant in that setting. When the dominant response is correct, performance can improve. When an incorrect or poorly learned response dominates, performance can worsen. The prediction concerns the response’s status, not a universal benefit of company or a permanent property of the person.1

The cases and data here are original teaching examples. Participants perform either a well-practiced symbol-matching task or a new reversed-rule version. Within each task, they are randomly assigned to work alone or with a silent observer. The outcome is mean correct responses out of 20; time available and instructions are matched.

TaskAloneObservedObserved minus alone
Practiced1819+1
New reversed rule129−3
Mean correct responses out of 20
Audience effects depend on the taskMean correct answers out of twenty: practiced task 18 alone and 19 observed; new reversed-rule task 12 alone and 9 observed. The lines compare group means, not individual trajectories.Correct responses /20051015201819129AloneObservedPracticedNew reversed rule
Original illustrative means. The audience contrast is +1 for the practiced task and −3 for the new rule; the table supplies every value.

The practiced-task difference is +1 response, while the new-task difference is −3. The descriptive difference between those audience contrasts is 1 − (−3) = 4 responses. Reporting only a pooled average could conceal this dependence on the task. With equal task weights, the means would be 15 alone and 14 observed: a one-response pooled decline that misses the improvement on the practiced task.

The pattern is consistent with the supplied dominant-response model, but no physiological arousal or attention measure is given. It does not prove which process caused the differences. The new task also changes a rule, not merely an arbitrary label of “hard.” A useful follow-up would confirm which errors dominate before the audience manipulation and retain comparable scoring and exposure.

Application: collective output versus individual effort

Social loafing concerns reduced individual effort when contributions are combined into a collective outcome. Reduced individual evaluability can contribute, but lower group output is not automatically loafing. A group may lose output through coordination problems, duplicate work, unclear instructions, or unequal opportunity to contribute.2

Consider four-person teams completing independent sorting trials at separate terminals. Each person receives the same number of equally difficult trials. In one condition, each contribution is evaluated separately; in another, only the pooled total is evaluated. Anonymous trial codes let researchers calculate per-person output without presenting individuals’ scores to an evaluator in the pooled condition.

Suppose mean correct items per person fall from 16 to 12 when evaluation becomes pooled, while available trials and error rates among attempted items remain comparable. The decline is 4/16 = 25% relative to individually evaluated output. Since people work on separate trials, lost coordination is less plausible than in a shared construction task. The result is compatible with reduced effort, although time engaged or attempted trials would make that inference more direct.

Keeping the same group and restoring individual evaluation would test the accountability explanation more closely than replacing the group with a different set of people. It would still be useful to examine evaluation anxiety and perceived importance of the task. Individual identification can alter more than one experience.

Changed case: who is expected to act?

A bystander effect is a difference in an individual’s likelihood or speed of helping when other potential helpers are present. Diffusion of responsibility is one possible process: the person expects someone else to act or feels less uniquely responsible. The original controlled work varied perceived co-witnesses; it did not establish that every crowd is indifferent or that help always decreases as a group grows.3

In a fictional online workshop, all attendees see the same request to flag a mislabeled file. Everyone has the same reporting button and knows what the error is. When the message is addressed to the group, 6 of 20 selected participants report it. When each selected participant is explicitly assigned the reporting role, 14 of 20 do so. These are 30% and 70%, a 40-percentage-point difference.

The assignment contrast weakens an inability-to-report explanation because the tools and demonstrated knowledge are held constant. It is consistent with responsibility allocation mattering. It does not prove diffusion alone: being singled out can also change attention or perceived evaluation. Measuring notice of the request and crossing private versus public assignment could address those alternatives.

Rule: anonymity does not determine the direction of behavior

Historical deindividuation accounts connect group immersion and reduced identifiability with changes in self-awareness or accountability.2 However, anonymity does not logically imply aggression. Research evaluating these accounts found that behavior depends substantially on norms made salient by the situation, challenging a simple loss-of-restraint story.4

Imagine two anonymous discussion groups: one emphasizes careful correction of errors, the other rewards rapid jokes. Different posting patterns could reflect those different local expectations. To test the effect of identifiability, compare identified and anonymous conditions within the same norm context; to test norm dependence, vary the norm separately. Changing both together would not tell you which produced the result.

Across these cases, use the specific comparison to name the supported claim. Observation can alter performance, pooled evaluation can alter effort, and shared responsibility can alter initiation of action. A group label alone establishes none of them.

Check your reasoning

Choose an answer before opening its feedback. Each option explains the reasoning behind that choice.

Check 1

A model predicts that an audience strengthens the currently dominant response. In a new sorting task, the familiar but now incorrect response occurs most often before training. What does the model predict if an audience is added at that point, with other conditions held constant?

A. Accuracy improves because every dominant response is correct.

Review this choice. The stem states that the dominant response is incorrect in the new task. Dominance and correctness are separate.

B. Accuracy is unchanged because an audience can influence only motivation.

Review this choice. The supplied model predicts a change in response likelihood; it does not restrict the outcome to a motivation report.

C. Incorrect familiar responses become more frequent.

Correct. Strengthening the response that already dominates increases the specified error. After learning, the dominant response could change.

D. Participants necessarily exert less effort because their scores are pooled.

Review this choice. No pooled evaluation or reduction in effort is described. That imports a different social process.

The principle to carry forward

Establish which response is dominant before predicting whether facilitation improves accuracy.

Check 2

Workers perform independent, matched trials. Output per worker is lower when only the team total is evaluated. Group size, trial access, and task instructions stay fixed. Which follow-up best tests reduced individual evaluability as an explanation?

A. Add harder trials only to the pooled condition.

Review this choice. A difficulty change adds an alternative cause of lower output.

B. Restore individual evaluation within the same group arrangement.

Correct. This changes evaluability while retaining the group and task structure, creating the most relevant test of the proposed explanation.

C. Replace the pooled group with highly experienced workers.

Review this choice. Experience and personnel change alongside the evaluation condition.

D. Use a shared task requiring substantially more coordination.

Review this choice. Additional coordination demands create a separate source of performance loss.

The principle to carry forward

Test the proposed effort mechanism while preserving opportunity, ability, and coordination demands.

Check 3

All participants notice a request, know how to respond, and have the same response tool. A responsibility-diffusion account predicts less action when responsibility is shared than when it is individually assigned. Which observation best distinguishes this account from a simple inability-to-respond account?

A. Action rises after individual assignment even though response skill and access are unchanged.

Correct. The changed responsibility with unchanged capability makes inability alone an inadequate explanation. Other effects of assignment still require consideration.

B. Action rises after participants receive a previously unavailable response tool.

Review this choice. The new tool directly changes ability to respond, so this pattern fits the inability account.

C. Action is unmeasured, but participants say they enjoy group work.

Review this choice. Preference for group work does not test the predicted action difference.

D. Participants fail when neither condition provides a working response tool.

Review this choice. Failure under physical inability does not distinguish how responsibility would affect capable participants.

The principle to carry forward

Hold capability constant when testing whether perceived responsibility matters.

Continue learning

Browse psychology & sociology · MCAT content library

MCAT preparation and roadmap · Aim High, Score Higher

Sources

1. Zajonc and Sales (1966) — Social Facilitation of Dominant and Subordinate Responses

2. OpenStax Psychology 2e — Conformity, Compliance, and Obedience

3. Darley and Latané (1968) — Bystander Intervention in Emergencies

4. Postmes and Spears (1998) — Deindividuation and Antinormative Behavior

5. AAMC — Content Category 7B