Postdoctoral Fellowship (M/F) – Integrating AI Recommendations into Safety-Critical Environments

Job Description

Missions

The objective of the postdoctoral fellowship is to design an interactive system to alert pilots and physicians to potential fixation issues and help them adjust their behavior appropriately, using the capabilities of the extended reality environment. By collaborating with our medical experts and pilots on one hand, and AI researchers on the other, you will identify the most effective methods for presenting diagnostic information to participants and helping them understand the relationship between the situation, their actions, and potential fixation errors. You will use structured comparative observation (Mackay and McGrenere, 2024) as well as other methods to collect qualitative and quantitative data to maximize user engagement with the system, so that the combined performance of the user and the system exceeds that of each actor acting separately.

Activity

In safety-critical situations, operators must make preliminary assessments based on partial and uncertain information in order to make decisions. This context is conducive to “fixation errors”—that is, failing to update one's assessment of a situation or persisting with a course of action that is no longer appropriate. These errors are a major cause of adverse events and accidents in the healthcare and aviation sectors. The IDEFIX project, funded by the ANR, aims to study the role of Artificial Intelligence and Virtual & Augmented Reality in helping operators in critical situations—such as airline pilots or healthcare professionals—avoid fixation errors.
We are developing a set of algorithms to analyze the activities of pilots and healthcare providers in virtual and augmented reality environments, using simulated and controlled critical situations. These algorithms are capable of identifying potential fixation behaviors based on environmental information and the operator's actions. The HCI challenge is therefore as follows: how to inform the operator of the AI's recommendations in an environment already saturated with information, without disrupting their work and while taking into account the fact that the AI assistant, just like the operator, may make errors in its analysis of the situation?

The work will focus in particular on studying:

* how to incorporate the operator's activity into the recommendation mechanism;
* how to present information to the operator while taking into account their activity, cognitive load, and the situation;
* how to enable the operator to interact with the AI in a way that maximizes decision quality and minimizes fixation times.

Your Profil

Skills

This position is intended for a candidate with a Ph.D. in human-computer interaction, computer science, or cognitive science, with a strong focus on activity modeling.

An interest in issues at the intersection of modeling, understanding human decision-making processes, and human-AI interaction will be particularly valued.
The ability to combine a strong theoretical contribution with validation within an interdisciplinary framework applied to critical situations will be a significant asset.

Your Work Environment

The Interdisciplinary Laboratory for Digital Sciences (LISN), located on the Paris-Saclay University campus, is a multidisciplinary research laboratory that brings together researchers and faculty members from various disciplines in engineering and information sciences, as well as life sciences and the humanities and social sciences.
It comprises approximately 350 people, including 160 researchers and faculty members, 54 engineers and technicians, and 145 doctoral students organized into five departments: Algorithms, Learning, and Computation; Data Science; Fluid Mechanics and Energy; Human-Computer Interaction; and Language Science and Technology.
The IDEFIX project is led by researchers from the Ex-Situ (Wendy Mackay) and CPU (Nicolas Sabouret, Céline Clavel, and Ève Fabre) teams in the Human-Computer Interaction Department. You will join the Ex-Situ team, whose work focuses primarily on human-computer interaction. This team explores the boundaries of interaction by studying “extreme” users in order to anticipate the needs of future applications, as well as to better understand the phenomenon of interaction itself and to create new forms of interaction. You will work in collaboration with researchers in computer science, cognitive psychology, and ergonomics.

Constraints and risks

none

Compensation and benefits

Compensation

Approximately 3000€ gross

Annual leave and RTT

44 jours

Remote Working practice and compensation

Pratique et indemnisation du TT

Transport

Prise en charge à 75% du coût et forfait mobilité durable jusqu’à 300€

About the offer

About the CNRS

The CNRS is a major player in fundamental research on a global scale. The CNRS is the only French organization active in all scientific fields. Its unique position as a multi-specialist allows it to bring together different disciplines to address the most important challenges of the contemporary world, in connection with the actors of change.

CNRS

The research professions