Many thanks to all coauthors for this great collaboration, and special thanks to Mona Irrmischer, Klaus Linkenkaer-Hansen, Lisa Luan, @christimmermann.bsky.social
@robincarhartharris.bsky.social
Our results thus invite the continued investigation of multi-dimensional consciousness and criticality across brain rhythms, particularly when characterizing the effects of psychedelics on brain dynamics.
Altered states may be distinct in some dimensions (e.g. phenomenal content) and alike in others (e.g. disruption of self-referential processing) due to different alterations of criticality across brain rhythms.
...of psychedelics with distinct phenomenal properties (e.g. DMT and 5-MeO-DMT), as well as other altered states such as deep meditation or anesthesia.
Moving beyond one-dimensional characterizations of consciousness and criticality, disentangling relations between multiple dimensions of subjective experience and criticality in different brain rhythms provides new potential hints on the neural correlates...
For example, frequencies involved in self-referential processing may become subcritical, as we observe here, while higher frequencies encoding intense phenomenal content may indeed be shifted towards critical or even supercritical regimes.
But here are the keys:
1) Different brain rhythms are likely to be involved in different brain functions.
2) Psychedelics may have different effects on criticality in different brain rhythms.
Shifts away from criticality, and particularly in the subcritical direction, may appear surprising since psychedelics have been hypothesised to bring the brain closer to criticality, or even push it beyond into supercritical regimes.
We found that DMT shifted the dynamics of normally dominant alpha oscillations away from criticality, and towards a more quiet subcritical state. Notably, these shifts away from criticality significantly correlated with intensity of disruptions to the sense of self.
Brain oscillations are thought to be proximal to criticality, a state characterized by high complexity and fluctuations occurring over many spatiotemporal scales. Here, we looked at how DMT changes the criticality of oscillations in different brain rhythms (i.e. temporal frequencies).
Great start of the year with a fantastic collaboration published in the Journal of Neuroscience! www.jneurosci.org/content/46/2...
Thank you Tom!
Special thanks to you @sergedumoulin.net for allowing this to happen!! As supervisor it would have been much easier for you to say 'no' to such a complex and potentially risky study. But you had the evidence-based open mindedness and courage to go through with it. I am very grateful for that 🙏
Thank you Evan!
Thank you Anirudh!
Thanks Tomas! Couldn't have done it without you :D
From founding the Amsterdam Psychedelic Research Association (APRA) in 2017 to leading the first-ever neuroimaging study with a classic psychedelic in Amsterdam, this paper represents the closing of a circle and truly means a lot to me. I hope you find it interesting :)
Thank you for reading! I am grateful to all coauthors (Gilles de Hollander, Nina Vreugdenhil, Tomas Knapen, Serge Dumoulin) and study participants, without whom none of this would have been possible.
Indeed, we speculate that the butterfly-effect cascading from such a relatively simple, but profound and pervasive computational change, may explain the broad variety and apparent paradoxicality of psychedelic effects.
Now, imagine what could happen to your subjective experience if a fundamental computation for contextual information-processing was altered throughout many, recurrently interacting levels of the brain's functional hierarchy, from visual, to temporal, to semantic processing...
Divisive normalization is not only present in vision; rather, it is considered a 'canonical neural computation' implementing contextual processing in a variety of sensory and cognitive domains. nature.com/articles/nrn...
In other words, we found that psilocybin altered a fundamental computational building block of contextual processing in the human brain: divisive normalization, particularly in the form of visual surround suppression.
Enter the divisive normalization population receptive field model. The parameter governing surround suppression systematically dropped with psilocybin! This could not be explained by noise or HRF changes alone, which are also included in the model. 🤖
This hinted at a potentially deep computational shift, occurring already at low levels of the brain functional hierarchy. So how can we capture it and make sure it’s not all noise or blood-flow changes?
To find out, we scanned participants with 7T fMRI while showing sweeping high-contrast bars. Key result: psilocybin systematically reduced surround suppression, a key mechanism to optimize contextual neural computations, in early visual areas! 🧠
This wasn’t because participants got worse at the task: perceptual noise was unchanged. The effect of psilocybin was specific to contextual perception! So what’s happening in the brain?
First, we looked at visual perception using the Ebbinghaus illusion, a classic example of how context shapes what we perceive. The contextual surround illusion got systematically stronger! It's not just a figure of speech: we literally see the world differently with psilocybin👁️
Our new paper is out in Nature Communications! nature.com/articles/s41...
We combined psychophysics, 7T fMRI, and computational modeling of vision with placebo, 5mg, and 10mg psilocybin, in the same group of participants, to clarify the computational mechanisms of psychedelics. 🧵
very much enjoyed being involved in this thought-provoking work led by @htmcgovern.bsky.social with @robincarhartharris.bsky.social and Selen Atasoy :)
A couple months ago I gave this talk for the QRI meetup in Amsterdam. It was a fun opportunity to make an overview of different research topics I've been working on in the past few years, including some (yet) unpublished new results. Hope you enjoy :)