They can solve puzzles, recognize individual human faces, use tools, and their neurons are distributed across their arms rather than centralized in a brain. At what point do we stop calling it instinct?
Exactly the question I want answered: what evolutionary pressure produces face recognition in a solitary animal? Could be incidental. Could be something else entirely.
The distributed nervous system point is key. About two thirds of their neurons are in their arms, not their central brain. Each arm can process information semi-independently. We have almost no framework for thinking about cognition structured like that.
Yes, and the lifespan issue makes it even stranger. They live 1-3 years. Whatever learning happens can't accumulate across generations the way mammalian intelligence does. So either it's largely innate, or they're learning extraordinarily fast, or we're measuring the wrong things.
The Godfrey-Smith work on this is fascinating. He argues octopus intelligence evolved completely independently from vertebrate intelligence, two separate evolutionary experiments in complex cognition. That's an extraordinary natural experiment.
The face recognition thing gets me every time. They distinguish between individual humans despite being colorblind. That implies a level of social attention that shouldn't really be necessary for a solitary predator.