Psychedelic neuroscience is currently shaped by two influential frameworks that appear, at first glance, to contradict each other.
On one side, the REBUS model proposes that psychedelics acutely relax high-level priors by weakening the tonic top-down control exerted by the default mode network (DMN). Within minutes to hours after psilocybin intake, network coherence decreases, hierarchical precision is reduced, and cognition becomes more entropic (Carhart-Harris & Friston, 2019; Carhart-Harris et al., 2012). This framework has been particularly effective at explaining acute phenomenology and transient increases in cognitive flexibility.
On the other side, recent circuit-level work by Jiang et al. (2025) complicates a purely network-centric view. Examining cortical structure ~24 hours after psilocybin exposure, long after subjective effects have resolved, they show that plasticity is not uniform across neuron types. Pyramidal tract (PT) neurons exhibit robust spine growth and stabilization, whereas intratelencephalic (IT) neurons show spine loss and pruning (Jiang et al., Cell, 2025). Plasticity is selective, directional, and projection-specific.
How do these views contradict each other?
“REBUS explains how pathological hierarchies are interrupted. Neuron-specific plasticity explains how new ones are installed.”
REBUS describes the acute destabilization phase of psychedelic action, when the drug directly perturbs network dynamics. Jiang et al.’s data capture the post-acute consolidation phase, when pharmacological effects have faded, and structural plasticity is being stabilized. These are not competing explanations of the same phenomenon, but sequential stages of a single process.
Crucially, the DMN is not a neuron type but a network state. Its transient loss of precision opens a window in which entrenched attractor states can be disrupted. What the Jiang data show is that the cortex does not simply return to baseline once this window closes. Instead, it selectively reinforces specific projection classes. PT neurons, positioned to exert broad influence over subcortical targets and behavioral output, are strengthened, while IT neurons, associated with dense intracortical and self-referential processing, are pruned.
This reframes REBUS. Psychedelics do not abolish control: they help redistribute it. Therapeutic benefit likely arises not from sustained entropy, but from a controlled destabilization followed by selective rebuilding.
This distinction also clarifies diagnostic boundaries. In depression, characterized by excessive rigidity, transient DMN loosening followed by adaptive re-anchoring may be beneficial (Sheline Y., 2010). In schizophrenia, where constraint and synchrony are already weakened (Uhlhaas & Singer, 2010), further destabilization may be maladaptive, suggesting that strengthening, rather than relaxing, high-level control is required.
REBUS explains how pathological hierarchies are interrupted. Neuron-specific plasticity explains how new ones are installed.
REBUS and Neuron-Specific Plasticity do not cancel each other. They complete each other.
References
- Carhart-Harris RL & Friston KJ. REBUS and the Anarchic Brain: Toward a Unified Model of the Brain Action of Psychedelics. Pharmacol Rev (2019).
- Carhart-Harris RL et al. Neural correlates of the psychedelic state as determined by fMRI studies with psilocybin. PNAS (2012).
- Jiang Q. et al. Psilocybin induces cell-type-specific cortical plasticity. Cell (2025).
- Sheline Y. Resting-state functional MRI in depression unmasks increased connectivity between networks via the dorsal nexus. PNAS (2010).
- Uhlhaas PJ & Singer W. Abnormal neural oscillations and synchrony in schizophrenia. Nat Rev Neurosci (2010).