Mícheál de Barra

I've a new blog on health and medicine: golgiapparatus.substack.com/p/2000-year-ol…

It's a myth that egg freezing doesn't work. It works extremely well for women who freeze young. It has low success rates for women in their 40s and late 30s, when fertility has already declined significantly. - Women who freeze enough of their eggs in their twenties have the same success rate using those eggs later as they would have had using them fresh in their twenties: 85-90%. -Women generally freeze too few eggs and too late (median age: 37). This is why overall success rates reported in papers are low. - Women's fertility does not drop off rapidly after age 35. That's a myth caused by faulty data. The decline is earlier and more linear. - Clinics in Spain are significantly cheaper but just as good or better than British or American ones in success rates. I got my eggs frozen in Valencia last week. - Clinic choice matters a lot. Average success rates can vary between 25% to more than 60% probability of live birth per embryo transfer for the worst and best clinics, respectively. worksinprogress.co/issue/were-fre… @_revoluzia_ and I are both in our late 20s, and both decided to get our eggs frozen, so that we could definitely have the number of children we wanted, regardless of where life takes us. Recent technological improvements make egg and embryo freezing an effective 'fertility insurance'. We share our lessons from the process in a new article for Works in Progress.

Today we’re launching the Prolific 100% Human Guarantee. If an AI agent is detected in your Prolific study, you'll get twice the cost of that participant response back. @Prolific’s bot authenticity checks have been doing AI detection work for a while now, tested against major detection methods and achieving 100% accuracy. But it's the years of investment and 50+ checks before participants even see your study that make us especially confident. Read about our guarantee here, I would love to hear any other feedback: prolific.com/100-human-guar…

I don't think the advice is awful, but it is worth noting that the extent to which it is 'evidence-led' is overblown. The recommendations are based off a literature review where most of the studies are non-causal, cross-sectional, and heavily confounded. Honestly, we really don't know how bad (or not) screen time is for kids. The advice itself is informed by the review, but it is highly precautionary. At every point, they err on the side of caution. In my view, when the Govt puts out advice they should be careful to communicate when we're dealing with a precautionary best-guess, and when we're dealing with 'vaccines stop this virus/smoking causes cancer' tier information.

Isofix killed the 3 child family

Italian Inquisition was so worried about people publishing bad science, they set up their own laboratories to repeat experiments and censor studies that failed to replicate. Some very contemporary concerns in the 1600s.

@NickJF75 The cars were visibly dusty this morning here in Oxford. I wondered if it might be from the Sahara. Wild!

High res visible satellite pass this morning shows #Saharandust plume across SE/E England & near continent vs much cleaner air across the north & west With rain moving north, there could be some dirty dusty deposits on cars in SE England, so hold off washing the car!

@doctorveera V cool. Have you spoken to anyone with the variant? Any idea how they think/feel about smoking? Is it nasty or just not addictive?

Link at article: Rajagopal et al. Nat Comm 2026 nature.com/articles/s4146…

Excited to share one of my favorite genetic discoveries made at the Regeneron Genetics Center. We went looking for genetic clues about why some people smoke more than others and found something in an unexpected place: the genomes of Indigenous Mexicans. 1/

Every time I post about falling fertility, someone replies: “Great for the planet.” I understand the intuition, but it gets the economics almost exactly backwards. To be clear: I am not arguing for explosive population growth. A gentle decline or stabilization would be my first choice. The problem is that we are not heading toward a gentle decline. We are heading toward a collapse. And a collapse changes everything. Environmental protection behaves like a luxury good. As countries become richer, citizens demand cleaner air, cleaner water, and stronger climate policy. Prosperity creates both the willingness and the fiscal capacity to pay for these goods. This is not a theoretical curiosity. The modern environmental movement was born in California in the 1960s, when the state was among the richest in the richest country on earth. That was no coincidence. You need to be prosperous before you start worrying about the spotted owl. A sustained fertility collapse works in the opposite direction. As populations age, pension and healthcare costs rise while the tax base shrinks. Governments under that kind of fiscal pressure protect mandatory spending first because that is what voters scream about (I am from Europe, and I can tell you this is the case with 100% certainty). Environmental investment, which is largely discretionary, is the easiest to postpone. And it will be postponed, particularly in middle- and low-income countries. Environmental policy is not a costless virtue. It requires administrative capacity, long planning horizons, and resources. Lots of resources. Decarbonization alone demands trillions in public and private investment over the coming decades. Where will that money come from if the working-age population is shrinking and the dependency ratio is exploding? If demographic collapse erodes prosperity and fiscal space, and the evidence strongly suggests it will, it will not increase environmental investment. It will make it harder to sustain. So, if you care about the environment, I am sorry, but what is happening with fertility right now is terrible news.

About 2% of children in the UK have elevated blood lead levels. I made an interactive map (with a little help from Claude) so you can see the % in your local area. lcrawfurd.github.io/lead-map/

@phl43 Yes, people seem very focused on issues like 'peer review being overwhelmed' rather than the opportunities to achieve years of progress much more quickly.

Based on my personal experience, using mostly Codex with GPT 5.3-Codex, AI is nowhere near the point where it could actually produce a good paper with that kind of prompt. But if you give it a detailed prompt explaining what you want it to do and iterate on the results for long enough, by asking questions on what it did and asking it to revise the parts you think are bad, it can already speed up the process of writing a paper considerably and take care of most of the grunt work involved in that. This is already incredible, but it's improving very rapidly and I think we're probably not very far from the point where it can one-shot a paper with minimal instructions, though I'm guessing humans will continue to play a crucial role for a while because they will still be needed to tell AI what to do. But I think the level of details required in the prompt will decrease quickly. I've read someone saying it's terrifying, but honestly I find that very exciting. Of course, it will have a seriously disruptive effect on academia and force it to reform in fundamental ways, but the eventual outcome seems more exciting than terrifying to me.

Ireland's love for a rural bungalow shows up well here:

Interesting convergence with @dansperber and Mercier's social models of human reasoning: cambridge.org/core/journals/…

@paulg @pablogguz_ Reminiscent of this famous graph in evolutionary anthropology, from unm.edu/~hkaplan/Kapla…

This graph from @pablogguz_ of Spanish fiscal contributions by age demonstrates an important fact. The 40 and 50 year olds are carrying both their kids and their parents. And not just economically. This is why it's hard to be middle-aged.

@BeatrizGietner And viagra is made in Ringaskiddy, Co. Cork.

i had no idea about this edition.cnn.com/2025/08/14/eur…

@simonsarris Yea, my two (soon three) kids have 9 cousins whom they adore. It's gives lots of freedom the parents as they are great at occupying one another.

They can go to my grandparents house in Maine, like I did. They can go to the same place, a farm which is in many respects quite magical. Only now its empty of children.

Same problem. I have three children, they have zero cousins. Not even second cousins. Zero. It's simply not possible to give them the Christmas party childhood I had.

@tomhfh @10DowningStreet Voters' economic misconceptions are perhaps the more important problem.

@10DowningStreet Please can someone at HMT teach the prime minister about this graph?

No more rip-off ticket prices. We will make it illegal to resell tickets for concerts, theatre, sport and other live events for more than their original cost. So fans have the chance to enjoy world-class events at the right price.

a biased bias illustration.

I had a go at improving it.

@acerbialberto Congrats Alberto!

🎉

Here are the X account for the new Leverhulme-funded project I’ve just started working on with colleagues at the University of Edinburgh. We will investigate, from a cultural evolution perspective, the spread of heritage-based hostility on social media.