Team ERCN – IMoPA

Neuroscience Clinical Research Team

Manager: Laurent KOESSLER

The ERCN team aims to advance our understanding of the healthy and pathological brain mechanisms underlying human cognition. The team adopts a multi-modal and multi-scale approach focused on clinical populations, particularly patients with drug-resistant epilepsy

Their research centers on two key cognitive functions:

Face recognition, which relies on the dominant sensory modality (vision) and on dense, complex, and lateralized brain networks.

Word recognition, a learned function acquired during development that enables written communication.

An Innovative Approach to Clinical Neuroscience Research

The ERCN team stands out through its unique methodological approach, centered on direct recordings of brain electrical activity across multiple scales (from single neurons to scalp EEG) often performed simultaneously. This rich dataset enables precise links to be drawn between neuronal activity and behavior (Ferrand et al., 2023; Angelini et al., 2024; Quian Quiroga et al., 2023).

One of the team’s key tools is the use of “frequency tagging” approach, a sensitive, rapid, and objective method of quantifying and tracking brain responses using fast periodic visual stimulation. It has an excellent signal-to-noise ratio. Initially developed for studying basic sensory functions, this technique is now used for exploring complex cognitive processes, such as visual recognition (Rossion & Boremanse, 2011; Norcia et al. 2015, ERC ‘facessvep’ – 2011 & ERC ‘humanface’ – 2022).

This figure illustrates the statistically significant contribution of deep brain electrical sources (here, the hippocampus) to the scalp EEG despite their depth relative to surface sensors, background brain noise (signal-to-noise ratio), and the geometric configuration of these sources (closed electric field)

Multi-scale EEG of Deep Brain Electrical Sources
(adapted from Koessler et al., 2015)
This figure illustrates the statistically significant contribution of deep brain electrical sources (here, the hippocampus) to the scalp EEG despite their depth relative to surface sensors, background brain noise (signal-to-noise ratio), and the geometric configuration of these sources (closed electric field).

A Synergy Between Research and Clinical Practice

Another major strength of the team lies in the strong integration of fundamental research with the clinical activities of the Neurology Department at the Nancy University Hospital (CHRU), a National and European Reference Center for Rare Epilepsies. The team, composed primarily of clinician-researchers, is based at the heart of the hospital (Pavillon Krug), in immediate proximity to care and diagnostic units.

Thanks to its strategic location, the team is able to conduct research that is directly aligned with clinical needs. This includes research into drug-resistant epilepsy, cognitive disorders related to conditions such as Alzheimer’s disease, and psychiatric illnesses, as well as neuropsychological assessments conducted before and after surgery. The team collaborates closely with several hospital units: epileptology (Pr. Maillard), functional surgery (Pr. Colnat-Coulbois), functional exploration (Pr. Jonas), psychiatry (Pr. Hingray) and neuropediatry (Dr. Kuchenbuch).

Research Axes

The ERCN team is centred around three scientific questions that range from fundamental science to clinical and methodological issues.

The uniqueness of this team lies in the study of cerebral mechanisms using intracerebral recordings in human beings in vivo and the study of a particular cerebral network: the ventral visual pathway.

Axis fundamental – Advancing Knowledge in Systems and Cognitive Neuroscience

Expand and refine the spatial mapping of face and written language recognition in the human brain, across both low- and high-frequency electrophysiological signals, and identify the optimal methodological parameters for these mappings.
Continue developing and optimizing original frequency-tagging paradigms using high-density EEG (128 channels, Biosemi) to measure cognitive functions related to visual recognition of faces and language, as well as attention.
Investigate the temporal dynamics of visual recognition activity within the ventral occipito-temporal network, using both low- and especially high-frequency electrophysiological signals.
Strengthen and extend the causal links between neural activity related to visual recognition, as identified through our paradigms, and behavior, using intracerebral electrical stimulation.
Reveal functional and effective brain connectivity through an innovative approach involving electrical stimulation of intracerebral electrodes while recording periodic electrophysiological activity at distant sites.
Study and decode, at the unit scale (recordings of population and single-neuron action potentials), the cellular mechanisms of face recognition.
Compare rhesus monkey and human models of face recognition across multiple modalities (fMRI, EEG, single-cell recordings), in collaboration with external partners B. Cotterau (CerCo, Toulouse) and T. Brochier (INT, Marseille), ANR 2023 (PREFER).
Investigate how spatial and selective attention modulates these neurofunctional networks.
Uncover the brain mechanisms underlying visual recognition of written language (letters, words), semantics, and naming, an effort involving collaboration between several PIs within the team (J. Jonas, L. Maillard, B. Rossion) with Luxembourg University (A. Lochy, C. Schiltz) and Cambridge MRC (M. Lambon Ralph; O. Hauk).

A. En haut : Le signal SEEG variant en temps et en fréquence (40-160 Hz, approche par ondelettes de Morlet) En bas : la modulation de l’amplitude haute fréquence (HF) dans le temps (l’enveloppe d’amplitude HF) est obtenue en faisant la moyenne des signaux temps-fréquence sur la bande de fréquences de 40 à 160 Hz. B. Cartes montrant l’amplitude lissée des basses fréquences sélective du visage affichée sur la surface corticale du cortex ventral occipito-temporal. C. Relation linéaire entre les cartes d’amplitude basse fréquence et HF présentées. — **Recording and Quantification of Low-Frequency (LF) and High-Frequency (HF) SEEG Signals Selective for Faces in the Ventral Occipito-Temporal Cortex** *(adapted from Jacques et al., 2022)*
A. Top: SEEG signal plotted over time and frequency (40–160 Hz, using Morlet wavelet transform). Bottom: High-frequency (HF) amplitude modulation over time (HF amplitude envelope), obtained by averaging the time–frequency signal within the 40–160 Hz frequency band.
B. Maps displaying the smoothed face-selective low-frequency amplitude projected onto the cortical surface of the ventral occipito-temporal cortex.
C. Linear relationship between low-frequency and high-frequency amplitude maps.

Axis clinical – Neuroscience Research

Map epileptic brain networks to define optimal strategies for intracerebral electrode implantation and surgical resection.
Investigate the relationship between epileptic networks, structural abnormalities, and functional impact in complex malformations of cortical development (MCD) (PI. L. Tyvaert).
Describe, understand, and predict cognitive outcomes in epilepsy, with a particular focus on visual recognition, naming, and reading, using connectivity analyses and electrical brain stimulation.
Develop predictive models using artificial intelligence to assess long-term outcomes in young children who have suffered perinatal brain injury.
Explore psychiatric comorbidities associated with neurological conditions, particularly epilepsy and psychogenic non-epileptic seizures (PNES).
Study language lateralization in patients with epilepsy.

image Axe 3 nelle Eq 10 — *Correlations Between Scalp EEG Discharge and Intracerebral Seizure-Onset Localizations (adapted from Ferrand et al., 2024)*

Axis Methodological – Technological and Methodological Advances in Human Neuroscience

Identify novel epileptic biomarkers, including those linked to signal amplitude reduction or originating from the insular cortex, to improve epilepsy diagnosis.
Discover new electrophysiological biomarkers of memory, particularly in Alzheimer’s disease and attentional processes.
Define optimal reference schemes and signal acquisition protocols based on the anatomical scale under investigation.
Characterize the biophysical relationships between signals recorded at different scales to support the development of automated biomarker extraction tools. In the long term, leverage our extensive database to test artificial intelligence algorithms (machine learning and deep learning) for improved epilepsy diagnosis and classification.
Understand the coding of electrical information, specifically how processes and biomarkers are organized and interconnected across anatomical levels—from single neurons to neuronal populations and brain networks.
Investigate the electrophysiological effects of electrical stimulation in vivo in humans using intracerebral EEG (ANR EpiThera, 2024).
Create functional connectivity maps to better understand epileptic and cognitive networks within the ventral visual pathway.
Develop new modes of brain stimulation, both intracerebral and surface-based, aimed at rehabilitating brain networks.
Model biophysical EEG current conduction, refining stimulation strategies for greater accuracy and efficiency (e.g., tissue conductivity, electrode configuration, stimulation thresholds).
Develop portable technologies for EEG recordings coupled with Fast Periodic Visual Stimulation (FPVS) outside clinical settings (1st Prize – Incubateur Lorrain 2024; ongoing framework agreement with Bioserenity–CNRS; ANR eTECH_Neuro 2024).

image Axe 3 Eq 10 — Simultaneous Coupling of Transcranial Electrical Stimulation and Stereo-Electroencephalography for In Vivo Measurement of Intracerebral Electric Fields in Humans

Our Technological Platforms

Multi-scale EEG Recording (Macro- and Micro-electrodes)

This platform enables an unprecedented in vivo exploration of human brain activity at multiple anatomical scales, from individual neurons to large-scale neural networks, through EEG recordings performed in drug-resistant epilepsy patients as part of their clinical care at Nancy University Hospital (CHRU).

A Unique Approach
We combine classical intracerebral recordings (SEEG) with unit-level measurements using implanted probes that include both macro- and micro-electrodes. This setup allows the observation of single-neuron and neuronal population activity with millisecond-level temporal precision.

Scientific Goals:

Assess the face recognition specificity of individual neurons, compared to their object detection capabilities, in the medial temporal lobe and ventral temporal cortex.
Understand the relationships between signals recorded at different anatomical scales.
Combine these recordings with fast periodic visual stimulation to generate measurable, objective brain responses.

Technology:
A 256-channel EEG system, synchronized with video, enables detailed analysis of signals from both single neurons and neuron populations across scales. Data is processed on a dedicated platform and can be securely shared between the CHRU and our research lab.

High-Density EEG Platform

This platform allows non-invasive recording of brain activity in healthy individuals and patients with neurological conditions, using a high-density EEG system (up to 256 channels).

Method:
We use rapid visual stimulation (e.g., 6 images per second), which elicits brain responses at the same frequency—a phenomenon known as frequency tagging. These precisely measured responses serve as objective biomarkers of visual recognition.

Scientific Goals:

Evaluate the quality of visual processing (faces, words, scenes) in adults and children, whether healthy or affected by neurological disorders.
Study the speed and localization of brain responses (e.g., recognition of familiar vs. unfamiliar faces).
Understand the development and disruption of cognitive functions.

Technology:
The BIOSEMI EEG system uses low-impedance active electrodes (<1 Ohm) to ensure excellent signal-to-noise ratios. EEG is synchronized with a visual stimulator, response keyboard, and eye-tracker to monitor participant attention.