⚙
AI-generated deep summary
by claude@2026-07, 2026-07-03
· read from full text
ⓘ
The paper develops a mathematical theory of agents using a single fixed encoding shared across multiple tasks, asking when optimization for performance yields internal representations that preserve true world-state structure versus only fitness-relevant distinctions. It finds that selection favors “ecological veridicality,” so any pair of world states that must be distinguished on tasks with positive probability under the task distribution will be separated in optimal encodings, while distinctions with zero aggregate fitness impact can collapse neutrally; the authors support this with static optimality proofs and evolutionary convergence results using tools including Price decomposition and quasispecies recursion (with additional global convergence claims under primitive mutation and Wright–Fisher finite-population approximation). A stated limitation is that the evolutionary convergence results depend on specific modeling assumptions (e.g., fixed encoding, mutation model, finite-population approximations, and fixed horizons). The paper does not explicitly discuss endometriosis or adenomyosis; it was included in the corpus via a keyword match in the upstream search index.
Abstract
When does optimisation for performance yield representations that track world structure? We develop a mathematical theory of agents with a single fixed encoding shared across tasks, and use it to resolve the broader debate over whether selection favors fitness-tuned interfaces or veridical perception. Selection favors ecological veridicality: preserving exactly those world-state distinctions required by the task ecology. The governing object is a separation condition on the task distribution $\mu$: if a pair of states is distinguished on tasks with positive $\mu$-measure, optimal encodings must separate it. In evolutionary terms, distinctions that systematically affect fitness across the ecology are selected to persist, while distinctions with zero aggregate fitness consequence can collapse neutrally without risk penalty. We prove static optimality results and deterministic evolutionary convergence (Price decomposition plus quasispecies recursion) to the best mutation-accessible optimum, with global convergence under primitive mutation and Wright--Fisher finite-population approximation on fixed horizons. As task diversity increases, resolved ecological complexity $k_T = |W/{\sim_T}|$ grows monotonically (graded cascade). The framework recovers both established poles of the debate: in the single-task limit, fitness-tuned interface encodings can dominate truth-tracking encodings, while in the fixed-encoding multi-task regime, selection favors ecological veridicality up to capacity limits.
Full text
2,353 characters
· extracted from
oa-doi-fallback
· click to expand
This is a Preprint and has not been peer reviewed. This is version 1 of this Preprint.
You must log in to post a comment.
There are no comments or no comments have been made public for this article.
This is a Preprint and has not been peer reviewed. This is version 1 of this Preprint.
Add a Comment
You must log in to post a comment.
Comments
There are no comments or no comments have been made public for this article.
When does optimisation for performance yield representations that track world structure? We develop a mathematical theory of agents with a single fixed encoding shared across tasks, and use it to resolve the broader debate over whether selection favors fitness-tuned interfaces or veridical perception. Selection favors ecological veridicality: preserving exactly those world-state distinctions required by the task ecology. The governing object is a separation condition on the task distribution $\mu$: if a pair of states is distinguished on tasks with positive $\mu$-measure, optimal encodings must separate it. In evolutionary terms, distinctions that systematically affect fitness across the ecology are selected to persist, while distinctions with zero aggregate fitness consequence can collapse neutrally without risk penalty. We prove static optimality results and deterministic evolutionary convergence (Price decomposition plus quasispecies recursion) to the best mutation-accessible optimum, with global convergence under primitive mutation and Wright--Fisher finite-population approximation on fixed horizons. As task diversity increases, resolved ecological complexity $k_T = |W/{\sim_T}|$ grows monotonically (graded cascade). The framework recovers both established poles of the debate: in the single-task limit, fitness-tuned interface encodings can dominate truth-tracking encodings, while in the fixed-encoding multi-task regime, selection favors ecological veridicality up to capacity limits.
https://doi.org/10.32942/X27S94
Applied Mathematics, Biostatistics, Cognitive Neuroscience, Evolution
Mathematical biology, Evolution of Perception, Perceptual representation, Multi-task learning, Evolutionary dynamics
Published: 2026-03-05 05:09
Last Updated: 2026-03-05 05:09
CC BY Attribution 4.0 International
Data and Code Availability Statement:
https://github.com/gvdr/evo_interface_veridicality
Language:
English
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.