Cognitive Computational Neuroscience


Our research uses computational and machine learning techniques to study functions of the human brain in health and disease. Our work focuses on three main axes:

When consciousness fades away we are not aware of our surroundings. However, our brains continue to process information from the environment, like sounds.
In our work, we are investigating how the human brain responds to stimuli of the environment when consciousness is lost. Moreover, we are combining electrophysiological measurements of brain activity with computational techniques, like measures of neural synchrony and complexity, to identify predictors of awakening from a coma.

Representative publications

Learning and predictions

Modeling associative learning Adapted from: Tzovara et al., 2018, PLoS Computational Biology, doi: 10.1371/journal.pcbi.1006243

The world around us is full of rich sensory experiences, which often follow repetitive rules. For example, temporal regularities, like the time of the day, might allow us to foresee and avoid subsequent traffic jams, or auditory events, like the sound of a siren, could help us predict the arrival of an ambulance, or a fire truck. Through exposure to reoccurring events we are able to learn patterns and eventually form predictions about future events before these occur.

To shed light in the neural computations that allow us to form predictions about future events, we are using machine learning and computational modeling techniques, to study neural activity during learning of environmental rules.

Representative publications

Machine learning for neuroscience

Deep learning pipeline *Figure from: Aellen et al., 2021

As the amount of data in the field of neuroscience and neurology increases, it becomes imperative to have powerful algorithms for analysing them. Machine learning algorithms have revolutionized several fields, but their use in electrophysiological data remains limited. In our work, we are developing deep learning pipelines for analysing electrophysiological data, with emphasis on interpretability. Moreover, we are evaluating the effects of algorithmic bias when applying machine learning techniques on medical data.

Representative publications