Representation, time, scale, and neural computation

A theoretical framework and computational methods are presented which address the need for a more mature and focused conception of representation in neural computing. Large-scale models of cognitive and neural machinery have eluded current methods. A means of generalising these methods is discussed, along with some new theoretical ideas, which should enable more inclusive and explanatory models in the future. Preliminary empirical tests, possible extensions, and psychological predictions are considered. The key insight behind the work presented is the place of meaning and time in the foundation of basic theoretical terms, including 'representation', and the importance of thoroughgoing and consistent materialism.