Exploring the Role of Conversational Agents in Immersive Cultural Heritage Serious Games: A User Study of Agent Interaction Styles
Year: 2026
Type: article
Abstract: Advances in artificial intelligence and digital interaction technologies have created new opportunities for serious games in cultural heritage communication and education. This study proposes a design framework for an immersive archeological serious game incorporating conversational agents to address limited interaction and uniform learning support in conventional cultural heritage games. The framework was implemented in the prototype Retrace: Ancient Zhongshan with four experimental conditions: no agent, Companion, Instructional, and Friendly Mentor. A between-subjects experiment with 48 participants was conducted using knowledge tests, the User Experience Questionnaire, the Cognitive Load Questionnaire, the Positive and Negative Affect Schedule, behavioral logs, and interviews. Results showed that all three agent conditions significantly outperformed the no-agent condition in overall user-experience scores and all six dimensions, while differences among agent styles were not consistently significant. Knowledge outcomes showed varying trends across conditions, with stronger mean differences observed for Companion and Friendly Mentor. Higher cognitive-load scores in these conditions may reflect additional cognitive processing rather than reduced cognitive load. Potential relationships between player types and agent styles were identified as exploratory findings. Overall, conversational agents enhanced the subjective experience of immersive cultural heritage learning and provided insights for interaction-oriented digital heritage design.
Source: Information
DOI: 10.3390/info17111117
Authors: Ting Tian, Shuo Tan, Xiaoyu Ren, Jingya Li
Institutions: Beijing Jiaotong University, Tianjin University
Language: English
Subfield: Developmental and Educational Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $1,930
OpenAlex ID: W7225620470
