Published on

Nexus

  • Call for projects: DEFFINUM
  • Total project budget: €3.631 million
  • CESI budget: €490,000
  • Project launch date: March 1, 2023
  • Project duration: 3 years

The NEXUS project focuses on developing an interoperable, open-source “Metaverse” platform for learning. It aims to facilitate the production and distribution of educational resources based on immersive reality. This new modality, harnessing the power of virtual reality, allows users to take full advantage of embodied cognition, a state known to be conducive to learning. One of the objectives of CESI’s work within the NEXUS project concerns the presence of users in the virtual reality environment for teaching and learning. It focuses on the effect that the embodiment of an avatar can have on users of immersive virtual environments. In the VR environment, interaction plays a central role: the learner is active and their actions create the possibility of regulating learning. In addition, the community is increasingly interested in the use of virtual reality because of the potential arising from the convergence of collective intelligence (CI), virtual reality (VR), and education. Thus, another objective of the work is to better understand the complex interactions between CI and VR in order to integrate them into an interface that promotes better communication and more effective collective decision-making.

Achievements as of December 31, 2024:

Contribution 1: The impact of avatar embodiment in virtual learning environments
The effect of avatar embodiment in virtual reality is a research topic that emerged with the democratization of immersive technologies, which have enabled users to embody a body completely different from their own in just a few moments. It is in this context that researchers discovered the existence of what they called the Proteus effect [1]. This phenomenon describes the tendency of users to adopt behavior and attitudes that conform to the expectations evoked by their avatar’s appearance. Despite numerous demonstrations of the Proteus effect in the literature, two aspects of this phenomenon remain insufficiently studied.

The first aspect concerns the theoretical explanation of the Proteus effect. To date, very few studies have sought to explore the cognitive mechanisms that could be behind the effect, and the literature still cites the explanations given in the first article on the subject, namely the self-perception theory [2]. In this context, the first contribution took the form of a literature review that was published by CESI. Initially, the aim of this review was to analyze current hypotheses on the underlying mechanisms of the Proteus effect (self-perception theory, deindividuation, and priming). Secondly, we proposed three new hypotheses to better understand the phenomenon (cognitive dissonance, embodiment, and perspective taking). Finally, after analyzing the various hypotheses, a model was proposed, incorporating the most credible hypotheses to explain the Proteus effect.

The second aspect of the Proteus effect that remains largely unexplored in the literature concerns its potential applications in real-world contexts. One of the few studies on the subject demonstrated that embodying different types of avatars in a virtual environment (VE) made it possible to modulate the creativity of a company’s employees during an idea generation task [3]. These results clearly highlight the potential benefits of applications of the Proteus effect. The second goal of this project is therefore to find innovative applications for the Proteus effect in an educational or professional setting. To this end, an initial experimental protocol was set up with a group of CESI students. This protocol aimed to reproduce the Proteus effect in an educational context. The students had to answer a series of statistics questions in the EV. This study showed that students who embodied an Einstein avatar scored higher on the test than those who embodied a generic avatar. Thus, this protocol successfully demonstrated that the Proteus effect can be applied in real educational contexts.

Based on these theoretical considerations and the potential applications of the effect, we conducted a series of experiments aimed at (1) understanding the respective impacts of the processes of deindividuation, embodiment, and identification on the emergence of the Proteus effect, (2) to highlight the implicit nature of the effect and its disappearance when made explicit, and (3) to delimit its transferability beyond expectations directly associated with the avatar’s appearance. The results and conclusions of all these studies are currently being formalized.

Contribution 2: Towards improving collective intelligence in a virtual environment
Collective intelligence (CI) is a measure that reflects a group’s ability to perform a wide range of tasks [4]. It is an essential indicator for predicting the effectiveness of a group’s functioning. While previous studies have examined the robustness of CI in various contexts, including face-to-face interactions and online settings [5,6], its transferability to virtual reality (VR) remains unexplored. This work aims to achieve a threefold objective: (1) to understand how collective intelligence influences group behavior (i.e., verbal and nonverbal signals) in an immersive virtual environment, (2) to use these signals to develop an interface that improves group dynamics in virtual reality, and (3) to apply these results to an immersive educational platform that enhances collaboration and group learning. This work will have implications for theoretical knowledge about collective intelligence, as well as for the possibilities offered by an immersive environment in an educational context.


The research focus for this work is defined by three main questions:

  1. Is a group’s potential consistent across the different media used for group tasks, including immersive VR environments?
  2. Are the nonverbal cues of collective intelligence identified in previous research observable in virtual reality (VR) environments? If not, how are they structured in virtual reality environments?
  3. Can the use of these signals in a virtual reality context lead to improvements in group performance and learning outcomes? If so, how can these signals be exploited? The research questions guide the planned studies. The first step is to translate the Collective Intelligence Test into French for broader applicability. The second step focuses on measuring behavioral signals in VR, testing the relevance of existing measurements and identifying new signals. The final step exploits the observed signals to develop a VR interface aimed at improving group performance and learning outcomes. All experimental work on these research questions is conducted in a collaborative BIM application in an immersive virtual environment.

Upcoming work:

  • Define the educational scenario that can be transformed into virtual reality-based learning. We will develop this scenario using UMI3D [7] / Intraverse, which allows the creation of 3D media in which users of any AR/VR device can collaborate in real time. The development will benefit from the results of ongoing research on the Proteus effect on user embodiment and the notion of collective intelligence in a virtual reality environment.
  • Upgrading the VR lab with equipment and infrastructure suitable for simultaneously distributing VR applications to multiple users.
  • References :
    [1] N. Yee and J. Bailenson, “The Proteus Effect: The Effect of Transformed Self-Representation on Behavior,” 2007, doi: 10.1111/j.1468-2958.2007.00299.x.
    [2] D. J. Bem, “Self-Perception Theory,” Adv. Exp. Soc. Psychol., vol. 6, no. C, pp. 1–62, Jan. 1972, doi: 10.1016/S0065-2601(08)60024-6.
    [3] S. Buisine, J. Guegan, J. Barré, F. Segonds, and A. Aoussat, “Using avatars to tailor ideation process to innovation strategy,” Cogn. Technol. Work, vol. 18, no. 3, pp. 583–594, 2016.
    [4] Woolley, A. W., Chabris, C. F., Pentland, A., Hashmi, N., & Malone, T. W. (2010). Evidence for a Collective Intelligence Factor in the Performance of Human Groups. Science, 330(6004), 686‑688.
    [5] Engel, D., Woolley, A. W., Jing, L. X., Chabris, C. F., & Malone, T. W. (2014). Reading the Mind in the Eyes or Reading between the Lines? Theory of Mind Predicts Collective Intelligence Equally Well Online and Face-To-Face. PLoS ONE, 9(12), e115212. https://doi.org/10.1371/journal.pone.0115212
    [6] Kim, Y. J. (2016). How Collective Intelligence Impacts Team Performance : Exploring Transactive Memory System as a Mechanism.
    [7] https://github.com/UMI3D/UMI3D-SDK