karihoffman.bsky.social
neuroscientist @vanderbilt studying neural circuit+population rules for creating and deploying knowledge. computational neuroethology.
84 posts
2,357 followers
181 following
Regular Contributor
Active Commenter
comment in response to
post
š
comment in response to
post
Many processes from industry are not appropriate or overkill for an academic setting, but we can and should be doing better than we currently are.
Our paper tries to give practical advice on how you can improve your scientific workflow, without being a coding guru.
www.nature.com/articles/s44...
comment in response to
post
Yeah, in fact, those would be welcome! We may yet need to reach out, weāre still stepping through how to implement it.
comment in response to
post
And just imagine, if itās possible to decode categories with LFP/iEEG, as here, what the possibilities are with neuronal ensemblesā¦(stay tuned!)
comment in response to
post
Yes! Psst, donāt forget ripple density increase during active retrieval, that paved the way: doi.org/10.1016/j.cu...
comment in response to
post
Iām sure thereās an only fans for that?
comment in response to
post
Show them: My student and I are trying to implement your Bayesian approach b/c of this post. Itās a very different substrate (not genes) so would not have easily come up in a lit search: bsky.app/profile/cian...
comment in response to
post
Not saying its the right path, just that itās my intuition
comment in response to
post
Iām not sure how sophisticated they are, and thatās a bit of the question? Biology comes up with stunningly simple computations for seemingly sophisticated behaviors? And carving a path seems (maybe deceptively) easier by starting with constrained bio bits.
comment in response to
post
Oh, no offense taken but computational sophistication wasnāt what I was after (and thought you were coming from the same place).
I want to know how the brain works, and with too many degrees of freedom, beginning with its quirky features and constraints can offer a starting point.
comment in response to
post
It seems like it ought to be the other way around. But in practice, how fine grained a level of bio description would be needed? We are still relatively early in being able to proscribe that, to build bottom-up (there are successes but many dead on roadside). Or?
comment in response to
post
Well, not to put anyone on the spot but thatās the response of many cognitive neuroscientists, too!
comment in response to
post
Or, hear me out, you could hop over and hijack your Leeds improvisational theatre production of āWhat Bothers Meā š You canāt script this stuff. www.leedsbeckett.ac.uk/events/perfo...
comment in response to
post
One of the fun parts of this gatekeeping is the fact that we are actually trying to empirically and theoretically test some of the assumptions that could delineate between some of the thoughts that stand between ecological and classical representational "neuro"science without taking any sides.
comment in response to
post
Congrats to the team! I foresee great things š®
comment in response to
post
Another one in the Krakauer et al., 2017 Neuron perspective: www.cell.com/neuron/fullt...
comment in response to
post
One motivation I have heard to cut/destroy #NIH and biomedical research is that AI will just replace it.
But that's folly. The people who say that can't have experience with real discovery science.
comment in response to
post
And also āthe hard part isāā¦discovery - knowing when to scratch that itch when data look funnyš¤. Ie steering off the hypothesis driven course and charting a new or overlooked path.ššŗļøš§
comment in response to
post
3/5) The real mechanism for credibility in science is simply the test of time:
Can someone reproduce the results?
Do others build on the findings?
Real rigour in science comes from waiting to see whether a result holds and leads to new results.
Period - that's it. It's not peer review.
comment in response to
post
Love this. And meeting and talking alone is powerful. 'What made ["righteous gentiles"] courageously agree to hide Jews while the majority went along with tyranny[?] ... Turns out there was one circumstance that determined almost everything ... you had to be asked.'
www.vox.com/future-perfe...
comment in response to
post
Neat- yes, I think Mike had emphasized the thru-line that T cells remain the ātrueā (and scalar) culprits even after recording populations. Both are important aspects of the results to consider.
comment in response to
post
My most disliked like š
comment in response to
post
In 2022 as in 1995! (Your group will know but others may notā¦) www.jneurosci.org/content/15/1...
comment in response to
post
FWIW within the structure of the task and model, I find the results are pretty cool š and make a clear story. Thatās probably why it came to mind.
comment in response to
post
Sure! Responses may be be observed to be lower -d in the brain not because they are IRL, across daily activities, but due to the low-d stereotyped nature of a given lab task. Might T cells do more? Possibly. Might decoding using non T/M cells outpace them if LIP is taxed to ādo moreā? Possibly.