A pattern recognition algorithm scoured 6850 accounts of people’s experiences with 27 drugs to learn more about how they alter consciousness
16 March 2022
Artificial intelligence has been used to analyse thousands of written reports of personal experiences with psychoactive drugs to gain a better understanding of their subjective effects and how they work in the brain.
Psychedelic drugs such as LSD, ketamine and psilocybin – the active compound in magic mushrooms – are being investigated as treatments for a range of conditions, including depression, addiction and post-traumatic stress disorder. The experiences they induce, which may be important for their therapeutic effects, are highly variable, and can include visual and auditory hallucinations, an altered sense of self and a distorted perception of time.
Danilo Bzdok at McGill University in Montreal, Canada, and his colleagues used a pattern-recognition algorithm to scour 6850 accounts of experiences submitted on the website Erowid, involving 27 different drugs.
They linked words used in the accounts for each drug, such as “euphoria”, “nausea” or “visuals”, with any of 40 receptors in the brain that the drug is known to interact with, and mapped drug effects onto areas of the brain where these receptors are most active.
The researchers, who weren’t available for interview, hope their work will help identify drugs that may induce particular subjective effects, and provide a framework for developing new treatments based on psychedelic drugs in the future.
Daniel Barron at Brigham and Women’s Hospital in Boston, who wasn’t involved in this study, says that while this tool isn’t yet ready for clinical use, it shows promise. “The core idea is that if you can determine that a given drug and brain function have a predictable relationship, this paves the way to understanding whether that change and therefore that drug is clinically useful,” he says.
“The cultural history suggests there’s a lot of potential tied up in these psychoactive drugs, but any clinical application will require that the right drug be prescribed at the right dose to the right patient at the right time,” he says.
Journal reference: Science Advances, DOI: 10.1126/sciadv.abl6989
More on these topics: