Effect of Overconfidence on Product Diffusion in Online Social Networks: A Multiagent Simulation Based on Evolutionary Game and Overconfidence Theory

• Published in: Complexity (New York, N.Y.) Vol. 2022; no. 1
• Main Authors: Wei, Xiaochao, Liao, Qi, Zhang, Yanfei, Nie, Guihua
• Format: Journal Article
• Language: English
• Published: Hoboken Hindawi 01.01.2022; Hindawi Limited; Hindawi-Wiley
• Subjects: Behavior; Consumers; Cooperation; Decision making; Diffusion; Game theory; Multiagent systems; Product development; Simulation; Social networks; System theory
• ISSN: 1076-2787; 1099-0526
• DOI: 10.1155/2022/1516419

The rapid development of online social media has significantly promoted product diffusion in online social networks (PDOSN). However, prior studies focusing on irrational behavior, such as overconfidence, in PDOSN are scarce. To investigate the effect of overconfidence on PDOSN, this study combined overconfidence and an evolutionary game to conduct a multiagent simulation on PDOSN. This combined method provided an effective reference to examine product diffusion in the context of irrational behavior. After careful consideration, this study identified three overconfidence scenarios, benefit, cost, and benefit and cost overconfidence, developed a multiagent simulation model for PDOSN using various overconfidence scenarios, and conducted a comparison with real-world cases to validate the model’s feasibility. The findings indicated that adoption benefits and betrayal penalties had a positive effect on the results in all models, while adoption costs had the opposite effect. When benefit and cost overconfidence occurred simultaneously, benefit overconfidence offset the negative effect of cost overconfidence. Moderate connectivity, a large number of core nodes, and high reconnection probability fully promoted product diffusion. Benefit overconfidence and cost overconfidence had a significant impact on the results in different networks. As such, this study combined psychological theory with simulation methods, providing insights for future research on product diffusion.