Biology subjects

Seeck, M.

Publications and source records attributed to Seeck, M..

4 recordsLinked to original sources

bioRxiv2026

How well does electrical source imaging reconstruct neural time series? Evidence from simultaneous scalp and intracranial EEG recordings

Electrical source imaging (ESI) estimates neural activity from scalp EEG, but its ability to reconstruct the temporal dynamics of spontaneous brain activity beyond source localization remains unclear. This is assessed using simultaneous high-density EEG and intracranial EEG (iEEG) recordings during spontaneous activity in 34 patients with drug-resistant focal epilepsy. Four ESI approaches (three distributed source models and one beamformer) are applied to estimate neural activity, which is then projected onto iEEG electrode locations to generate virtual iEEG signals for comparison with measured recordings. Reconstruction fidelity is quantified using Pearson correlations between actual and virtual iEEG signals, while spatial accuracy is assessed through peak localization error and spatial deviation derived from point-spread and cross-talk functions. Analyses are performed in broadband and delta, theta, alpha, and beta bands. Temporal reconstruction remains limited, with the highest estimated correlation reaching 0.126, localization errors of approximately 3 cm, and substantial spatial leakage. Reconstruction quality decreases with electrode depth and increasing frequency, while improving with higher epileptic spike rates. No single method is optimal across all metrics, although beamforming shows the weakest temporal reconstruction. These findings indicate that ESI only partially captures spontaneous intracranial dynamics and define spatial and temporal constraints for interpreting EEG-based source reconstructions.

neuroscience↗

bioRxiv2022

Reinstatement of emotional associations during human sleep: an intracranial EEG study

The scientific literature suggests that emotional memories benefit from a privileged consolidation over neutral memories. This effect extends to consolidation processes that occur during sleep. Indeed, during sleep, a complex set of oscillations (namely slow-oscillations, theta rhythm and spindles) mediates the communication between brain regions involved in the long-term integration of memories. However, whether sleep oscillations may contribute to the reactivation and consolidation of emotional memories in humans is still unclear. Because non-invasive electroencephalography (EEG) has limited access to deep brain regions implicated in memory and emotion (e.g., hippocampus, amygdala, orbitofrontal cortex), here we recorded EEG signal from these brain regions using intracranial electrodes placed in medically-resistant epileptic patients in the context of presurgical investigation. During wakefulness, we presented the patients with emotional (i.e., humorous) vs emotionally neutral pictures paired with a sound. Then, we tested for the reinstatement of emotional-associations by delivering the sound during a subsequent period of sleep. We found that the reactivation of emotional (compared to neutral) memories during sleep enhanced slow-oscillation and spindle activity in the orbitofrontal cortex, paralleled with an increase in theta connectivity between the hippocampus and the orbitofrontal cortex. In addition, we observed that the theta response to emotional memories reactivated at subsequent wake was different than for neutral memories, suggesting a change in memory traces with targeted memory reactivation. These data suggest that consolidation of emotional events during sleep is due to a larger expression of sleep features (in the slow-oscillation, theta and sigma frequency bands) and that the mechanisms of brain plasticity also take place in emotional brain regions during NREM sleep.

neuroscience↗

bioRxiv2018

Audio-tactile and peripersonal space processing around the trunk in human parietal and temporal cortex: an intracranial EEG study

Interactions with the environment happen by the medium of the body within ones peripersonal space (PPS) - the space surrounding the body. Studies in monkey and humans have highlighted a multisensory distributed cortical network representing the PPS. However, electrophysiological evidence for a multisensory encoding of PPS in humans is lacking. Here, we recorded for the first time intracranial electroencephalography (iEEG) in humans while administering tactile stimulation (T) on the trunk, approaching auditory stimuli (A), and the combination of the two (AT). To map PPS, in AT trials, tactile stimulation was delivered when the sound was far, at an intermediate location, or close to the body. We first identified electrodes showing AT multisensory integration (i.e., AT vs. A+T): 19% of the recording electrodes. Among those electrodes, we identified those showing a PPS effect (30% of the AT electrodes), i.e., a modulation of the evoked response to AT stimulation as a function of the distance between the sound and body. For most sites, AT multisensory integration and PPS effects had similar spatiotemporal characteristics, with an early response (~50ms) in the insular cortex, and later responses (~200ms) in pre- and post-central gyri. Superior temporal cortex showed a different response pattern with AT multisensory integration at ~100ms without PPS effect. These results, representing the first iEEG delineation of PPS processing in humans, show that PPS processing happens at neural sites where also multisensory integration occurs and at similar time periods, suggesting that PPS representation (around the trunk) is based on a spatial modulation of multisensory integration.

neuroscience↗

bioRxiv2017

Focal Versus Distributed Temporal Cortex Activity For Speech Sound Category Assignment

Percepts and words can be decoded from largely distributed neural activity measures. The existence of widespread representations might, however, conflict with the fundamental notions of hierarchical processing and efficient coding. Using fMRI and MEG during syllable identification, we first show that sensory and decisional activity co-localize to a restricted part of the posterior superior temporal cortex. Next, using intracortical recordings we demonstrate that early and focal neural activity in this region distinguishes correct from incorrect decisions and can be machine-decoded to classify syllables. Crucially, significant machine-decoding was possible from neuronal activity sampled across widespread regions, despite weak or absent sensory or decision-related responses. These findings show that a complex behavior like speech sound categorization relies on an efficient readout of focal neural activity, while distributed activity, although decodable by machine-learning, reflects collateral processes of sensory perception and decision.

neuroscience↗