> as you transition to death those barriers are less necessary as the body diverts increasingly scarce resources to surviving just a few moments longer.
The more I think about survival based evolutionary explanations for NDEs, the less sense they make.
Obviously evolution is true, and there's an obvious relationship between the physical degradation of the body and the brain and hallucinations. I'm not trying to make a cheap appeal to mysticism and deny these things, but NDEs are profoundly weird and difficult to explain when you think about them from that angle.
Why would a body motivated to survive at all costs waste resources creating comforting hallucinations with some kind of internal coherence during catastrophic failure? Wouldn't a more logical and theoretically sound failure mode for a body trying to survive at all costs be some kind of increasingly incoherent descent into something like TV static as resources get diverted from sense making to repairing systems critical to survival? Or just pure unconscious blackness as with general anesthesia? If consciousness is purely computational, then any coherent internal experience implies the brain is spending biological resources maintaining the physical integrity of something, despite being increasingly severed from the sense input it needs to actually navigate the world to survive. And if the body is going through the trouble of maintaining some level of internal consciousness as it nears death, why wouldn't it simply create a hellish ever increasing amount of fear and pain until the moment of complete physical death to create the strongest possible motivation to avoid ever repeating the experience?
Many people who experience NDEs and survive report craving a repeat of the experience and losing their fear of death. That's profoundly counterproductive from a survival standpoint. There's an argument about group related benefits and a need to offset communal panic due to our knowledge of our own mortality that's easier to ground in a purely survival based explanation, and while that definitely fits better, I increasingly get the sense we're trying to overfit evolutionary explanations that assume a purely survival oriented computational theory of consciousness to things we don't actually understand nearly as well as we think we do, and that the fear of not knowing can just as easily be used to argue motivated reasoning for appealing to things we basically understand like computation and biology as the fear of death can be used to argue motivated reasoning for appealing to things we don't.
The history of all human knowledge is defined by an increasing ability to transcend and expand our theories to incorporate ever more detailed knowledge about previously unseen things. First we were convinced the world was made of unseen animal spirits, then we were convinced it was made of unseen combinations of the four elements in a world governed by a pantheon of superhuman deities, then we were convinced it was made of a hierarchy of unseen forces interacting with seen forces through God given, rationally discoverable natural law, and now we are convinced it is made of purely physical rules which may or may not be fully comprehensible given what we can observe, and those seen and unseen purely physical forces created complex biological systems that can model the world through different types of computation, some of which we understand, and some of which we don’t.
I think it's extremely unlikely that we've figured out the final and most comprehensive framework for understanding reality, and I think there’s a lot about conscious experience and our ability to meaningfully perceive and categorize things that are still deeply mysterious/poorly understood.
EDIT: Didn’t like my previous wording of this/changed it, and still don’t feel like I’m doing justice to what I’m trying to get across. Iain McGilchrist’s “The Matter with Things”, John Vervaeke talks about “Relevance Realization”, Freeman Dyson’s lectures/books about the importance of “heretical views” for the expansion of knowledge, Donald Hoffman’s work claiming evolved models of reality in every kind of environment never create accurate maps, and just observing how difficult the alignment/verification problem is in AI are all pointing in this same direction, and make a more compelling case for what I’m trying to say than I can.