PhD in Computer Science: Building a computational model to modulate human time perception (# of pos: 2)

We are heading towards a hybrid society of AI-enabled agents cooperating with humans in a variety of tasks. Human decision making and perception are however notably prone to cognitive biases, which means that humans deviate from rational decision making or have subjective perception of their environment. In applications supporting the human, the integrated AI should account for these biases when making suggestions or taking actions.

Within the context of this PhD position, we focus on the bias in human time perception. Sometimes time flies, whereas at other moments, time feels expanded. This subjective judgment of duration influences the optimality of human decisions.  In psychological experiments, the subjective judgement has been shown to be a function of the individual's environment, the task at hand, the allocation of the individual's attention, etc.

Intriguingly, psychological experiments have shown that humans estimate time differently when they are presented with particular stimuli, like flickering dots or beeps. It seems thus to be possible to modulate the duration of a perceived interval by presenting well-chosen stimuli. In this PhD, we aim to develop a set of computational models that can predict and modulate human time perception in less-controlled environments. This requires to build a model that takes as input particular environmental features as well as physiological signals, estimates the current human subjective time perception (does the user feels time going faster than objective time or slower?) and to determine what stimuli could be present to adjust this time perception in the desired direction.

The work involves building a computational model and building experimental set-ups, such as elementary games, to evaluate this computational model.  Hence, we are seeking for students interested in cognitive architectures from a mathematical perspective and who are capable of programming algorithms into a workable prototype. 

The selected candidate will work on this topic in the context of a prestigious FET-OPEN project, ChronoPilot (project website:

https://chronopilot.eu

), in collaboration with project partners from Greece, Luxemburg, Germany and Belgium. This position is seeking for a student who will be responsible for the computational models. In collaboration with experts on VR/AR and in psychology, you will integrate this in a working prototype. Hence: we are mostly looking for students that have a solid understanding of (applied) mathematics (e.g. generative modelling, Bayesian reasoning, etc...) and that have coding skills with interest in psychology. We are 

not

 seeking for data science engineers or for psychologists.

Main tasks:

• Design, development, and implementation of a computational model of subjective time experience. These models take as input environmental stimuli, task-related stimuli as well as physiological state measured via sensors, and output a subjective time. 
• Integration of this model into a working prototype, taking multimodal input (audio, video), estimating in real-time the current subjective time perception, and then taking corrective actions to adapt the subjective time perception in the desired direction. 
• Validation of these models, initially in simulation, but mostly through user experiments, and interpretation of the results in tight collaboration with psychologists.
• Close international and interdisciplinary collaboration (especially psychology) with project partners from Germany, Greece, and Luxembourg
• Independent work on a diverse set of tools and methods (mainly machine learning)
• Collaboration on publications and presentations at international conferences
• Supervision of master theses related to the subject of this PhD

The PhD position is a fully funded position. After successful evaluation of the progress in the first year, the contract can be extended with 3 more years (4 years in total).