Search for: All records

Debugging process plays a crucial role in helping students pinpoint their specific learning weaknesses, allowing them to modify their strategies for enhanced academic performance. Notably, changes in pupil dilation serve as an indicator of arousal associated with confronting learning challenges. This physiological response acts as a “physiological footprint” that reflects cognitive engagement, facilitating internally focused cognitive processes such as insight generation and mind-wandering. In this study, we proposed that pupil dilation could be a valuable predictor of students’ metacognitive awareness throughout the debugging process, specifically within an augmented reality (AR) learning environment. The findings revealed significant differences in pupil dilation among students categorized by their varying levels of debugging, which represents a specific dimension of the Metacognitive Awareness Inventory. 

This study investigates the method for measuring cognitive workload in augmented reality-based biomechanics lectures by analyzing pupil dilation. Using Dikablis Glasses 3 and Microsoft HoloLens, we recorded physiological and subjective data across learning and problem-solving phases. Pupil dilation was normalized and segmented, enabling a comparison of cognitive demands between phases. The results indicated significant correlations between pupil dilation and NASA TLX cognitive demand, particularly in lectures that primarily involved procedural knowledge. These findings suggest that instructional design and content complexity have a significant impact on cognitive load, providing valuable insights for optimizing AR-based learning environments to support cognitive efficiency and student engagement. 

This study explores the application of slouching scores to assess ergonomic posture in augmented reality (AR) environments. Employing Microsoft HoloLens 2 with Xsens motion capture technology, participants engaged in interactive biomechanics tasks, including a practical luggage-lifting exercise. Real-time feedback guided users towards safe posture, emphasizing spinal alignment and reducing physical strain. Slouching scores functioned as quantitative measures of posture quality, establishing a connection between unsafe postures and the requisite postural adjustments. The results illustrate how AR-integrated systems can enhance posture awareness, improve user ergonomics, and promote active learning in both educational and professional settings. 

In the context of student learning, investigating effective feedback mechanisms within augmented reality (AR) learning systems is crucial for better understanding and optimizing study behaviors. This study examines the influence of metacognitive monitoring feedback within an AR setting. Our hypothesis suggests that regularly providing students with feedback on their metacognitive monitoring within an AR learning environment has a beneficial effect on their metacognitive state. The results of the study confirm that frequent exposure to such feedback significantly improves scores on the Metacognitive Awareness Inventory. Essentially, there was a marked increase in the inventory scores of participants who received ongoing feedback, compared to those who only received metacognitive monitoring feedback once after the lecture, particularly in the areas of planning, monitoring comprehension, and debugging strategies. This enhancement is achieved by influencing student calibration by directly impacting their metacognitive state. 

With the growing need for augmented reality (AR) technology, understanding and optimizing study behaviors in AR learning environments has become crucial. However, one major drawback of AR learning is the absence of effective feedback mechanisms for students. To overcome this challenge, we introduced metacognitive monitoring feedback. Additionally, we created a location-based AR learning environment utilizing a real-time indoor tracking system to further enhance student learning. This study focuses on the positive impact of metacognitive monitoring feedback in a location-based AR learning environment. Our hypothesis posits that regularly providing students with feedback on their metacognitive monitoring within this new AR learning system positively influences their metacognitive awareness. The study's findings confirm that frequent exposure to such feedback significantly enhances the Metacognitive Awareness Inventory (MAI) scores. Participants who received continuous feedback demonstrated a significant increase in MAI scores compared to those who received feedback only once after the lecture. This improvement is achieved by influencing student calibration and directly enhancing their metacognitive awareness. 

There is an increasing demand for developing new metrics that can effectively measure the physical demand experienced by users in augmented reality (AR) environments. In this study, we evaluated one of the recent metrics, called “slouching score,” in an AR-based biomechanics lecture. This study aims to uncover the correlation between the AR interaction and the physical demand of users in a different setup compared to the earlier study. The slouching score, which evaluates posture changes that may indicate fatigue during AR interactions, is measured using Xsens motion capture equipment. These calculated scores are compared with responses to physical demand assessments surveyed using NASA-TLX questionnaires. One of the key differences between the current study and earlier ones is that participants had to physically move to access the next AR module in earlier studies. In contrast, this time, participants simply needed to click a virtual arrow button to view the next AR modules, eliminating the need for physical movement. Our preliminary findings show correlations between the slouching score from some modules and the NASA-TLX physical demand ratings. 

Abstract Global climate models (GCMs) and Earth system models (ESMs) exhibit biases, with resolutions too coarse to capture local variability for fine-scale, reliable drought and climate impact assessment. However, conventional bias correction approaches may cause implausible climate change signals due to unrealistic representations of spatial and intervariable dependences. While purely data-driven deep learning has achieved significant progress in improving climate and earth system simulations and predictions, they cannot reliably learn the circumstances (e.g., extremes) that are largely unseen in historical climate but likely becoming more frequent in the future climate (i.e., climate non-stationarity). This study shows an integrated trend-preserving deep learning approach that can address the spatial and intervariable dependences and climate non-stationarity issues for downscaling and bias correcting GCMs/ESMs. Here we combine the super-resolution deep residual network (SRDRN) with the trend-preserving quantile delta mapping (QDM) to downscale and bias correct six primary climate variables at once (including daily precipitation, maximum temperature, minimum temperature, relative humidity, solar radiation, and wind speed) from five state-of-the-art GCMs/ESMs in the Coupled Model Intercomparison Project Phase 6 (CMIP6). We found that the SRDRN-QDM approach greatly reduced GCMs/ESMs biases in spatial and intervariable dependences while significantly better-reducing biases in extremes compared to deep learning. The estimated drought based on the six bias-corrected and downscaled variables captured the observed drought intensity and frequency, which outperformed state-of-the-art multivariate bias correction approaches, demonstrating its capability for correcting GCMs/ESMs biases in spatial and multivariable dependences and extremes. 

Abstract A catastrophic Mw7.8 earthquake hit southeast Turkey and northwest Syria on February 6th, 2023, leading to more than 44 k deaths and 160 k building collapses. The interpretation of earthquake-triggered building damage is usually subjective, labor intensive, and limited by accessibility to the sites and the availability of instant, high-resolution images. Here we propose a multi-class damage detection (MCDD) model enlightened by artificial intelligence to synergize four variables, i.e., amplitude dispersion index (ADI) and damage proxy (DP) map derived from Synthetic Aperture Radar (SAR) images, the change of the normalized difference built-up index (NDBI) derived from optical remote sensing images, as well as peak ground acceleration (PGA). This approach allows us to characterize damage on a large, tectonic scale and a small, individual-building scale. The integration of multiple variables in classifying damage levels into no damage, slight damage, and serious damage (including partial or complete collapses) excels the traditional practice of solely use of DP by 11.25% in performance. Our proposed approach can quantitatively and automatically sort out different building damage levels from publicly available satellite observations, which helps prioritize the rescue mission in response to emergent disasters.