PhD proposition M/F: Towards Developmental AI in Robotics: Modeling the Learning of Joint Attention in Children

Job Description

Thesis Subject

Context :
Even before acquiring language, young children participate in complex communicative interactions based on the coordination of actions and the sharing of attention. Joint attention, which involves sharing an interest in an object with another person, plays a central role in these exchanges. In children with autism spectrum disorder (ASD), these abilities are impaired, particularly due to difficulties maintaining eye contact. Digital technologies, and robots in particular, have been widely explored to support these children, whether as learning tools, social mediators, or research instruments. Furthermore, approaches in artificial intelligence, especially through neurocomputational models, remain largely theoretical and poorly integrated into robotic platforms. Research in developmental robotics on joint attention does not clearly establish a link with the specific characteristics of autism.

Thus, this project aims to use robotics as an experimental tool to analyze child-robot interactions and as a lever to better understand the mechanisms underlying autism, by formulating testable hypotheses on the development of joint attention.

Objectives:

The objectives of this project are threefold. (1) First, it aims to clarify whether young children with typical development or those with autism spectrum disorder (ASD) perceive biological and non-biological agents in the same way; (2) Second, it seeks to determine what behaviors a robot must adopt to be recognized as a social interlocutor by the child, by systematically comparing children's responses to a robot exhibiting or not exhibiting socially relevant behaviors; (3) Finally, the project aims to deepen our understanding of the first manifestations of joint attention by proposing an original computational model, directly implemented and tested on a robot.

Research program:

Two studies will be conducted: the first will aim to clarify the mechanisms underlying joint attention in interactions among children aged 9 months to 6 years. The second study will focus on modeling the development of joint attention using a robotic agent.

Study 1: The importance of gaze in joint attention with a biological versus a non-biological agent.
The direction of gaze is a key socio-communicative cue (Senju et al., 2008), but it is still unclear whether its importance is specific to human interactions or extends to non-biological agents. Studies show that infants follow the gaze of both a human and a robot (Senju et al., 2008; Okumura et al., 2013). Furthermore, a robot that has interacted socially with an adult is perceived as a communicative agent, influencing children's attention more strongly (Meltzoff et al., 2010). We will explore whether young children follow a robot's gaze, whether older children are sensitive to its social behaviors, and whether children with autism spectrum disorder (ASD) react differently from typically developing children.

Study 2: Modeling the Joint Attention Mechanism.
We will seek to model the mechanisms of joint attention in a robot during interactions with a social partner, allowing it to learn directly from its social environment. Our modeling is based on multi-layered artificial neural networks, designed using bio-inspired principles, with the unique ability to learn in real time. Our goal is therefore to propose a neural model based on data from psychology and neurobiology, testable on robots, in order to explore how this skill can emerge and develop autonomously through interaction with humans.

This project is fundamentally multidisciplinary, integrating knowledge, concepts, and methodologies from psychology and developmental robotics. Let us recall that the central objective of the project is to propose models capable of explaining the development of joint attention, to understand its limits through a predictive approach, and then to use these predictions to design new experiments in psychology.

Requirements
For this position, you should meet the following requirements:
• Master degree in computer science, robotics or artificial intelligence
• Excellent programming skills
• knowledge of psychology and cognitive science
• Proficiency in spoken and written English
• Scientific curiosity and desire to work in an interdisciplinary project related to cognitive science
• Intellectual rigor and proactive attitude
An experience related to the topic of the PhD project will be a plus.

Hiring process
All applications must be submitted via this portal (i.e. Portail emploi CNRS) and should include:
• CV
• Motivation letter including a short description of your background, a statement of your
interests and motivation for this position, and why you think you would be a good fit (1 page)
• Master diploma
For full consideration, all the items listed above should be included or the absence thereof
justified.
Informal inquiries via email prior to full applications are welcome.
Shortlisted candidates will be invited for an interview.
Applications will be accepted at least until June 30 otherwise until the position is filled.

Your Work Environment

This position is a 36-month doctoral contract within the framework of the CNRS MITI research call. This doctorate is thus part of a broader scientific project including internal and external collaborations from which the candidate will also benefit. The doctoral project will be carried out between the ETIS laboratory (UMR8051) at CY Cergy Paris University, under the supervision of Dr. Sofiane Boucenna, and the CLLE laboratory at the University of Toulouse, under the supervision of Professor Bahia Guellaï. The successful candidate will be affiliated with the ETIS laboratory, a joint laboratory between CY Cergy Paris University, ENSEA, and the CNRS. ETIS is a leading research unit in computer science in France.

Compensation and benefits

Compensation

2300 € gross monthly

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