skip to main content

Title: Modified Playback of Avatar Clip Sequences Based on Student Attention in Educational VR
We demonstrate a system that sequences teacher avatar clips considering student eye tracking. We are investigating subjective suitability of avatar responses to student misunderstandings or inattention. Three different avatar behaviors are demonstrated to allow a teacher pedagogical agent to behave more appropriately to student attention or distraction. An in-game mobile device provides an experiment control mechanism for 2 levels of distractions.
Authors:
; ;
Award ID(s):
1815976
Publication Date:
NSF-PAR ID:
10168891
Journal Name:
2020 IEEE Conference on Virtual Reality and 3D User Interfaces (VR)
Page Range or eLocation-ID:
850 to 851
Sponsoring Org:
National Science Foundation
More Like this
  1. We study student experiences of social VR for remote instruction, with students attending class from home. The study evaluates student experiences when: (1) viewing remote lectures with VR headsets, (2) viewing with desktop displays, (3) presenting with VR headsets, and (4) reflecting on several weeks of VR-based class attendance. Students rated factors such as presence, social presence, simulator sickness, communication methods, avatar and application features, and tradeoffs with other remote approaches. Headset-based viewing and presenting produced higher presence than desktop viewing, but had less-clear impact on overall experience and on most social presence measures. We observed higher attentional allocation scores for headset-based presenting than for both viewing methods. For headset VR, there were strong negative correlations between simulator sickness (primarily reported as general discomfort) and ratings of co-presence, overall experience, and some other factors. This suggests that comfortable users experienced substantial benefits of headset viewing and presenting, but others did not. Based on the type of virtual environment, student ratings, and comments, reported discomfort appears related to physical ergonomic factors or technical problems. Desktop VR appears to be a good alternative for uncomfortable students, and students report that they prefer a mix of headset and desktop viewing. We additionally providemore »insight from students and a teacher about possible improvements for VR class technology, and we summarize student opinions comparing viewing and presenting in VR to other remote class technologies.« less
  2. We investigated student perceptions of cold calling on their feelings of anxiousness and how graduate teaching assistants (GTAs) alleviated these feelings when students shared their ideas publicly in the context of tutorial and laboratory sessions. Physics and chemistry GTAs who led active-learning tutorials and labs practiced cold calling paired with error framing with avatar-students in a mixed-reality simulator at the beginning of the semester. Then, we observed the GTAs teaching real students in their actual classroom. We recruited eleven students from sections led by GTAs who were observed to use cold calling in their classroom to participate in semi-structured interviews. Several students reported that cold calling increased their feelings of anxiousness. However, students also reported that GTAs used strategies paired with cold calling that reduced their feelings of anxiousness, such as acknowledging student responses as valuable and remembering student names. We discuss implications for professional development on active learning strategies.
  3. Knowledge distillation is a popular technique for training a small student network to emulate a larger teacher model, such as an ensemble of networks. We show that while knowledge distillation can improve student generalization, it does not typically work as it is commonly understood: there often remains a surprisingly large discrepancy between the predictive distributions of the teacher and the student, even in cases when the student has the capacity to perfectly match the teacher. We identify difficulties in optimization as a key reason for why the student is unable to match the teacher. We also show how the details of the dataset used for distillation play a role in how closely the student matches the teacher --- and that more closely matching the teacher paradoxically does not always lead to better student generalization.
  4. Policy distillation, which transfers a teacher policy to a student policy has achieved great success in challenging tasks of deep reinforcement learning. This teacher-student framework requires a well-trained teacher model which is computationally expensive. Moreover, the performance of the student model could be limited by the teacher model if the teacher model is not optimal. In the light of collaborative learning, we study the feasibility of involving joint intellectual efforts from diverse perspectives of student models. In this work, we introduce dual policy distillation (DPD), a student-student framework in which two learners operate on the same environment to explore different perspectives of the environment and extract knowledge from each other to enhance their learning. The key challenge in developing this dual learning framework is to identify the beneficial knowledge from the peer learner for contemporary learning-based reinforcement learning algorithms, since it is unclear whether the knowledge distilled from an imperfect and noisy peer learner would be helpful. To address the challenge, we theoretically justify that distilling knowledge from a peer learner will lead to policy improvement and propose a disadvantageous distillation strategy based on the theoretical results. The conducted experiments on several continuous control tasks show that the proposed frameworkmore »achieves superior performance with a learning-based agent and function approximation without the use of expensive teacher models.

    « less
  5. Teacher responses to student mathematical thinking (SMT) matter because the way in which teachers respond affects student learning. Although studies have provided important insights into the nature of teacher responses, little is known about the extent to which these responses take into account the potential of the instance of SMT to support learning. This study investigated teachers’ responses to a common set of instances of SMT with varied potential to support students’ mathematical learning, as well as the productivity of such responses. To examine variations in responses in relation to the mathematical potential of the SMT to which they are responding, we coded teacher responses to instances of SMT in a scenario-based interview. We did so using a scheme that analyzes who interacts with the thinking (Actor), what they are given the opportunity to do in those interactions (Action), and how the teacher response relates to the actions and ideas in the contributed SMT (Recognition). The study found that teachers tended to direct responses to the student who had shared the thinking, use a small subset of actions, and explicitly incorporate students’ actions and ideas. To assess the productivity of teacher responses, we first theorized the alignment of different aspectsmore »of teacher responses with our vision of responsive teaching. We then used the data to analyze the extent to which specific aspects of teacher responses were more or less productive in particular circumstances. We discuss these circumstances and the implications of the findings for teachers, professional developers, and researchers.« less