Dimos Karamanlis @dimokaramanlis@qoto.org

🚨Pre-print alert 🚨Yes that’s right, we’re publishing not one, but two, pre-prints! Here we use the *electrophysiology data* from the Brain Wide Map as well as *widefield imaging data* to investigate the question “Where in the brain is prior knowledge represented?” Recordings in 267 brain regions from 121 mice (1/8)
#neuroscience
https://www.biorxiv.org/content/10.1101/2023.07.04.547684v1

Rats use #eyebrows to sense the wind! Animals use wind sensing (#anemotaxis) for navigation & survival. @annmclemens @BrechtLab &co use structural, ephys & behavioral analyses to reveal that supra-orbital #whiskers act as wind antennae in rats #PLOSBiology https://plos.io/3riXfQV

New study by Jacob L. Yates et al. introduces new tools enabling natural #behavioranalysis in untrained subjects, revealing accurate #visual #receptivefields and tuning curves in marmoset monkeys, highlighting the potential of free viewing for studying #neuralresponses and natural #behaviordynamics.

📔 Yates, Coop, Sarch, et al., "Detailed characterization of neural selectivity in free viewing primates", Nat Commun 14, 3656 (2023), https://doi.org/10.1038/s41467-023-38564-9

#Neuroscience

In an alternative history of the world, quantum mechanics could have been discovered by chemists following up on the theories of Clebsch and Gordan.

We now use their math to understand the funny way angular momentum 'adds' when we combine two quantum systems. We use this in chemistry.

But they were already suggesting to use this math for chemistry back in the 1890s. They were ignored by chemists! But it was very hard, back then, for people to believe that atoms were governed by the math of linear algebra and invariant theory - now called group representation theory.

According to Wikipedia:

Like James Joseph Sylvester, Paul Gordan believed that invariant theory could contribute to the understanding of chemical valence. In 1900 Gordan and his student G. Alexejeff contributed an article on an analogy between the coupling problem for angular momenta and their work on invariant theory to the Zeitschrift für Physikalische Chemie. In 2006 Wormer and Paldus summarized Study's role as follows:

"The analogy, lacking a physical basis at the time, was criticised heavily by the mathematician E. Study and ignored completely by the chemistry community of the 1890s. After the advent of quantum mechanics it became clear, however, that chemical valences arise from electron–spin couplings ... and that electron spin functions are, in fact, binary forms of the type studied by Gordan and Clebsch."

I learned this amazing piece of history from James Dolan. The Wikipedia quote is actually from the biography of the mathematician Eduard Study:

https://en.wikipedia.org/wiki/Eduard_Study#Valence_theory

On explanations in brain research:

A thread of the same idea comes up again and again in brain research. It's the notion that identifying the biological details (such as the brain areas/circuits or neurotransmitters) associated with some brain function (like seeing or fear or memory) is not a complete explanation of how the brain gives rise to that function (even if you can demonstrate the links are causal). To paraphrase:

Mountcastle: Where is not how https://www.hup.harvard.edu/catalog.php?isbn=9780674661882
Marr: How is not what or why http://mechanism.ucsd.edu/teaching/f18/David_Marr_Vision_A_Computational_Investigation_into_the_Human_Representation_and_Processing_of_Visual_Information.chapter1.pdf
@MatteoCarandini: Links from circuits to behavior are a "bridge too far" https://www.nature.com/articles/nn.3043
Krakauer et al: Describing that is not understanding how https://www.cell.com/neuron/pdf/S0896-6273(16)31040-6.pdf
Poppel: Understanding brain maps does not formulate "what about" the brain gives rise to "what about" behavior https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3498052/

Any other explicit references to add to this list? @Iris, @knutson_brain, Anyone?

Also, I imagine that some form of the opposite idea must also be percolating: the notion that 'algorithmic' descriptions of the type used to build AI will be insufficient to do things like treat brain dysfunction (where we arguably need to know more about the biology to, e.g., create drugs). Any explicit references of that idea? @albertcardona @schoppik, @cyrilpedia, Anyone?

Traditional computer software tools resemble the standard mathematical concept of a function 𝑓:𝑋→𝑌: given an input 𝑥 in the domain 𝑋, it reliably returns a single output 𝑓(𝑥) in the range 𝑌 that depends on 𝑥 in a determinstic fashion, but are undefined or give nonsense if fed an input outside of the domain. For instance, the LaTeX compiler in my editor will take my LaTeX code, and - provided that it is correctly formatted, all relevant packages and updates have been installed, etc. - return a perfect PDF version of that LaTeX every time, with no unpredictable variation. On the other hand, if one tries to compile some LaTeX with a misplaced parenthesis or other formatting problem, then the output can range from compilation errors to a horribly mangled PDF, but such results are visually obvious to detect (though not always to fix).

#AI tools, on the other hand, instead resemble a probability kernel μ:𝑋→Pr(𝑌) rather than a classical function: an input 𝑥 now gives a random output sampled from a probability distribution μₓ that is somewhat concentrated around the perfect result 𝑓(𝑥), but with some stochastic deviation and inaccuracy. In many cases the inaccuracy is subtle; the random output superficially resembles 𝑓(𝑥) until inspected more closely. On the other hand, such tools can handle noisy or badly formatted inputs 𝑥 much more gracefully than a traditional software tool.

Because of this, it seems to me that the way AI tools would be incorporated into one's workflow would be quite different from what one is accustomed to with traditional tools. An AI LaTeX to PDF compiler, for instance, would be useful, but not in a "click once and forget" fashion; it would have to be used more interactively.

I wrote a longer-form piece over at post.news about the problems with using the verb "hallucinate" to describe AI chatbots that make things up.

Here's the link for those that what it to read it formatted there.

https://post.news/article/2Lr1Pj6ITLA0LGxLi2CQIHUI1WB

I'll serialize it here as well, below.

Trained in an academic setting, I have often felt a little uneasy conveying science beyond my expertise.

@markdhumphries writes that this is not only okay, but actually highly encouraged for good popular science writing. He advocates doing research on the topic and conveying your own journey through learning.

https://medium.com/the-spike/its-a-good-thing-when-i-don-t-know-what-i-m-writing-about-6d0ce69d5629

#writing #science #PopularScience

#migration#re-Introduction.

I'm a theoretical neuroscientist at U Mainz Medical Center and co-affiliated with U Bonn Medical center. Primary focus on cortical circuits, their network activity, synaptic plasticity and protein dynamics in dendrites. Broadly interested how circuits learn e.g. neuro/AI interface and want to understand how neural networks compute, both algorithmically and intracellularly. Occational posts about societal and academic issues.

@jacklerner I would remind students: Whenever you read a productivity tip or lessons from someone's process, ask yourself: Who's taking care of the house and kids? Who manages their finances? Who's booking their flights and tracking their expenses and picking up their laundry? Do they have any disabilities, health issues, other challenges?

Do not beat yourself up if you're not able to be as productive as someone with significantly more privileges re: time. We don't all have the same 24 hours.

My little book is now officially published by Cambridge Univ Press and available for free here https://www.cambridge.org/core/elements/attending-to-moving-objects/0914ABF2EF7D03676124F7250874071A . I draw lessons from object tracking research to illuminate the nature of the bottlenecks on human visual processing.

The official version is available for free for two weeks. The version I published with bookdown (https://tracking.whatanimalssee.com/intro.html#summary) will be available for free at least until I die, or become too ashamed of the book. #attention #perception
#openaccess @cognition

Naive question (maybe): Is there a definition of 'computation' akin to the mathematical definition of information (entropy/MI)? I don't mean Turing machines. e.g. something that could determine the extent to which a group of neurons/synapses are signalling versus computing?

Job alert: tenure-track Assistant Professorship in Applied Psychology! 📢​

My department at Leiden University is recruiting an Assistant Professor in Applied Cognitive Psychology, with tenure after 1.5 years. Research topics include behavior change and sustainability, and the position includes teaching and coordinating our MSc program. Deadline is 10 February, apply here. 

Visual cortex is affected by sounds. Does this mean it is multisensory? Does it represent sounds?

Today in Nature #Neuroscience (https://doi.org/10.1038/s41593-022-01227-x) Célian Bimbard and team find otherwise.

Sound-evoked activity in visual cortex was similar across neurons. Even neurons in hippocampus. It persisted after we cut fibers from auditory cortex!

This activity was predicted by… subtle body movements evoked by the sounds. It reflects brainwide state and behavior signals.

#NeuroNewPaper

We explain how responses to a mismatch between locomotion and optic flow can arise in V1. We propose they are a result of a cooperation between motor and visual areas to explain the perceived optic flow. Now out in Neural Computation:

https://www.biorxiv.org/content/10.1101/2022.04.07.486697v1
(free)

https://doi.org/10.1162/neco_a_01546 (paywall)

An interesting question: Does cortex integrate different modalities already in the early levels of processing? We think yes, by distributing representations over multiple areas.

A big thanks to @kendmiller + for helping the #neuroscience community get its hashtag act together!

https://neuromatch.social/@kendmiller@qoto.org/109645950754401219

For neuro paper threads: sigmoid.social/about/more has already claimed #PaperThread and #NewPaper (the latter announcing a paper without a thread) for the AI community. I enjoy seeing their papers too, but we need a distinct tag for neuro papers. For a thread, ... maybe we want something simple like #NeuroPaperThread and #NeuroNewPaper?

Great idea Ken - let's do this.

➡️​​EVERYONE: BOOKMARK THIS! ⬅️​ and follow those hashtags. And then don't hold back. After all, it's what we all show up here for: to hear what you've figured out and learn from you.

#neuroscience #neuroAI #psychology

There's a generational gap which I'm starting to experience more and more, where nearly everything I talk to my kids about — be it nuclear fusion, be it war in Ukraine, be it GPUs — they source with YouTube videos.

But I simply cannot ingest these. Watching a 1½ hours sequential display of maybe even valid statements but which you can't easily scan for useful information feels like watching TV in 1990's, especially that YT is frequently interrupted by ads. And specifically for that reasons I haven't had a TV at home since 2000's.

Videos are kind of OK fo a lecture, where the lecturer sequentially guides you through an unfamiliar topic. But a rather lengthy video to present a single point, that would otherwise fit on half of a printed page, which you can easily scan and locate relevant paragraphs as a matter of seconds?

Feels very much like a downgrade, sorry...

Across a bunch of scientist biographies, I noticed that a lot of scientists were part of a club to discuss science. It's usually something mostly informal that meets about once a month. Some scientists that I know were part of such clubs and developed ideas there are Norbert Weiner, Ramon y Cahal, Alan Turing, Pyotr Kapitsa, among others. Counting down to the new year, I'll be posting about the societies that have fascinated me.

#science #sciencehistory

#introduction 👋
I’m a neuroscientist & artist. I got into the brain biz from making art—it’s like running #vision experiments on yourself. #Color is an obsession, it connects the physical world with #perception, #cognition, society, so I think it’s a great tool to understand us. I taught #neuroscience at #wellesley, ran a lab at #MIT & now at #NIH #NEI #NIMH. #Science happens in teams & diversity improves teams, so I really value #diversity in science, at all levels, of all kinds. #watercolor