Impact of user engagement on virtual reality behavioral response from the human–computer interaction perspective toward a service context: A hybrid statistics and machine learning approach

Abstract

Owing to recent uncertainty in the service sector, new attractive ways, such as virtual reality, have been adopted by online sites to provide products and services to individual users in the tourism context. Based on human–computer interaction user engagement and the domain segmentation innovativeness theory, we develop and test a theoretical framework to examine the impact of key user engagement drivers on in-person tour intentions as behavioral responses using the partial least squares structural equation model and a neural network with activation function analysis as the new two-stage approach. According to a survey sample of visitors' responses as the case, the results demonstrate that a) user engagement, including aesthetic appeal, focused attention, perceived usability, and reward experience, improves attitudes toward virtual reality; b) product-possessing innovativeness positively moderates the relationships between attitude toward virtual reality and in-person tour intentions; c) information-possessing innovativeness negatively moderates the relationships between the attitude toward virtual reality and in-person tour intentions; d) focused attention and reward experience show both linear high performance and nonlinear importance; and e) activation function analysis could further improve the accuracy of prediction.