@axoaxonic@synapse.cafe
@axoaxonic@synapse.cafe avatar

axoaxonic

@axoaxonic@synapse.cafe

Studying mathematical/theoretical/computational neuroscience & data science in dᶻidᶻəlal̓ič (Seattle) -- more interested in studying nervous systems through abstraction than studying neurons to improve artificial systems

⚧ ND (ASD/mTBIs/cPTSD/ET/probably more letters) ⬛

Long time musician, now slowly inching my way thru academia as first in my close family to do so. Grew up unschooled, which gave me a heavy autodidactism habit

CW: Posts about disease, links to research involving animals

This profile is from a federated server and may be incomplete. Browse more on the original instance.

whitequark, to random
@whitequark@mastodon.social avatar

should i get a centrifuge for 13x75 tubes y/n

axoaxonic,
@axoaxonic@synapse.cafe avatar

@whitequark Here's an open source design that uses a hard disk motor https://hackteria.org/wiki/Hacked_Hard-disc_Centrifuge

NicoleCRust, to random
@NicoleCRust@neuromatch.social avatar

Summer reading recs?!

I’m looking to compile a big pile of IRL books for summer beach reading.

Big fan of scifi, specfic, murder mysteries. Strong female characters = bonus points. Can’t deal with darkness. Nothing high brow or tedious. Nonfiction lover but that list is already bursting at the seams.

I also love long series - 3 body problem, wheel of time, foundation … those waves all have good memories attached.

If that brings anything to mind, please send the rec!!

axoaxonic,
@axoaxonic@synapse.cafe avatar

@NicoleCRust Kiini Ibura Salaam and Nisi Shawl wrote some of the most vivid sci fi stories I've read -- I remember their scenes like dreams. They're both published by Aqueduct Press, a feminist sci fi publishing company from Seattle http://aqueductpress.com/
I think they both have released short stories mostly, but Nalo Hopkinson is also published by Aqueduct and her books are really good

Ones you probably already know about: Ursula Le Guin, Octavia Butler, Ted Chiang, Samuel Delany, Stanisław Lem

Not sci fi:
Djuna Barnes, Ana Castillo, Maryse Condé, Clarice Lispector, Jorge Luis Borges

[sorry it's a huge list of authors, I'm terrible at picking favorites]

axoaxonic, to random
@axoaxonic@synapse.cafe avatar

So much of my PTSD from 10 years of being involved in protests and direct actions is from how the the cops and counterprotestors reacted, but the memories of the actions themselves are the opposite of trauma memories yet equally indelible.

Stay safe out there, take care of yourselves and others, don't let them break your spirit and shred your message with cowardly brute strength.

axoaxonic, to Neuroscience
@axoaxonic@synapse.cafe avatar

Can't find the original post thanks to Mastodon's lack of search, but this is happening in ten days: http://tchumatchenko.de/Workshop/workshop_schedulePage.html

Dendritic computation,
synaptic plasticity & molecular mechanisms

#Neuroscience #Dendrites

ttpphd, to random
@ttpphd@mastodon.social avatar

My sir/ma'am ratio is way off lately.

axoaxonic,
@axoaxonic@synapse.cafe avatar

@ttpphd Could visit the Philippines for balance https://en.wiktionary.org/wiki/mamsir

timnitGebru, to random
@timnitGebru@dair-community.social avatar

Has anyone does any investigations on the so-called human "super recognizers" used by police? The first time I heard about it was in a German article discussing how German police are using that to identify Eritrean protestors they want to target.

axoaxonic,
@axoaxonic@synapse.cafe avatar

@timnitGebru Long reply, sorry, but a couple weeks ago I was looking into them. This is one of the only videos I found that wasn't a pop news segment https://youtube.com/watch?v=E1UPdvUb_9w, a talk by Dr. Meike Ramon of the Applied Face Cognition Lab at University of Lausanne. The lab has produced a few papers on SRs https://afclab.org/publications

"Only 15 empirical studies currently exist" since the original study on four subjects in 2009 by Russell, et al https://link.springer.com/article/10.3758/PBR.16.2.252

They are used by law enforcement agencies (notably, SR Kelly Hearsey said on the Crawlspace Podcast that she was hired for the Jamal Khashoggi case to investigate the use of a body double), but Dr. Ramon calls the employment of them by multiple police agencies "dubious", where hype about their skills is manufactured by "fee based associations" while scientific knowledge about the phenomenon is limited.

The cops promote their abilities to recognize people in general, but the academic basis is about face recognition, studying real people with superior face recognition ability, on the opposite end of the spectrum from those with face-blindness/prosopagnosia. Selection for employment is often based on mass online testing, which can be faked or done by bots etc.

Her lab's work has shown that most people with exceptional skill in one area like face recognition does /not/ actually correlate with exceptional skill in related tasks like connecting the same person's face at different angles or times in their life, although there are a few who excel at every task.

Unfortunately in the talk she doesn't go into the media claims that super recognizers can pick out people who are wearing facial and head coverings (like protestors) like with the person in this BBC clip https://youtube.com/watch?v=PuPfQ8UZTGQ It only uses an example of someone who's face is mostly visible though. While one study provided evidence that they're better at recognizing obscured faces than average recognizers, mostly when eyes are still visible, https://gala.gre.ac.uk/id/eprint/19273/3/19273%20DAVIS_Masters_of_Disguise_2017.pdf (nothing about gait/body language), I don't think there's any research on whether or not they can out perform computer vision

axoaxonic, to random
@axoaxonic@synapse.cafe avatar

A lot of people are posting a video by Sabine Hossenfelder right now, and I'm not going to comment on the video or the points and discussion, but I wanted to post this video detailing her problematic views on trans issues, how she promoted (in a biased-centrist way) the harmful, TERF-associated, and unfounded view that gender affirming care for trans kids is a social contagion leading to "rapid onset of gender dysphoria," while making claims that transitioning before puberty is harmful https://www.youtube.com/watch?v=r6Kau7bO3Fw Also there's a line where she offhand belittles people who say early gender affirming care saves lives

"Whether or not she meant to, she repeatedly used a misleading rhetorical device to elevate transphobic talking points to the same level as scientific evidence, and she did that in front of a very large audience"

axoaxonic,
@axoaxonic@synapse.cafe avatar

@mrcompletely There's definitely a correlation between scientists doing that more and increasing traffic to and revenue from their content. Their integrity in basing what they say publicly on strong evidence definitely seems to sharply fall once they start posting vids and podcasts about viral topics that get them tons of engagement. They seem to stop caring that they're not even experienced with those topics

kjhealy, to random
@kjhealy@mastodon.social avatar

Kerry Howley out there on the front lines, exploring the current outer limits of Californized Celebrity-Academic Self-Help Podcast Psychopaths. It’s a relief that she’s such a good writer because honestly these people are just batshit weirdos who insist on talking about themselves like space aliens. https://nymag.com/intelligencer/article/andrew-huberman-podcast-stanford-joe-rogan.html

axoaxonic,
@axoaxonic@synapse.cafe avatar

@kjhealy It really surprised me when I found out Huberman's postdoctoral advisor was Ben Barres

axoaxonic, to Seattle
@axoaxonic@synapse.cafe avatar

Yesterday I met the final boss version of cars pulling into the bike lane: construction downtown at Westlake redirected all car traffic to move around it by going into the bike lane. No warning signs, just cones and a stream of cars suddenly in front and behind me for two blocks

axoaxonic, to random
@axoaxonic@synapse.cafe avatar
axoaxonic, to random
@axoaxonic@synapse.cafe avatar

In this society that emphasizes competition and mastery to create value, so many people say you have to focus on one thing to be successful, but no one really says how to do that.. When I was younger I heard that advice, I decided to focus on neuroscience, but even within this field there's actually millions of potential things to focus on.

If anyone wants to share their method/s of deciding what to devote serious time to, feel free to reply

axoaxonic,
@axoaxonic@synapse.cafe avatar

@NicoleCRust I believe I'll always be deeply interested in a lot of different things and weave them together, I 100% can't turn that part of my mind off, nor do I want to. I'm mostly amazed at people's ability to focus on one research topic, even if it's just for a year or two before going to the next, making some really solid work happen. Even DaVinci finished paintings.

When everything's interesting or all the champagne on the shelf looks good, usually the best thing to do is to pick whatever one you're drawn to and go and enjoy it. That's the most important part anyways in "no wrong choices" situations, the unique experience the choice results in.

axoaxonic, to random
@axoaxonic@synapse.cafe avatar

Some research suggesting that traveling waves exist in the spinal cord as well:

"Synergy temporal sequences and topography in the spinal cord: evidence for a traveling wave in frog locomotion", Saltiel, et al, 2016
https://dspace.mit.edu/bitstream/handle/1721.1/105252/429_2015_Article_1133.pdf?sequence=1

"Propagation of Sinusoidal Electrical Waves along the Spinal Cord during a Fictive Motor Task", Cuellar et al, 2009
https://www.jneurosci.org/content/29/3/798 [cat study]

jonny, to random
@jonny@neuromatch.social avatar

The thing about that "the papers will disappear if the journals do" article is that they wont and the only reason is piracy.

axoaxonic,
@axoaxonic@synapse.cafe avatar

@UlrikeHahn @jonny @tdverstynen Sci-Hub/LibGen scrape journals using access credentials to get pdfs, and add more when a document is searched for but not found in it's database. Peer-to-peer is just giving remote access to your documents or whichever folder to anyone with a program like Soulseek or whatever people use nowadays. Neither of these are affected by citation count, and I've found papers that have 0 cites according to goog scholar

"as of March 2017, Sci-Hub’s database contains 68.9% of the 81.6 million scholarly articles registered with Crossref and 85.1% of articles published in toll access journals" 10.7554/eLife.32822
It's missing some articles but way less than the amount without citations

axoaxonic, to random
@axoaxonic@synapse.cafe avatar

I tried uploading three different pictures of me to a modern automatic photo captioning app that uses AI, and it referred to me as a "woman," a "person," and a "man"

This paper from 6 years ago is still relevant:
"The Misgendering Machines: Trans/HCI Implications of Automatic Gender Recognition", Os Keyes, 2018
https://dl.acm.org/doi/10.1145/3274357

axoaxonic, to random
@axoaxonic@synapse.cafe avatar

Went to the Shahdmehr lab youtube page too look if there were any vids, and was surprised to find a whole series of hour-long lectures from the Johns Hopkins Cerebellum Seminars

https://www.youtube.com/@shadmehrlab1352

Invidious version to avoid trackers and allow adblockers:
https://vid.puffyan.us/channel/UCSifaCZt1HmnZJ3440AOf8g

icastico, to vinyl
@icastico@c.im avatar

The Glow Part 2 by The Microphones #vinyl @vinylrecords #postRock #ChamberPop #DreamPop #NowPlaying

axoaxonic,
@axoaxonic@synapse.cafe avatar

@icastico @vinylrecords

I played bar trivia with Phil Elverum once. Our team was named Slime

petersuber, (edited ) to academia
@petersuber@fediscience.org avatar

From a survey of @penn_state : "72% …reported they had not purchased a course’s required , and 33% said they had not registered for a specific course because of its cost of required course materials. In addition, 33% said they had earned a poor grade because they couldn’t afford to purchase a course textbook, and 17% dropped a course because of the cost of materials required for the course."
https://www.psu.edu/news/university-libraries/story/students-may-avoid-paying-textbooks-expense-academic-success/


@academicchatter

axoaxonic,
@axoaxonic@synapse.cafe avatar

@petersuber @penn_state @academicchatter One time I took a class while I could barely afford food, so I bought the textbook, went to my car and took pictures of every page for two hours, then returned the book. It was either that, not taking the class, or not eating

NicoleCRust, (edited ) to random
@NicoleCRust@neuromatch.social avatar

Delightful mystery

People in Tampa can hear a low frequency sound across the Florida peninsula; even in their homes. The mystery: where is it coming from? One expert thinks it’s coming from under water: black drum fish mating season.

https://www.npr.org/2024/02/01/1228286349/south-tampa-mystery-where-is-the-sound-coming-from-neighbors-investigate

https://ocr.org/sounds/black-drum/

axoaxonic,
@axoaxonic@synapse.cafe avatar

@NicoleCRust there's a similar phenomenon in West Seattle https://www.knkx.org/other-news/2012-09-06/mysterious-hum-keeping-west-seattle-up-at-night

People think it's probably from a fish called the Midshipman fish

NicoleCRust, to random
@NicoleCRust@neuromatch.social avatar

Thoughts on these provocative ideas (about how research in psychology should proceed)?

The last author tipped me off to this one. Curious to hear impressions.

Beyond Playing 20 Questions with Nature: Integrative Experiment Design in the Social and Behavioral Sciences

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4284943

(also here, behind the BBS paywall: https://www.cambridge.org/core/journals/behavioral-and-brain-sciences/article/abs/beyond-playing-20-questions-with-nature-integrative-experiment-design-in-the-social-and-behavioral-sciences/7E0D34D5AE2EFB9C0902414C23E0C292)

The dominant paradigm of experiments in the social and behavioral sciences views an experiment as a test of a theory, where the theory is assumed to generalize beyond the experiment’s specific conditions. According to this view, which Alan Newell once characterized as “playing twenty questions with nature,” theory is advanced one experiment at a time, and the integration of disparate findings is assumed to happen via the scientific publishing process. In this article, we argue that the process of integration is at best inefficient, and at worst it does not, in fact, occur. We further show that the challenge of integration cannot be adequately addressed by recently proposed reforms that focus on the reliability and replicability of individual findings, nor simply by conducting more or larger experiments. Rather, the problem arises from the imprecise nature of social and behavioral theories and, consequently, a lack of commensurability across experiments conducted under different conditions. Therefore, researchers must fundamentally rethink how they design experiments and how the experiments relate to theory. We specifically describe an alternative framework, integrative experiment design, which intrinsically promotes commensurability and continuous integration of knowledge. In this paradigm, researchers explicitly map the design space of possible experiments associated with a given research question, embracing many potentially relevant theories rather than focusing on just one. The researchers then iteratively generate theories and test them with experiments explicitly sampled from the design space, allowing results to be integrated across experiments. Given recent methodological and technological developments, we conclude that this approach is feasible and would generate more-reliable, more-cumulative empirical and theoretical knowledge than the current paradigm—and with far greater efficiency.

axoaxonic,
@axoaxonic@synapse.cafe avatar

@beneuroscience @NicoleCRust @MolemanPeter @UlrikeHahn @jonny

It might be related to spatiotemporal scale, since nonequilibrium systems exhibit local thermodynamic equilibrium (sec 3.2 of this https://courses.physics.ucsd.edu/2020/Fall/physics210b/Non-Eqbrm%20Thermo_Demirel%20and%20Gerbaud.pdf says it's common in simpler nonequilibrium systems, though it is also found in more complicated active matter systems such as bird flocks https://doi.org/10.1038/nphys3846 ). Even though the whole nervous system must be a dissipative nonequilibrium system, at the microscopic scales of smaller networks this might be okay to ignore without too much consequence, but idk if that's a good idea if the aim is biological realism.

This paper shows that some biological dynamics only appear in ANNs if they are in nonequilibrium states due to asymmetric connections, which are common in bioNNs. https://www.pnas.org/doi/10.1073/pnas.1310692110 These dynamics include oscillations (traveling waves?) and free energy minimization, which aren't present in the symmetric Hopfield networks they tested

NicoleCRust, to random
@NicoleCRust@neuromatch.social avatar

Simple RNN models that capture neural network criticality / edge of chaos?

I'd like to play around with simple (tutorial-like) recurrent neural network models that capture the phenomenon of criticality. Something like the smallest possible number of recurrently connected model neurons that can recapitulate phenomenon (like information processing peaks for intermediate coupling weights).

Any leads?

axoaxonic,
@axoaxonic@synapse.cafe avatar

@NicoleCRust @derstrudel @albertcardona Although reservoir computing works great at predicting chaos https://pubs.aip.org/aip/cha/article/31/1/013108/341924/On-explaining-the-surprising-success-of-reservoir
I just found this illustrating that they might not work with the same optimality at the edge of chaos like many dynamical systems do, as it's more a transition from stability to instability https://ar5iv.labs.arxiv.org/html/2012.01409
Haven't looked at the above links posted by @tyrell_turing yet though

axoaxonic,
@axoaxonic@synapse.cafe avatar

@tyrell_turing @NicoleCRust @derstrudel @albertcardona I think the difference is the nonexistence of a chaotic attractor when RCs lose stability.

"In many reservoir computers, depending on the node type, there is no actual chaotic behavior. Instead, when the largest Lyapunov exponent for the reservoir becomes positive, the reservoir network becomes unstable; in simulations, the reservoir signals diverge to positive or negative infinity. It is more accurate to call the point where the Lyapunov exponent becomes positive the edge of stability, so I will use that term instead of edge of chaos."

But they don't seem to elaborate on the distinction

axoaxonic, to random
@axoaxonic@synapse.cafe avatar

Applying information theory to cellular ion concentration gradients, these authors derive a principle where cells optimally code responses to environmental perturbations -- incl input from other cells -- by minimizing the cross entropy (Kullback-Leibler divergence) between intracellular and extracellular ion concentrations.

"We demonstrate the ion dynamics in neuronal action potentials described by Hodgkin and Huxley (including the equations themselves) represent a special case of these general information principles."

"Cellular information dynamics through transmembrane flow of ions", Gatenby and Frieden 2017 https://www.nature.com/articles/s41598-017-15182-2.pdf

axoaxonic,
@axoaxonic@synapse.cafe avatar

The greater the KL-divergence, the longer the codeword of Shannon information, the longer the response time. Natural selection would choose rapid responses, so the KL-divergence is minimized: the probability density of the fluctuating intracellular ion concentration is kept as close to the mostly-constant probability density of the extracellular concentration as possible.

The huge range of different perturbations leading to intracellular ion concentration changes would increase the evolutionary pressure for complex and diverse ion channels under this principle

axoaxonic,
@axoaxonic@synapse.cafe avatar

All cells process information, but nerves/neurons are specialized to do so extremely fast in order to respond accurately, in chain reactions, in real time to inputs from perception and memory. Spikes should be really good at this, they're both very short codewords and strong electrical perturbations for neighboring neurons, but they aren't the sole information carriers that would fit the digital computer metaphor. Analog abounds

"transmembrane ion gradients potentially may encode in excess of 10^14 bits of
Shannon information"

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • normalnudes
  • hgfsjryuu7
  • magazineikmin
  • thenastyranch
  • Youngstown
  • slotface
  • everett
  • ngwrru68w68
  • mdbf
  • kavyap
  • tsrsr
  • Durango
  • PowerRangers
  • DreamBathrooms
  • Leos
  • InstantRegret
  • khanakhh
  • osvaldo12
  • vwfavf
  • tacticalgear
  • rosin
  • cubers
  • cisconetworking
  • GTA5RPClips
  • ethstaker
  • tester
  • modclub
  • anitta
  • All magazines
    •