Qualitative Cusp Catastrophe Multi-Agent Simulation Model to Explore Abrupt Changes in Online Impulsive Buying Behavior

Abstract

We develop a qualitative catastrophe (a nonlinear sudden violent change) multi-agent simulation model to investigate the evolution of group behavior, specifically abrupt changes in online impulsive buying (OIB) behavior. Studies have rarely investigated the mechanism of abrupt changes in OIB at the group level. To address the research gaps and advance this area of research, we employed a sequential multiple-methods approach. First, we designed a questionnaire to obtain and analyze consumer data to identify OIB drivers. Second, we built a qualitative catastrophe model based on empirical findings to describe sudden changes in the OIB behavior of individuals by merging catastrophe theory (CT) and qualitative simulation (QSIM). Finally, grounded in the qualitative catastrophe model, we constructed an agent-based model (ABM) to simulate group-level OIB behavior. The empirical findings revealed the following. (1) Sudden changes in group-level OIB occur as consumers’ sense of quantified self increases when self-control is low. (2) The greater the number of consumers with a proving preference (who prefer to prove their competence and performance to others) in the group, the higher the possibility of catastrophe; the scale of catastrophe increases significantly with the enhancement of product information features. (3) We identify optimal gamification for sudden increases in group-level OIB; the larger the degree of social networking is, the higher the likelihood of catastrophe behavior. Our proposed combination of a survey study, QSIM, and an ABM is a plausible solution for behavioral research on consumers, and the integration paradigm could help maintain market stability and promote products.