Truncation selection¶
Truncation selection selects every breeding candidate whose measured or predicted trait value lies beyond a fixed cutoff, giving equal reproductive eligibility within the selected tail and none outside it.
Core Idea¶
Truncation selection is a breeding or evolutionary-computation rule that ranks a population by a selection criterion and permits only individuals above a fixed cutoff—the best specified proportion or all exceeding a threshold—to contribute to the next generation. Within the selected set, reproductive contribution may be equal or governed by another rule. The sharp boundary creates strong, simple selection without requiring proportional fitness differences among the accepted individuals. In quantitative genetics, breeders may rank animals or plants by phenotype, estimated breeding value, or an index combining traits.
Scope of Application¶
-
Animal breeding. Phenotype, estimated breeding value, or a multi-trait index defines the retained parents.
-
Plant breeding. Strong selection concentrates desired traits while mating and population design manage diversity.
-
Quantitative genetics. Selected proportion, selection differential, heritability, and response are related under stated assumptions.
-
Selection experiments. Repeated cutoff application tests trait response, correlated change, and loss of variance.
-
Liability-threshold traits. Binary observations can be analyzed through an underlying continuous liability model.
Clarity¶
Truncation selection ranks candidates and permits only those above a fixed threshold or within a chosen top proportion to reproduce. It differs from proportional selection because rank differences within the accepted group need not alter contribution unless another rule is added. Clarity requires the measured phenotype or breeding-value index, cutoff, selected proportion, sex or family structure, and diversity controls.
Manages Complexity¶
Truncation selection compresses reproductive choice to a ranked criterion and one cutoff. The breeder tracks the selected proportion, threshold, selection differential, heritability, family structure, and diversity cost instead of assigning a separate reproductive weight to every candidate. Threshold- and proportion-based branches behave similarly once the population distribution is known; equal or further-weighted reproduction within the selected set creates another branch.
Abstract Reasoning¶
Threshold move. Rank candidates by a measured trait and retain only those beyond a chosen cutoff for reproduction. Response move. Predict the next generation's mean from selection intensity, heritability, and available variation while distinguishing phenotype from breeding value. Variance move. Anticipate reduced diversity and increased relatedness when a small selected tail supplies parents. Optimization move. Set the cutoff by balancing genetic gain, population size, inbreeding, cost, and correlated traits. Boundary move.
Knowledge Transfer¶
Within the home domain. Truncation selection transfers across animal breeding, plant breeding, quantitative genetics, and experimental evolution when candidates are ranked on a trait and only those beyond a threshold reproduce. Phenotype, breeding value, cutoff, selection intensity, heritability, response, and inbreeding retain genetic roles. Beyond the home domain (B — shared abstract mechanism). Hiring and algorithmic filters also admit only cases past a threshold, sharing hard cutoff selection. Reproduction, inheritance, linkage, and genetic variance remain home-bound. Any threshold classifier is not biological truncation selection, and strong phenotypic selection does not guarantee equivalent genetic gain.
Relationships to Other Abstractions¶
Current abstraction Truncation selection Domain-specific
Parents (1) — more general patterns this builds on
-
Truncation selection is a kind of Selection Prime
Truncation selection is a domain-specific kind of Selection: Truncation selection selects every breeding candidate whose measured or predicted trait value lies beyond a fixed cutoff, giving equal reproductive eligibility within the selected tail and none outside it.
Hierarchy path (1) — routes to 1 parentless root
- Truncation selection → Selection
Neighborhood in Abstraction Space¶
Truncation selection sits in a sparse region of the domain-specific corpus (65th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Selection, Speciation & Experimental Evolution (22 abstractions)
Nearest neighbors
- Disassortative mating — 0.85
- Hardy-Weinberg Principle — 0.84
- Wallace Effect — 0.84
- Vicar of Bray (scientific hypothesis) — 0.84
- Genetic Load — 0.84
Computed from structural-signature embeddings · 2026-10-08