We invite You to participate in a study that explores the capability of artificial neural networks to model psychedelic experiences. More precisely, we explore how well the inner activity of computer vision models captures the (closed- or open-eye) visuals experienced under the influence of classical psychedelic substances, such as DMT, LSD, mescaline, or psilocybin.

We invite everyone who has at least one experience with a classical psychedelic to participate in the study! Filling out the questionnaire takes about 20 minutes, no personally identifiable information (PII) is gathered.

You can find the survey here: https://forms.gle/eQ9VBbhCubJmbe9R6.

The study is conducted in the Computer Science and Psychology Institutes of University of Tartu, Estonia.

Thanks in advance! Every participant is greatly appreciated.

Leave a comment or write to me, if you're interested to hear about the results as well.

This one seems to have slipped beneath the radar, at least in terms of direct responses here.

Have any Nexus members taken this survey?

naqkvay: the survey is apparently still active - how long will this remain the case? When do you anticipate being able to publish the results, and what do you envisage doing with them subsequently? Do you think there are other aspects of the psychedelic experience that can be modelled? If not, why not?

And finally, what attracted you to the topic in the first place? How strong would you say the presence of psychedelics is in Estonian culture? What are the prevalent attitudes towards them in Estonia?

Hope you see this reply, we'd be fascinated to hear more - thanks for sharing, and (a rather belated) welcome to the Nexus!

Aww, looks like I'm too late to participate. The survey is closed. I'd love to see the results, though. Please share them here when convenient.

In my limited experience, machine-generated images sometimes come close to visuals caused by LSD, 4-Aco-DMT and especially ketamine (not a classic psychedelic, but the one most closely approximated by these images IMO) but I have never seen an artificially generated image that comes close to my DMT or mushroom visuals.

I also second downwardsfromzero's questions, especially about other aspects of the psychedelic experience that can be modeled. I've never done DiPT, which is allegedly the auditory equivalent of other classical psychedelics' visual hallucinations, but there are auditory distortions inherent to other psychedelics which I feel are well simulated by certain kinds of electronic music. Aspects such as the headspace are of course a lot more difficult to reproduce directly in any physical sort of medium.

What have you learned from this survey, and did the results surprise you in any way? Were replies more or less uniform than you expected them to be?