Lightnews — Scholar-powered news

Luke Sjulson

@lukesjulson.bsky.social

4K followers 2.3K following 390 posts

MD/PhD neuroscientist/psychiatrist. Decision-making, addiction, multi-region ephys and imaging in vivo, novel optical methods for spatial transcriptomics https://sjulsonlab.org

sjulsonlab.org

Posts Media Videos Starter Packs

Pinned

Luke Sjulson @lukesjulson.bsky.social · Dec 13

Neuroscience students asked us to teach a PRACTICAL course on experimental methods, and it is now on YouTube!

Please like and repost to help us get the word out!

www.youtube.com/playlist?lis...

Lecture 1: Signals and data acquisition
Focusing on hardware, digital/analog I/O, synchronization
🧵

Neuroscience methods - YouTube

Nanocourse: Approaches to Study Neural Circuits This course was taught by Anita Autry, Tiago Gonçalves, and Luke Sjulson at Albert Einstein College of Medici...

www.youtube.com

19 200 420

Luke Sjulson @lukesjulson.bsky.social · 3h

at the time

Luke Sjulson @lukesjulson.bsky.social · 3h

that was available at the time. In some cases the clinician makes an obvious error, but in a case like this where the correct diagnosis was made and it's a choice between two different treatment options, there can be a lot of gray area, and wrong-in-retrospect does not necessarily mean it was wrong

1 1

Luke Sjulson @lukesjulson.bsky.social · 3h

I'm glad to hear that your mom is doing well. Without knowing anything about the specific details of your mom's case, I would say that in situations like this Monday morning quarterbacking is easy, and it's not always clear cut what the "right" and "wrong" choices were with the information

1 1

Luke Sjulson @lukesjulson.bsky.social · 4h

Yeah, clinicians (including myself) are wrong sometimes, no question. Medicine is intrinsically hard, and it isn’t easy to stay up to date. That doesn’t mean that professional clinicians don’t have knowledge and insight that laypeople don’t

2 2

Luke Sjulson @lukesjulson.bsky.social · 4h

Often the worst examples are people who’ve experienced an illness first hand and think they understand the entire scope of that illness better than professional clinicians

1 3

Luke Sjulson @lukesjulson.bsky.social · 4h

You’re right, and I think a key difference is that professionals see thousands of people, and laypeople (even those with direct personal experience) tend to overgeneralize from small N examples and underestimate the complexity and heterogeneity of mental illness

1 7

Reposted by Luke Sjulson

Dr. Jonathan N. Stea @jonathanstea.bsky.social · 5h

It can be very easy for people who have never professionally sat with people (and their families) who experience the devastation of mental illness to pontificate about its existence and treatment from an armchair.

4 15 99

Luke Sjulson @lukesjulson.bsky.social · 5h

I think everyone in my lab wants to do that with the entire building we’re in

Luke Sjulson @lukesjulson.bsky.social · 7h

I think so. But I’m not sure that PMTs have changed that much since then, other than GaAsP electrodes

Luke Sjulson @lukesjulson.bsky.social · 8h

Good question, I don’t know. I suspect it’s one of the PMTs from the Super K neutrino observatory

1 1

Luke Sjulson @lukesjulson.bsky.social · 8h

Thus must have been made by @hamamatsucorp.bsky.social, right? “Photon is our business” indeed

Luke Sjulson @lukesjulson.bsky.social · 8h

Here’s a better image with my hand in front of it for scale

1 3

Luke Sjulson @lukesjulson.bsky.social · 8h

Look at the size of this photomultiplier tube in the Nobel Prize museum. Imagine doing two photon imaging with one of those!

3 1 17

Luke Sjulson @lukesjulson.bsky.social · 9h

It makes me sad there is an entire generation that has not seen Office Space

3 1 12

Luke Sjulson @lukesjulson.bsky.social · 10h

On one hand, i agree that we know exactly how LLMs work, and on the other hand I don’t. “How something works” can mean a lot of different things. But we can be confident that LLMs don’t have “consciousness” or whatever. I’m not sure whether that will be true for models 10 yrs from now

Luke Sjulson @lukesjulson.bsky.social · 10h

That is precisely what has been happening over the past several years, with many of the top ML people leaving academia

Luke Sjulson @lukesjulson.bsky.social · 21h

Language is how other people convey their thoughts to us, so we are naturally prone to take language as evidence of thought. But evolutionarily, thought is much older and deeper than language

Frames @oxinabox.bsky.social · 3d

as a girl with a PhD in natural language processing and machine learning it's actually offensive to me when you say "we don't know how LLMs work so they might be conscious"

I didn't spend 10 years in mines of academia to be told ignorance is morally equal knowledge.

We know exactly how LLMs work.

1 5

Reposted by Luke Sjulson

manfredic.bsky.social @manfredic.bsky.social · 6d

I’m pleased to share our new paper, “Hippocampal ripple diversity organizes neuronal reactivation dynamics in the offline brain”, out in @cp-neuron.bsky.social !

With @vitorlds.bsky.social and David Dupret, we show that diversity in ripple current profiles shapes reactivation dynamics

4 26 64

Reposted by Luke Sjulson

Kris Jensen @kristorpjensen.bsky.social · 14d

I’m super excited to finally put my recent work with @behrenstimb.bsky.social on bioRxiv, where we develop a new mechanistic theory of how PFC structures adaptive behaviour using attractor dynamics in space and time!

www.biorxiv.org/content/10.1...

9 85 200

Luke Sjulson @lukesjulson.bsky.social · 14d

Congrats, Blake!

Luke Sjulson @lukesjulson.bsky.social · 14d

Congrats, Kurt, and good luck! R01s get much easier over time, whether you want them to or not

1 3

Reposted by Luke Sjulson

furby hancock @furbyhancock.bsky.social · 15d

Too late, Batman. Once this Tylenol floods the city's water supply, my wiki won't run out of editors ever again

49 5.6K 19K

Reposted by Luke Sjulson

Julie Charlton @juliecharlton.bsky.social · 16d

🛑 Attention researchers who train freely-moving rodents!

Jess Breda and I developed a protocol for training center-port nose fixation 61% faster than a previous curriculum while keeping violation rates low.

Preprint here: www.biorxiv.org/content/10.1...

FixGrower: An efficient and robust curriculum for shaping fixation behavior in rodents

Center-port fixation is a common prerequisite for many freely-moving rodent tasks in neuroscience and psychology. However, typical protocols for shaping this behavior are non-standardized and ineffici...

www.biorxiv.org

1 9 35

Luke Sjulson @lukesjulson.bsky.social · 15d

I was texting someone about the Tylenol/autism announcement, and for the first time, I accidentally typed “ducking” and my phone successfully autocorrected it to “fucking”

Luke Sjulson @lukesjulson.bsky.social · 20d

Congrats on the paper btw! Looks like a tour de force