Lex Kravitz
Neuroscientist studying obesity, feeding, and dopamine
- Well folks. I bought a new big chainsaw. This means the triangle will be fine. You're welcome.
- Any truck news?
- [Not loaded yet]
- [Not loaded yet]
- Strange story! Hopefully you can walk the line and teach them about remixing permissions while also not inhibiting their interests in 3d design, printing, and even business! If possible I'd try communicating with their parents instead of the 10yo directly.
- [Not loaded yet]
- [Not loaded yet]
- Yeah let's send the findings through their tool!
- If it's the group from U Chicago I did it. It was fine, but none of the suggestions made sense as actual experiments to complete, and I responded as such. I can't seem to find the results in my email but they did not make enough of an impression on me to save them or think about them again.
- 2025 was a rough year for science worldwide (and in general), but, despite everything, I am optimistic. Why? Because of @qedscience.bsky.social! This crazy endeavour actually has a chance of changing how academia works.. 👇
- Thanks for your work on this, it's really really really impressive! I look forward to the updates!
- @superkash.bsky.social Hey Tom any truck news lately?
- [Not loaded yet]
- Haha thanks! I love how you're rocking the DIY build!
- This paper contains some good arguments about an issue that concerns me a lot when I hear my colleagues talking about LLM use in developing their research: Whose ideas are you presenting as your own? (Though the fatalist argument the authors make at the end of paper is disappointing/bizarre.)
- It’s good to see papers start to address LLMs as structural plagiarism — provenance, more hidden than the original words or training data. www.nature.com/articles/s42...
- [Not loaded yet]
- I totally agree it was needless on a short piece like this!
- This statement gave me a hall of mirrors feeling.
- Just got home from a trip and received this book by @juliabelluz.bsky.social and @kevinh-phd.bsky.social!
- This was a really fun project: We blocked obesity in mice by making them do a tiny bit of work for their food. www.sciencedirect.com/science/arti...
- [Not loaded yet]
- Thank you Sung-Yon!
- [Not loaded yet]
- Thank you for your kind reply!
- [Not loaded yet]
- I've found rotary joints unusable for photometry, we've tried a few types (doric, Thor) but they introduce a lot of artifacts when they rotate. So we don't use rotary joints, instead use long 3 meter cables and loop them up high so there's room for the fibers to twist up some without tangling.
- Thrilled to announce that my lab is now open at Ohio State! We study astrocytes as gene therapy targets for brain repair after ischemic stroke and in vascular dementia, developing new viral tools along the way. If you're interesting in joining or collaborating, please reach out! gleichmanlab.com
- Congratulations!!
- I found Jimmy Carter's Environmental Program (fittingly?) in the dumpster in the alley behind my house. "The sight of earthrise was awesome. It was also sobering. From that moment we could no longer avoid understanding that all life must share this one small planet and its limited resources."
- It's funny that ChatGPT does this shit too
- [Not loaded yet]
- The fit line looks like poverty is rising robustly, but I'm not seeing that in the data points themselves. Maybe end the fit line at the min and max of the data instead of extending past it?
- 😂
- It can be hard to say no to service requests in academia. We sought advice on how to ensure researchers don’t get stuck doing unpaid labour go.nature.com/3FsSD1p
- guys I tried ChatGPT for drafting some summary sections in a grant application and it was just... really lame. Like vague and bland and not something that was ever going to engage a grant reviewer at 10.30pm on a Thursday night while they had a glass of red wine in their hand
- I've tried using that logic to explain to PhD applicants why it is counter productive to use ChatGPT at all in their essays. It can do a passable job, putting out middle of the pack prose. But admissions committees are looking for the top ~10% so.....
- [Not loaded yet]
- I love how you're making the og FED3 work!
- [Not loaded yet]
- [Not loaded yet]
- I thought of your posts today when trying to look up a fairly simple fact on Google.
- Working on a new gaming system for mice over the holidays
- Remarkable! Check out figure below. “We find that when ants work in groups, their performances rise significantly. Groups of people do not show such improvement and, when their communication is restricted, even display deteriorated performances. “ IG Nobel nomination?
- Researchers challenged longhorn crazy ants and humans with the same task: maneuvering a T-shaped object through two consecutive open doorways. Single humans always outperformed single ants, but ant groups could beat human groups. In PNAS: www.pnas.org/doi/10.1073/...
- I love this figure
- I received a >50 page response to reviewers that I need to understand before I can start re-reviewing a paper I already spent many hours reviewing the first time. What are we doing???? Props to @elife.bsky.social for trying something different!
- [Not loaded yet]
- I agree, it's even helpful for talking to scientists in your own discipline!
- I took this photo of Jupiter + moons from inside my house in St Louis, aiming the telescope at the bright spot in the sky and using my phone pointed down the eyepiece to capture it. Technology is amazing, this is not a fancy scope. I turned the lights off in the room but otherwise here's the setup!
- Congrats to Dr. Alex Legaria, the first PhD student to graduate from my lab! Lots of feelings as his PI, but mostly I'm grateful that I didn't eff it up 😄😅😁
- [Not loaded yet]
- It's cold here!
- [Not loaded yet]
- It doesn't make sense but 3d printers are now more reliable than 2d
- I am old so I can say this: innovation in neuroscience research can be accelerated if we did not waste so much resources repeating what we already know from NHP and rat research in Cre mouse lines
- Then what would I do with all these lasers?
- [Not loaded yet]
- The elusive D1/D2 co-expressor!
- If you are an undergrad student in the St Louis area and you're interested in a summer 2025 research program please check out ASSURE! assure.wustl.edu
- [Not loaded yet]
- 61 hours?? Can you turn it up to 11?
- This seems punitive. I hope eLife can survive while having a different publishing model than the 50 bazillion other journals
- [Not loaded yet]
- [Not loaded yet]
- Basal googlia
- As we're entering grad school application season, I have a thought on why applicants should completely avoid AI in personal statements. The goal is to stand out from the hundreds of other passionate and accomplished applicants, and any text written by AI is pretty much guaranteed not to stand out.
- Modern food environments are driving obesity, but what is it about them that does this? Is it the content and types of food in modern diets? Or how easy it is to obtain low cost unhealthy food? We address these questions in mice in an updated pre-print: www.biorxiv.org/content/10.1...
- Does replying re-up this post now that I have some followers?
- Our work is consistent with experiments in humans, in which small “nudges” such as changing the placement or convenience of snack foods greatly reduces intake of those foods. However, human studies are typically short in duration and cannot evaluate changes in body weight.
- In this way, we found that a small action – one nose-poke – can reduce food intake and diet-induced weight gain. We conclude that easy access to unhealthy foods may be a critical contributor to the human obesity epidemic. Please check out the pre-print! www.biorxiv.org/content/10.1...
- In this way, we held the nutritional content and diet composition constant, while very lightly manipulating how hard it was to obtain. Mice gained weight when the diet was free (Blue bars), but not when they had to nose-poke *just one time* (Red bars, FR1)!
- This suggests that beyond its specific nutritional content, high-fat food has to be very easy to get to induce weight gain in mice. Critically, the nose-poke is an incredibly easy action to complete, and there was no limit on how much high-fat diet they could earn each day.
- We tested this in mice, giving them a diet that we know drives obesity - 60% fat high-fat diet - under two conditions. In the first, the diet was freely available, while in the second we made it slightly harder to obtain by making mice "nose-poke" to access it.
- However, it is still unclear *why* people are eating more than they did a few decades ago. Is it due to changes in the macronutrient content of food, such as more sucrose or fat?
- Or is it that calorie-dense foods have become extremely easy to obtain, often being low in cost and ubiquitously available in schools, hospitals, hardware stores, clothing stores, and other places that are not primarily sellers of food.
- Multiple converging lines of evidence have linked the obesity epidemic to increases in how much food people eat, which is estimated to have increased ~20% since 1970.
- [Not loaded yet]
- In the study by Unger et al they noted that a ~5°C difference in housing temp can account for large (~30%) changes in metabolic rate and food intake. I agree with their first recommendation: "Monitor, record, and publish the housing temperatures exhibited throughout your experimental protocol."
- Temperature is a critical experimental variable and it is super easy to record. There are lots of temperature data loggers online. Or check out Pallidus.io for more complete solutions for monitoring in-cage environment and behavior!
