New Views of Madness

This book describes the development of an information- and control-theoretic framework for embodied cognition and cognitive failure, unifying work in computational psychiatry, bounded rationality, and embodied/extended AI around a rate-distortion-centric formalism. Cognition is modeled as uncertainty reduction by a dual information source coupled to a regulator via Rate Distortion Theory and the Data Rate Theorem, yielding a unified Rate Distortion Control Theory (RDCT). This framework derives canonical phenomena such as the Yerkes–Dodson inverted-U between arousal and performance from first principles, then generalizes to multiscale cognition–regulation dyads and culture-bound pathologies of institutions and AI systems. The work extends information-theoretic bounded rationality models by fusing rate distortion with the Data Rate Theorem, so that environmental topological information appears as distortion and cognitive phase transitions are driven by bandwidth limits, delay kernels, and noise. Besides that, an important purpose is to provide a new class of statistical tools for the analysis of observational and experimental data on embodied cognition. This book should be of interest to graduate students, researchers, and practitioners faced with the imposing dynamics of embodied cognition and its manifold dysfunctions at and across a vast range of scales and levels of organization.