That's cool. Could it also be that when they pursue long-term goals the goals are highly unrealistic, dream-like even, because otherwise, they do not feel the subjective value of delayed rewards strongly enough?

... This latter statement can be viewed to refer to an *asymptotic* property of the CI viewed as a *decision rule* as long as the critically important "in such cases" part is interpreted as capturing the hypothetical infinite process of making claims by a hypothetical (& immortal) sampling agent.

... We cannot say:

"the true value is in the CIs with probability 9%"

but we can say:

"the statement 'the true value is in the CIs' is true with probability 95% *in such cases*" ...

... it is not something we can just express in a formal language. One can view the strange tension between how we want to view the CIs and what we can say about the CIs as long as we stick to the technical language as a special case of this. In particular: ...

... I say *the agent doing the reasoning is the golem* (!). Not only should we not believe in the statistical models we use, we also *cannot* know if a prior distribution captures our uncertainty because our *actual* uncertainty is a *mystery we can only attempt to study* (by reasoning about it);..

... about a hypothetical agent doing the reasoning in a hypothetical (and extremely simple) world. That's because we know that the model is an (extreme) idealization. @rmcelreath.bsky.social coined the great term "golem" to emphasize this aspect of statistical inference; ...

I knew we were on the same page, so I hadn't read the blog post before, but now that I did (which was fun!), I would like to share the thoughts it inspired: One can think of all forms of statistical inference as an exercise in *meta*-reasoning since statistical inference is virtually always ...

Universal first-name basis is a great and wondrous gift that the Nordics are offering to the world, that we've been impatiently waiting for like half a century for them to pick up.

It was fun, thanks!

You would have to do much more to connect what I am saying to some notion of truth. Being a product of behavior is logically independent from being a sentence, let alone a true sentence. And yes, psychology is the only science that studies itself. Because it is the fundamental science.

This is what I am saying: Science *as such* and the contents of scientific theories *as such* are a product and an instrument of behavior.

This is neither here nor there. It is one thing to be foundational and quite another to be well-developed. Psychologists are mostly quite dumb. This does not change the fact that what they are *trying* to study is the only foundation of all science.

Self-awareness is not the most important psychological notion, *behavior* is.

Thermodynamic constraints on biological architectures is also something *we can talk about*. Psychology is the fundamental science.

There is a more general point to be made here. There is no such thing as an ontological property that does not have an epistemological aspect simply because a property is already something we can label as property, which means it is something we, humans, can consider.

It wasn't physics, it was reality. Physics is a scientific discipline. It did not exist before someone started thinking about it and it does not operate in any literal sense.

Can you draw some hard to accept consequence from my claim? If this is indeed a mistake to claim that then it should lead to some kind of absurdity, no?