Can analogue minds live forever? Introducing the 'Step-Structure Principle' as a testable limit on the Church Turing thesis for brain emulation

Progress in neuroscience and AI has renewed interest in whether human-level consciousness could be instantiated in non-biological systems, including via whole-brain emulation. A common assumption underlying such proposals is that any function sufficient for consciousness can be implemented on a Turing-equivalent digital computer. This paper challenges the strength of that assumption by introducing the Step-Structure Principle: while Turing-equivalent architectures can emulate the input–output behaviour and causal logic of any computable function, they do not in general preserve the step-structure of a computation: the number and organisation of inputs and outputs involved in the architecture’s computational steps, as well as subsequent aggregate metrics, such as total steps and information processing density. I argue that this distinction matters if specific architectural features of computation are relevant to phenomenal consciousness. Three independent motivations are developed: (i) constraints arising from computational cost and irreducibility in complex systems, (ii) phase-transition-like thresholds suggested by empirical neuroscience, and (iii) the problem of phenomenal binding in conscious experience. Each motivation is uncertain, but any one would be sufficient to undermine the assumption that digital architectures automatically suffice for human-style consciousness. The paper outlines empirically testable research pathways to assess whether step-structure is relevant, drawing on advances in neuroscience, computational modelling, and adversarial collaborations in consciousness research. The results bear directly on AI consciousness, digital sentience, and the safety of proposed trajectories toward artificial or uploaded minds.