Dedre Gentner

What happens when you suddenly get hold of a difficult-to-grasp idea? When you finally “get it?” Science says that flash of clarity often arrives through comparison. Electricity becomes “like” water flowing through pipes. An argument is “like” a tug-of-war. A complex organization becomes a “family tree.” According to Northwestern University cognitive scientist Dedre Gentner, what we typically dismiss as figures of speech actually reveal the computational systems of the mind. Our mental computers are constantly running comparisons, borrowing structure from something familiar to make sense of something new. 

This is the central tenet of Gentner’s body of work, which has revolutionized how cognitive scientists understand the way we learn and reason. We tend to view metaphors and analogies as decoration, when, really, they are evidence of the deep machinery, the mental equipment that helps us piece together an understanding of the world. 

For much of cognitive science’s history, analogy lived in the “soft” category, useful for teaching, perhaps, but too slippery to study rigorously. Gentner helped change that by treating analogy as a process that can be analyzed, tested, and formalized. In Gentner’s structure-mapping theory, the strength of an analogy isn’t that two things share the same surface features. It’s that they share the same relationships, the same pattern of connections among parts.  

To better reveal the power of analogy to reveal deep, abstract principles, Gentner has documented the major role that analogy has played in scientific discovery. For example, the French scientist Sadi Carnot (1796-1832), a pioneer of modern thermodynamics, discovered the basic principles of heat engines by means of an extended analogy between heat and water. In 1596, Johannes Kepler used an analogy with light to propose that the sun emits a motive spirit or force that causes the planets’ motion.  Like the light from a lamp, this motive force spreads and becomes weaker with distance, accounting for why the outer planets move slower than the inner planets. This discovery fed into Newton’s fuller explanation of gravity some 80 years later.  

Gentner also emphasized something that great analogies have in common: they are not just a list of matched parts. They are systems. A strong analogy carries an interlocking web of relationships—cause and effect, constraints, hierarchies, and feedback. This is why the best comparisons don’t merely help you remember; they help you predict. If the structure holds, you can reason forward in the new domain using what you already understand in the old one.  

Once you see analogy as a cognitive tool, a new question emerges: how do we learn to use it? Gentner’s analogy research has followed children learning basic concepts as well as adults applying knowledge creatively. One of her key insights is that comparison itself can teach. When learners line up two examples side by side, shared relations become easier to notice. This allows children to extract a more abstract template that can be applied to new situations. In classrooms, even brief prompts to compare can help learners notice relations they would otherwise miss.  The language we use with children aids in this process. When children hear the same word applied to two different things, they naturally compare them, and in so doing they can discover the shared meaning. The word then acts as a way label that meaning. In this way language can make complex relations easier to carry from one context to another. A child who learns the language of “above,” “between,” “inside,” “cause,” and “effect” is not merely acquiring vocabulary; they are acquiring handles for relationships. These ideas have become vital underpinnings for a science of learning, in part because they point toward practical ways to support reasoning rather than to simply measure it.  

Gentner’s influence extends beyond psychology and education into artificial intelligence, because once you can describe a cognitive process precisely, you can begin to build tools that emulate parts of it. Her theory helped inspire the Structure-Mapping Engine (SME), a computational system created with Ken Forbus and other collaborators that implements analogical matching in software. The SME is different from modern AI tools, like large language models (LLMs). Their machinery bears little resemblance to that of the human mind, but the SME, running protocols derived from Gentner’s structure-mapping theory, is useful for cognitive simulation. It creates analogies and learns in the same way we do.  

In other words, Gentner’s work does not only explain human reasoning; it helps make analogy a legitimate target for computation with applications in machine learning. 

Her work returns us to something intimate, the way a human mind reaches outward. We are not limited to what we have already seen. We learn by mapping, by carrying structure from the known to the unknown, from the concrete to the abstract. In classrooms, it inspires new teaching methods focused on comparison rather than rote recall. In science, it can mean the leap that turns a puzzling phenomenon into a solvable problem. And in everyday life, it can mean the skill of seeing past surface differences to recognize a deeper pattern that matters. Gentner has helped show that this is not a poetic extra. It is one of humankind’s most fundamental powers. 

Dedre Gentner was born in San Diego, California. She earned a bachelor’s degree in physics from the University of California, Berkeley, did graduate work in physics at the University of Chicago, and then completed a Ph.D. in psychology at the University of California, San Diego. She later joined Northwestern University, where she became a leading figure in both psychology and cognitive science. 

Information as of March 2026