Our lab members

1. Principal investigator
2. Post-doctoral researchers
   • Irene Cogliati Dezza
   • Pieter Verbeke
3. PhD researchers
   • Jonas Simoens
   • Haopeng Chen
   • Xiaoyu Zhang
4. Visiting researchers
5. Internship students

Principal Investigator

Tom Verguts

Personal statement
Humans are biological agents. Thus, behavior is generally adaptive, meaning that it aims to optimise some goal function (e.g., to minimise error or to maximise reward). At the same time, humans live in rapidly changing environments, so this ideal will often not be reached.

From this perspective, human behavior is often not optimal, but humans learn to gradually achieve their goal (function). In summary, humans learn adaptive behavior, and this defines the topic of our research group.

Although this is a fairly broad topic, there are a few recurring themes, such as learning to use inter-areal synchrony for neural communication; learning what to store in declarative memory; and learning the (meta-)parameters of decision processes. For this research, we use behavioral, computational, and neural methods.

Post-doctoral researchers

Irene Cogliati Dezza

I am a postdoctoral researcher funded by the FWO-postdoctoral fellowship at Ghent University (BE) and University College London (UK). I am also associate editor at In-Mind Italy and co-founder of BeBrain. I hold a BA in Biology, a MA in Neurobiology, a university certificate in data science and a PhD in computational cognitive neuroscience. My research focuses on understanding how people decide what they want to know and how they explore novel and unknown courses of action. I conduct my research in adults and children in both healthy and clinical populations. My approach combines state-of-the-art methods from diverse disciplines including psychology, neuroscience and computer science.

Pieter Verbeke

My research focusses on how humans and artificial agents can balance shared versus separated task representations to optimize continual learning. Here, separated representations are useful to avoid (catastrophic) interference and shared representations are useful to speed up learning via generalization. We argue that humans do this via hierarchical learning. At the hierarchically higher level, relations between tasks are learned and used to decide which lower-level modules get control over behavior. The appropriate modules can be bound via the synchronization of oscillations (in biological agents) or via multiplicative gating (in artificial agents). To investigate this, we use multiple tools such as computational modelling, EEG, fMRI and behavioral studies.

PhD researchers

Jonas Simoens

Psychologists tend to be interested not only in understanding, but also in improving, human cognition and behavior. Correspondingly, it has already been extensively demonstrated that concrete behaviours can indeed be modulated by selectively rewarding certain behaviours more than others. Inspired by computational models of cognitive control, I investigate whether, in the same way, it is also possible to modulate abstract task execution parameters, such as learning rate, as described by computational models of learning and decision making. Moreover, I investigate whether these parameters can be adapted to multiple environments (in terms of reward contingencies) simultaneously, guided by associated contextual features.

I conduct this research in collaboration with Tom Verguts and Senne Braem, using a combination of computational modeling, behavioural and neuroimaging techniques.

Haopeng Chen

My research topic is the “behavioral and neural nature of the testing effect”. The testing effect refers to the phenomenon that testing can help people reinforce the learned materials better than restudying. Although the Testing effect has been demonstrated in many studies and holds major educational implications, its origin has remained unclear. Based on earlier empirical work and theory formation, we currently postulate that the testing effect derives from reward prediction error (RPE). To be specific, during testing, people will calculate their confidence in their answers and get feedback, which will trigger the RPE (feedback-confidence). Therefore, it might be the RPE triggered by testing that leads to the testing effect. We will try to investigate this postulation at both behavioral and neural (fMRI) levels.

Xiaoyu Zhang

My research interest is how human attention works in learning different tasks and how to model this procession with neural networks and oscillation. Humans can learn new tasks with transferable knowledge and adapt to different environments quickly. Among these activities, attention plays an important role, which has been proposed to be implemented via oscillation in the brain. We build models with neural networks and oscillations and investigate how the models implement cognitive flexibility, particularly generalization with well-designed complex psychological tasks. Then, we will compare the model performance with the EEG results from human behavioral experiments. Using computational modeling, EEG, and behavioral methods, I am collaborating with Tom Verguts and Mehdi Senoussi for this research.

Visiting researchers

Currently, we have no visiting researchers working in our lab.

Internship students

Currently, we have no visiting researchers working in our lab.

---

Please get in contact if you are interested in a research stay or an internship in our lab. Want to see who used to work at our lab? Find a list of our previous lab members here.