Opinions reflect reception, acceptance, and sampling of elite cues.
Zaller's 1992 model. Its central prediction is counterintuitive: the most politically aware are most influenced by elite cues, because they receive more of them while resisting only those conflicting with prior commitments.
In practice
- A voter who reads three papers a day shifts on a policy the week their party's leadership does; a voter who reads none does not shift at all.
- A well-informed partisan changes position when their side does, and reports the change as reasoning, which is what the model predicts.