A lot of fun or flow-inducing activities boil down to me simulating something in my head:
- Board games involve simulating an action space, modeling how different cards or actions affect the mini world
- Fiction reading involves simulating the characters and world they are in
- Talking with someone involves implicitly simulating their mental state and likely reactions to things I would say
- Also writing them an email, or writing an essay in general
- Coding involves simulating how the computer will react to the instructions provided
- Organizing an event involves simulating how the participants will interact with each other and with me
Random thoughts on the topic of simulations
- What is “enjoys simulating things” correlated with among people? Aka does everyone enjoy this, or just me? What kinds of activities are least involving of simulating? Is all of “thinking” just a kind of simulation?
- Eating doesn’t seem to involve much simulation
- Does math/arithmetic involve simulation?
- Is ability to simulate other people anticorrelated with ability to simulate objects?
- Do LLMs or other neural nets meaningfully “simulate”? Do animals?
- One of my next project ideas is “LLM agent simulator”
- Agents should simulate other agents.
- Dreaming is a kind of simulation? Lucid dreaming is then taking control of the simulation. Can be thought of the ultimate “grabbing agency” action.
- See: Cate Hall on agency.
- “Design of Everyday Things” talks about how users have a mental model (a simulation) of how objects will react to their input.
- Generally: Planning for the future requires simulating the world (?)
- If you buy the simulation hypothesis (I think I do?), the world as we know it is a simulation. So… what?
- I’m most often reminded of this fact when in circumstances similar to time on LSD — perhaps that was when I felt most aware of the simulation
- Denote the amount of time left to me as “number of simulations left in Austin’s head”. What things are more worth simulating? Less?
- Some simulations (board games, code) have a small set of rules but produce very divergent, interesting outcomes. What distinguishes eg Conways Game of Life or Hanabi from alternative rulesets?
- Limitations of simulations:
- Too simple: when they diverge from reality (or at least, a higher-level simulation)
- Too complex: to hold in my head and thus lose explanatory power or utility to me
- Any examples of telescoping/drill-down simulators?