Stefan Schubert
Effective Altruism and the Human Mind (with Lucius Caviola) is available for free at: academic.oup.com/book/56384
For physical and audiobook versions, see: stefanschubert.substack.com/p/physical-and-audi…
- [Not loaded yet]
- [Not loaded yet]
- In my opinion, people should be saying "Cheers!" when they have ice cream
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- Kidney stones really are a bitch. Wouldn't wish them on my worst enemy. Well maybe my worst but not on anyone I merely disliked.
- [Not loaded yet]
- [Not loaded yet]
- I want to talk population panic & understanding demography. New estimates from @ncfmr-bgsu.bsky.social on completed fertility in the US show that, among women 45-50 in 2024, about 15% were childless. Two-thirds of women had 2 or more children. 1/ doi.org/10.25035/ncf...
- In 2024, the average number of children ever born (CEB) for women was 1.92 at ages 40-44 & 2.00 at ages 45-50 (not in the profile, something I calculated using the same Current Population Survey (CPS) data). For comparison, the Total Fertility Rate (TFR) in 2024 was 1.63. Why the difference? 2/
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- 🚨 REPLICATION REPORT UPDATE: One year ago, a tweet by John Holbein alerted me, @ollefolke.bsky.social, and @jopieboy.bsky.social to a paper with a shocking result about Sweden’s law criminalizing the purchase of sex.🧵
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- Paper finally published in JEP General! Individual variability in mental imagery vividness does not predict perceptual interference with imagery: A replication study of Cui et al., 2007: osf.io/preprints/os... A study cited over 500 times couldn't be replicated, even in extreme imagery... a 🧵 1/8
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- Teens & 20-somethings think 30-40 year olds are uncool? Wow this is a brand new phenomenon.
- the Champions League semi-finals are the highlight of the football season for me. In terms of quality and drama you can’t beat it. And this is the most exciting quartet of sides at this stage for many years. It’s going to be excellent.
- Four really great games. There’s a huge amount of negativity about the current era of football (in terms of off-field stuff, much of it justified) but the technical/tactical level of the top-level matches is outstanding.
- [Not loaded yet]
- [Not loaded yet]
- Is there a half-life for the success rates of AI agents? I show that the success rates of AI agents on longer-duration tasks can be explained by an extremely simple mathematical model — a constant rate of failing during each minute a human would take to do the task. 🧵 1/
-
View full threadIn summary, the METR data on agent performance for different duration tasks is well fit by an exponential survival curve and a constant hazard rate per unit of human-equivalent time. This is also suggestive of the underlying process being a sequence of filters. 20/
- Note that the usual caveats of the METR data apply: it is not clear whether their task suite is truly representative of the kind of tasks needed to to AI research, nor whether their results hold for other kinds of tasks. 21/
- Book marking this to check in November. When I will either feel smug or silly.
- [Not loaded yet]
- [Not loaded yet]
- A 2021 paper in a pay-to-publish #Nature journal claimed #mindfulness tasks (notice 3 things around you or differences between 2 similar images) reduced some cognitive #biases: doi.org/10.1057/s415... The mindful #debiasing was undetected in a re-analysis and 2 replications: doi.org/10.1007/s126...
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- what a complete idiot.
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- [Not loaded yet]
- Aspiring writers! If you've thought about starting or restarting a blog, want to be part of a writing community, get help with editing & learn from other authors, how about applying to Roots of Progress's fellowship? Friends who took it last year highly recommend it, and I'm one of the advisors!
