Predicting cybersecurity behaviors in the metaverse through the lenses of TTAT and TPB: a hybrid SEM-ANN approach

Purpose: The Metaverse holds vast amounts of user data, making it essential to address threats to its confidentiality, integrity and availability. These threats are not purely technological, as user actions and perceptions, shaped by psychological factors, can influence cybersecurity challenges. Thus, a holistic approach incorporating technological and psychological dimensions is crucial for safeguarding data security and privacy. This research explores users’ cybersecurity behavior in the Metaverse by integrating the technology threat avoidance theory (TTAT) and the theory of planned behavior (TPB). 

Design/methodology/approach: The model was assessed using data collected from 746 Metaverse users. The empirical data were analyzed using a dual structural equation modeling-artificial neural network (SEM-ANN) approach. 

Findings: The main PLS-SEM findings indicated that cybersecurity behavior is significantly affected by attitude, perceived behavioral control, subjective norms, perceived threat and avoidance motivation. The ANN results showed that perceived threat with a normalized importance of 100% is the most significant factor influencing cybersecurity behavior. The ANN results also showed that perceived severity with a normalized importance of 98.79% significantly impacts perceived threat. 

Originality/value: The novelty of this research stems from developing a unified model grounded in TTAT and TPB to understand cybersecurity behaviors in the Metaverse. Unlike previous Metaverse studies that solely focused on measuring behavioral intentions or user behaviors, this study takes a step further by evaluating users’ cybersecurity behaviors. Alongside its theoretical insights, the study offers practical recommendations for software developers, decision-makers and service providers.