17 April 2025 in Journal of Computing in Higher Education
Considering both the transformative opportunities and challenges presented by generative AI (GenAI) in academic writing, effectively integrating GenAI into the academic setting becomes a significant need requiring prioritization. Yet, there is limited understanding regarding the nature of interactions between different types of students, what behavioral patterns students exhibit during a student-GenAI interaction (SAI) on a given task, and how these different SAI patterns relate to the actual writing task performance. This study, therefore, aimed to identify SAI patterns of academic writing tasks depending on students’ level of AI literacy and examine the differences in academic writing performance between the identified SAI patterns. Drawing from the combination of three data sources, including think-aloud protocols, screen-recordings, and chat histories between 36 Chinese graduate students and a GenAI writing system, epistemic network analysis (ENA) was used to reveal the distinctive SAI patterns of students with different levels of AI literacy. The study found that students with a high level of AI literacy exhibited a collaborative approach to SAI, actively accepting GenAI’s suggestions and engaging GenAI in meta-cognitive-related activities such as planning, whereas students with a low level of AI literacy demonstrated much less interaction with GenAI in completing their writing tasks, instead choosing to ideate and evaluate independently. In addition, the Wilcoxon rank-sum (Mann-Whitney U) test was conducted to assess the writing task performance of the two AI literacy groups. Findings revealed statistical differences in all evaluation rubrics (content, structure/organization, expression). This study offers implications for the design and implementation of GenAI agents in writing tasks and the pedagogy of GenAI-assisted instruction.
27 May 2022 in IEEE Transactions on Games
Virtual Reality (VR) technology has progressed rapidly and is used in various domains, particularly games. Simulator Sickness (SS) still represents a significant problem for its wider adoption. The most common way to detect SS is using the Simulator Sickness Questionnaire (SSQ). SSQ is a subjective measurement and is inadequate for real-time applications such as VR games. This research aims to develop a model to predict SS in real time using in-game characters’ movement and users’ eye motion data during gameplay in VR games. To achieve this, we designed an experiment to collect such data with three types of games. We trained a Long Short-Term Memory neural network with the eye-tracking and character movement data to predict SS. Our model can predict SS in real time with an accuracy of 83.4% for players who suffer from severe sensitivity to SS. Our results indicate that, in VR games, our model is an accurate and efficient method to predict SS in real time.
4 March 2024 in IEEE Transactions on Visualization and Computer Graphics
Users' perceived image quality of virtual reality head-mounted displays (VR HMDs) is determined by multiple factors, including the HMD's structure, optical system, display and render resolution, and users' visual acuity (VA). Existing metrics such as pixels per degree (PPD) have limitations that prevent accurate comparison of different VR HMDs. One of the main limitations is that not all VR HMD manufacturers released the official PPD or details of their HMDs' optical systems. Without these details, developers and users cannot know the precise PPD or calculate it for a given HMD. The other issue is that the visual clarity varies with the VR environment. Our work has identified a gap in having a feasible metric that can measure the visual clarity of VR HMDs. To address this gap, we present an end-to-end and user-centric visual clarity metric, omnidirectional virtual visual acuity (OVVA), for VR HMDs. OVVA extends the physical visual acuity chart into a virtual format to measure the virtual visual acuity of an HMD's central focal area and its degradation in its noncentral area. OVVA provides a new perspective to measure visual clarity and can serve as an intuitive and accurate reference for VR applications sensitive to visual accuracy. Our results show that OVVA is a simple yet effective metric for comparing VR HMDs and environments.