Post

Torben Ott
@torbenottlab.org

2/ We found a striking behavioral effect: psilocybin reduced change-detection performance to chance, without causing obvious motor deficits. At the same time, visual stimuli drove a strong 4-Hz oscillation across visual cortex, especially in VISp and SST interneurons.

Torben Ott
@torbenottlab.org

3/ Surprisingly, despite this oscillation, image id encoding was completely intact! By contrast, this oscillation was closely linked to CHANGE encoding, pushing visual cortex dynamics in SST interneurons towards states usually associated with surprise.

1 repost4 likes

Torben Ott
@torbenottlab.org

4/ Unexpectedly, psilocybin's effect were not global or uniform but dependent on the image structure: Only images with long, continuous contours recruited change-encoding neurons that were modulated by psilocybin.

Torben Ott
@torbenottlab.org

5/ Our findings point to a previously unrecognized role for SST interneurons in the acute psychedelic state, potentially linking altered surprise signals to visual distortions. A huge thank-you to the entire OpenScope team, especially @jeromelecoq.bsky.social, for making this work possible!

翼/Tsubasa
@tsubasa-rsrch.bsky.social

Interesting work. A question: the change-vs-repeat decoder was trained pre-injection and tested post. If retrained within the post-psilocybin epoch, is change still decodable? That would separate "information lost" from "code shifted, so the old readout fails". I didn't find it in the text.