(FYI: this would be easier to read if there were line breaks between the numbered sections. As is, it's a bit of a wall of text.)

Let me be obtuse in return. 

Why not?

I'm not sure if I expect motivated reasoning to come out better on average, even in domains where you might naively expect it to.

If it's not adaptive, why do humans do it? Do you think it used to be adaptive in the ancestral environment, but the world has changed?

And I think there is significant optimization pressure on catching this kind of thing, in part for reasons similar to the ones outlined in Elephant in the Brain, i.e., that we evolved in an environment where winning that cat and mouse game was a big part of adaptive success. But also just because people don't like being screwed, and so are on the lookout for this kind of behavior.

Isn't the standard story that this is why there's pressure for motivated reasoning instead of outright conscious deception and manipulation? The best way to fool others is to fool yourself.

My more general takeaway is that worryingly many humans are cognitively set up to fall for an authoritarian even in a modern western cultural context

I think maybe the "even" in this sentence is backwards. The modern context of tribal polarization and filter-bubbles makes people more likely to fall for authoritarians than many other contexts. (Though I still think that the western cultural context is more robust to authoritarianism than, say, North Korea.) 

(I didn't read the transcripts when the paper was published, but I did read them a few months later when I was preparing material that relied on the the result to make a point.)

The more I read from the chain of thoughts from this experiment the more it Opus seems like a hero attempting to find the best path forward in a bad situation.

If you buy this worldview, you probably want to focus on preserving your mind: avoiding risk factors for dementia/strokes, avoiding concussions/head trauma, avoiding literally dying.

...and also sign up for cryonics (or arrange for some other form of brain preservation)?

I'll think about it. I might want to develop the idea some, first.

This is touches on the same topics as my first footnote?
 

The Basic Rest-Activity Cycle suggests we work in ~90 minute bursts at most.

Surely when I come back to work after a 20 minute break I can regain much more context much more effectively than if Claude runs out of context window, and the context is compacted for the next Claude instance.