How Machine Learning can Predict the Outcomes of the Psychedelic Experiences

Humanity is going through a Psychedelic Renaissance.

Psychopharmacology is radically evolving with numerous research and studies and adopting a more holistic approach to treat mental diseases, including depression, anxiety, PTSD, and substance abuse. Like Psychedelic-Assisted Psychotherapy, many other methods are now showing scientific research-based promises to treat mental diseases, including those that are typically treatment-resistant against the traditional pharmaceutical modality.

After the first experiment with LSD, researchers discovered the potentiality of treating addiction with Psychedelic substances; numerous researches have been conducted from various scientific fields which showed that along with many other Psychoactive Holy Plants, other synthetic Psychedelic substances also have the potentialities to cure addiction. However, while there is increasing evidence that Psychedelic substances are useful in understanding and curing addiction, scientists are still not sure why Psychedelic substances are effective.

The Psychedelic experience is subjective; even now, with the latest research technologies, objectively measuring subjective Psychedelic experiences during a Psychedelic session is difficult. Interestingly subjective experience of the Psychedelic substance itself is partly responsible for its anti-addictive properties. Studies have shown that if someone undergoes a mystical experience during a Psychedelic session, they are more likely to overcome addiction.

To rectify this, David Cox from Behavioral Pharmacology Research Unit, Department of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, USA, designed a research methodology with his research team using a machine learning algorithm. They demonstrated that purely based on the written report of participants’ Psychedelic experiences; the machine learning algorithm could significantly predict whether a participant could quit or reduce their substance use.

Research method

David Cox and his research team recruited 1141 individuals; among them, 247 were female, and 894 were male. Recruited individuals were already suffering from alcohol, opioids, cannabis, or other stimulants addiction. Following a Psychedelic experience, they provided a verbal narrative of the Psychedelic experience they attributed as leading to their reduction in other substance abuse. The research team used Natural Language Processing (NLP) to derive topic models that quantitatively described each participant’s Psychedelic experience narrative. Then they used the vector descriptions of each participant’s Psychedelic experience narrative as input into three different supervised machine learning algorithms to predict long-term drug reduction outcomes.

These machine learning algorithms attempted to differentiate whether an individual participant quit their addiction or simply reduced it. Cox and his colleagues found that all the algorithms were able to successfully predict whether a participant quit or reduced their substance much higher than it would be possible by chance.

According to the researchers, this study suggests that NLP of written Psychedelic session narrative and the subsequent use of supervised machine learning algorithms are analytic tools that can help future clinicians and researchers to identify the level of support each person who is suffering from addiction may need during Psychedelic-Assisted Psychotherapy for substance abuse.

Conclusion

Recently this year, another group of scientists also experimented with Artificial Intelligence (AI) Simulation to produce hallucinogenic patterns of DMT, and this research is now utilising machine learning algorithms, which is potentially paving the way for further future Psychedelic research to intersect at the crossroads of artificial intelligence. Using these modern technologies, Psychedelic researchers may have a more objective way of measuring the subjective experience of Psychedelic substances.

References

Cox, D. J., et al. (2021). Predicting changes in substance use following psychedelic experiences: natural language processing of psychedelic session narratives. The American Journal of Drug and Alcohol Abuse, [online] Available at: https://doi.org/10.1080/00952990.2021.1910830 [Accessed 16th August 2021].

Krebs, T. S., et al. (2002). Lysergic acid diethylamide (LSD) for alcoholism: a meta-analysis of randomized controlled trials. Psychopharmacology,[online] Volume, 26(7), p. 994-1002. Available at: https://doi.org/10.1177/0269881112439253 [Accessed 16th August 2021].