I was attending a webinar with Robin Carhart-Harris on the comparative phenomenology of 5-MeO-DMT and DMT when something struck me. DMT builds worlds. 5-MeO-DMT dissolves them. I found myself wondering what neural mechanisms, or at least what correlates, could explain such a fundamental divergence in experiential architecture. What follows are several hypotheses worth considering.
Under DMT, experiencers report vivid entities, geometries, and spatial relationships that feel fully present and internally consistent, a structured environment that the self navigates and inhabits. Under 5-MeO-DMT, experiencers consistently report the opposite: not a populated world but the dissolution of any boundary between self and environment. How come DMT induces world-building while 5-MeO-DMT, a single methoxy group away, induces world dissolution?
The first place to look is the default mode network, but the relationship is not straightforward. Robust DMN disruption also occurs with LSD and psilocybin, yet both compounds produce rich perceptual content, imagery, narrative, visual elaboration (Nour et al., 2016). Therefore, DMN dysregulation alone cannot be what separates world-building from world-dissolution. What may matter is the degree of that disruption. During the webinar, Carhart-Harris offered an analogy: a jet of water from a hose travels in a tight arc before gravity takes over and it falls. The psychedelic state may follow a similar curve. At moderate levels of DMN disruption, the brain's generative machinery runs freely. With the DMN's suppressive influence on sensory and associative networks loosened, prefrontal and visual association circuits integrate top-down predictions with subcortical signals from hypothalamic and limbic sources, and elaborate them into structured worlds. Push disruption too far and that architecture collapses entirely. DMT, LSD, and psilocybin appear to land on the ascending arc. 5-MeO-DMT may go past the peak. Too little perturbation and nothing opens. Too much and everything dissolves.
Theta oscillations are rhythmic electrical patterns in the 4-8 Hz range that coordinate activity across brain regions during active cognitive and spatial processing. They earn their place in this discussion because of their specific role in spatial cognition. But the picture involves two distinct mechanisms operating in different circuits, not a single unified frequency band.
In the hippocampus, theta rhythms are the oscillatory signature of active navigation, memory encoding, and cognitive map construction, the operations that underlie the experience of inhabiting a structured, navigable space. In visual cortex, a separate mechanism operates: 5-HT2A agonism increases spontaneous and evoked 5-Hz oscillatory bursts in primary visual cortex and retrosplenial cortex, consistent with strengthened top-down cortical dynamics that could drive internally sourced visual content (White et al., 2026). These bursts fall within the theta frequency range but are distinct from sustained hippocampal theta, and the two systems are functionally dissociable, with cortical oscillations emerging independently of hippocampal drive. Separating their contributions in human EEG recordings is difficult given the limits of scalp-level source localization, and intracranial or high-density approaches will be needed to resolve this properly.
The two tryptamines pull these systems in opposite directions with respect to theta activity. Under DMT, Timmermann et al. (2019) found a transient increase in theta activity precisely at the moment of peak visual intensity, consistent with the idea that world-building depends on this oscillatory state. Given the limits of scalp EEG, this signal likely reflects contributions from both hippocampal and cortical sources. Under 5-MeO-DMT, rodent studies show a shift toward sleep-like spectral signatures in hippocampus and prefrontal cortex, consistent with collapse of the active processing that hippocampal theta supports. When both systems go offline, there is no spatial map and no visual content to populate it with. The conditions for world-building are absent at both levels simultaneously.
Receptor pharmacology may explain why 5-MeO-DMT drives the system toward that collapse. Both compounds are potent 5-HT2A agonists, so shared 2A activity cannot by itself explain their divergence. The key difference is that 5-MeO-DMT carries substantially higher 5-HT1A affinity relative to its 2A activity. These two receptors pull in opposite directions on relevant circuits: 5-HT2A activation in visual and association cortices drives the elaborative, content-generating dynamics described above, while 5-HT1A is inhibitory, with strong subcortical and raphe expression, and serotonin acting at 1A receptors is known to suppress hippocampal theta. Under DMT, the 2A signal runs with relatively little 1A opposition, and hippocampal theta is transiently sustained at the visual peak. Under 5-MeO-DMT, the additional 1A load may simultaneously brake cortical elaboration and suppress the hippocampal spatial scaffolding on which world-building depends, pushing the system past the threshold where content generation is possible at the receptor level, before those downstream network dynamics can unfold.
Carhart-Harris's REBUS framework maps onto this picture in a single sentence: DMT may free the brain's generative model enough to construct vivid worlds from internal signals, while 5-MeO-DMT may overwhelm that model entirely, leaving no architecture to build from. The pharmacological, oscillatory, and network-level mechanisms described here are likely all contributing to which side of that threshold a compound lands on, and together they suggest that world-building is not a default property of the psychedelic state, but a condition that requires the right perturbation, at the right level, in the right circuits.
References
- Nour, M.M., Evans, L., Nutt, D., & Carhart-Harris, R.L. (2016). Ego-dissolution and psychedelics: Validation of the Ego-Dissolution Inventory (EDI). Frontiers in Human Neuroscience, 10, 269.
- Timmermann, C., Roseman, L., Schartner, M., Milliere, R., Williams, L.T.J., Erritzoe, D., Muthukumaraswamy, S., Ashton, M., Bendrioua, A., Kaur, O., Turton, S., Nour, M.M., Day, C.M., Leech, R., Nutt, D.J., & Carhart-Harris, R.L. (2019). Neural correlates of the DMT experience assessed with multivariate EEG. Scientific Reports, 9, 16324.
- White, C.M., Azimi, Z., Staadt, R., Song, C., Knöpfel, T., & Jancke, D. (2026). Psychedelic 5-HT2A agonist increases spontaneous and evoked 5-Hz oscillations in visual and retrosplenial cortex. Communications Biology, 9, 52.
