Artificial intelligence (AI) systems offer effective support for online learning and teaching, including personalizing learning for students, automating instructors’ routine tasks, and powering adaptive assessments. However, while the opportunities for AI are promising, the impact of AI systems on the culture of, norms in, and expectations about interactions between students and instructors are still elusive. In online learning, learner–instructor interaction (inter alia, communication, support, and presence) has a profound impact on students’ satisfaction and learning outcomes. Thus, identifying how students and instructors perceive the impact of AI systems on their interaction is important to identify any gaps, challenges, or barriers preventing AI systems from achieving their intended potential and risking the safety of these interactions. To address this need for forward-looking decisions, we used Speed Dating with storyboards to analyze the authentic voices of 12 students and 11 instructors on diverse use cases of possible AI systems in online learning. Findings show that participants envision adopting AI systems in online learning can enable personalized learner–instructor interaction at scale but at the risk of violating social boundaries. Although AI systems have been positively recognized for improving the quantity and quality of communication, for providing just-in-time, personalized support for large-scale settings, and for improving the feeling of connection, there were concerns about responsibility, agency, and surveillance issues. These findings have implications for the design of AI systems to ensure explainability, human-in-the-loop, and careful data collection and presentation. Overall, contributions of this study include the design of AI system storyboards which are technically feasible and positively support learner–instructor interaction, capturing students’ and instructors’ concerns of AI systems through Speed Dating, and suggesting practical implications for maximizing the positive impact of AI systems while minimizing the negative ones.

Self-regulated learning (SRL) is crucial for helping students attain high academic performance and achieve their learning objectives in the online learning context. However, learners often face challenges in properly applying SRL in online learning environments. Recent developments in artificial intelligence (AI) applications have shown promise in supporting learners’ self-regulation in online learning by measuring and augmenting SRL, but research in this area is still in its early stages. The purpose of this study is to explore students’ perceptions of the use of AI applications to support SRL and to identify the pedagogical and psychological aspects that they perceive as necessary for effective utilization of those AI applications. To explore this, a speed dating method using storyboards was employed as an exploratory design method. The study involved the development of 10 AI application storyboards to identify the phases and areas of SRL, and semi-structured interviews were conducted with 16 university students from various majors. The results indicated that learners perceived AI applications as useful for supporting metacognitive, cognitive, and behavioral regulation across different SRL areas, but not for regulating motivation. Next, regarding the use of AI applications to support SRL, learners requested consideration of three pedagogical and psychological aspects: learner identity, learner activeness, and learner position. The findings of this study offer practical implications for the design of AI applications in online learning, with the aim of supporting students’ SRL.

In the midst of the COVID-19 pandemic, the use of non-face-to-face information and communication technology (ICT) such as kiosks has increased. While kiosks are useful overall, those who do not adapt well to these technologies experience technostress. The two most serious technostressors are inclusion and overload issues, which indicate a sense of inferiority due to a perceived inability to use ICT well and a sense of being overwhelmed by too much information, respectively. This study investigated the different effects of hybrid technostress—induced by both inclusion and overload issues—on the cognitive load among low-stress and high-stress people when using kiosks to complete daily life tasks. We developed a ‘virtual kiosk test’ to evaluate participants’ cognitive load with eye tracking features and performance features when ordering burgers, sides, and drinks using the kiosk. Twelve low-stress participants and 13 high-stress participants performed the virtual kiosk test. As a result, regarding eye tracking features, high-stress participants generated a larger number of blinks, a longer scanpath length, a more distracted heatmap, and a more complex gaze plot than low-stress participants. Regarding performance features, high-stress participants took significantly longer to order and made more errors than low-stress participants. A support-vector machine (SVM) using both eye tracking features (i.e., number of blinks, scanpath length) and a performance feature (i.e., time to completion) best differentiated between low-stress and high-stress participants (89% accuracy, 100% sensitivity, 83.3% specificity, 75% precision, 85.7% F1 score). Overall, under technostress, high-stress participants experienced cognitive overload and consequently decreased performance; whereas, low-stress participants felt moderate arousal and improved performance. These varying effects of technostress can be interpreted through the Yerkes-Dodson law. Based on our findings, we proposed an adaptive interface, multimodal interaction, and virtual reality training as three implications for technostress relief in non-face-to-face ICT.

Artificial intelligence (AI) offers promising opportunities for enhancing personalized learning experiences for students, leading to improved educational outcomes by reshaping traditional teaching ...

Background: With the global rise in Alzheimer disease (AD), early screening for mild cognitive impairment (MCI), which is a preclinical stage of AD, is of paramount importance. Although biomarkers such as cerebrospinal fluid amyloid level and magnetic resonance imaging have been studied, they have limitations, such as high cost and invasiveness. Digital markers to assess cognitive impairment by analyzing behavioral data collected from digital devices in daily life can be a new alternative. In this context, we developed a “virtual kiosk test” for early screening of MCI by analyzing behavioral data collected when using a kiosk in a virtual environment. Objective: We aimed to investigate key behavioral features collected from a virtual kiosk test that could distinguish patients with MCI from healthy controls with high statistical significance. Also, we focused on developing a machine learning model capable of early screening of MCI based on these behavioral features. Methods: A total of 51 participants comprising 20 healthy controls and 31 patients with MCI were recruited by 2 neurologists from a university hospital. The participants performed a virtual kiosk test—developed by our group—where we recorded various behavioral data such as hand and eye movements. Based on these time series data, we computed the following 4 behavioral features: hand movement speed, proportion of fixation duration, time to completion, and the number of errors. To compare these behavioral features between healthy controls and patients with MCI, independent-samples 2-tailed t tests were used. Additionally, we used these behavioral features to train and validate a machine learning model for early screening of patients with MCI from healthy controls. Results: In the virtual kiosk test, all 4 behavioral features showed statistically significant differences between patients with MCI and healthy controls. Compared with healthy controls, patients with MCI had slower hand movement speed (t49=3.45; P=.004), lower proportion of fixation duration (t49=2.69; P=.04), longer time to completion (t49=–3.44; P=.004), and a greater number of errors (t49=–3.77; P=.001). All 4 features were then used to train a support vector machine to distinguish between healthy controls and patients with MCI. Our machine learning model achieved 93.3% accuracy, 100% sensitivity, 83.3% specificity, 90% precision, and 94.7% F1-score. Conclusions: Our research preliminarily suggests that analyzing hand and eye movements in the virtual kiosk test holds potential as a digital marker for early screening of MCI. In contrast to conventional biomarkers, this digital marker in virtual reality is advantageous as it can collect ecologically valid data at an affordable cost and in a short period (5-15 minutes), making it a suitable means for early screening of MCI. We call for further studies to confirm the reliability and validity of this approach.

Background: Early detection of mild cognitive impairment (MCI), a transitional stage between normal aging and Alzheimer disease, is crucial for preventing the progression of dementia. Virtual reality (VR) biomarkers have proven to be effective in capturing behaviors associated with subtle deficits in instrumental activities of daily living, such as challenges in using a food-ordering kiosk, for early detection of MCI. On the other hand, magnetic resonance imaging (MRI) biomarkers have demonstrated their efficacy in quantifying observable structural brain changes that can aid in early MCI detection. Nevertheless, the relationship between VR-derived and MRI biomarkers remains an open question. In this context, we explored the integration of VR-derived and MRI biomarkers to enhance early MCI detection through a multimodal learning approach. Objective: We aimed to evaluate and compare the efficacy of VR-derived and MRI biomarkers in the classification of MCI while also examining the strengths and weaknesses of each approach. Furthermore, we focused on improving early MCI detection by leveraging multimodal learning to integrate VR-derived and MRI biomarkers. Methods: The study encompassed a total of 54 participants, comprising 22 (41%) healthy controls and 32 (59%) patients with MCI. Participants completed a virtual kiosk test to collect 4 VR-derived biomarkers (hand movement speed, scanpath length, time to completion, and the number of errors), and T1-weighted MRI scans were performed to collect 22 MRI biomarkers from both hemispheres. Analyses of covariance were used to compare these biomarkers between healthy controls and patients with MCI, with age considered as a covariate. Subsequently, the biomarkers that exhibited significant differences between the 2 groups were used to train and validate a multimodal learning model aimed at early screening for patients with MCI among healthy controls. Results: The support vector machine (SVM) using only VR-derived biomarkers achieved a sensitivity of 87.5% and specificity of 90%, whereas the MRI biomarkers showed a sensitivity of 90.9% and specificity of 71.4%. Moreover, a correlation analysis revealed a significant association between MRI-observed brain atrophy and impaired performance in instrumental activities of daily living in the VR environment. Notably, the integration of both VR-derived and MRI biomarkers into a multimodal SVM model yielded superior results compared to unimodal SVM models, achieving higher accuracy (94.4%), sensitivity (100%), specificity (90.9%), precision (87.5%), and F1-score (93.3%). Conclusions: The results indicate that VR-derived biomarkers, characterized by their high specificity, can be valuable as a robust, early screening tool for MCI in a broader older adult population. On the other hand, MRI biomarkers, known for their high sensitivity, excel at confirming the presence of MCI. Moreover, the multimodal learning approach introduced in our study provides valuable insights into the improvement of early MCI detection by integrating a diverse set of biomarkers.