the implicit assumption is that the current state of affairs is far from optimal, not that my particular definition of optimal is correct. In fact part of the point is to explore what values we might hold (and still hold after reflection on those values) that would value fiction.

Again, this doesn't feel like it relies on any attribute specific to fiction. You could say about almost any aspect of the world or our activities that it's unlikely to be optimal (whatever your goals may be), and so it's useful to question things - I agree with that. But the rest of your argument does try to be specific to fiction.

I feel this is a valuable exercise largely because when I do reflect on it, arguments to the effect that fiction is something I reflectively value are rather difficult to make.

Speaking for myself, I like consuming (reading, watching) fiction because it's enjoyable in the moment. I'm quite sure I reflectively endorse this as a positive value; that doesn't mean there aren't other things I could be doing with even greater value, but I don't know what they are, and I don't think I can find out by questioning the value of fiction.

I feel like this remark ignores the part before the text you quote ("there's an actual person") which is very much not a fully general argument, but rather an argument against solitary activities which are misleading superstimulus telling system 1 it's achieving things it's not.

Thanks for the correction.

Why or how do you think fiction misleads system 1? When I read a book, I don't feel like I'm imagining being one of the characters, it feels like I'm watching them from the side. When I suffer from loneliness or sadness or depression, it doesn't help to read or watch fiction about happy socially fulfilled people; on the contrary, it sometimes causes me pain because it forcibly reminds me about my problems, and the disconnect between me and the heroes is too great.

I do enjoy "escapism" in the sense that fiction can help me forget, while I'm reading it, about my cares and troubles (except in cases like the above). This simply feels like focusing intently on one thing prevents me from thinking about the other in the background. It's a similar experience, in that sense, to playing a game, holding a mentally challenging conversation, or focusing on a programming problem.

I realize, of course, that I'm describing my personal experiences, nothing more. Yours are different.

There's also a big difference between creative activities (spending solitary time writing a book, say) and consumptive activities. It's certainly possible to spend alone time without the activity being "isolated" in the sense that I mean.

A friend of mine really enjoys drawing. From a young age, whenever she had a minute free she'd sketch something, and she's gotten very good at it by now. She doesn't care much about giving drawings to people, or keeping most of them; she enjoys the process of drawing, composition, etc.

Is that "creative" in your sense? She values the act of creation more than the finished work. But you think the main value lies in sharing the finished work with others. It's not surprising that two people happen to have different values. I'm sure there are others who feel like you, and others who feel like her, or like me. But do you think you're presenting arguments for others to adopt your values? Because if so, I'm not sure I understand them at this point.

fiction feels disproportionately like a good thing (because it's designed to).

That's a feature, not a bug! The value of fiction, and the immediate reason for consuming it - for me and I think for most people - is the direct enjoyment of it, not any second-order effects. Since fiction exists in a market, and consumers choose the fiction they enjoy most, and it's produced by smart people, of course it evolves towards feeling like a good thing. It's not disproportionate because there's nothing to be proportionate to - fiction's stated goal is exactly to make people feel good.

If there is any distinction to be made between what feels immediately valuable and what I'd find valuable on reflection, fiction will tend to optimize for the first.

This tends to be true of any market product, because people are impulsive and irrational buyers who don't reflect much. But for people like me who value fiction mostly for the fun experience of consuming it, reflection doesn't necessarily suggest something else. The conflict may arise mostly if you think you're consuming fiction for purposes other than having fun - because, indeed, it's not optimized for those other purposes.

I think people (not just me) generally develop a kind of "memetic immune system" defense against the types of superstimulus which are present in abundance in their culture.

That's true. I should point out that fiction is a much more ancient superstimulus than e.g. modern supertasty food or superattractive actors. Audiovisual special effects are new, but fiction as text and fiction as theater have existed for millenia.