Machine Learning Algorithm Maps Psychedelic Effects to Specific Regions of the Brain

Image credit: Milad Fakurian via Unsplash
Scientists are beginning to discover which parts of the brain are responsible for certain hallucinogenic sensations, with a little help from artificial intelligence.
Published earlier this month in the journal Science Advances, a new study used a pattern-learning algorithm and individual drug use testimonials to identify the brain regions that are likely responsible for the visual hallucinations, ego dissolution, and altered time perceptions experienced during a psychedelic trip.
Identifying the parts of the brain that cause these feelings is crucial, the researchers say, as this could lead to the development of new drugs that are specifically designed to target these areas and provide benefits to those who suffer from complex mental health disorders.
What can AI teach us about the human brain?
Using artificial intelligence to unravel the mysteries of the human brain might seem odd at first, but the pattern-recognition capabilities provided by these algorithms can be a powerful tool for researchers.
In this new study, the researchers examined testimonials collected by Erowid Center, a not-for-profit organization that operates a library of first-hand accounts of the experiences that were elicited by psychoactive drugs. This library contains 6,850 testimonials relating to 27 different psychedelic compounds, including MDMA, LSD, and ketamine.
From these, the researchers identified key words associated with each drug experience. Employing a 3D map of the brain generated using thousands of gene transcription probes, they were then able to link these subjective experiences to the brain regions where relevant receptors are commonly found.
“We transformed these testimonials into what’s called a bag-of-words model, which breaks down a given text into individual words and counts how many times each word appears,” two of the study authors, Galen Ballentine, MD, and Sam Friedman, PhD, wrote in an article for the Conversation.
“We then paired the most commonly used words linked to each psychedelic with receptors in the brain that are known to bind to each drug. After using an algorithm to extract the most common subjective experiences associated with these word-receptor pairs, we mapped these experiences onto different brain regions by matching them to the types of receptors present in each area.”
Serotonin and adrenergic receptors could explain ego dissolution
This study identified new links between psychedelic experiences and the brain, as well as confirming other theories already present in the scientific literature. For example, the 5-HT2A serotonin receptor is known to be heavily involved in mediating the effects of psychedelics. This study found that the feeling of ego dissolution – a commonly reported sensation during psychedelic trips where the user loses much of their sense of self – was predominantly associated with the 5-HT2A receptor. However, other serotonin receptors (5-HT2C, 5-HT1A, 5-HT2B), as well as the D2 dopamine receptor and the adrenergic receptors Alpha-2A and Beta-2, were also linked to this feeling.
The 5-HT2A receptor, which is widely distributed throughout the central nervous system, was additionally associated with many of the changes in sensory perception reported by users. This could be due to the presence of the receptor in the visual cortex of the brain, the researchers explained, where it binds to a molecule that helps to regulate mood.
“Feelings of transcendence were connected to dopamine and opioid receptors in the salience network, a collection of brain regions involved in managing sensory and emotional input,” the authors wrote. “Auditory hallucinations were linked to a number of receptors spread throughout the auditory cortex.”
The researchers found that many of these key receptor combinations were located in the deepest layers of the brain’s information processing layers. This aligns with the current leading theory explaining psychedelics’ action in the brain, which states that psychedelics may temporarily relax the normal high-level executive functions and thereby liberate a more ‘bottom-up information flow’.
Untangling the effects of psychedelics could lead to new medicines
With the growing interest in psychedelic-assisted therapy clinics, scientists are eager to learn more about the neurobiological mechanisms at play when psychedelics are ingested.
“Hallucinogenic drugs may very well turn out to be the next big thing to improve clinical care of major mental health conditions,” said Professor Danilo Bzdok, the study’s lead author, in a statement.
“Our study provides a first step, a proof of principle that we may be able to build machine learning systems in the future that can accurately predict which neurotransmitter receptor combinations need to be stimulated to induce a specific state of conscious experience in a given person,” Bzdok added.
Creating such a machine learning system could lead to the rapid development of new drugs, ones specifically designed to activate the receptors in the brain that would promote the most potent therapeutic effects.