How Our Minds Make Connections: An Introduction to Associationism

Imagine you’re walking home from school and you smell fresh chocolate chip cookies. Suddenly, you’re thinking about your grandmother’s kitchen, even though she lives hundreds of miles away. Or maybe you hear a particular song, and instantly you’re back at summer camp, feeling exactly the way you felt that summer.

How do these jumps happen? Your mind didn’t decide, logically, “I smell cookies, and my grandmother used to bake cookies, therefore I should think about her kitchen.” The thought just arrived.

This happens all the time—one thought leads to another without any deliberate reasoning. Philosophers and psychologists have a name for this: association. And some of them have argued that association isn’t just one small part of how we think—it’s the whole story.

That’s a pretty wild idea. But it raises a fascinating question: Could all thinking really just be one thought triggering another, like dominoes falling?

What Is Associationism?

Associationism is the view that the way we learn and think is basically about forming connections between ideas based on our experiences. If you’ve often seen salt together with pepper, then thinking about salt will tend to make you think about pepper. If you’ve been scared by a dog, then seeing a dog might make you feel scared again.

This sounds simple enough. What makes associationism radical is that some of its supporters have tried to explain everything about the mind using just this one principle. Not just simple connections like salt and pepper, but complex thoughts, learning new things, even reasoning itself.

The basic idea goes back to philosophers like David Hume (1711–1776), who argued that our minds work mostly by associating ideas that have been connected in our experience. If two things have appeared together in your experience—like thunder and lightning—then when you think of one, the other tends to follow.

But here’s the key question: Is that really all there is to thinking? Or do we have something extra—some special mental machinery that lets us go beyond mere association?

How Associationism Explains Learning

One of the strongest claims associationists make is about how we learn. The idea is that learning is basically just forming new associations through repeated experience.

Pavlov’s dogs provide the most famous example. Ivan Pavlov noticed that dogs would start salivating when they saw the person who usually fed them, even before the food arrived. He rigged up an experiment: he would ring a bell right before giving dogs food. After a few times of hearing the bell and then getting food, the dogs started salivating at just the sound of the bell, even when no food came.

The dog had formed an association: bell → food. Nothing complicated, no reasoning involved. Just two experiences repeatedly happening together, creating a mental connection.

Edward Thorndike took this further with cats in puzzle boxes. He put hungry cats in boxes with levers. If a cat accidentally pushed the lever, the door would open and the cat could escape to get food. At first the cats just scrambled randomly. But after a few successes, they learned to push the lever faster. Thorndike called this the Law of Effect: behaviors that lead to good outcomes get strengthened; behaviors that lead to bad outcomes get weakened.

So the basic story is: we learn by having experiences that get linked together. The more often things happen together, and the more the outcomes matter to us, the stronger the associations become.

What Makes an Association an Association?

This might sound straightforward, but philosophers and psychologists have spent a lot of time trying to figure out what exactly an association is. Here’s the key distinction they’ve discovered:

An association is NOT the same as a proposition.

What’s the difference? Consider these two things:

The concepts bird and fly being associated in your mind
The thought “birds fly”

They sound similar, but they’re actually quite different. If bird and fly are just associated, then when you think about birds, the idea of flying pops into your head—and vice versa. But you haven’t actually said anything about birds. You haven’t claimed that birds fly, or wondered whether they do, or denied it. You just have two concepts that tend to activate each other.

A proposition like “birds fly” does something more. It predicates the property of flying onto birds. It says something about the world that could be true or false.

This difference matters because if associationism is going to explain all of thinking, it needs to explain how we get from simple associations (bird ↔ fly) to full thoughts and judgments (“birds fly”). And that’s actually really hard to do with just associations alone.

Here’s another weird feature of associations: they’re symmetric. If salt is associated with pepper, then thinking about salt should make you think of pepper just as easily as thinking about pepper makes you think of salt. But we don’t always experience things that way—if you always hear “salt AND pepper” but never “pepper AND salt,” the association might run more strongly in one direction.

How Do You Break an Association?

Once you’ve formed an association, how do you get rid of it? This turns out to be surprisingly difficult, and the answer reveals something important about how associations work.

According to associationists, there are only two ways to break an association:

Extinction: Repeatedly experience one thing without the other. If the bell used to mean food, but now the bell rings and no food comes, eventually the association weakens.

Counterconditioning: Replace the association with an opposite one. If you associate something with a good feeling, you can try to link it with a bad feeling instead.

But here’s the crucial point: You cannot break an association just by reasoning about it. If you’ve associated a food with getting sick (maybe you got food poisoning from a particular dish), no amount of telling yourself “that food didn’t actually cause the sickness” will make the association go away. The feeling of disgust persists even when you know it’s irrational.

This is one of the most important and interesting things about associations: they don’t listen to reason. They’re shaped by experience, not by logic or argument.

Is Association Really All There Is?

Many philosophers and psychologists have argued that associationism can’t be the whole story about the mind. Here are some of the biggest challenges:

The Problem of Predication

Remember the difference between an association (bird ↔ fly) and a proposition (birds fly)? How could you get from one to the other just by forming more associations? An association between concepts doesn’t tell you which one is the subject and which is the predicate. It doesn’t tell you anything about the structure of the thought.

This problem was first raised by the philosopher Immanuel Kant (1724–1804). He argued that our minds must have some built-in structure that organizes our experiences—some rules that aren’t themselves learned through association. Otherwise, we’d just have a jumble of associated ideas without any way to make judgments or form coherent thoughts.

The Problem of Learning Words

Children learn words incredibly fast. By age six, most kids know about 6,000 words. And they often learn a word after hearing it just once. This is called fast mapping.

In one famous study, preschoolers were shown two trays—one red, one olive-green—and asked: “Can you get me the chromium tray, not the red one, the chromium one?” The children picked the right tray. And when tested later, many of them still remembered what “chromium” meant—even though they’d only heard it once.

This is a problem for associationism, which predicts that learning should be gradual and require many repetitions. How can one-shot learning happen if all learning is just forming associations through repeated experience?

The Problem of Content

If I show you a red triangle, many things are present at the same time: redness, triangularity, the specific shade of red, the exact shape, maybe the location on the screen, the lighting conditions, and so on. Which of these gets associated with what follows?

This is called the coextensionality problem. When lots of properties are present together, which ones become the basis for associations? You need some principle to pick out which features matter. But associationism, on its own, doesn’t seem to have a good way to do this.

The Problems of Contiguity

Associationism traditionally assumes that things need to happen close together in time to get associated. But there’s good evidence that this isn’t always true.

The Garcia effect demonstrates this. Rats that drank flavored water and then got sick up to 12 hours later still formed an association between the taste and the sickness, even though the two events were far apart in time. But here’s the weird part: the rats didn’t form associations between a light-and-sound signal and getting sick, even when they happened close together. This suggests that animals are prepared to associate some things (tastes with sickness) but not others—which contradicts the idea that association learning is completely general and works the same way for everything.

Similarly, blocking experiments show that not everything that happens together gets associated. If you’ve already learned that a light predicts a shock, and then a sound starts happening together with the light, you won’t learn that the sound also predicts the shock. The existing association blocks the new one.

Where Associationism Still Lives

Despite these criticisms, associationism hasn’t disappeared. It’s had a major revival in two areas of psychology:

Implicit attitudes are the unconscious evaluations we have about things—like automatically associating certain groups of people with positive or negative feelings, even if we consciously reject those associations. Many psychologists think these attitudes work associatively. You can’t change them just by reasoning; you’d need to change the experiences that created them.

Dual-process theories propose that we have two thinking systems. System 1 is fast, automatic, and associative. System 2 is slow, deliberate, and logical. On this view, associationism isn’t the whole story about the mind—but it does describe how a big part of our thinking works.

The Big Question

So where does this leave us? The associationist idea—that thinking is just one thought triggering another, shaped by our experiences—is simple and powerful. It explains some things really well, like why certain smells or songs bring back vivid memories, or why habits are so hard to break.

But it also faces serious challenges. It struggles to explain how we form complex thoughts with structure, how we learn things from a single experience, and how we figure out which features of our experience matter.

Most philosophers and psychologists today think associationism is part of the story about how our minds work, but not the whole story. The interesting debate is about how big that part is, and how it interacts with other mental processes.

Nobody really knows the answer yet. But that’s what makes the question worth thinking about.

Appendix: Key Terms

Term	What it does in the debate
Association	A mental connection between two ideas or concepts, where activating one tends to activate the other
Proposition	A structured thought that says something about the world and can be true or false (like “birds fly”)
Classical conditioning	Learning that one stimulus predicts another (like a bell predicting food)
Law of Effect	The principle that behaviors leading to good outcomes get strengthened
Extinction	The process of weakening an association by experiencing one thing without the other
Counterconditioning	Breaking an association by replacing it with an opposite one
Fast mapping	Learning a new word after hearing it just once or twice
Coextensionality problem	The problem of figuring out which features of an experience get associated when many features are present
Blocking	When an existing association prevents a new one from forming
Implicit attitudes	Unconscious evaluations that work associatively and resist logical correction

Appendix: Key People

David Hume (1711–1776) — Scottish philosopher who argued that all thinking is based on associations between ideas formed through experience
Immanuel Kant (1724–1804) — German philosopher who argued that our minds have built-in structure that can’t come from association alone
Ivan Pavlov (1849–1936) — Russian physiologist who discovered classical conditioning by studying dogs’ salivation
Edward Thorndike (1874–1949) — American psychologist who developed the Law of Effect through experiments with cats in puzzle boxes
Jerry Fodor (1935–2017) — American philosopher who argued forcefully that associationism can’t explain the structure of thought

Appendix: Things to Think About

If associations can’t be broken by reasoning, what does that mean for changing bad habits or irrational fears? Can you think of an association you have that you wish you could get rid of?
Children learn words incredibly fast—but they also learn which things are important to learn words for. Is learning what to pay attention to something that could be explained by association, or does it require something extra?
If we have both associative thinking (System 1) and logical thinking (System 2), how do they interact? Can one override the other? Which one should we trust more?
The Garcia effect suggests that animals are prepared to associate certain things (like tastes with sickness) but not others. Does this mean our minds come with some built-in biases about what can be associated? If so, how does that fit with the idea that all learning comes from experience?

Appendix: Where This Shows Up

Advertising and marketing exploits your associations—connecting products with good feelings, attractive people, or happy memories
Phobias and anxiety seem to work associatively: a traumatic experience creates a lasting association that doesn’t respond to logical argument
Implicit bias tests (like the Implicit Association Test) measure your automatic associations between different social groups and positive or negative words
Habits—good and bad—operate largely through associations, which is why they’re so hard to change just by deciding to change them
Machine learning systems use associative principles to learn patterns from data, though the most successful ones also use structured algorithms