How Information Evolves: Stability-Driven Assembly and the Emergence of a Natural Genetic Algorithm

TL;DR

Stability-Driven Assembly (SDA) is presented, a framework in which stochastic assembly combined with differential persistence biases populations toward longer-lived motifs, motivating an evolutionary ladder hypothesis where persistence-driven selection precedes genetic replication.

Abstract

Information can evolve as a physical consequence of non-equilibrium dynamics, even in the absence of genes, replication, or predefined fitness functions. We present Stability-Driven Assembly (SDA), a framework in which stochastic assembly combined with differential persistence biases populations toward longer-lived motifs. Assemblies that persist longer become more frequent and are therefore more likely to participate in subsequent interactions, generating feedback that reshapes the population distribution and implements fitness-proportional sampling, realizing evolution as a natural, emergent genetic algorithm (SDA/GA) driven solely by stability. We apply SDA/GA to chemical symbol space using SMILES fragments with recombination, mutation, and a heuristic stability function. Simulations show hallmark features of evolutionary search, including scaffold-level dominance, sustained novelty, and entropy reduction, yielding open-ended dynamics absent from equilibrium models with fixed transition rates. These results motivate an evolutionary ladder hypothesis where persistence-driven selection precedes genetic replication.

Figures (13)

• Figure 1: Symbolic SDA/GA loop in ABC space. Base elements are replenished while unstable motifs expire, yielding a population skewed by stability ($S=50,30,1$). Roulette-wheel sampling proportional to abundance selects motifs for interaction (concatenation in SDA, recombination in GA), generating new assemblies. This feedback of stability$\rightarrow$persistence$\rightarrow$population skew produces emergent fitness-proportional selection without an explicit fitness function.
• Figure 2: Pattern distribution in the SDA system with concatenation. Left: with no persistence bias yields a near-uniform distribution. Right: stability-driven persistence produces dominance of a small set of high-stability motifs.
• Figure 3: Pattern distribution in the generalized SDA/GA system with recombination. Left: with no persistence bias yields a near-uniform distribution. Right: persistence-driven selection yields dominant stable motifs with a broader low-frequency tail maintained by recombination.
• Figure 4: Entropy dynamics under different operators. In both panels, the upper blue curve shows the unconstrained control without persistence bias. (a) Concatenation-based SDA exhibits oscillatory boom–bust entropy cycles due to synchronized expiration of long motifs. (b) Recombination-based SDA/GA shows smoother entropy decline, as recombination desynchronizes expirations.
• Figure 5: Pattern diversity under different operators. In both panels, the upper blue curve shows the unconstrained control without persistence bias. (a) Concatenation-based SDA exhibits oscillations reflecting synchronized motif turnover. (b) Recombination-based SDA/GA shows steadily increasing diversity, indicating continuous generation of low-frequency variants.