Abstract

What can a retro arcade game tell us about the architecture of the human brain? Traditional neuroscience often views decision-making as a series of isolated choices, but in this talk, I propose a shift in perspective. Using Pac-Man, we investigate problem-solving as a structured, hierarchical process. Our analysis of human and macaque gameplay reveals that both species “speak” a language of problem-solving, though with a divergence in complexity. Monkeys utilize simple grammar structures, whereas expert humans deploy complex “sentences” to navigate dynamic environments. Using fMRI, we map this ability to a right-lateralized network that strikingly mirrors the left hemisphere’s language centers. This suggests the brain encodes strategic grammar using the same computational principles found in language. By leveraging complex environments like Pac-Man, we can go beyond standard models of decision-making to reveal how the brain organizes behavior at a higher, more abstract level.