Synthesis of Self

This paper formalizes the Synthesis of Self, an information-theoretic framework integrating the Free Energy Principle and functional hemispheric lateralization into a unified computational neuroscience architecture. I model the brain as an asymmetric dual-processor system, quantifying the dynamics between a discrete tokenization engine (the language-dominant Manager network) and a continuous, analog field-processor (the relational Architect network). Crucially, the framework formalizes the longitudinal ontogeny of the self, providing a computational account of how human identity architecture stabilizes over developmental time. By evaluating the transcallosal Coupling Coefficient (C), computed via a non-linear sigmoidal transfer mapping function , the model maps how early relational inputs program the system's baseline network and how subsequent fractional variations dictate macro-structural integration. A mathematical degradation in C forces an uncoupling of the lateralized processors, culminating in a critical processing latency spike  defined as a Complexity Stall. Demonstrating profound abductive consilience, this network geometry systematically retrodicts the neurobiological topologies of six major clinical phenotypes: Schizophrenia, Anorexia Nervosa, Bipolar Disorder, Borderline Personality Disorder, Pathological Narcissism, and Obsessive-Compulsive Disorder. By anchoring the dynamic spectrum of C within triangulated biophysical layers (VMHC, DCM, and paired-pulse TMS), this framework translates descriptive syndromic classifications into an objective, coordinate-based quantization of network processing failures.