A dynamic message-passing approach for social contagion in time-varying multiplex networks

¹ College of Computer Science and Technology, Chongqing University of Posts and Telecommunications Chongqing 400065, China
² Chongqing MII Key Lab. of Computer Networks & Communications - Chongqing 400065, China
³ Potsdam Institute for Climate Impact Research - Potsdam D-14473, Germany

Received: 13 July 2018
Accepted: 17 September 2018

Abstract

The information about social behavior diffusion on a communication network and the social behavior contagion through a physical-contact network are two related dynamical processes. We investigate here an asymmetrically interacting, double-layer network to elucidate the interplay between information diffusion on a time-varying network and the social contagion on the counterpart layer. We uncover that the time-varying character affects the final adoption size. We also find that the behavior spreading in the physical-contact network introduces the adoption size from first to second order. A maximizing area of the difference between the final infected size and the adoption size appears in a suitable range of the behavioral information transmission rate and the initial seed size. We develop a dynamic message-passing theory to understand the interplay between both types of spreading dynamics.