The referential dynamics of cognition and action

Referential behavior theory (RBT), a general dynamical approach to psychological and related systems that operate through a control or referencing process, is introduced. A review of existing evidence shows that this approach can apply to a variety of human and animal systems and tasks, whether the framing language is that of homeostasis, error correction, coupled biological oscillators, motor control, adaptive change, cognitive goal-setting, evaluation and refinement, or neural network learning. Thus, RBT provides a path for reconciliation of dynamical and information-processing accounts of action and cognition. RBT generates a class of mathematical equations, one of which, the discrete control equation (DCE), forms the basis for more detailed investigation. The primary focus here is on the application of the DCE to the temporal structure of regular human movement. Given certain conditions, the equation produces various standard (and new, more general) forms of the circle map class that governs relative phase in motor coordination and, hence, generates well-documented nonlinear ''dynamical" motor phenomena such as behavioral attractors, phase transitions, critical slowing, and so on. Under certain other conditions, the DCE produces the linear stochastic timing models often associated with motor program traditions, accommodating hierarchical effects, open- and closed-loop conditions, and unilimb and multilimb movement A number of new predictions are identified from the approach, and a review of experimental evidence gives support for the claim that the current formulation is an integrated generalization and improvement of several aspects of existing motor program and dynamic approaches.