I am a cognitive neuroscientist interested in the neural mechanisms that support cognitive control and decision making. Much of my career has focused on figuring out the function of anterior cingulate cortex, a poorly understood brain area that is involved in many cognitive tasks. My laboratory is presently funded by an Advanced Grant from the European Research Council to answer this question. Our working hypothesis is that the ACC uses hierarchically-organized world models to evaluate the reward value of long-term plans against the immediate costs of their execution . To test this hypothesis we follow a "converging methods" approach involving electroencephalography, functional magnetic resonance imaging, lesion studies, and computational modeling.
I hold master’s degrees in software engineering (EFREI) and cognitive artificial intelligence (Utrecht University),
and an interdisciplinary doctorate in cognitive psychology/machine learning (Plymouth University) on the topic of
insight problem-solving.
I’m interested in the learning of temporal structure for decision-making. What are the
purposes of learning temporal structure (preservation of context, reduction of cognitive costs, improved
exploration...)? How do we learn it, according to what principles and techniques? To answer these questions,
I take inspiration from machine learning – in particular recurrent neural networks and (hierarchical) reinforcement
learning.
At Ghent University, my work focuses on the Anterior Cingulate Cortex (ACC), a brain region thought
to play an important role in sequential decision-making. I design artificial recurrent neural networks and train
them on sequential decision-making tasks. With my colleagues at the Learning and Cognitive Control Lab led by Clay
Holroyd, we compare the representations that emerge from these models to the fMRI activation patterns for participants
performing the same task (using Representation Similarity Analysis techniques), enabling us to compare the models
relative to one another and relative to the brains of participants. Thereby we seek to pinpoint the function of the
ACC, and to better understand human sequential decision-making.
Before coming to Ghent as post-doc, I completed my Cognitive Neuroscience research
master and PhD at the Donders Institute for Brain, Cognition, and Behavior (Nijmegen,
the Netherlands). My PhD was supervised by Dr. Bernd Figner and Prof. Dr. Karin
Roelofs, and focused on the role of time ambiguity in intertemporal (i.e., short- vs.
far-sighted) decision-making. More broadly, I wanted to understand which psychological,
neural, computational, and hormonal mechanisms are involved when making intertemporal
or risky decisions with and without ambiguity. To study this question, I used a
combination of methods including functional neuroimaging, behavioral experiments,
computational modeling, Bayesian mixed models and hormonal assessments.
In the LCCL
lab I am part of the ERC-funded team that investigates the precise (computational)
function of the anterior cingulate cortex, or ACC. More specifically, we test whether
the ACC selects and motivates high-level, temporally extended behaviors according to
principles of hierarchical reinforcement learning, using uni- and multivariate fMRI
methods in combination with recurrent neural network (RNN) models. In my current
project, I investigate how cognitive load may impact the encoding of task-presentations
in the ACC.
I studied Electronics and Telecommunications Engineering at Universidad Técnica Particular de Loja (UTPL) in Ecuador, and Biomedical Engineering (MSc) at University of Lübeck and University of Applied Sciences Lübeck, in Germany. I joined the LCC Lab in 2020 as a PhD student, and I work in Decision Making, Reward, and Cognitive Control. My interests also include Reinforcement Learning, Memory, Contextual Learning, and Music Reward. My current research focuses on how the Anterior Cingulate Cortex (ACC) regulates shifts between strategies during foraging behavior (exploring vs exploiting), for that we use computational models to implement our hypothesis and then test these models against fMRI experiments in human subjects.
My PhD project in the laboratory of Senne Braem is about investigating the neural implementation of cognitive flexibility. Together with my co-promotors Clay Holroyd and Marcel Brass, we aim to unravel the neural mechanism which enables human to engage in flexible behaviours. Unlike previous studies which focus mainly on certain brain region or network, we are striving to build a comprehensive computational model in both temporal and spatial domain by leveraging various brain imaging techniques (EEG, fMRI), psychophysiological measurement (pupillometry), and behavioural paradigms. Before joining Clay’s lab, I obtained my Master degree on Cognitive Neuroscience at Donders Institute, Nijmegen. Outside of academia, I enjoy cooking, practicing calligraphy, and training on power lifting (beginner level).
I have completed my bachelor in Biology at University of Crete in Greece and my master
in behavioral biology at Wageningen University in The Netherlands. My interest in
cognitive neuroscience started with working on stress during my master thesis at the
Donder Institute for Brain, Cognition & Behavior in Nijmegen. Then, I joined Prof. Erno
Herman’s group to investigate, under the supervision of Dr. Vassena Eliana, the effects
of acute stress on decision making and motivation with functional neuroimaging.
My main research interest lies on decision making, cognitive control and reinforcement
learning. Here, at Ghent University, in the LCCL lab, I am part of a team aiming to
develop and test a new formal, unified theory of ACC function. I will mainly focus on
the investigation of the effects of reward on the ACC patterns of activity during
sequential behavioral paradigms using functional neuroimaging (fMRI).
I graduated from the University of Ghent in Medicine in June 2020 and I am currently doing a Master after Master, in which I am training to become a neurologist. I am combining this with a PhD in the Interdisciplinary Program that mainly focuses on the field of stroke and epilepsy, under the supervision of Prof. Dr. Veerle De Herdt (neurologist at the UZ Ghent) and Prof. Clay Holroyd. In the LCCL lab, I am part of a team aiming to develop and test a new formal, unified theory about the exact function of the anterior cingulate cortex. I am responsible for two studies: one study that includes patients who have had a stroke in the frontal lobe (more specifically; the anterior cingulate cortex), and one study that includes patients with refractory epilepsy who are admitted to the Center for Neurophysiological Monitoring at the UZ Ghent for invasive video-EEG monitoring.
I have a Bachelor degree in Psychology from the University of Groningen, and a Master degree in Cognitive Neuroscience from the University of Amsterdam. During my theses, I have mainly focused on computational and statistical modeling, specifically linking computational models of cognition and behaviour to both EEG and fMRI data. In the LCC lab, I aim to investigate how humans can learn and leverage hierarchical structure behaviourally, and what neural signatures we can find of such hierarchies. Specifically, we believe the ACC might attribute obtained rewards to learned hierarchical representations, which can be reflected in the EEG as a 'Reward Positivity' locked to temporally abstracted events. We also believe the ACC might hold distributed representations of plans, strategies, and (sub)goals that imply the presence of hierarchy, which we hope to identify with fMRI and RSA.
I obtained master's degree in basic psychology in South China Normal University. I joined the LCC Lab in 2022 as a PhD student. I am interested in reinforcement learning, decision making, reward processing etc. My research will mainly focus on localizing cortical reward prediction error signals using RSA (Representational Similarity Analysis) to figure out how ACC works in the reward processing.