It's amazing that one can make a whole career of only spouting pessimistic interpretations a field that is clearly advancing to amazing capabilities.
@GaryMarcus
criticizing is a lot easier than coming up with something better. I urge you to get out of your comfort zone and say
๐ค The news is the price and speed, not radically new capabilities
Given the cutoff date, and what Altman said when we were at Senate in May โฆ my gut tells me that GPT-4 Turbo is a relabeling of what was supposed to be called GPT-5, and not enough a cognitive breakthrough merit
Here I was browsing a new (for me) site of human stories (
@Narratively
) when I saw my dear friend and superwoman
@Ocasio2018
who's gonna kick ass in NY-14's primary in a couple of weeks. Yeehaw!
I wrote about a dilemma I've been struggling with: loving animals while confining and killing them in the name of basic
#scientific
#research
, in which utility of outcomes is unknown ahead of time. Hope this
#scicomm
can start a productive conversation...
Just had a 2-hour lab meeting w/
@Datta_Lab
about systemic racism in America, science, and
@Harvard
. It's heartwarming to see support and leadership from a scientist like Bob but also terrifying that he's one of the few PIs I've heard...
My first pre-print is up! We made 64-channel carbon fiber electrode arrays for recording neural activity. Carbon fibers are a nice alternative to microwires (e.g. tungsten or nichrome) because of their small cross-section (4.5 micron diameter!) and flexibility.
Excited to share that Iโve started a new role
@FlagshipPioneer
focusing on venture creation in AI+biology!
Black turtleneck b/c Boston is cold, not from illusions of grandeur!
Iโm a doctor! Itโs been a long seven years but how else could I have gotten this much whiskey? I couldnโt have done it without the patient support of friends and family!
My first pre-print is up! We made 64-channel carbon fiber electrode arrays for recording neural activity. Carbon fibers are a nice alternative to microwires (e.g. tungsten or nichrome) because of their small cross-section (4.5 micron diameter!) and flexibility.
Anyone interested in brains and behavior checkout this release of MoSeq. It's a powerful tool to understand behavior and now comes with beautiful code.
We are excited to announce the release of a plug-and-play version of Motion Sequencing (MoSeq), a tool for characterizing mouse behavior (see new website here: )
I love birds, esp corvids. Neuroscientists should def study corvids. I respect Nieder for all the work he's done on crows. The framing of this--that neural activity in crows demonstrates that they have consciousness--is just bonkers. Someone explain this?
@leslievosshall
@HHMINEWS
This sounds like a great start. I canโt help but imagine that this is another โthe rich will get richerโ scenario in which HHMI labs will attract better talent than they already do, and non-HHMI labs and postdocs will be left in the dust. The NIH needs to adopt similar standards
I had the pleasure to voice my opinion about Natureโs new open access policy, which is incredibly expensive and will likely lead to further inequalities in science. [a thread...]
Opinion | By levying an article processing charge in excess of $11,000, argues Grigori Guitchounts, Springer Nature promises to do irreparable harm to the already crippled scientific publishing system.
I tell my fellow graduate students that if theyโre tweeting less than 60 hours per week, theyโre tweeting less than the full professors, and are therefore not real scientists.
I tell my graduate students and post-docs that if theyโre working 60 hours per week, theyโre working less than the full professors, and less than their peers.
Hereโs prime evidence that low pay, poor TT job prospects, idiotic incentive structure, etc, is leading to a massive brain drain in academia. And weโve been wondering why academia is full of toxic assholes who make arguments like these:
Anyone who has been a young PI starting a new lab with postdoc hires that are coming from a wide variety of backgrounds with limited experience in YOUR field knows EXACTLY why they are called โtraineesโ and why they are not payed six figure salaries.
Hey
#cosyne2022
, if youโre into predictive coding, embodied cognition, visual cortex, movement, and neural networks, come check out my talk this afternoon at 3:30pm in Sala XVI!
So inspired after meeting with Alexandria Ocasio-Cortez. On June 26th we have a chance to prove that real Democrats who only answer to their constituents, can win against Dems beholden to corporate money. If you believe in Democracy, learn more & donate at
Vision, too, is an active sense. Animals make all sorts of movements that affect neural dynamics in visual areas, but we rarely study vision in freely moving animals...
โGet a life,โ thinks the seventh-year graduate student as he passes a group of Pokรฉmon Go players on his way back to lab with his takeout dinner, past 8pm ๐ค
#ohtheirony
#phdchat
Do you like vision? Do you like movements? Do you like movement signals in visual cortex? If so, come see my talk at
#cosyne19
at 12:45 pm today!
#science
I'm super excited about this work because it changed the way I think of vision. While vision is normally studied as a passive sense, IRL animals interact with their sensory environments. The sense of touch involves whisking and palpation; olfaction involves sniffing; etc...
@guitchounts
holding it down for PiN at
#cosyne19
! Predicting head movements in rats from V1 activity โall while rocking pineapples in style, nicely done!
This is absolutely insane. The academic publishing system is broken. Predatory publishers are relying on taxpayer money and free reviewer labor to beef up their pockets. When will we rid ourselves of this cancer?
Had a fantastic time at the
#MCNWorkshop18
@HHMIJanelia
!! Many thanks to Vivek, Hannah, Misha, Alla, Josh, and Gowan for organizing and provoking amazing discussions! It was great to spot some spooky corvids @ Bobโs!
@DPankaew
This would be killer if it also explained the paper and went into details on the background and what the results mean. Something like the
@hubermanlab
podcast (not that we should replace Huberman with AI just yet)
@GaryMarcus
Totally agree we shouldn't overhype. but I think the issue is that predictions alone are meaningless. If you can build or even propose an alternative system, that would be amazing, even if it doesn't work. You can make a prediction that e.g. embodied agents or predictive coding
Do you like visual cortex? Want to hear how you can decode head direction and movement direction in V1 of freely moving rats? Wake up at the crack of dawn and come to my talk Monday at 8:15am!
#SfN18
#SFN18
Where do these signals come from? Many studies pointed to a secondary motor cortex (M2), which projects to V1 directly. So we lesioned M2 and... presto all the direction signals disappeared...
Olfactory sensory neurons (OSNs) use transcription to make flexible predictions about the environment and dynamically change odor responses, now in our new paper led by the amazing
@TatsuyaTsuka0
and
@davidhbrann
! Thread about what we found ๐1/n
Individual cells in V1 were tuned to either to the direction of head motion or direction of visual flow. We think these signals allow the visual cortex to distinguish between self-generated sensory signals and those that come from the outside world
Re:
@nature
's exorbitant
#OpenAccess
fees: in interviews w/ low/middle-income-country researchers,
@paimadhu
notes in
@Forbes
the $11k fee is >1 year's salary in many parts of the world. Only the wealthiest researchers will afford
#OA
in glam journals.
Opinion | By levying an article processing charge in excess of $11,000, argues Grigori Guitchounts, Springer Nature promises to do irreparable harm to the already crippled scientific publishing system.
...talking about critical issues like this and showing genuine care for his employees. Talking about problems is a small but necessary thing to do and I hope other lab heads can initiate conversions in their own groups
After four eternal years, what a huge relief. Proud and happy for
@JoeBiden
and
@KamalaHarris
! Now, lots still left to do to heal and rehabilitate our suffering and broken country!
[only a year later, but]
#cosyne19
talks are out! Check me out here talking about head direction and orienting signals in the visual cortex of rats! This work has evolved quite a bit since last year! Read the latest here:
Iโm having a blast on this terrible day reading neuroscientists react to this. The fact that we donโt know much about the brain isnโt that surprising, is it? Like what, you guys think itโs pretty much solved? โMission Accomplishedโ?
A prior track record of exceptional achievement in engineering *is* required, but *no* prior experience working on the brain or human physiology is required. We will teach you whatโs known about the brain, which is not much tbh.
So we recorded V1 neural activity and movement in freely moving rats, and found that unlike locomotion, which increases V1 activity, orienting movements of the head (HOMs, if you will) actually suppressed V1 on really fast timescales...
Ever wonder what happens in your brain when youโre in nature? Follow my attempts to answer this question through evolutionary biology, psychology studies (hikers w/ EEG caps!) and single neuron recordings. Listen at
@curioio
or read on
@NautilusMag
:
Are we wired to be outside? ๐ตโโ๏ธ๐
Neuroscientist
@guitchounts
contemplates in this fascinating piece for
@NautilusMag
.
Discover more now โฌ๏ธ๐
As a neuroscientist, I am definitely pro science, pro gathering knowledge for its own sake. But I also care about the animals in my lab and hate to see them imprisoned and killed. Any
#scientists
out there feel the same way? How do you deal with such moral conundrums?
I was also inspired to open the laboratory doors to the public, much in the same way that the wonderful
@AOC
shares the inner workings of Congress with everyone on Instagram (I'm no good with insta stories, so I put my thoughts down on paper instead)
Many neuroscientists are anal about measurements of neural signal but extremely sloppy with task designs and behavioral measures. But great signals combined with poor behavioral measures will yield garbage, often misleading garbage. don't ask me to elaborate.
Lots of previous studies have shown locomotion signals in visual cortex (V1), but they were mostly done in restrained animals, which limits the types of movements you can examine. We wanted to know how movements in freely moving animals affect V1...
Iโm sorry but this makes perfect sense and shouldnโt be news to anyone. How many data scientists are in academia? Salaries in academia are notoriously low. Everyoneโs code sucks.
Iโm sorry โ this doesnt make sense given that most tenured faculty make less than this.
What alien could afford this?
But if this is true or even close to true,
@NIHDirector
may consider that the size of a modular R01 precludes us from having data scientists.
Nothing like a complete failure of your MacBook a day before you leave town for postdoc interviews. I guess they didn๏ฟฝ๏ฟฝ๏ฟฝt want to make it too easy. Thanks,
@apple
@neurojaym
and I noticed an alarming trend among academic scientists: how silent they are when it comes to speaking truth to power. Perhaps some folks think that politics doesn't affect them, or that their lab work is more interesting than fighting racism...
@guitchounts
and I teamed up to write this piece, 'Truth Out Loud', for
@harvardneurosci
@harvardbrainsci
. In academic communities, the responsibility to fight racism falls on everyone's shoulders, and especially our faculty.
Let's get to work. Pls RT!
Dunno if twitter is targeting such tweets to me, but seems the general mood among postdocs is ~negative~ these days. Are things getting worse or is it all the same as always?
Career checklist: Publish 2x in Nature for PhD โ ; get multiple federal fellowships, including K99 in postdoc โ ; (probably) publish again in CNS for postdoc โ ; (probably) lose K funding, despite a global pandemic... โ ?
Just... fuck this whole, ill-conceived game
@kaymtye
letโs petition
@NIHDirector
to raise everyoneโs salaries to a livable wage. That will slow the brain drain from academia, improve lives, improve code, etc.
I wonder why Azeen Ghorayshi did not interview the female first author of this paper--
@DanaRubiLevy
, a brilliant scientist who worked hard on this--which came out a day before International Women's Day.
Important new work from
@Datta_Lab
@ShanskyLab
Overturns the dogma that females are too "complicated" to include in biomedical studies. This dogma has led to a huge disparity in clinical studies, many of which exclude women
1/2
So rare to see someone with so much grace and moral clarity. I'm proud of
@AOC
<3 <3 <3
Rep. Alexandria Ocasio-Cortez spoke from the House floor on the incident with Republican Rep. Ted Yoho
Portugal is super lovely, but Iโm exhausted from
#cosyne19
and ready to be back in Boston, where Iโm sure spring has by now sprung ๐๐ธ๐บ๐ฟ๐ฑ๐ฒ @ Sintra, Portugal
I am immensely grateful to
@michaelpollan
for early guidance and encouragement on this piece, as well to Kevin Berger
@NautilusMag
for compassionate and prudent editing!
Yes, birds have consciousness. This shouldn't be surprising at this point. By the framing of this study though, some single-celled organisms should be considered conscious.
@emilylmullin
@darrenlogan
@wudanyan
Ah that makes more sense. From a basic science perspective, there have been plenty of breakthroughs in mouse (and rat, cat, fly, primate, and bird) work! Translation takes a while though, and you never know where useful discoveries will come from :)
Decoding striatal neurodynamics during natural behaviors in mice. Important bench work from
@guitchounts
to improve brain computer interfaces for individuals with paralysis using AI. Brimming with translational potential at Gordon Research Day
@SpauldingRehab
&
@harvardmed
Everyone in
#NY14
go out and vote today for
@Ocasio2018
!! I have known her since we were wee lads starting college
@BU_Tweets
- she is an amazing human with ethics as defined as
@TheRock
's muscles. Go vote!