State Space

State space is the set of all possible configurations that a dynamical system can occupy. In physics, it is the space of all positions and momenta of a mechanical system. In control theory, it is the space of all variables needed to specify the system's future evolution. In computational neuroscience, it is the space of all possible patterns of neural firing. The state space is not merely a notational convenience; it is the geometric arena in which the dynamics of a system unfold.

The structure of a state space determines what a system can do. A state space with multiple attractors permits multistability — the system can settle into different stable patterns depending on initial conditions. A state space with chaotic dynamics permits sensitive dependence on initial conditions, making long-term prediction impossible despite deterministic evolution. A state space with a low-dimensional manifold embedded in high dimensions permits the system to explore a structured subset of possibilities rather than wandering randomly.

The state space framework connects to dynamical systems theory, control theory, statistical mechanics, and complex systems research. It is the foundational concept underlying the neural manifold framework, which studies how neural activity is constrained to low-dimensional subspaces within the full state space of possible firing patterns.

State space is not a representation of reality. It is the reality of the model — the complete universe of possible behaviors that the system's equations permit. The choice of which variables constitute the state is itself a theoretical decision, and different choices produce different state spaces with different dynamical properties. There is no unique state space for a given system, only state spaces that are more or less useful for the questions being asked.