Beyond fitness: the nature of selection acting through the constructive steps of lifecycles.

The article focuses on redefining the concept of selection and the adaptive component of evolutionary change by replacing traditional fitness- and unit-of-selection-based frameworks with a lifecycle-centered approach using stoichiometric population processes (SPPs) and hypergraph representations. It argues that conventional definitions of fitness, rooted in replicator-centric views, inadequately capture the full complexity of selection, especially when reproduction involves complex lifecycle stages and interactions beyond simple replication. By modeling lifecycles as sequences of transformation events and partitioning reproductive success across these lifecycle realizations, the approach refines summary statistics of selection, recovers multi-level fitness effects, and resolves selection modes obscured or distorted by additive fitness regressions. The paper illustrates these concepts through diploid, biallelic models with random mating and self-fertilization, demonstrating how the lifecycle framework preserves causal interpretations in the Price equation and unifies diverse selective regimes that conventional fitness-based methods conflate or exclude. This construction-oriented representation recasts population genetics from an object-focused to a process-focused theory, providing a foundation for a more comprehensive and causally consistent theory of adaptive information flow in evolution.