Analogical Reasoning: What It Is and How to Train It

Analogical reasoning is the ability to recognize that two structurally different situations work the same way. It is how Darwin connected Malthus's ideas about human population pressure to animal populations competing for resources. It is how Kepler described planetary motion by comparing the sun to a sea captain pushing sailors with his force. It is not a metaphor for insight — it is the mechanism of insight.

When researchers study creative thinkers across science, business, and design, analogical reasoning appears as a consistent differentiator. The ability to transfer structure from a known domain to an unsolved problem separates expert problem-solvers from novices, regardless of field.

What Is Analogical Reasoning?

Analogical reasoning is the process of identifying a structural match between two things that appear on the surface to be unlike. The operative word is structural: not surface similarity (two things that look the same) but relational similarity (two systems that operate the same way).

Cognitive scientist Dedre Gentner at Northwestern University developed the structure-mapping theory of analogy. Her research showed that people reason analogically by mapping relationships from a source domain they understand well onto a target domain where they face a problem. A student who understands water flowing through a pipe can apply that relational structure to reason about electrical current — not because pipes and wires look alike, but because the underlying relationship between pressure, flow, and resistance holds in both.

Surface similarities are easy to spot and often misleading. A sports car looks aggressive like a lion. Structural analogies are harder and far more useful. When Rutherford proposed the atom as a miniature solar system, the model was wrong in its details — but the structural analogy was productive enough to direct decades of physics research.

Why Analogical Reasoning Matters for Creativity

Most original ideas come from transferring a solution from one domain into another. Velcro came from examining how burr seeds cling to animal fur and asking what the structural principle was. The architecture of the internet drew on ideas about how the brain forms associative networks. SONAR was developed by modeling how bats navigate in the dark.

This is analogical transfer: a problem in domain A gets solved by recognizing that domain B already solved a structurally equivalent problem. Without the ability to see across domains, every solution you generate must come from within the domain where you're already stuck.

Analogical reasoning also underlies associative thinking — the ability to connect ideas that aren't obviously related. Where associative thinking generates loose connections, analogical reasoning tests whether those connections hold structurally. One generates candidates; the other evaluates them.

Analogical Reasoning vs. Surface Similarity

The most common failure mode in analogical reasoning is surface seduction: being drawn to cases that look similar and treating that resemblance as structural. Two companies are both tech startups that went public in the same year — but the structural conditions governing their markets may be completely different.

Gentner's research distinguishes three types of comparisons:

• Mere appearances: same features, different relations (a lemon and a tennis ball are both yellow)
• Analogies: same relational structure, different features (the atom is like the solar system)
• Literal similarity: same features and same relations (a poodle is like a labrador)

Creative breakthroughs come almost exclusively from the middle category. Analogy examples from science and history confirm this: the productive ones are always cases where the surface is misleading and the structure is what carries the insight.

False analogies are persuasive precisely because they exploit surface similarity. Spotting a bad analogy requires the same cognitive skill as building a good one — you have to test whether the structural mapping holds.

How Analogical Reasoning Is Measured

The Remote Associates Test (RAT) is one established measure of analogical reasoning. It presents three unrelated words and asks for a single word that connects all three — which requires finding the structural relationship that links three separate conceptual domains. Read more about what the RAT measures and its origins.

The Analogical Encoding task is a more direct measure: you are given two different scenarios and asked to identify their shared deep structure. Research by Loewenstein, Thompson, and Gentner (2003) showed that this comparison task — rather than studying examples in isolation — produced significantly stronger transfer to novel analogical problems. The act of comparison forces you to articulate the relational mapping, which is what encodes the structure rather than the surface details.

How to Improve Analogical Reasoning

Compare cases deliberately. Gentner's research consistently shows that comparing two structurally similar cases side by side produces much stronger analogical transfer than studying each case alone. The comparison forces you to extract the relational structure and sets it apart from the surface details of each case.

Read across domains. The wider your knowledge base, the more source domains you have available to draw from. Domain specialists often get trapped because they only know one set of solutions. Reading in history, biology, engineering, and economics gives you more structural patterns to work with when you encounter a new problem.

Ask "what is this structurally like?" When you face a new problem, pause and explicitly ask what familiar structure it resembles — not superficially, but relationally. What roles do the parts play? What drives the dynamics? What constraints govern the system? The question forces you past surface features.

Practice with analogical reasoning exercises. The Analogical Encoding exercise presents pairs of scenarios and asks you to find their shared structure. Unlike passive reading, active comparison forces you to articulate the relational mapping — and that articulation is what builds the skill over time.

Study where analogies break down. Analyzing false analogies — cases where a structural mapping fails — sharpens your sensitivity to what counts as genuine relational similarity versus what just looks similar. Every failed analogy teaches you something about the conditions that make structural transfer valid.

The Research

Gentner and Markman (1997) reviewed the evidence that analogical reasoning is central to both learning and creative problem-solving. Their conclusion: surface similarity triggers initial recognition, but structural similarity is what produces insight and useful transfer to novel problems.

Holyoak and Thagard (1995) proposed that successful analogical reasoning requires three constraints simultaneously: structural consistency (the mapping holds internally), semantic similarity (the domains share enough vocabulary to communicate), and pragmatic relevance (the analogy is actually useful for the problem at hand). Most failed analogies fail on the third constraint — they are structurally coherent but solve the wrong problem.

Kurtz and Loewenstein (2007) showed that people who practiced comparing structurally similar cases significantly outperformed controls on novel analogical transfer problems. The skill transfers — practicing with one pair of domains improves performance on entirely different domains.

Why Domain Insularity Is the Real Enemy

Expert problem solvers actively look outside their field because the most useful analogies come from unexpected places. This is not a personality trait — it is a deliberate practice. Nobel laureate Herbert Simon argued that the scientific method itself is essentially analogical: you build a model of one phenomenon and test how far that model transfers.

The constraint is that you can only draw on what you know. Expanding the base of domains you understand — even shallowly — is a direct investment in analogical reasoning capacity. This is one strong argument for the creative process deliberately including an incubation phase: time away from a problem allows your brain to search for structural matches across memory rather than staying fixated on the domain where the problem lives.

Ready to train your creativity? Try science-backed exercises that measure and improve your creative thinking. Start a Free Exercise