Computers in Human Behavior最新文献

Despite the recent growth in the integration of artificial intelligence (AI) into second/foreign language (L2) education, its emotional side has been ignored, to date. In order to address this gap, the present qualitative study aimed to explore the typology of emotions that Chinese English as a foreign language (EFL) teachers had experienced in their AI-based L2 classes. Drawing on the technology acceptance model (TAM) and control value theory (CVT), a sample of 50 EFL teachers were interviewed individually. The results of thematic analysis showed that Chinese EFL teachers, in this study, had experienced a variety of positive and negative emotions due to AI technologies. The most frequently experienced positive emotions were ‘enjoyment’, ‘excitement’, ‘motivation’, and ‘satisfaction’. Conversely, the participants had most repeatedly experienced negative emotions of ‘anxiety’, ‘stress’, ‘worry’, and ‘frustration’ in their AI-based classes. The findings are discussed in light of prior research and suggestions and implications are presented to EFL teachers and educators.

Many, if not most, mobile applications tend to elicit personal information from users to offer personalized services. However, users may not be comfortable with such intrusiveness and therefore hesitate to download or adopt a new app even when its use could be beneficial to their health and well-being. To overcome this friction and help users make informed decisions, we propose message interactivity as a solution. Guided by privacy calculus, the theory of interactive media effects (TIME), and the elaboration likelihood model (ELM), we conducted a 4-condition, pre-registered online between-subjects experiment (N = 305) to assess the effect of message interactivity on attitudes and behavioral intentions pertaining to information disclosure in mobile health apps. Data indicate a significant positive effect via three serial mediators, including perceived contingency, elaboration, and perceived benefits. Theoretical and practical implications are discussed.

Social virtual reality (SVR) attempts to allow for connections akin to face-to-face communication (Ftf). Yet, it is unclear whether the technology successfully mimics Ftf or more closely resembles other mediated communication channels. This study empirically compares SVR and other communication channels in terms of perceived social affordances, privacy, and trust through a between-subjects online survey (n = 743). Findings indicate that SVR and Ftf are similar regarding some perceived affordances (e.g., personalization) but differ with respect to others (e.g., anonymity, presence). Additionally, SVR is perceived as significantly distinct from one or multiple mediated channels for almost every measured social affordance. While SVR is seen as offering relatively greater levels of affordances that benefit interpersonal interaction, privacy concerns and a lack of trust in other users were found to often characterize the current user experience. This study provides theoretical insights for affordance research and practical implications for SVR designers.

Video games offer a unique platform for players to engage interactively with morally challenging topics and dilemmas. Despite the growing popularity of games that offer such content, there is a paucity of research on the player experiences and the specific game mechanics that facilitate moral decision making. To address this gap, this research identifies key game mechanics that support moral decision making through a comprehensive review of related literature and qualitative survey responses from players (n $=$ 30). The effects of these mechanics on players’ decision making processes and their overall impact on player experience were further explored through semi-structured, video-elicitation interviews (n $=$ 11). This research develops a theoretical framework based on the findings from these two exploratory studies, culminating in a set of design guidelines to inform the future development of moral decision making games.

Studies have shown that integrating Artificial Intelligence into robot-assisted language learning ushers in an immersive situation to establish empathy for communication competence and enjoyment. To investigate the usefulness of Artificial Intelligence-driven robots in learning English for Medical Purposes oral skills, this quasi-experimental study was conducted through an embedded mixed methods design in the 2024 academic year. One thousand and forty male (n = 398) and female (n = 642) students of different academic disciplines from the Isfahan University of Medical Sciences were included in the study and grouped under three categories. They were randomly assigned to control (n = 520) and experimental (n = 520) groups. In the qualitative phase, students’ communication enjoyment was debriefed through semi-structured interviews. Simultaneously, in the quantitative phase, once the participants watched simulcast lectures and joined in conversation with Artificial Intelligence-driven (non)-robots, their academic and professional oral skills were assessed. The collected data were the participants’ responses to the interviews and formative assessment of their progress and performance through Objective Structured Video Examinations and Mini-Clinical Evaluation Exercises. The interview results suggested that the participants had empathy in joining conversations with robots. The statistical analysis indicated that in performing role-play to teach oral skills to the Artificial Intelligence-driven robot, the participants achieved significantly greater communication competence than those who did role-play for virtual agents. Students of Medicine were great academic and professional achievers as they were significantly successful in establishing empathy and communication. The findings could open up further prospects for using Artificial Intelligence-driven robots in discipline-specific language learning contexts.

The potential of Generative Artificial Intelligence (AI) in language education has been widely recognized. However, there has been limited attention given to the emotional experiences of language teachers using AI and its relationship with AI-enabled productivity. By investigating 1,683 pre-service language teachers’ experiences of using generative AI in their teaching practicum or learning, this study explored how teachers’ emotional responses to AI use in teaching and learning are related to their AI-enabled productivity through the mediation of appraisal and coping. We uncovered several key findings: (1) achievement, challenge, and loss emotions were directly and/or indirectly related to AI-enabled productivity, while deterrence emotions were not; (2) achievement and challenge emotions were positively correlated with challenge appraisal and negatively correlated with hindrance appraisal, whereas loss and deterrence emotions showed the opposite pattern of correlation; (3) challenge emotions were positively related to approach-oriented coping, while loss and deterrence emotions were positively associated with avoidance-oriented coping; (4) among the coping strategies, only positive reinterpretation was positively associated with AI-enabled productivity; and (5) challenge appraisal and positive reinterpretation were significant mediators in the relationships between emotions and AI-enabled productivity, either separately or sequentially. These findings provide valuable insights for future research and practice, aiming to support the application of generative AI in the context of language education.

The current study aimed to examine the bi-directional relationships between pre-sleep electronic media use and sleep, and to assess the relative predictive strength of different types of electronic media use in the evening on subsequent sleep. Eighty-eight young adults (20.57 ± 0.99, 27 males) completed an online intake survey and received notifications to complete daily diaries for seven consecutive days. Results of dynamic structural equation models (DSEM) revealed that the total screen time, sleep onset time, sleep latency, sleep duration, and sleep quality showed significant positive autoregressive parameters. In contrast, the results revealed no statistically significant cross-lagged associations between the total screen time in the evening and sleep outcomes. Inspection of the predictive strength of various types of screen activities at the within-person level revealed that increased active social media use is linked to delayed sleep onset, while internet browsing is associated with shorter sleep latency. At the between-person level, screen time of T.V. watching, video streaming, social media use, and shopping correlated with delayed sleep onset. Shopping was also connected to longer sleep latency. Additionally, more time spent on T.V., video streaming, non-work internet browsing, and shopping was linked to reduced total sleep duration. These findings indicated a screen activity-dependent association between pre-sleep screen time and subsequent sleep. More objective assessments of electronic media use (e.g., app recording screen time) and sleep (e.g., actigraphy) and longitudinal monitoring over a relatively long period (i.e., several weeks or months) are warred to further elucidate (cross-lagged) relationships between electronic media use and sleep.

Health conditions, encompassing both physical and mental aspects, hold an influence that extends beyond the individual. These conditions affect personal well-being, relationships, and financial stability. Innovative strategies in healthcare, such as digital phenotyping, are strategic to mitigate these impacts. By merging diverse data sources, digital phenotyping seeks a comprehensive understanding of health, well-being, and behavioral conditions. Machine learning can enhance the analysis of these data, improving the comprehension of health and well-being. Therefore, this paper presents a systematic literature review on machine learning and digital phenotyping, examining the research field by filtering 2,860 articles from eleven databases published up to November 2023. The analysis focused on 124 articles to answer six research questions addressing machine learning techniques, data, devices, ontologies, and research challenges. This work presents a taxonomy for mapping explored areas in digital phenotyping and another for organizing machine learning techniques used in digital phenotyping research. The review found increased publications in 2023, indicating a growing interest in the field. The main challenges arise from the studies’ small participant samples and imbalanced datasets, limiting the generalizability of the results to broader populations and the choice of ML methods. Furthermore, the reliance on self-reported data can introduce potential inaccuracies due to recall and reporting biases. Beyond self-reports, authors explored different data types, including physiological, clinical, contextual, smartphone-based, and multimedia. Despite using video recordings in controlled experiments, studies have yet to investigate this method within intelligent environments. Researchers also analyzed neurophysiological phenotypes, suggesting the potential for interventions based on these characteristics.