Skip to content

Preimage

Version
v1 · 2026-08-24 · History
Prime #
1080
Origin domain
Mathematics
Subdomain
set theory and functions → Mathematics

Core Idea

A preimage is the set of all inputs that map to a given output under some mapping: it reverses the arrow while honouring the mapping's many-to-one character, so backward reasoning honestly yields a set of explanations rather than a unique cause.

How would you explain it like I'm…

Who Made the Footprints

If you see wet footprints on the floor, you ask 'who could have made these?' Maybe it was your brother, or your sister, or the dog. The Preimage is the whole list of who could have left those prints. You're working backward from what you see to everything that might have caused it.

Working the Rule Backward

A Preimage is the set of all the inputs that could have produced a given result. A regular rule goes forward: 'put in 3, get out 9' by squaring. The Preimage runs it backward: 'I got 9 — what could I have put in?' The answer is both 3 and -3, so it's a set, not just one number. This happens whenever different inputs can lead to the same output: working backward honestly gives you all the possibilities, not a single answer. The more inputs that land on the same output, the more guessing you'd have to do to figure out which one really happened.

All the Possible Causes

A Preimage is the set of all inputs that map to a given output (or set of outputs) under some mapping. Where a function asks 'given this input, what comes out?', the Preimage asks 'given this output, what inputs could have produced it?' The structural move is to reverse the arrow of a mapping while respecting that it's many-to-one — which forces the answer to be a set, not a single element, whenever the mapping isn't one-to-one. That's the honesty at the heart of it: reasoning backward through a many-to-one mapping yields a whole class of possible explanations, not a unique cause. Three facts always travel with it: the Preimage of an output that actually occurs is never empty; it's a single element only if the mapping is one-to-one there; and it partitions the input space when you sweep across all outputs. A Preimage that's just one element means the cause is pinned down; a large Preimage means it's underdetermined and you need more evidence to narrow it.

 

A Preimage is the set of all inputs that map to a given output, or set of outputs, under some mapping. Where a function says 'given this input, what comes out?', the Preimage operation says 'given this output, what inputs could have produced it?' The structural move is to reverse the arrow of a mapping while honouring its many-to-one character — which forces the answer to be a set, not a single element, whenever the mapping is not injective. That is the heart of the concept: backward reasoning under a many-to-one mapping yields equivalence classes of explanations, not unique causes, and the Preimage names exactly that set. The skeleton has three parts: a mapping of any kind (a deterministic function, a causal mechanism, an observation pipeline, a query projection); a target subset of outputs of interest; and the Preimage, the complete set of upstream sources consistent with the target. Three facts travel with it: the Preimage of an output in the image is never empty; it is exactly one element only if the mapping is injective there; and it partitions the input space when the target ranges over the codomain. Where the forward mapping directs attention from cause to effect, the Preimage directs attention backward, and its distinctive contribution is making the cardinality of ambiguity explicit — a singleton Preimage is fully identified, while a large one is underdetermined and needs more evidence to narrow.

Broad Use

  • Mathematics: the preimage of every open set being open defines continuity, and it distributes over union, intersection, and complement.
  • Diagnostic reasoning: the preimage of a symptom is the differential diagnosis, the suspect list, the fault tree.
  • Epidemiology: the preimage of an infected case is the set of prior contacts who could have been the source.
  • Databases: a filter clause computes a preimage — given a target predicate, return the inputs satisfying it.
  • Cryptography: a hash's security is the intractability of its preimage — given an output, find any input mapping to it.
  • Inverse problems: tomography and seismic inversion recover the source configurations consistent with an observed signal.

Clarity

Forces the question how many things map here?, making the size of the explanation-set explicit and distinguishing "we found a cause" from "we found the cause."

Manages Complexity

Compresses "what could have caused this?" into one algebraic operation, and frames the whole inverse problem as a single narrowing of the explanation-set toward a singleton.

Abstract Reasoning

Makes identifiability — is the preimage a singleton? — the master question of every backward-inference problem, with preimage size measuring how much information the forward mapping destroys.

Knowledge Transfer

  • Topology to engineering: "what is the preimage of failure?" is the same operation as fault-tree analysis, with the same closure properties.
  • Databases to epidemiology: a filter and contact tracing compute one operation and share preimage-pruning heuristics.
  • Inverse problems to policy: recovering a source from a signal ports to finding the policy mixes that could have produced an outcome.

Example

A patient's elevated liver enzymes have a large diagnostic preimage — several hepatitides, drug injury, autoimmune disease — and the workup is preimage-narrowing: each test is chosen because it has different preimages under each candidate, shrinking their intersection toward one.

Relationships to Other Abstractions

Local relationship map for PreimageParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.PreimagePRIMEPrime abstraction: Function (Mapping) — presupposesFunction(Mapping)PRIMEDomain-specific abstraction: Perceptual Constancy — is part ofPerceptualConstancyDOMAINDomain-specific abstraction: Isochron — is a kind ofIsochronDOMAINDomain-specific abstraction: Kernel — is a kind ofKernelDOMAIN

Current abstraction Preimage Prime

Parents (1) — more general patterns this builds on

  • Preimage presupposes Function (Mapping) Prime

    'The preimage RIDES ON a forward mapping but runs its arrow backward...

Children (3) — more specific cases that build on this

  • Isochron Domain-specific is a kind of Preimage

    The proposed strict upward parent is prime:preimage.

  • Kernel Domain-specific is a kind of Preimage

    A Kernel is the Preimage specialized to the singleton containing the target homomorphism's distinguished zero or identity element.

  • Perceptual Constancy Domain-specific is part of Preimage

    Perceptual constancy contains a preimage computation that runs a learned distal-to-proximal generative map backward from sensory signal to candidate distal properties.

Hierarchy path (1) — routes to 1 parentless root

Not to Be Confused With

  • Preimage is not Function Mapping because the preimage runs the arrow backward to recover sources, whereas a function mapping runs it forward from input to output.
  • Preimage is not an Inverse Function because the preimage honestly returns a set, whereas an inverse exists only when every preimage is a singleton.
  • Preimage is not a Transformation because the preimage reverses an existing mapping and distributes over Boolean operations, whereas a transformation maps representation to representation forward.