Phd Position F/m Augmentingworkspace Awareness with Passive Bci: From Monitoring to Enhancing Team Coordination

Le descriptif de l’offre ci-dessous est en Anglais_

**Type de contrat**: CDD

**Niveau de diplôme exigé**: Bac + 5 ou équivalent

**Fonction**: Doctorant

**Niveau d'expérience souhaité**: Jeune diplômé

**Contexte et atouts du poste**:
The PhD will be supervised by two different research
groups. Firstly, the Potioc project-team, part of the Inria Center at the University of Bordeaux, focuses on the
design, development, and evaluation of novel approaches that leverage multimodal interaction to enhance user
experience. This team has specific expertise in the development of Brain-Computer Interfaces (BCIs) (systems
evaluation. The first two supervisors, Fabien Lotte and Sébastien Rimbert, are members of the Potioc projectteam,
which specializes in Brain-Computer Interfaces (BCIs). Both are recognized experts in the field of EEGbased
interaction and neuroadaptive systems, with complementary expertise in machine learning for BCI (Lotte)
and applied cognitive neuroscience and BCI evaluation in clinical and collaborative contexts (Rimbert). The
third supervisor, Arnaud Prouzeau, is affiliated with the ILDA team at Inria Paris-Saclay, which specializes in
collaborative digital environments and the design and development of visualization tools. He is an expert in collaborative
human-machine interaction, notably in Virtual Reality (VR) environments. A collaboration between
these two teams has already been initiated through a preliminary project, which resulted in a recent joint publication
and poster presentation in an international conference [17]. This PhD project will provide an opportunity
to build on that foundation and further strengthen the collaboration between the two groups. In particular, the
two teams have the ideal and complementary set of skills to study markers of collaboration (the speciality of
ILDA) in brain activity, using BCIs (the speciality of Potioc).

**Mission confiée**:
We hypothesize that specific patterns of EEG-based inter-brain synchrony are predictive of collaborative success
or breakdown, and can be used to support real-time adaptation of shared workspaces. In other words, we
assume that measurable neural correlates of team alignment—or misalignment—exist, and that they can inform
the design of neuroadaptive systems aimed at enhancing human collaboration.

The thesis will address the following research questions:
RQ1 : What EEG markers of collaboration can be found depending on the nature of the collaborative context,
such as loose coordination, tight coordination, or mutual misunderstanding?
RQ2 : How robust and stable are these neuromarkers across different tasks and individuals, and to what extent
are they influenced by known sources of variability in the EEG and BCI field?
RQ3 : How can we design neurofeedback based on these markers in order to enhance workspace awareness and
avoid misalignment in collaboration?
The overarching thesis objective is to determine whether EEG-based BCIs can provide actionable insights into
the cognitive states of collaborating individuals, and ultimately support more fluid and effective teamwork.

**Principales activités**:
This project first research question will address whether we can identify neurophysiological markers that are
specific to different types of collaborative coupling. In particular, we aim to identify neurophysiological EEG
markers of collaboration across various situations: in loose collaboration, where collaborators work independently,
in tight collaboration, where they engage in the same task and in situation of mutual misunderstanding,
where there is a breakdown in the collaboration. A second, follow-up, research question will be: are these neuromarkers of collaboration stable and robust across tasks and context. We will notably study the robustness
and stability of these markers throughout different tasks such as validated-collaboration task performed with
mice and keyboard (e.g., as in Prouzeau et al. [16], from the ILDA team) as well as during collaborative (active)
BCI tasks, where two BCI users collaborate in a computer task fully controled using a BCI - i.e., with EEG only
(e.g., as in Bonnet et al. [3], from the Potioc team). We will also study the robustness of these neurophysiological
known EEG and BCI performance variability [6]. Building on this, a third research question could be considered: how can we design neurofeedback based on these markers of collaborative coupling, in order to enhance
workspace awareness, reduce misalignment and improve mutual understanding during collaborative tasks? If
successful, this approach could benefit high-stakes collaborative environments where failures in coordination
can have critical consequences. Applications include air traffic control, remote medical procedures, or digital
collaborative educational platforms, where detecting and

**Compétences**:
Methodological contribution: We aim to develop a method to measure workspace awareness and its deeper
cognitive process, e.g., focus