We developed a generative AI framework that learns to simulate impaired consciousness from massive datasets of neural activity, revealing hidden mechanisms of disorders of consciousness. It predicted selective damage to the basal ganglia indirect pathway, abnormal inhibitory cortical wiring and promising treatments, which were confirmed in patient tissue, brain scans and clinical data.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 print issues and online access
$259.00 per year
only $21.58 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to the full article PDF.
USD 39.95
Prices may be subject to local taxes which are calculated during checkout
References
Thibaut, A., Schiff, N., Giacino, J., Laureys, S. & Gosseries, O. Therapeutic interventions in patients with prolonged disorders of consciousness. Lancet Neurol. 18, 600–614 (2019). This review article presents the current landscape of therapeutic interventions for disorders of consciousness while highlighting the ongoing scarcity of established treatment options.
Schiff, N. D. Recovery of consciousness after brain injury: a mesocircuit hypothesis. Trends Neurosci. 33, 1–9 (2010). This paper shows presents a seminal ‘mesocircuit’ hypothesis of anterior forebrain contributions to the collapse and recovery of consciousness following severe brain injury.
Toker, D. et al. Criticality supports cross-frequency cortical-thalamic information transfer during conscious states. eLife 13, 86547 (2024). Our previous paper used genetic optimization to tune the same model of the basal ganglia–thalamocortical system to simulate anaesthetized states.
Albada, S. J. & Robinson, P. A. Mean-field modeling of the basal ganglia-thalamocortical system. I: Firing rates in healthy and Parkinsonian states. J. Theor. Biol. 257, 642–663 (2009). This paper describes the brain-wide mean-field model that served as a generator in our adversarial optimization.
Selten, M. et al. Regulation of PV interneuron plasticity by neuropeptide-encoding genes. Nature 643, 173–181 (2025). This paper demonstrates that upregulation of Vgf and Scg2 drives synapse formation between cortical parvalbumin-expressing interneurons.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This is a summary of: Toker, D. et al. Adversarial AI reveals mechanisms and treatments for disorders of consciousness. Nat. Neurosci. https://doi.org/10.1038/s41593-026-02220-4 (2026).
Rights and permissions
About this article
Cite this article
AI predicts how brain injuries disrupt consciousness — and how to restore it once it’s disordered. Nat Neurosci (2026). https://doi.org/10.1038/s41593-026-02223-1
Published:
Version of record:
DOI: https://doi.org/10.1038/s41593-026-02223-1
