Question-order effects in national surveys follow an a priori 'QQ equality' derived from quantum probability

The empirical anchor for applying quantum probability to human judgment is a 2014 PNAS study (published as "Context effects produced by question orders reveal quantum nature of human judgments") by Zheng Wang, Tyler Solloway, Richard Shiffrin, and Jerome Busemeyer. It analyzed 70 national representative surveys (roughly 651–3,006 participants each) in which pairs of attitude questions were asked in both orders across respondents.

The distinctive move is that quantum probability theory yields an a priori, nonparametric prediction — the authors' "QQ (question-order) equality" — about how the two orderings must relate. Classical probability imposes no such constraint, but the quantum-derived equality held across the survey corpus. This gives quantum probability a replicated foothold in survey methodology, not merely in decision-theory lab tasks: the order in which "is Clinton honest?" and "is Gore honest?" are asked shifts responses in a lawful, predictable way.

The result is the empirical basis for the non-commuting-operator representation that a later network model builds on (claim-quantum-opinion-model-reduces-to-friedkin-johnsen), and it is one of the phenomena a 2016 experimental critique tried — and, for a different task, failed — to reproduce (claim-quantum-like-models-fail-conjunction-fallacy). That measured attitudes depend on measurement order also bears on how much stable introspective access people have to their own opinions (cf. introspection-access-problem).