Markov blanket

A Markov blanket is the minimal set of variables that renders a target variable conditionally independent of all other variables in a system. In the context of the free energy principle, the Markov blanket of a system is the boundary that separates its internal states from its external environment — the set of sensory and active states through which the system interacts with the world. The concept, borrowed from Bayesian networks and graphical models, becomes in Friston's framework a metaphysical claim: the boundary between an organism and its environment is not a physical membrane but a statistical one. A system that maintains its Markov blanket is a system that maintains its identity against entropy. Yet the blanket is not a given; it is a construct of the system's own generative model. The Markov blanket is not a wall but a hypothesis — and like all hypotheses, it can be wrong.