artificial selection
Enable an AI agent to recognise artificial selection, assess a programme's evidence and outcomes, and identify justified changes to how organisms contribute to subsequent generations.
Research draft, second pass
A second pass drafted this model: the structure a model of this thing needs, and what is known about it in the world. The line under this one says how the second half was obtained - researched against sources, or recalled without web access, in which case nothing here was read anywhere and every claim is a lead to verify. Unreviewed either way.
Researched by: Codex + Grok
Purpose and description
Enable an AI agent to recognise artificial selection, assess a programme's evidence and outcomes, and identify justified changes to how organisms contribute to subsequent generations.
Artificial selection is human-imposed, non-random differential reproduction among individuals on the basis of heritable traits, so that those traits change in frequency across generations.
It can be Determine whether a proposed intervention constitutes artificial selection and identify its selected population.; Compare candidate contributors against trait objectives, inheritance evidence and exclusion criteria.; Propose reproductive or propagative contributions subject to diversity, welfare and fitness constraints.; Trace selected contributors into descendants and compare intended with realised contributions.; Assess whether observed changes support a selection response or require further controls and measurements.; Recommend continuing, revising or pausing selection when objectives or constraint thresholds are reached..
Distinguishing features
Humans deliberately favour particular organism differences when deciding reproductive or propagative contribution; environmental survival differences alone do not establish artificial selection.
The selected differences are intended to persist through inheritance or clonal propagation; improvement confined to a treated or trained individual does not qualify.
Candidate selection changes expected representation in subsequent generations; sorting organisms solely for consumption or display does not qualify.
Direct genetic modification is distinguishable from selecting among organisms carrying variants, although both activities can occur within one programme.
A documented selection decision can establish the activity even before a response is demonstrated; an improved population alone does not establish how that improvement arose.
Scope
+ Human-defined breeding objectives and eligibility criteria for candidate organisms
+ Evidence connecting selected traits to transmissible variation
+ Decisions about which organisms reproduce or propagate and how much they contribute
+ Selection response across generations, including unintended correlated changes
+ Diversity, welfare and fitness constraints that limit acceptable selection
- Natural selection operating without human-directed reproductive choices
- Genetic engineering as the direct introduction or alteration of genetic material
- Training, conditioning or husbandry changes that alter individuals without selecting inherited differences
- Domestication as a broader historical relationship between humans and populations
- Routine reproduction or propagation without preferential contribution based on selected differences
- Commercial production and distribution of the resulting organisms or products
Characteristics
- Selection objective
- Named target traits, desired directions or ranges, priorities and unacceptable trade-offs Makes improvement assessable against the actual breeding purpose.
- Selection stage
- Proposed, candidates assessed, contributors selected, reproduction underway, descendants assessed, paused or ended Separates intended decisions from executed selection and demonstrated outcomes.
- Inheritance evidence
- Unknown, hypothesised, pedigree-supported, experimental, genomic or supported by multiple evidence types Indicates whether selecting the measured difference has a credible route to changing descendants.
- Reproductive system
- Sexual outcrossing, self-fertilisation, clonal propagation or mixed Determines how selected variation is transmitted and which contribution controls are possible.
- Selected proportion
- Selected contributors divided by eligible candidates, reported by cohort and contributor role Describes how restrictive selection is without assuming equal reproductive contribution.
- Selection differential
- Difference between selected-contributor and eligible-population trait means, in trait units, with weighting stated Quantifies the phenotypic contrast imposed by a selection decision.
- Realised contribution
- Contributor-to-descendant links with offspring counts, propagated-unit counts or estimated contribution shares Shows whether planned selection actually changed representation in the next generation.
- Observed selection response
- Change in specified trait or breeding-value estimate per generation or unit time, with comparison basis and uncertainty Supports judging progress while exposing environmental and measurement confounding.
- Population diversity status
- Unassessed, within programme limits, approaching a limit or outside a limit, with supporting indicators Makes loss of future breeding options visible alongside target-trait gains.
Also called
Where this came from
wikidata · CC0 1.0
Drafted structure
Bundle to layer to finding to question, as the second pass will find it: 6 bundles · 11 layers · 17 findings · 30 questions.
Selection objectives and boundaries Establishes what humans are selecting for and where the selection episode begins and ends.
Artificial selection is recognised through directed reproductive choices, so a model must connect an objective to a defined candidate population.
Breeding objective
Captures the desired inherited changes and the reasons for preferring them.
Target trait priorities
Record the traits, desired directions or ranges, and trade-offs that govern selection.
- Which inherited traits should change, in what direction or toward what range? definition
- When target traits conflict, which priorities and unacceptable outcomes govern contributor selection? action
Candidate population
Defines the organisms and generations subject to a particular selection decision.
Selection episode boundary
Record the eligible cohort, selection authority and distinction between intentional choices and incidental reproductive differences.
- Which organisms are eligible candidates, and what defines the cohort or generation boundary? boundary
- Who set the selection criteria, and what records show that those criteria influenced reproductive or propagative choices? provenance
Trait evidence and inheritance Connects measured candidate differences to a credible mechanism of transmission.
A desirable phenotype may reflect environment or treatment, so selecting it requires evidence about what descendants can inherit.
Candidate trait assessment
Establishes how candidate performance was observed and made comparable.
Comparable phenotype evidence
Record trait measurements, assessment conditions and factors that could distort candidate rankings.
- How was each target trait measured, at what life stage, and with what repeatability or uncertainty? measurement
- Which differences in husbandry, environment, age or treatment could explain apparent superiority among candidates? boundary
Transmission basis
Records evidence that selected differences can persist through the relevant reproductive system.
Heritable selection signal
Distinguish observed performance from estimates or evidence of transmissible merit.
- What pedigree, progeny, experimental or genomic evidence supports transmission of the selected differences? provenance
- How do sexual reproduction, selfing or clonal propagation affect which differences are expected to persist? definition
- What uncertainty remains in the inheritance evidence for this population and its intended environment? measurement
Selection and reproductive contribution Records how candidate preferences become differential representation in descendants.
Ranking organisms is insufficient: the defining intervention is a change in who contributes to subsequent generations.
Contributor selection
Captures the rule used to retain, exclude or weight candidates.
Applied selection rule
Record selection criteria, candidate assessments, selected contributors and departures from the intended rule.
- Which thresholds, rankings, family comparisons or combined trait scores determined contributor selection? definition
- What proportion of eligible candidates was selected, and how did selected trait values differ from the eligible cohort? measurement
- Which exceptions changed the selected set, and why were they accepted? provenance
Contribution realisation
Connects selected organisms to actual mating, seed production or clonal multiplication.
Planned and realised descendants
Record pairing or propagation plans, actual contributions and unintended contributors.
- Which matings or propagation allocations are intended, and how much should each selected organism contribute? action
- What descendant evidence establishes actual contributions, including failed reproduction or unintended parentage? provenance
Response and attribution Assesses change across generations and the strength of its connection to selection.
Executing a selection programme does not establish success; observed changes need a comparison basis and an account of competing explanations.
Generation comparison
Makes descendant performance comparable with the programme baseline and relevant controls.
Realised trait change
Record change in target traits using stated generation boundaries, measurement conditions and uncertainty.
- How much did each target trait change relative to the baseline or comparison population? measurement
- Are assessment environments and methods comparable across generations, and what adjustments were made? provenance
Causal confidence and revision
Evaluates explanations for the response and their implications for the next selection cycle.
Selection response interpretation
Record how strongly the evidence supports a response to selection and what would justify changing the programme.
- What evidence separates a response to selection from environmental change, migration, drift or changes in measurement? boundary
- What result would justify continuing the selection rule, revising it or collecting stronger evidence first? action
Breeding constraints and side effects Tracks costs of selection that can make apparent progress unacceptable or unsustainable.
Target-trait gains can coincide with concentrated ancestry, reduced reproductive performance or harmful correlated traits.
Diversity and relatedness
Assesses how contributor choices affect ancestry concentration and future selection options.
Ancestry concentration limits
Record relatedness, contribution imbalance and programme-specific limits on loss of diversity.
- How related are selected contributors, and how concentrated will ancestry or clonal representation become? measurement
- What limits require changing pairings, retaining additional lines or broadening the contributor pool? action
Correlated traits and organism condition
Tracks non-target changes in health, welfare, fertility and performance in the intended environment.
Unacceptable correlated response
Record observed or suspected side effects and the thresholds that constrain further selection.
- Which non-target traits, health outcomes or reproductive functions changed alongside the selected traits? measurement
- Which observed harms or performance losses require excluding a contributor, revising an objective or pausing selection? action
- What evidence links these changes to selection rather than to management or environmental conditions? provenance
Evidence and external alignment What the world already says about this thing, gathered so the model can be checked against it.
A model that cannot be lined up against existing standards, identifiers and practice cannot be adopted by anyone who already uses them.
Reported evidence
Findings from the breadth pass, kept separate from the structural claims.
Kinds and varieties
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- methodical (intentional) selection to a predetermined standard
- unconscious selection by keeping valued individuals without a breeding plan
- mass or phenotypic selection on observed trait values
- pedigree and estimated-breeding-value (BLUP) selection
- marker-assisted and genomic selection
- straight-breeding within a closed population or breed
- crossbreeding between lines, breeds, or populations
- experimental laboratory selection lines
- Which of these kinds and varieties hold for the sense of artificial selection this model covers, and on what evidence? provenance
Standards and regulation
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Global Plan of Action for Animal Genetic Resources - Food and Agriculture Organization of the United Nations (adopted 2007)
- Breeding Strategies for Sustainable Management of Animal Genetic Resources, FAO Animal Production and Health Guidelines No. 3 - FAO (2010)
- Genomic characterization of animal genetic resources, FAO Animal Production and Health Guidelines No. 32 - FAO (2023), endorsed by the Commission on Genetic Resources for Food and Agriculture
- Which of these standards and regulation hold for the sense of artificial selection this model covers, and on what evidence? provenance
Real-world use
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Livestock genetic-improvement programmes that rank animals and decide matings for milk, meat, growth, or egg traits.
- Public and private plant-breeding pipelines that produce grain, vegetable, fruit, and ornamental cultivars, including the Brassica vegetables derived from wild mustard.
- Closed studbooks and kennel or pigeon-fancy breeding to a written morphological or performance standard.
- Laboratory selection experiments that reshape a trait over successive generations as a model of evolutionary change.
- Darwin's public argument in On the Origin of Species, which used familiar stock and pigeon breeding to introduce natural selection.
- Which of these real-world use hold for the sense of artificial selection this model covers, and on what evidence? provenance
Typical measurements
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- narrow-sense heritability (h²) - 0.1-0.5 for many production and morphological traits - dimensionless
- selection intensity (i) - about 0.5-2.7 (selected fraction from roughly half the candidates down to about 1%) - phenotypic standard deviations
- generation interval (L) - about 0.3-1 in poultry and annual crops; about 2-7 in cattle, horses, and many trees - year
- inbreeding coefficient (F) - near 0 in a large outbred base; often 0.2-0.5 in closed pedigree dog breeds - dimensionless
- effective population size (Ne) - tens to a few hundred in many commercial poultry and pig lines; conservation programmes commonly target at least 50-100 - individuals
- Which of these typical measurements hold for the sense of artificial selection this model covers, and on what evidence? provenance
Failure modes and hazards
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Inbreeding depression and loss of genetic diversity in closed lines and breeds.
- Breed-associated disease and conformational welfare harm, including extreme morphology in some companion breeds and metabolic or musculoskeletal load in high-yielding livestock.
- Correlated decline in unselected traits such as fertility, flavour, or disease resistance.
- Replacement of locally adapted landraces by a small set of industrial genotypes.
- Epidemic risk in genetically uniform flocks, herds, or crop stands.
- Reduced survival of selected forms if they return to unmanaged conditions.
- Which of these failure modes and hazards hold for the sense of artificial selection this model covers, and on what evidence? provenance
Regional variation
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Evolutionary biology and school biology use "artificial selection"; animal-science and plant-breeding professions usually say "breeding" or "genetic improvement".
- UK and many Commonwealth syllabuses teach the synonym "selective breeding".
- Darwin's methodical/unconscious split is used more in domestication research than on farm.
- East Asian educational usage calques the Darwinian pair (for example Mandarin 人工选择 / 人择 versus 自然选择 / 天择).
- Industrial genomic programmes are concentrated in high-input livestock and seed sectors; many smallholder systems still practise informal, often unconscious, selection.
- Which of these regional variation hold for the sense of artificial selection this model covers, and on what evidence? provenance
Neighbouring kinds and how to tell them apart
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- natural selection - The agent of differential reproduction is the environment, not a human chooser; the same population-genetic mechanism, a different fitness criterion.
- sexual selection - Mates within the species choose; humans are not the choosing agent.
- genetic engineering / transgenesis - DNA is altered directly, including across species barriers; artificial selection only chooses which individuals reproduce.
- genome editing - Specified nucleotides are changed in situ; artificial selection re-sorts standing and mutational variation through mating.
- domestication - A multi-trait evolutionary process of adaptation to a human niche; artificial selection is one mechanism inside it, not the whole process.
- genetic drift - Allele-frequency change that is random with respect to trait value, not preferential breeding of chosen phenotypes.
- Which of these neighbouring kinds and how to tell them apart hold for the sense of artificial selection this model covers, and on what evidence? provenance
Sources
- Artificial Selection - Definition as human identification and perpetuation of desirable plant and animal traits; examples in pigeons, dogs, livestock, and Brassica crops.
- Artificial Selection and Domestication: Modern Lessons from Darwin's Enduring Analogy - Darwin's split of artificial selection into methodical and unconscious selection, and the shared mechanism with natural selection (non-random differences in reproductive success).
- What is selective breeding? - Equivalence of selective breeding and artificial selection; parent-choice mechanism and long agricultural use.
- 5.7: Artificial Selection - Breeding-based definition; distinction of animal breeds from plant varieties/cultivars; Darwin's Origin of Species argument from domestication.
- Breeding Strategies for Sustainable Management of Animal Genetic Resources (FAO Animal Production and Health Guidelines 3) - How states and breeders actually organise straight-breeding and cross-breeding programmes.
- Guidelines | Animal genetics - FAO Global Plan of Action framework and later genomic-characterization guidance as the intergovernmental standard-setting track for livestock selection.
- Selective breeding - Synonymy with artificial selection; contrast with natural selection; continued use of mating-based selection after the rise of genetic engineering.
- Molecular genetic variation of animals and plants under domestication - Domestication versus later diversification under human selection; plant domestication genes versus more polygenic animal domestication.
- artificial selection - Coinage by Darwin in 1859 and the synonym selective breeding; common-language equivalents.
What the second pass must settle
- Should this registry entry include unintentional human-mediated selection, or should its boundary require deliberate reproductive preference?
- Should artificial selection among clonally propagated organisms use the same model boundary as selection through sexual reproduction?
- How should repeated selection of edited organisms link to a genetic-engineering model without duplicating responsibility for the intervention?
- What inheritance and comparison evidence is sufficient to label an outcome a demonstrated selection response across different organism groups?
- Which neighbouring models should own organism-specific welfare standards, acceptable diversity limits and environmental release constraints?