Advertisement · 728 × 90

Posts by Will Gervais

Once upon a stranger:
The science of how small talk can add up to a big life

Dr. Gillian Sandstrom

Once upon a stranger: The science of how small talk can add up to a big life Dr. Gillian Sandstrom

HEY EVERYONE, DROP WHAT YOU'RE DOING!

The amazing @gilliansocial.bsky.social 's book is out in the UK today!

Go get it: uk.bookshop.org/p/books/once...

📚💙

3 days ago 4 1 1 0

This "big-N quant research in Nature/Science journals" pattern is a direct manifestation of The Matthew Effect in scientific publishing (beyond misinformation lit). Groups with massive resources dominate glamor pubs burying complexity and nuance under size and impact. Metascience is a great example.

5 days ago 52 13 1 0

Tapping the sign

1 week ago 6634 1881 1 64

My own take on power analyses: I'm good with "we had xx power to detect effects as big as xx" but find "we made a pseudo-random guess about an effect size & determined that we need a sample of N=[some highly specific number]" to be too close to cosplay for me.

4 weeks ago 43 7 3 4

3.14159265358979323846264338327950288419716939937510582097494459230781640628620899862803482534211706798214808651328230664709384460955058223172535940812848111745028410270193852110555964462294895493038196442881097566593344612847564823378678316527120190914564856692

🧵 1/∞

4 weeks ago 1606 304 78 32

I'm also not worried for the same reason that so many scientists don't choose to p-hack or fabricate data just because they can. Literally everyone can do these things. But we're capable of making choices, learning things, having integrity, prioritizing some values over external incentives, etc.

4 weeks ago 22 6 1 0

oh fuck off

1 month ago 4416 555 230 346
Advertisement

Science needs a culture of trust much more than a culture of interpersonal skepticism and suspicion. Verification is important but we cannot base verification systems on extreme distrust. Scientists' cynicism about their colleagues' values, priorities, and integrity leaves no room for social growth.

1 month ago 63 14 3 1

THAMES Time, MFers!!

Episode 3 of the blog series, all about the SIX SECRET LETTERS FOR CRITICAL THINKING THAT THE SCIENTISTS DONT WANT YOU TO KNOW

willgervais.com/blog/2026/3/...

#psychscisky 🧪 #philsci #methodology #teaching

1 month ago 12 7 2 0

The single best horror-comedy Ive ever seen. Its in the history books for the one of the longest one-shots in a film (37 min) and for making 1,000x its budget (cost $25,000 USD, made $30.5 million worldwide)

Masterclass of a film. And do not read anything about it going in.

1 month ago 12 2 1 0

Please do! Hope it's useful!

1 month ago 0 0 1 0

THAMES Time, MFers!!

Episode 3 of the blog series, all about the SIX SECRET LETTERS FOR CRITICAL THINKING THAT THE SCIENTISTS DONT WANT YOU TO KNOW

willgervais.com/blog/2026/3/...

#psychscisky 🧪 #philsci #methodology #teaching

1 month ago 12 7 2 0
Parts of a kiwi bird as fruit

Parts of a kiwi bird as fruit

7 months ago 308 54 10 2

1. New in-press paper on religion's functions, co-authored with amazing lab manager James Nesbit! Full text here, more info in thread below: columbiasamclab.weebly.com/uploads/5/9/....
#AcademicSky #PsychSciSky #SocialPsyc #DevPsyc #CogPsyc @socphilpsych.bsky.social 🧪

1 month ago 16 7 1 1
Video

holy shit

1 month ago 14242 3096 239 607
Video
1 month ago 13954 5271 92 259

Cheers!

1 month ago 1 0 0 0
Advertisement
Preview
Godless Polemics: Atheist Pamphleteering and the Specter of Emancipation in the United States Godless Polemics examines the role of pamphleteering as a medium for radical critique in the intellectual and cultural history of American atheism, focusing on its function as oppositional and counter...

Succinct and a very interesting open access book on atheist pamphleteering in the United States: "Godless Polemics" (2026) by Florian Zappe. www.routledge.com/Godless-Pole...

1 month ago 7 3 1 0

Wilde

1 month ago 1 0 0 0

Surveys that find British young adults have high levels of Christian belief & practice have two things in common:
1. They survey opt-in panels that accept volunteers.
2. They don't make the underlying survey datasets publicly available for researchers to analyze.

1 month ago 140 34 4 4

My Beautiful Bullshitty Baby:

Episode 2 of our blog series on methods renovations. Diving into my Science and Critical Thinking module - layout, rationale, & the first few weeks.

willgervais.com/blog/2026/2/...

Read! Share! Tell us what you think!
#psychscisky 🧪 #philsci #methodology #teaching

1 month ago 18 7 0 1

NEW: Valeria Rainero, Jörg Stolz, Ruud Luijkx, The Faith Factor. How Scholars’ Religiosity Biases Research Findings on Secularization sociologicalscience.com/articles-v13...

2 months ago 10 5 0 3
Preview
The billionaires' eugenics project: how Epstein infiltrated Harvard, muzzled the humanities and preached master-race science Edge - Jeffrey Epstein's favourite intellectual salon - was sold to me as a gathering of the world's finest minds, writes Virginia Heffernan. The files reveal it was something far darker: a decades-lo...

NEW: Read this terrific piece @virginiaheffernan.bsky.social for @thenerve.news in which she unpicks Edge.org, the hugely influential 'salon' funded by Epstein & threaded through with race science & eugenics.
1/

www.thenerve.news/p/epstein-bi...

1 month ago 1861 1061 55 127

On the publication bias discourse, I regret that metascience has become a source of decontextualized, low-res, bean-counting-focused `science is in crisis' narratives. It is largely uncurious abt science, desperately lacking in theory & measurement. I'll quote a few takes I liked & add my thoughts🧵

2 months ago 123 32 3 12
Preview
Petition - Free Civil Legal Support for Survivors of Sexual Violence Support free civil legal support for sexual violence survivors. Help create a fair, accessible path to justice when the system fails to act.

97% of r@pists walk free in the UK. Survivors don't. Let's change that!

This petition calls for free civil legal support for survivors, creating a real route to accountability where criminal justice often fails - and helping prevent repeat harm over time.

therightwordsproject.org/petition-fre...

2 months ago 4 6 0 0
Post image
2 months ago 486 144 4 15

stop using AI to do your research. it hallucinates too often. if you want an answer to something, post something arrogant on the appropriate subreddit. something like: "this item performs 10% better than everything else. only idiots deny this." this will bait nerds into doing your research for you.

2 months ago 17788 3595 145 167
Advertisement

Ooooof. Miss the phenomenon but not so much the experience.

Stay hunkered & safe!

2 months ago 1 0 0 0

Ice storm?

2 months ago 2 0 1 0