Detecting Leadership Opportunities in Group Discussions Using Off-the-Shelf VR Headsets

Chenghao Gu,Jiadong Chen,Jiayi Zhang,Tianyuan Yang,Zhankun Liu,Shin'ichi Konomi

DOI: https://doi.org/10.3390/s24082534

IF: 3.9

2024-04-16

Sensors

Abstract:The absence of some forms of non-verbal communication in virtual reality (VR) can make VR-based group discussions difficult even when a leader is assigned to each group to facilitate discussions. In this paper, we discuss if the sensor data from off-the-shelf VR devices can be used to detect opportunities for facilitating engaging discussions and support leaders in VR-based group discussions. To this end, we focus on the detection of suppressed speaking intention in VR-based group discussions by using personalized and general models. Our extensive analysis of experimental data reveals some factors that should be considered to enable effective feedback to leaders. In particular, our results show the benefits of combining the sensor data from leaders and low-engagement participants, and the usefulness of specific HMD sensor features.

engineering, electrical & electronic,chemistry, analytical,instruments & instrumentation

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to use the sensor data of off - the - shelf VR devices (such as head - mounted displays) to detect leadership opportunities in group discussions in virtual reality (VR) environments, especially to identify those suppressed speaking intentions, thereby supporting leaders to organize and promote group discussions more effectively. Due to the lack of certain forms of non - verbal communication in VR environments, it makes VR group discussions difficult even when leaders are assigned. Therefore, this paper proposes a new detection method. By analyzing the sensor data collected by VR devices, personalized and general deep - learning models are constructed to detect the suppressed speaking intentions of group members, and then provide leaders with information on when to intervene in the discussion to promote more effective communication.

Specifically, the paper focuses on the following research questions:
- RQ1: Is it possible to detect opportunities to promote VR group discussions by constructing personalized and general models using sensor data from off - the - shelf VR devices?
- RQ2: Do the sensor characteristics of leaders help to detect the suppressed speaking intentions of members?
- RQ3: Which type of sensor characteristics of leaders contributes the most to detection?

The main contributions of the paper include:
1. Proposed general and personalized models for detecting leadership opportunities, which only require data collected from off - the - shelf VR devices and no additional sensing devices.
2. Demonstrated that through integrating the sensor data of leaders and members, the personalized model can improve the detection performance and revealed the usefulness of leader sensor data in detecting the speaking intentions of low - participation members.
3. Studied the contributions of different types of leader sensor characteristics (including head - mounted displays, controller positions and rotations, etc.) to performance improvement. The results show that for most low - participation members, the head - mounted display sensor characteristics contribute the most to the performance improvement of the personalized model.
4. Through video analysis of the behavior patterns of VR group discussions, explored the reasons why leader sensor data can improve the detection performance of low - participation members and discussed the implications of these findings for designing VR group discussion support mechanisms.