Yeah, it's a mouse study, but there are tons of human studies backing the whole gut-brain connection. There are even a bunch of books on it [1][2].
What's really cool is that the paper used low-dose capsaicin (just 5 μg/kg injected), and it completely restored hippocampal FOS activity and memory in older mice. Basically, that's the same stuff you get in cayenne pepper supplements - pretty easy to get your hands on.
I only have marginal knowledge about neuroscience, but one of my neuroscience professors in class would tell us
"You can cure anything in mice."
I don't know the mechanism why, but you can find tons of papers with incredibly strong results for curing of mitigating dementia, cognitive decline, addiction, etc in mice, but these almost never seen to work on people.
They're human specific ailments. We create a fake version of them in mice, then we fix the fake version. The basic problem with these issues is we don't understand the root cause. So we can replicate the symptoms in a mouse model then fix the symptoms, but that doesn't work in humans because the root cause is still there.
I guess it's because most major disorders and diseases have so many pathways at play that figuring out which one's actually causing the problem at the individual level is just too tricky.
The other thing concerns how potent the effect is to be therapeutic. In many cases, the effect is just marginal to be meaningful.
I've long regarded the great variety of chilis as its own distinct food group. But wonderful as they are for flavoring food, quite often in my home, I'm not sure how much of an effect orally consumed capsaicin has on memory functioning.
Conceivably parenteral capsaicin has different effects on hippocampal integrity or physiology than achievable with ingestion. I'm not familiar enough with disposition of capsaicin in the gut to comment further. My question is whether capsaicin passes from gut into the circulation in any appreciable quantity. I suspect it doesn't but I couldn't say I know for sure. I'll have to add it to the already long list of things I need to look up.
No 2 is a fascinating study! My mother was 'taking' (eating) large amounts of raw chilli as she found it a very efficient cure for her rheumatoid arthritis - she had brilliantly reduced joint pain. BUT after a few years it caused painful and disfiguring rosacea.
She stopped the chilli and moved to acupuncture for the arthritis, which worked pretty much as well, but not something she can do herself at home for 'free'.
Since she has v low BML, I'm now pleased to see she stopped eating too much chilli!
> What's really cool is that the paper used low-dose capsaicin (just 5 μg/kg injected), and it completely restored hippocampal FOS activity and memory in older mice.
There are countless papers published where simple ingredients produce miracles in mice. Most of them don’t replicate.
If you look up most food ingredients you can find someone, somewhere claiming to have used it to produce amazing outcomes in mice. After you read a lot of those you learn not to take individual papers seriously if the claims seem too good to be true.
The rational mind should not be seeing singular papers and assuming they’re correct. There are a lot of incentives for researchers to publish amazing results that benefit their career. They find ways to publish these through small sample sizes, p-hacking, or worse like faking results.
The amazing results usually disappear in larger studies by more rigorous researchers. There are so many papers showing amazing things in a handful of mice in a lab or even human volunteers that do not appear again in properly powered studies.
I never said we have sufficient evidence to act. But "too good to be true" + "singular paper" together can become an unfalsifiable dismissal - by that logic, every important result looks suspicious before it replicates. The interesting question is what priors should update our confidence here.
Stanford/Arc Institute and published in Nature + mechanistic grounding + prior research on gut-brain axis gives me way more confidence than average, but you're right, that's not nearly enough for most, but quite sufficient for me, and surely others with informed priors or a strong motive.
> by that logic, every important result looks suspicious before it replicates
Every important result should look suspicious before replication. This is the rational way to interpret early research.
You should not allow your mental probability distribution to be anchored around the first claim you see that is proposed as a paper. In the modern publishing environment, a heuristic of assuming singular results will not replicate would be accurate more often than assuming they’re true.
This isn’t intuitively obvious until you’ve read a lot of papers. It’s unfortunate but true.
Even some of the widely accepted findings like the benefits of fish oil supplementation are having a hard time replicating in large scale studies. Go back 10 years and it was almost universally accepted that those early fish oil studies must be true.
Hi! I'm a backend engineer (~8 YOE) with strong backend & DevOps experience and decent frontend skills. Looking for a backend or backend-leaning full-stack role.
I worked at Automattic (US), the company behind WordPress; fully remote, async teams across the globe.
I've built and maintained time-sensitive, high-throughput, distributed services (millions of ops daily) and owned features and small- to mid-sized projects end-to-end from design to deployment. I do my best working autonomously, and I like to think of myself as a generalist.
I've been part of teams focused on speed and rapid iteration, and I've also worked on high-quality systems where long-term maintainability and reliability matter the most.
Even worse, I got contacted through YC Jobs (workatastartup.com) with a message that was basically: "Star, fork, and submit PRs to our open-source repo and we'll review you for a contract."
I immediately realize it's engagement farming + free labor. I said "No thanks."
Got this reply: "(...) I'm looking forward to reviewing your PRs. Feel free to share me any of your questions. (...)"
Apparently, no one read my reply - not even AI. They are automating this shit. It's sad that many fall for it (check their Github repo)
Google is terrible at marketing, but this feels like a big step forward.
As per the announcement, Gemini 3.1 Pro score 68.5% on Terminal-Bench 2.0, which makes it the top performer on the Terminus 2 harness [1]. That harness is a "neutral agent scaffold," built by researchers at Terminal-Bench to compare different LLMs in the same standardized setup (same tools, prompts, etc.).
It's also taken top model place on both the Intelligence Index & Coding Index of Artificial Analysis [2], but on their Agentic Index, it's still lagging behind Opus 4.6, GLM-5, Sonnet 4.6, and GPT-5.2.
Gemini consistently has the best benchmarks but the worst actual real-world results.
Every time they announce the best benchmarks I try again at using their tools and products and each time I immediately go back to Claude and Codex models because Google is just so terrible at building actual products.
They are good at research and benchmaxxing, but the day to day usage of the products and tools is horrible.
Try using Google Antigravity and you will not make it an hour before switching back to Codex or Claude Code, it's so incredibly shitty.
That's been my experience too; can't disagree. Still, when it comes to tasks that require deep intelligence (esp. mathematical reasoning [1]), Gemini has consistently been the best.
What if the most interesting finding ends up buried under a vague title? Aside from the "self-generated skills" aspect, there isn't much there that meaningfully warrants deeper discussion.
I chose a title that directly reflects an interesting finding - something that offers substantial insight to the community. I think the rule should be applied with some nuance; in this case, being explicit is a net positive.
I have no interest in linkbait, and I hope that's evident from my previous submissions
Thanks @dang for moderating! This is indeed not our original findings and this is a sub conclusion for an ablation we did to remove the confound of LLMs internal domain knowledge. Thanks for submitting for us @mustaphah here's a little bit more details on how we approach this:
> I would frame the 'post-trajectory generated skills' as feedback-generated skills, so is Letta: https://www.letta.com/blog/skill-learning. We haven't seen existing research or hypothesis debating whether the skills improvement might come from the skill prompt themselves activated knowledge in LLMs that can help itself. So that's why we added an ablation of 'pre-trajectory generated skills' because we have that hypothesis and this seems a very clean way to test it. Also it is very logical that feedback generated skills can help, because it most certainly contain the failure mode of agents on that specific tasks.
Yeah, I got your point when I read the paper. You're essentially controlling for "latent domain knowledge."
I might have been a bit blunt with the title - sorry about that, but I still think it was a good title. From what I've observed, a lot of Skills on GitHub are just AI-generated without any feedback or deliberative refinement. Many thought those would still be valuable, but you've shown evidence otherwise.
no worries it's totally fine! there is indeed work needs to be done on the feedbacks generated skills. Thanks for helping us submitting on HackerNews. And for
> a lot of Skills on GitHub are just AI-generated without any feedback or deliberative refinement. Many thought those would still be valuable, but you've shown evidence otherwise.
we do find most skills on the internet to be useless, and thanks to the generosity of https://skillsmp.com/ author, we were able to get all the meta data of the 99k skills indexed on his website. We did a lot of filtering and deduping and we discovered ~40k+ skills were relevant at the time we did the study.
Yes, I appreciate that, and yes there is room for nuance. But I think you went too far in this case, meaning that the delta between the article title and the submission title was too large. For example, the word "useless" appears nowhere in the article abstract nor in the article body. That's a big delta.
I was starting to type out a longer explanation but I ran out of time - however, I probably would just be repeating things I've said many times before, for example here: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que... - perhaps some of that would be helpful.
You're a fine HN contributor and obviously a genuine user and I hope I didn't come across as critical! From our side it's just standard HN moderation practice. The way we deal with titles has been stable for many years. It isn't entirely mechanical, there are many subtleties (back to the nuance thing) but the core rules have served the site really well. THe main thing we want to avoid is having the title field be a mini-genre where whoever makes the submission gets to put their spin on the article.
https://www.psypost.org/scientists-just-discovered-that-a-hi...
reply