Forecasting collective trends in human opinion

Marlon Ramos1,2∗, Jia Shao3∗, Saulo D. S. Reis1,4∗, Celia Anteneodo, José S. Andrade, Jr , Shlomo Havlin, Hernán A. Makse 1 Levich Institute and Physics Department, City College of New York, New York, NY 10031, USA 2 Departamento de F́ısica, PUC-Rio, 22451-900 Rio de Janeiro, Brazil 3 Bloomberg LP, New York, NY 10022, USA 4 Departamento de F́ısica, Universidade Federal do Ceará, 60451-970 Fortaleza, Ceará, Brazil 5 Minerva Center and Physics Department, Bar-Ilan University, Ramat Gan 52900, Israel ∗ These authors contributed equally to this work Abstract We investigate how extreme opinion trends arise in society by analyzing polls that inquire about a wide range of issues such as religion, economics, politics, abortion, extramarital sex, books, movies, and electoral vote. The surveys lay out a sharp statistical predictor for the rise of extreme views. The precursor is a nonlinear relation between the fraction of individuals holding a certain extreme view and the fraction of individuals that includes also moderates, e.g., those who are “very conservative” in politics versus “moderate to very conservative”. The onset of nonlinearity forecasts the transition from moderate to extreme opinion before it actually occurs. It is an early-warning signal of an abrupt bootstrap-percolation transition where extreme opinions cascade through society due to people’s “stubbornness”, as suggested by modeling. By means of a phase diagram we classify societies from moderate to extreme in terms of critical fractions of extremists and people’s ties. The nonlinear analysis can be extended to study investors sentiment or competition of innovations of any kind, allowing to forecast sudden market events or adoption of new products.