A nice shift in perceived colour between central and peripheral vision. The fixated disc looks purple while the others look blue. The effect presumably comes from the absence of S-cones in the fovea. From Hinnerk Schulz-Hildebrandt: arxiv.org/pdf/2509.115...

Ted Chiang: "The task that generative A.I. has been most successful at is lowering our expectations, both of the things we read and of ourselves when we write [...]. It is a fundamentally dehumanizing technology because it treats us as less than what we are: creators and apprehenders of meaning."

To me, this is theoretical (=scientific) *psychology* at its best. Thanks to these authors, we now know some new, fairly universal (abstract), undeniable (proved), and useful facts about *affordances*.

The term "gold standard" is especially popular in fields like psychology, where researchers are often unable to articulate valid reasons for taking their claims seriously. Mainly because they are mathematically illiterate. It is essentially an appeal to the "authority" of the research community.

Social scientists are funny people exhibit 669

Aka "I think (about myself), therefore I am."

There seems to be a deep, meta-structural analogy between the notion of thing-in-itself (ding an sich) and the notion of a set that contains all sets as its elements, I think; in both cases, any attemp at an explication breaks at the metalinguistic level. In particular, ...

facts

For Monty Hall problem aficionados: a #causal DAG, with explanation in the alt text. Who knew that the M in M-bias stood for Monty?

I’m gonna keep saying this until i run out of breath: nobody ends up a fascist because that’s where the best arguments lead the brightest minds. People end up fascist because it gives them the permission structure to engage in the cruelty and domination they already like!

The True Blood pilot aged well!

New paper out with @boryslaw.bsky.social 🥳 In which we sketch out how to rethink measurement invariance causally for applied researchers. And provide a causal definition of measurement invariance! www.sciencedirect.com/science/arti...

The authors "conclude that psychometric standards must be sufficiently rigorous to distinguish genuine constructs and associations from methodological artifacts that can otherwise pose a serious validity threat." Which sounds great, but is typically *impossible* to achieve in psychology 1/5

Recently came across @filiplajszczak.bsky.social's series formalizing a 1967 math textbook in #LeanLang. Designed for those "with no prior experience with formalization" - nice bridge for newcomers to #LeanProver! Read the intro post here: filip.lajszczak.dev/lean-4-with-...

This is probably as good an opportunity as any to point out that ultimately, value judgments are informative only about the person who holds them.

Science is about freedom of exploration within the bounds of physical and conceptual logics.

An introduction to Lean 4. ~ E. Cosme Llópez, L. Gong. www.uv.es/coslloen/Arx... #ITP #LeanProver #FunctionalProgramming

We need to stop using a framework of 'ChatGPT killed someone', it feeds into the myth of the system's agency & OpenAI's framing. The right framework to use: OpenAI is continuing to perpetuate an indefinite industrial disaster. The oil pipe is not at fault when it breaks. The company that built it is

If it turned out that, in addition to all his other amazing traits, Trump is also a pedophile, it would be difficult to even imagine choosing a worse person for president.

That's literally every institution ever.

All statistical models are wrong, but some are stupid

To me, this looks like a book that most scientists, almost regardless of the field they represent, could benefit greatly from reading!

That's because a lot of them are fine with fascism. Being right-wing is a dangerous form of intellectual disability, strongly related to racism, conspiracy theories, and anti-intellectualism. It may lead to fascism and genocide. Right-wing people need our help, and we need to help them.

Peer review is meaningless, period. A few people have a brief exchange or two, and a decision is made to label a paper as "scientific enough" or not. This is silly. It keeps happening bc researchers are silly. And they are silly because they are people. We are at best locally and temporarily smart.

2 years ago today

Do you also roll your internal eyes whenever you see "neuroscience of X," where X is some high-level functional notion like decision making, memory, motivation, decision making, thinking, attention, or utter annoyance?

Check out these religion resources: Size of religious groups (interactive table) www.pewresearch.org/religion/fea... How the global religious landscape changed recently www.pewresearch.org/religion/202... The process of religious decline www.nature.com/articles/s41...

As a totally separate point: the effect of choice didn't replicate in humans in a large-scale preregistered replication! And the Cohen's ds in the original paper range from 1.3 to 4.17. journals.sagepub.com/doi/full/10....

I must have missed consciousness summer! And autumn, and spring. I would love to learn about ONE thing that we discovered about consciousness proper and not (functionally defined) awareness of this kind or another through research so far.

This article just reminded me of this old tweet

About that - there is no subject. + unlike e.g. memory, consciousness is not a functional notion: There are many memory tasks, but there is no task that *by definition* requires consciousness. So, being conscious is a *physical* property, out of reach of AI or psychology in general. You're welcome.

I looked at the paper that the members of a lab I am also in were supposed to discuss today. I thought it might be fun to share one observation. The first sentence of the "In brief" section immediately tells you it would be great if the authors found a different hobby: ...

Actually in psychology, the real reason we do most replication studies is to waste public money.

You may see a glimmer of hope; I see mere thousands.

Even when it is adaptive, dependence on past events is not yet memory, and if it is memory, the notion is very weakly defined. Just like mere resonance is not selective attention.

For all of the educators trying to deal with AI, I did a deep dive in early June and these are the four articles I decided to assign to my students. 1. This is the longest of the bunch, but so wonderfully comprehensive. www.technologyreview.com/2025/05/20/1...

I love how tons of people suddenly stumbled into the philosophy of language, completely unprepared. Not everyone dealt with it well, but overall it's an overlooked golden era, isn't it? Like, wtf is language, and _why does it work_? What does it encode, under the surface?? HOW?...

Perhaps the most shocking thing I've learned these last four years is that text is this occult spooky medium that becomes creepier and more mind bending the longer you think about it. It's clear what audio and video are, they have clear intuitive physical interpretations, but what the hell is text?

That's exactly right! And the same applies to causal inference.

Folks, folks! Just like thinking, science always was, and always will be metascience.

It’s always been tremendously important but I’d argue that, in the era of ChatGPT, the value of being able to actually write well will be even greater. Clarity of thought and clarity of argument are different but related, and often the best physicists are also among the best explainers of physics.

My autoethnograpic paper on conversations with my 3 yo daughter about whether she's allowed to watch "just one more episode" of Gabby's Dollhouse is finally published

What is the fundamental limitation of attempts to replicate the results of most studies, and only requires thinking for a moment to become obvious, for 100 points.

18. Many academics underestimate the degree to which their political tolerance for immigration is based on it being low-skilled. When there was talk of getting foreign academics to Germany these days many were immediately against it saying we have not enough funding for our own. Figures.

How are people still discussing free will here? (Yeah, I know how) Just because we do not know how something is determined does not mean it is not determined. It is thus impossible to know that something is not determined. You're welcome.

This paper is making the rounds: arxiv.org/abs/2506.21734 A tiny (27M) brain-inspired model trained just on 1000 samples outperforming o3-mini-high on reasoning tasks. #MLSky 🧠🤖

How is fascism possible in the USA? How can wars still happen? We are forever blind to our own nature.

What a fantastic accomplishment -- and what a fantastic story! www.quantamagazine.org/at-17-hannah...

One thing you can count on now is it’s going to get worse before it gets even worse.