We develop quantitative methods on neurophysiological large datasets to investigate brain information processing during cognition and disease. In particular, we are interested in the mechanisms by which stimuli, behavioral responses, and pathological states are encoded and distributed through the simultaneous activity of multiple brain areas. In our studies, we mainly analyze single-cell (spike train) and neural population (human intracranial EEG) data.

Epilepsy: Temporal and spatial characterization of epileptic networks

We study the emergence and maintenance of the pre-ictal state (the brain state prior to epileptic seizures) by means of dynamic functional connectivity analysis of long-lasting periods (~12 hours) of intracranial data from epileptic patients [1]. We are also interested in spatially determining the main brain areas where seizures begin (“focus”). With this regard, we have recently developed quantitative tools to predict seizure focus localization for pre-surgical diagnosis [2,3].

Selected publications

Cognition: Neural coding and neural communication

We are interested in the neural coding problem and in particularly, in relating this problem to that of information transmission. We have mainly studied both problems in the context of reward-driven perceptual tasks in monkeys [1,2] to characterize the thalamo-cortical and cortical-cortical directed functional paths that are activated during these tasks.

Selected publications

Computational models and methods

We have contributed to develop non-parametric statistical methods for non-linear [1] and linear models [2] for the inference of functional connectivity pathways using multiple and simultaneous brain area recordings.

Selected publications