I am currently a post doctoral researcher at IMN, University of Bordeaux in France within the "Network dynamics for procedural learning" lab led by Arthur Leblois and Nicolas Mallet.
My academic career has been guided by the ambition to use theory and computational tools to get more insights into biology. First, I had the opportunity to go to a Maths and Physics preparatory school at Lycée Masséna in Nice, France. Then, I joined the engineering school of Mines de Saint Etienne where I could learn more about Biology, Mathematics and Computational languages. Finally, I went to École des Ponts et Chaussées where I did the Master ANEDP of University Paris 6 (UPMC).
I completed my PhD at Inria Sophia-Antipolis in the "TOSCA" and "MathNeuro" teams under the supervision of Etienne Tanré and Romain Veltz. My PhD focused on modelling and analysing (mathematically and numerically) neural networks in which neurons interact through plastic synaptic weights. Rigorously analysing the coupled dynamics (of neurons and synapses) is a non trivial task. In order to overcome this difficulty, I use probabilistic tools to simplify such complex models : at the mesoscopic scale, I make a time-scale separation when there is some biological time-scale differences (synaptic plasticity much slower than the neurons dynamics for example) and at the macroscopic scale, I performed a mean field analysis when the network considered possesses a large number of neurons.
For my first postdoc, I joined the NeuroLogic group of Arvind Kumar at KTH Royal Institute of Technology in Stockholm. As part of the dBRAIN project, I've been working on Parkinson’s disease (PD) and in particular how brain imaging/activity can give us new insights in PD by analysing them with a modelling perspective. I've been working in close collaboration with people from the NatMEG facility (MEG data).
My current postdoctoral research project is a continuation of my previous postdoc, this time bridging data and models from animals (rodent and non-human primate) to humans with PD. Using computational and experimental approaches, I aim to better understand the mechanisms of deep brain stimulation (DBS) and optimise its parameters for a range of brain disorders, ultimately improving therapeutic efficacy and patient quality of life.
Mail : pascal.hellson@gmail.com
8. Mean-field analysis of a neural network with stochastic STDP (with Etienne Tanré and Romain Veltz)
(October 2025). arXiv
7. Biomarkers of brain diseases (with Arvind Kumar)
(September 2025). arXiv
6. Non-equilibrium dynamics of the neocortex in Parkinson’s disease (with Elias Benyahiya, Sreekanth Manikandan, Josefine Waldthaler, Mikkel C. Vinding, Daniel Lundqvist, Per Svenningsson, Dhrubaditya Mitra and Arvind Kumar)
(August 2025). medRxiv
5. Renormalization group analysis of noisy neural field (with Jie Zang, Arvind Kumar and Dhrubaditya Mitra)
(March 2025). arXiv
4. Structural constraints on the emergence of oscillations in multi-population neural networks (with Jie Zang and Arvind Kumar)
eLife (March 2024). arXiv
3. Cortex-wide topography of 1/f-exponent in Parkinson’s disease (with Per Svenningsson, Daniel Lundqvist, Mikkel Vinding and Arvind Kumar)
npj Parkinson's Disease (July 2023). arXiv
2. A Mathematical Analysis of Memory Lifetime in a Simple Network Model of Memory
Neural Computation (July 2020). arXiv, HAL
1. A new STDP Rule in a neural Network Model (with Etienne Tanré and Romain Veltz)
Thesis : Plasticity in networks of spiking neurons in interaction, defended (slides in French) on March 29th, 2021.
INRIA Sophia-Antipolis PhD Seminar (2017) : Spike-Timing Dependent Plasticity (STDP) models or how to understand memory
CNOD 2017 : A simple spiking neuron model based on stochastic STDP
Second year PhD Colloquium EDSFA (2018) : A Mathematical approach on memory capacity of simple synapses models
Neuromod Seminar (2020) : A Mathematical approach on memory capacity of simple synapses models
CCNSv2 (2021) : A new stochastic STDP Rule in a neural Network Model
ICMNS (2021) : Slow-fast and long time behaviour analysis of a neural network with stochastic STDP
GdR ISIS meeting (2022) : Estimating the brain-wide distribution of excitation-inhibition balance in Parkinson's disease
Dive deep - Digital Futures (2023): Cortex-wide topography of 1/f-exponent in Parkinson’s disease
WINQ workshop (2024) : Graph signal processing to get insights into Parkinson’s disease
MEG Nord (2024) : Graph signal processing on MEG for Parkinson’s disease (NATMEG) + Aperiodic activity in MEG-DBS from PD patients (CFIN)
ICTP Workshop (2024, Simons foundation travel grant): Mean Field Analysis of a Stochastic STDP model
ICMNS (2024) : Mean Field Analysis of a Stochastic STDP model
Bordeaux Neurocampus - Seminar (2025) : Aperiodic activity relationship with E-I balance and its alteration in Parkinson’s disease
Virtual SMB MathNeuro Mini-conference (2025): Mean Field Analysis of a Stochastic STDP model
ICMNS 2017 : A simple spiking neuron model based on stochastic STDP
ICMNS 2018 : A mathematical approach on memory capacity of a simple synapses model
ICMNS 2019 : A Mathematical Analysis of Memory Lifetime in a simple Network model of Memory
MEG Nord 2023 : Cortex-wide excitation/inhibition topography in Parkinson’s disease
Neural Traces 2024 : Integration and segregation analysis of resting-state MEG in Parkinson’s Disease
MEG Nord 2024 : On the neural dynamics role of excitation-inhibition (EI) balance in Parkinson's disease
SNUFA 2025: Mean Field Analysis of a Stochastic STDP model
NeuralNet 2025: Graph Signal Processing on MEG for Parkinson’s disease
BrainNet 2023: video recordings on youtube !
Interpretable Brain Data (IBD) 2023: video recordings on youtube!
BrainNet+ 2024: video recordings on youtube !
BrainNet 2025: video recordings on youtube !
MEG-DBS and LFP-DBS: Andreas Højlund (CFIN, Aarhus University, Denmark) and Erik Johnsen (Aarhus University Hospital). Two projects started thanks to respectively the Nordic MEG Hub Mobility Grant and Parkinson Fonden Travel Grant. The first project is on the DBS effects on MEG aperiodic activity in Parkinson’s disease, and the second on the aperiodic activity in LFPs recorded during STN-DBS surgery.
Mean field for networks with STDP: Quentin Cormier (Inria de Saclay) and Milica Tomasevic (École polytechnique). We study the mathematical foundations for proposing a mean field approximation of models of neuronal network with STDP. More info here.
You can find most of my teaching material here. Please do not hesitate if you have questions or comments.
CAMP 2025 Summer School at IISER Pune, India: Some courses on network dynamics (introduction to mean-field: Wilson-Cowan derivation and Brunel 2000 paper mean-field insights), STDP, RL hands-on on Hopfield&Amari models. Amazing Summer School where I've met amazing people: researchers, TAs and students! Thanks again to all of you.
Excitation-Inhibition balance and STDP: Courses given as part of the Advance Topics in Brain Science at KTH, Stockholm. Thanks to Arvind Kumar.
Simulations and ANNs: Course given as part of the Neuronal circuits course at Stockholm University. Thanks to Christian Broberger.
Introduction to Statistics: Course given to L1 students at ISEM (first year Bachelor) during my ATER in 2020. (FRENCH)
Probability: Exercises given to L3 Maths students (last year Bachelor) during my ATER in 2020. (FRENCH)
Master's Students
2025: Zoheir Baghdadli and Jeremy El Grabli, ENSICAEN, Caen, France
Parkinson’s disease through the lens of subspace communication in MEG signals
2024: Elias Benyahia, ENS PSL, Paris, France
Estimating the rate of entropy generation of the brain at a macroscale, using MEG
2023: Valter Lundegårdh, ETH Zurich, Zurich, Switzerland
Graph Signal Processing for Parkinson’s Disease
2022: Archishman Biswas, IIT Bombay, Bombay, India
Analysis and modelling of LFP signals to study Parkinson’s Disease
2021: Michela Santariello, Sapienza University of Rome, Rome, Italy
Dynamical analysis of resting-state functional connectivity in patients with Mild Cognitive
Impairment
2021: Constantin Lührmann, University of Göttingen, Göttingen, Germany
Origin of time scales in networks of spiking neurons
Bachelor's Students
2023: Axel Nilsson and Karl Lindblad, Computer Science at KTH, Stockholm, Sweden
What can a network with STDP learn and how it resists attacks?
2023: Gustav Bressler and Sigge Dackevall, Computer Science at KTH, Stockholm, Sweden
Effects of including neuromodulation and noise in a network with STDP?