Rabbit hole · 3 connected questions
How do assumptions about what counts as an observer (observer-moments), and whether consciousness is substrate-independent, concretely affect probabilistic inferences about being simulated, about cosmological selection effects, and about personal identity across duplicates?
How these converge
All three topics hinge on the same specific issue: when we reason probabilistically about worlds, simulations, and cosmology we must decide which entities to count as observers and how to assign measure to their experiences. That decision depends on a substantive metaphysical claim about consciousness (is it realized by any sufficiently similar computation or only by particular physical substrates?) and technical choices about the unit of observation (observer-moments, persons, or physical systems). Those choices are not mere verbal preferences or abstract Bayesianism; they change numerical probabilities (e.g., of being in a simulation, of doomsday-style conclusions, or of which duplicate is ‘you’) and thus link the simulation hypothesis, anthropic reasoning, and debates over the physical basis of consciousness into one concrete inferential problem.
Where these converge
Observer-moments as the probabilistic unit
All three debates require choosing a unit to count for probability assignments: whole organisms, momentary experiences, computational instantiations, or physical systems. Whether we weight observer-moments, continuous person-lives, or substrate-specific instantiations changes anthropic updates and simulation odds (e.g., many short simulated observer-moments vs fewer long biological lives).
Substrate-independence determines which physical/virtual systems qualify as observers
If consciousness is substrate-independent (realized by computation), then simulated entities plausibly count as observers and contribute to anthropic measure—raising the probability we are simulated. If consciousness depends on specific physical properties not present in typical simulations, then simulated brains may not generate genuine observers and should be excluded from counting. Thus metaphysical stances in philosophy of consciousness directly alter the force of the simulation argument and anthropic inferences.
Reference-class and measure sensitivity ties anthropic paradoxes to simulation credences
Anthropic reasoning’s sensitivity to reference class, priors, and duplication affects whether we treat simulations as massively increasing observer numbers (and so dominating posterior probabilities). The same technical choices that produce divergent anthropic conclusions (doomsday-like results, fine-tuning updates) also drive divergent assessments of simulation probability—and those choices presuppose positions about what counts as an observer from consciousness debates.
The chain
Keep going: open any topic above to find its own related questions.