Towards a therapeutic interaction game: developing an adaptable artificial agent

Host Institution: Aix-Marseille University

Department that hosts the PhD: Institute for Movement Sciences

Contact: Prof. dr. Bootsma (use this email contact only to request additional information, do NOT use this email for applications)


Project description

The aim of this project is to develop an adaptable artificial agent, capable of taking on the role of one of the players in the doubles-pong task (see below), that can ultimately be used in therapy for persons with autistic spectrum disorder (ASD). For this purpose, in this project the manual interception behaviour of individuals with and without ASD will be studied and modelled at the level of the individual agent-environment system. Adapting the model-generated interception behaviour by variation of model parameters will be used to mimic the behavioural characteristics observed in standard and ASD populations, allowing evaluation of how human-like, and perhaps how ASD-like, the behaviour is being perceived. Because previous research suggested that the behaviour of the agent-agent-environment system emerges from the dynamics of the two coupled agent-environment systems, these aspects of the artificial agent might prove beneficial in developing the doubles-pong task into a therapeutic tool for ASD. Working towards a doubles-pong tool for therapeutic use, this project will focus on interception by a single participant, without but also with a partner present, the partner being diagnosed with ASD or not. Combined with a detailed model of the operative informational coupling, established in the partner (ESR 5) project, the agent-environment model of individual interception arrived at in this project will be used to build a full doubles-pong agent-agent-environment model, including an artificial agent.


Doubles pong (task employed in Projects 5,6, 7 and 8)

The doubles pong setup (from Benerink et al 2016)

In the doubles-pong task, two individuals have to coordinate their actions in such a way that one of them intercepts an approaching virtual ball. Each player controls a separate paddle that moves along the bottom of a shared computer screen. The ball that falls under an angle from the top of the screen needs to be intercepted by one of the paddles, without the two paddles colliding. Previous research has suggested that the division of interception space by the two players emerges from their interactions rather than being fixed a priori. In the cases that a ball is directed in between the two individuals’ starting positions, both start moving and one player subsequently abandons this movement, presumably when (s)he sees that the other is on an interception course. ESRs 4-8 will use the doubles-pong task.


People involved

Prof. dr. Bootsma

Dr. Casanova

Dr. Zaal

Dr. Farella


Key publications

Benerink, N. H., Zaal, F. T. J. M., Casanova, R., Bonnardel, N. & Bootsma, R. J. (2016). Playing ‘pong’ together: Emergent coordination in a doubles interception task. Frontiers in Psychology7, 1910.

Bootsma, R. J., Ledouit, S., Casanova, R. & Zaal, F. T. J. M. (2016). Fractional-order information in the visual control of lateral locomotor interception. Journal of Experimental Psychology: Human Perception & Performance42(4), 517–529.

Ledouit, S., Casanova, R., Zaal, F.T.J.M., & Bootsma, R.J. (2013). Prospective control in catching: The persistent angle-of-approach effect in lateral interception. PLoS ONE 8(11): e80827. https://10.1371/journal.pone.0080827


Specific required skills of PhD student

Skill area
Language (writing) English
Language (speaking) English
Programming Matlab or Python
Statistics ANOVA

Non-linear regression

Statistical programs R
Background Human Movement Sciences, Behavioural Sciences, Biomedical Engineering, or any other field pertinent to behavioural experimentation and modelling
Project specific knowledge Mathematical modelling

Dynamical systems


Project characteristics

Primary focus project Behavioural experiments with humans

Able-bodied children and adults without disorders

Children and adults with ASD (Autism Spectrum Disorder)

Methods End-effector kinematics

Differential equations