ǝʞɐlq ǝʌɐp

The 50% of the biomedical science community that survives the next 3 years will celebrate!

No. If you talk to folks that run large labs, the really impressive thing is the number of grants "submitted" per year. They may have marginally higher success rates, but they submit a HUGE number per year.

The right amount of animal pain and suffering is "as little as possible", but regulated use of mammals in agriculture and research rests on similar ethical grounds. It is OK to raise purpose bred animals to benefit the human condition through research provided we minimize suffering.

2/2

In context of other use, purpose-bred animals in research is a non-issue. I mean, in the US, we eat 33 million cows per year, over 100 million pigs, euthanize over 300k dogs and 300k cats, shoot 6m deer, and use about 70k NHPs and 100m rodents in research.

[rodent estimate from my friend LC]

1/

a couple of men are standing next to each other and one of them is wearing gloves and the other is wearing a bandana ALT: a couple of men are standing next to each other and one of them is wearing gloves and the other is wearing a bandana

Very nicely done!

And sometimes the reviewer who finds the impact much lower than the others just sticks to their position. Because stubborn.

The end score is almost always nearly approximated by the WORST score of the three reviewers, after discussion. No matter the discussion, nearly always closest to the worst.

Following the discussion, the difference of
opinion remained, and the panel expressed moderate enthusiasm for this application.

Impact score: 51
From sheets, estimated prelim scores: 1 to 2, 1, 5 to 7, 2

2/2

From a Summary Statement this year:

During the discussion there was a difference of opinion and some of the reviewers considered the application high impact whereas another reviewer weighed the weaknesses more negatively and considered the application low impact.
1/

Exactly.

Inflexible perma-grumps.

That is neither what I said, nor what I implied. I am 100% behind reviewers applying due diligence, and well stated logically defensible critiques behind their scores.

But some reviewers are high frequency outliers, and do not have well stated logically defensible critiques matching their scores.

I think there can be clear differences drawn between people who are occasional outliers, and people who are outliers ALL THE TIME, and I know I have seen both.

I generally think highly of the other reviewers on applications I am assigned.

Sometimes it is clear they did not use due diligence.

Sometimes it is clear they are scoring based on things they will not discuss.

Those can be rated.

I don't think it is bad when a reviewer is an outlier.

I do think it is bad when a reviewer is an outlier at a frequency MUCH higher than other reviewers.

I ALWAYS check whether I am an outlier, and for underlying differences in critiques when I review. I am not an outlier often.

I don't. In fact, any decent reviewer should be an outlier sometimes, we all have different perspectives and knowledge.

But I would maintain that some reviewers are high frequency outliers and they are either poorly calibrated, malicious, not using due diligence, or gaming the system.

I am of the opinion that CSR should, at the end of each study section, have each reviewer rate the other reviewers on their applications. These, and the tendency to be an outlier when scores are spread, should be used to alter the process of selecting reviewers.

Due diligence is under-rated.

As someone who does not live and die in this ecosystem, I wonder if you see what is happening. My perception is that reviewer selection is much more haphazard than 2 years ago and that this is a direct result of massive turnover at NIH. As a result, review outcomes are far more random.

My quote of the day

Science is the search for truth, that is the effort to understand the world: it involves the rejection of bias, of dogma, of revelation, but not the rejection of morality.

Linus Pauling

I find that nothing challenges American exceptionalism like a trip to China to meet with the scientists there.

This is regulatory speed and production cost. But, with a dramatic increase in human trials (China easily has done 10X the US in the last 5 years) there will be inevitable increases in engineering also.

Once they have approved working prototypes, they will get a revenue stream and it's all over.

China's brain-computer interface industry is racing ahead | TechCrunch China’s brain-computer interface industry is rapidly scaling from research to commercialization, driven by strong policy support, expanding clinical trials, and growing investor interest.

The BCI market in China is ahead of the market in the US, and is accelerating away. The brain computer interfaces that ultimately improve the lives of people in the US will be developed and made in China, and the intellectual property profits will also go to China.
techcrunch.com/2026/02/22/c...

I absolutely see your point.

If you calculate how many trials AI takes to learn an association and compare to a human, the AI takes orders of magnitude more to learn. LeCun has written extensively on this. If AI could learn like humans, it would be a huge advance. Until then, AI will use all the energy and water we have.

Before March 3.

She made an inaccurate statement about the US stance in regards to Taiwan and China at an international conference. One that does not reflect US historical policy. And JD Vance pounced on it.

AFAICT, the limits on charge capacity for the metal (not the tissue) still hold (50-150 uC/cm^2). Which, for 4 nC, means the surface area limit is 2.6e4 microns^2 or a square 160x160 microns. For platinum.

Other exotic metals can carry and deliver more charge.

2/2

As a followup, a DIFFERENT guidance is typically used for very small area microelectrodes, that of 4 nC per phase, which is 40 microamps for 100 microseconds.

1/

YouTube video by skolander Green Eggs and Ham narrated by the Reverend Jesse Jackson

The Rev Jesse Jackson.
Oct 8 1941 - Feb 17 2026
May he rest in peace knowing the world is a better place from his actions. My favorite bit from him is comedy, something his children encouraged him to do.

youtu.be/A1mqg4C0awA?...

Weaponizing the FDA Against Vaccines By rejecting Moderna’s application for an mRNA influenza vaccine, RFK Jr.’s FDA is advancing his anti-vaccine agenda.

open.substack.com/pub/pauloffi...
A closer look at the FDA's recent kill shot against an mRNA influenza vaccine.

I don't think he made this decision without running it up the flagpole first.

The FDA does indeed prospectively approve trial design and outcome metrics, and provide feedback on steps necessary for approval.

Halting retrospectively breaks trust in the process.