Neuroscientists have recreated the state of psychedelic trips in deep neural network algorithms to understand how human brains process drugs
AI has been used to demonstrate how psychedelic drugs like LSD, DMT and psilocybin change the functions of serotonin receptors in human nervous systems.
The study was conducted by Michael Schartner, a member of the International Brain Laboratory at Champalimaud Centre for the Unknown in Lisbon and his colleague, Christopher Timmermann from Imperial College London.
Schartner’s previous work found that powerful psychoactive substances produced a non-stoppable increase in neural signal diversity while Timmermann’s research showed LSD reduced the neural response to sudden unfamiliar stimuli but expanded for familiar stimuli. Both findings shed light on human brain dynamics in altered states of consciousness.
“The ventral visual stream in human brains seems key for visual experiences but is certainly not sufficient,” said Schartner. “Also, the exact role of serotonin in the gating of sensory information is still to be explained. Another big open question is how exactly the feedback and feed-forward flows of neural activity need to be arranged to bring about any experience.”
Schartner and Timmermann embarked on an unusual plan of “feeding” virtual versions of psychedelic substances to neural network algorithms and observing what happened. The “feeding” process involved a continuous introduction of noise to the image-generating neural networks so that their outputs became extremely distorted or blurred.
Once these algorithms got ‘high’, they would model the neural basis of psychedelic trips on how people would describe themselves having their DMT trips. This provides Schartner and Timmermann with a better channel to study how human brains process drugs.
“Deep neural networks – the workhorse of many impressive engineering feats of machine learning – are the state-of-the-art model for parts of the visual system in humans,” Schartner added. “They can help illustrate how psychedelics perturb perception and can be used to guide hypotheses on how sensory information is prevented from updating the brain’s model of the world”.
Schartner and Timmermann believe the structural similarities between neural networks and human visual cortices offer them a new way to understand the impact of psychedelics. Ultimately, they hope their research findings will lead to promising results in the treatment of depression and anxiety.
The full study can be read, here, in the Neuroscience of Consciousness