The computational self: location in space, time, and possibility

L.A. Paul (Yale, Philosophy and Cognitive Science)



To think for yourself, you need to be able to solve new and unexpected problems. This requires you to identify the space of possible environments you could be in, locate yourself in the relevant one, and frame the new problem as it exists relative to your location in this new environment. Combining thought experiments with a series of self-orientation games, I will discuss collaborative work that explores the structure of how we perform this computational task. In particular, we propose that the way humans exploit their ability to center on themselves as physical agents in the real world mimics the way we solve what we call "the avatar problem" in a First-person game. In real life, human agents perform the very same computational task, triangulating their perceptual and proprioceptive inputs with different third person representations of themselves to represent themselves as an embodied agent in the world.


Understanding this computational feat is relevant for AI researchers trying to build an artificial agent that can truly think for itself and to philosophers interested in centered worlds epistemology.