biased because I just saw Diego Bohórquez give a very impressive talk but any of his work on gut-brain connections (e.g. www.science.org/doi/full/10....)

for broader terms like "intelligence", I'm not sure. it's why I avoid these sorts of terms completely. it probably changes depending on the field, but if these terms emerged out of human-focused work then there will always be people that don't think they apply to non-human processes

I would say yes... based on the logic that what we want to understand is what humans do, so we come up with a term to describe a process, then we come up with measures that capture that process (imperfectly). I think it matters when using that term that we are describing the same process.

I think you can get at different aspects of the process by looking at neural activity vs. behavioral paradigms. But I think something could do well at certain episodic memory tests with a process that varies significantly from what people that study it in humans mean when they use the term.

I guess I use episodic memory as an example because I believe the tests are a particularly lazy proxy of the process that are good at picking up major dysfunction but miss a lot else (which is why many people studying episodic memory as a process don't use these basic tests in their work).

I think we have a range of measures of outcomes and a range of measures of process, and some combination of those (depending who you ask) would comprise episodic memory. I do think that the process is part of our definition of these terms (although we've often forgotten this in human research too).

I sort of agree, but is the operationalization of a concept all it is scientifically? I wouldn't use "intelligence" when studying humans or AI (too broad and poorly defined), but if AI can do episodic memory tasks, does it have episodic memory? I would say no...

Too much whiskey and rye

I think it was February, I remember it made me shiver

Music literally died that year

Really interesting work! How do you think this relates to "events" (rare, high fluctuation time-points) in fMRI? I know there was a back-and-forth debate a few years ago about what types of signals were driving fMRI FC:
www.pnas.org/doi/full/10.... www.sciencedirect.com/science/arti...

@inmice.bsky.social

Just continuing the long-standing scientific tradition of prioritizing cool-sounding names for things over clear science communication 😅

Something like this? Source: jtr13.github.io/cc21fall2/ra... but then it doesn't look like a raincloud... (original raincloud plot source pmc.ncbi.nlm.nih.gov/articles/PMC...)

This is a ready-made advert for raincloud plots