I did a quick Google search and there were quite a few results that were in support of (specifically) by-product theory (see here, here, or here). So, I don’t know — your work was the first time when I saw religion presented as primary evolutionary adaptation.

I agree with you that by-product theory as you presented it (especially that analogy with appendix) is not what’s going on. Therefore, I liked your ideas about metarepresentations and that’s also why I pointed out that paper on superior pattern processing (SPP).

My current understanding is that evolution of brain led to emergence of SPP which gave rise to symbolic thinking and then metarepresentations (maybe the two or even all three can be considered synonymous). These, when combined with relics of animal ritual behaviors and intellectual attempts to come to terms with new knowledge (such as awareness of mortality), resulted in what I would call “superstitious habits”. This appears to took place sometime during Middle Pleistocene (400–300k BP — first burials of Homo naledi) and culminated with emergence of art and complex (modern) language (see oldest cave paintings that are at least 65k years old — Neanderthal art discovered in Spain — and laryngeal anatomy of H. sapiens).

The first problem I see is that all current explanations dealing with roots of religion (without fail) label these pre-religious “superstitious habits” as bona-fide religion, while this is clearly not the case (I elaborated on this in my previous replies — you may believe that animism is a religion, but it’s an incorrect understanding nevertheless). The second major problem is that everyone is more or less focused on the era “around 50,000 years ago, [when] Homo sapiens made a cognitive leap that gave us an edge over other species.” Yet, it’s obvious that there was neither a sudden “cognitive leap” (it was rather a prolonged process — archaeology merely reveals the end results) nor was this development restricted to Homo sapiens (at least Neanderthals were in the same boat too).

Therefore, it’s not surprising that it took another c. 40k years and a neolithic revolution for religions to appear on the historic scene.

So as it seems, while there is absolutely no evidence that religion is a primary evolutionary adaptation, there is plenty of suggestions that point to religion being a higher level (social) construct of relatively recent origin. And there is also another very strong argument for this being so, something that I still didn’t mention — the very existence of atheists and non-religious people. Sure, their existence says nothing about the origins of religiosity, but if religion was, indeed, a primary evolutionary adaptation, atheists are (if nothing else) evidence that this adaptation run its course (thus “evolutionary jury” ruled). Then again, I’m inclined to think that it’s a rather strong evidence in support of religion being a construct emerging from more fundamental cognitive processes.