As artificial intelligence is increasingly being integrated into daily life, understanding the factors that drive individuals to seek AI-related information becomes more important. This study employs the Planned Risk Information Seeking Model as a theoretical framework to explore AI information-seeking behaviours from the perspective of individual differences. Apart from the effects of individual differences, the findings were generally in line with the theoretical model. However, negative affect negatively predicted information insufficiency and was not significantly related to information seeking intention. Furthermore, examination of the effects of individual differences revealed that I-type epistemic curiosity positively predicted both information insufficiency and information seeking intention, yet D-type epistemic curiosity was not significantly related to information insufficiency and information seeking intention. Moreover, information innovativeness was found to be negatively related to information insufficiency but positively related to information seeking intention. Theoretical and practical implications are discussed.

In recent years, videoconferencing has become the primary mode of communication among employees. While it facilitates social interaction and enhances connectivity, studies have highlighted its negative implications such as videoconference fatigue and technostress. Most existing research has employed quantitative methods, revealing a positive relationship between videoconferencing use and stress. This study advances the discourse by offering qualitative insights into the stressors, resultant strains, and coping mechanisms adapted by employees at different hierarchical levels. Through 30 interviews and 7 diary studies in Singapore, the study found new insights on the strains experienced and coping measures adopted among different employees, highlighting a need for videoconferencing etiquette and norms, and showing paradoxical ways of coping by using more technology to counter technostress and more social interaction to counter strains. This study extends the literature on stress and strain from a qualitative perspective, and revisits the existing technostress framework in the current context.

This study examines how chatbots respond to emotional expressions in consumer inquiries within online Q&A communities, where questions often convey confusion, frustration, or anger. Grounded in Affect-as-Information (AAI) theory, it compares chatbot and human responses across different stages of the consumer journey. Data from Reddit included consumer-generated questions, human replies, and responses from two large language model (LLM) chatbots with varying NLP capabilities. The findings reveal that positive emotions in consumer questions reduce response cognitivity, while negative emotions in post-purchase contexts trigger reverse emotional contagion. Chatbots showed higher emotionality for pre-purchase and greater cognitivity for post-purchase questions, unlike human responders who remained consistent. These insights refine AAI theory by showing how positive emotions can disrupt information exchange and highlight key differences between human and AI responses, informing improved chatbot design.

While much AI adoption research emphasizes technological factors, few studies explore social determinants or consumer purchase intention. This study addresses this gap by examining how social influence affects Chinese consumers’ intention to purchase AI products. Conducted in the rapidly evolving Chinese market, an online survey of 538 respondents was analyzed using social influence theory and the value-based adoption model. Results show that social influence positively impacts perceived usefulness, enjoyment, and ease of use, while negatively affecting perceived cost. In turn, perceived usefulness and enjoyment significantly increase purchase intention, whereas perceived cost reduces it. Moreover, personal innovativeness moderates these relationships: it strengthens the effect of social influence on perceived ease of use and weakens its effect on perceived enjoyment, but shows no moderation effect on perceived usefulness or cost. These findings highlight the importance of integrating social and individual factors when investigating consumer behavior in emerging AI markets.

Drawing upon self-determination theory, this study investigates whether the effects of competition interact with individuals’ prior knowledge to influence the motivations for and antecedents to their pro-environmental behaviour. Using a virtual reality serious game about plastic waste, we conducted a 2 (Game environment: Competition vs. Non-competition) × 3 (Prior knowledge about plastic waste: Low vs. Medium vs. High) between-subjects experiment with 61 participants (Mage = 23.31, SDage = 2.77). Results indicated that competition had differential impacts depending on individuals’ prior knowledge. Competition had negative effects on motivation and antecedents to pro-environmental behaviour for players with low levels of prior knowledge and positive effects for players with medium levels of prior knowledge. As the first study to investigate prior knowledge as a moderator for the effects of competition in a virtual reality serious game, our research contributes to the literature by clarifying the conditions under which competition could promote pro-environmental behaviour and offers suggestions on customised use of competition for communication practitioners.

Exergames offer both enactive and vicarious learning through the graphical representations of the self and virtual coach. This study established and tested a model of exergame motivation with Social Cognitive Theory as the foundation. A 2 (User Avatar: Absent versus Present) × 2 (Virtual Coach: Absent versus Present) between-subjects experiment was conducted with 137 high school students. Results supported a model in which the user avatar led to identification, with the relationship mediated by self-presence. Playing with a virtual coach increased social presence. Both identification and social presence were significantly related to future exercise intention, with the relationships mediated by in-game competence. These findings suggest notable theoretical and practical implications of using self-presence with avatars and social presence with virtual agents to enhance exergame outcomes through enactive and vicarious learning.