Public agreement may be a poor estimate of what group members actually believe in private.
Pluralistic ignorance is not a single famous paradox but a mature social-psychological construct. People may publicly follow a norm while privately disagreeing because they mistakenly believe that others do endorse it.
Public agreement may be a poor estimate of what group members actually believe in private.
In classic examples, individuals underestimate how many others share their private objections. Because everyone reads others’ public behaviour as evidence of their private beliefs, a norm persists that is widely rejected in private.
The phenomenon is mature in social psychology and connects to employee voice and organisational silence. LLM-agent results are preliminary and must not replace human evidence.
Private belief → expectations about others → public conformity → perceived norm. The loop can reinforce itself without the public norm being widely endorsed in private.
AI transfer to investigate: AI may create consensus-first summaries and thereby influence the perceived norm. The social mechanism existed long before AI.
Proposed test, not yet conducted: Compare consensus-first AI with anonymous pre-elicitation and distribution-aware AI summaries. Measure the mismatch between private preference, perceived norm, public statement and final choice.
This case provides the scientific core underlying many Abilene-like stories.
Measure private belief, perceived group norm and public expression separately. The proposed diagram is not a dataset and does not assume actual unanimity.
Prentice & Miller (1993), Pluralistic ignorance and alcohol use on campus — pp. 243–244 and 252–254
The discussion of Prentice and Miller (1993) uses a text copy of selected passages. Their findings do not validate results from language-model agents or organisational AI summaries.
Help improve this case
We review your contribution before updating the text or sources.