bioRxiv (Cold Spring Harbor Laboratory)
July 13, 2022
Andrea I. Luppi, Justine Y. Hansen, R. Adapa et al.
5 citations
preprint
Psychoactive drugs reshape brain function by engaging multiple neurotransmitter systems simultaneously. By mapping the distribution of 19 neurotransmitter receptors and transporters (via PET) and the connectivity changes caused by 10 drugs (anesthetics, psychedelics, and stimulants), the study shows that drug effects are organized along hierarchical gradients of brain structure and function. Additionally, brain regions susceptible to drug-induced changes are also vulnerable to structural alterations from brain disorders. These findings reveal systematic links between molecular neurochemistry and large-scale functional reorganization.
Andrea I. Luppi, Pedro A. M. Mediano, Fernando E. Rosas et al.
5 citations
preprint
Consciousness can be understood not as a single unified thing but as composed of distinct information-theoretic elements. A new approach called Integrated Information Decomposition (ΦID) shifts from measuring how much integrated information a system has to analyzing its composition. This provides a formal way to determine whether consciousness is an emergent phenomenon based on that composition. Two organisms can have the same amount of integrated information yet differ in its composition. A new measure, ΦR, and the ΦR-ing rate quantify how efficiently an entity uses information for conscious processing. This decomposition identifies qualitatively different 'modes of consciousness,' enabling mapping between phenomenology and information-theoretic structure, starting with selfhood.
British Journal of Anaesthesia
April 1, 2025
Milan Van Maldegem, Jakub Vohryzek, Selen Atasoy et al.
3 citations
Ketamine, at anesthetic doses, produces a state where people are unresponsive yet often report vivid inner experiences, separating conscious awareness from behavioral responsiveness. Using connectome harmonic decomposition on fMRI data, researchers found that brain signals during ketamine-induced unresponsiveness show increased fine-grained spatial patterns, indicating higher neural granularity. This harmonic signature aligned with those of LSD-induced and ketamine-induced psychedelic states, but misaligned with signatures from unconscious individuals due to propofol sedation or brain injury. The method can track changes in conscious awareness even when behavior is absent, offering a tool for consciousness and anesthesia research.
Neuropsychopharmacology
September 12, 2025
Jakub Vohryzek, Andrea I. Luppi, Selen Atasoy et al.
2 citations
The brain's function depends on its structural wiring, and psychedelics alter this relationship. Using connectome harmonic decomposition, a method linking brain activity to the network of white matter pathways, the authors show that under N,N-dimethyltryptamine (DMT), the brain's harmonic repertoire shifts similarly to that seen with psilocybin, LSD, and ketamine. Repertoire entropy—a measure of the diversity of brain states—increases under DMT. For the first time, the energy spectrum difference and repertoire entropy of connectome harmonics track the intensity of subjective experience in real time, indicating a close coupling between brain network dynamics and conscious experience.
bioRxiv (Cold Spring Harbor Laboratory)
June 25, 2024
Milan Van Maldegem, Jakub Vohryzek, Selen Atasoy et al.
2 citations
preprint
Ketamine, a dissociative anesthetic, produces different brain dynamics at anesthetic versus sub-anesthetic doses. Using connectome harmonic decomposition (CHD) to analyze resting-state fMRI data from volunteers during ketamine-induced unresponsiveness, the study found increased prevalence of localized harmonics, similar to patterns seen in psychedelic states induced by LSD or psilocybin. This contrasts with traditional GABAergic sedation (e.g., propofol), where global harmonics increase with higher doses. The results indicate that ketamine-induced unresponsiveness does not necessarily suppress conscious experience and influences connectome harmonics oppositely to GABAergic hypnotics. CHD may track alterations in conscious awareness rather than behavioral responsiveness.
Elife
July 18, 2024
Andrea I. Luppi, Pedro A. M. Mediano, Fernando E. Rosas et al.
1 citation
Loss of consciousness significantly disrupts the brain's ability to integrate information. In a study involving functional MRI analysis, it was revealed that gateway regions in a 'synergistic global workspace' correspond to the default mode network, while broadcaster regions align with the executive control network. This integration breakdown occurs during general anaesthesia or disorders of consciousness, with recovery restoring functionality. The findings enhance understanding of consciousness by bridging Global Neuronal Workspace and Integrated Information Theory, highlighting the critical role of brain networks in maintaining conscious experience.
bioRxiv (Cold Spring Harbor Laboratory)
Hanna M. Tolle, Andrea I. Luppi, Timothy Lawn et al.
1 citation
preprint
A geometric deep learning model called graphTRIP predicts post-treatment depression severity from pretreatment clinical and brain imaging data. Trained on a clinical trial comparing psilocybin and escitalopram, it achieves strong predictive accuracy (r = 0.75) and generalizes to an independent dataset. The model links better outcomes to reduced functional coupling within serotonin systems and broader serotonergic integration with sensory-motor networks. Causal analysis shows a group-level advantage of psilocybin over escitalopram but identifies individuals with specific stress-related neuromodulatory profiles who may benefit more from escitalopram, advancing precision medicine and biomarker discovery in depression.
bioRxiv (Cold Spring Harbor Laboratory)
January 13, 2026
Iván Mindlin, Carlos Coronel-Oliveros, Jacobo Sitt et al.
A biologically grounded inhibitory homeostatic plasticity rule embedded into the Dynamic Mean Field (DMF) model creates a Homeostatic Dynamic Mean Field (HDMF) model that dynamically tunes local excitation-inhibition balance. The HDMF reproduces statistical observables of brain activity as well as the original DMF, can sustain neuromodulatory perturbations without overhead computations, and generates unprecedented sleep-like slow-wave activity that can coexist with wake-like asynchronous dynamics, permitting modeling of dissociated states of consciousness such as parasomnias. A single homeostatic rule broadens the stability and expressiveness of the DMF, providing a unified platform for studying how local adaptive processes shape the diverse global dynamics of the human brain.
bioRxiv Preprint Server
April 12, 2021
Peter Coppola, Lennart R.b. Spindler, Andrea I. Luppi et al.
preprint
The diversity of brain dynamics within small-world network topology, measured as sample entropy (dSW-E), consistently predicts levels of awareness across sedation and disorders of consciousness, even after accounting for underlying functional connectivity dynamics. Both subcortical and cortical areas show predictive value, but subcortical regions exhibit higher and more robust effect sizes. The dynamic reorganization of the functional information architecture, especially in the subcortex, emerges with awareness and offers explanatory power beyond the complexity of dynamic functional connectivity alone.
bioRxiv Preprint Server
June 7, 2026
Andrea I. Luppi, Dragana Manasova, Justine Y. Hansen et al.
preprint
Functional connectivity in the awake human brain is shaped primarily by cognitive co-activation—the tendency of brain regions to work together during mental tasks—more than by structural or molecular constraints. This predominance is systematically lost across five datasets involving pharmacological and pathological perturbations of consciousness (chronic disorders of consciousness; anesthesia with sevoflurane, propofol, or ketamine), when cognition is disconnected from the environment or abolished. During such states, the predictors of functional architecture shift away from cognitive co-activation and toward anatomical and molecular constraints.
Brain Communications
January 1, 2026
Dorottya Szocs, Dian Lyu, Andrea I. Luppi et al.
The pulvinar nucleus of the thalamus shows the strongest functional connectivity change with loss of consciousness under anesthesia in healthy volunteers, while the ventral-latero-ventral nucleus shows the strongest change in patients with disorders of consciousness. These nuclei exhibit distinct connectivity patterns with higher-order brain networks such as the default mode and executive control networks. In patients, the neural connectivity biomarker mirrored behavioral changes, suggesting potential clinical relevance for targeted deep brain stimulation therapy.
Neuroimage
February 11, 2023
Andrea I. Luppi, Pedro A. M. Mediano, Fernando E. Rosas et al.
High-level brain functions are thought to arise from coordinated activity across neural systems, but this has been hard to test empirically. Using a framework called Integrated Information Decomposition, which quantifies emergence in dynamical systems, the authors analyzed functional MRI data and found that emergent and hierarchical neural dynamics are significantly reduced in chronically unresponsive patients with severe brain injury. Emergence capacity was positively correlated with hierarchical organization in brain activity. Combining network control theory and whole-brain modeling, the authors show that reduced emergent and hierarchical dynamics in these patients can be explained by disruptions in the structural connectome. The results suggest that chronic unresponsiveness after severe brain injury may stem from structural damage to neural infrastructure needed for emergent brain dynamics.
Communications Biology
September 8, 2022
Angeliki Zarkali, Andrea I. Luppi, Emmanuel A. Stamatakis et al.
People with Parkinson's disease who experience hallucinations spend more time in a brain state where regions are functionally segregated from one another and make fewer transitions between brain states. The shift from an integrated to a segregated state requires less energy in those who hallucinate, making that state potentially preferable. The regional energy needed for this transition correlates with neurotransmitter density and gene expression for serotoninergic, GABAergic, noradrenergic, and cholinergic receptors, but not dopaminergic receptors. The findings suggest that neurochemistry and brain structure together shape the dynamic brain states that underlie hallucinations.