The Law of Effect Explained: How Thorndike's Cats Shaped Learning Theory
TL;DR: The law of effect says that a behaviour followed by a satisfying consequence becomes more likely to be repeated in that situation, while one followed by discomfort becomes less likely. Edward Thorndike formulated it from experiments in which cats learned to escape from puzzle boxes. It is the ancestor of operant conditioning and of most modern thinking about rewards and habits. Below: the cat experiments, what Thorndike got wrong, and where the idea shows up now.
What is the law of effect?
The law of effect is one of the founding principles of the psychology of learning. Thorndike stated it in his 1911 book Animal Intelligence in these terms: of several responses made to the same situation, those that are accompanied or closely followed by "satisfaction to the animal" will, other things being equal, be more firmly connected with the situation, so that they are more likely to recur when it recurs. Those followed by discomfort will have their connections weakened. The greater the satisfaction or discomfort, the greater the strengthening or weakening of the bond.
The American Psychological Association's dictionary gives the modern, plainer version: the principle that consequences of behaviour act to modify the future probability of that behaviour.
Thorndike and the puzzle boxes
Edward Thorndike completed his PhD at Columbia University in 1898 under James McKeen Cattell. His dissertation, Animal Intelligence: An Experimental Study of the Associative Processes in Animals, rested on a simple method. He put a hungry animal, usually a cat, in a wooden box. To get out, the cat had to operate a mechanism.
The mechanisms varied. In one box (Box A), a string ran over a pulley from the door bolt to a wire loop; clawing or biting the loop opened the door. In another (Box C), a wooden button held the door shut and had to be turned. In a third (Box K), three separate acts were needed. According to Wikipedia, the boxes were about 20 by 15 by 12 inches.
What happened next is the heart of the story. The cat did not work out the mechanism and then apply it. It clawed, bit, squeezed and thrashed. Eventually it hit the mechanism by accident and escaped. Put back in the box, it took a little less time. Over many trials the useless movements dropped away and the successful one appeared sooner. In the chapter on his cat experiments, Thorndike lists twelve cats, aged roughly three to nineteen months. For one of them, Cat 12 in Box A, the first trial took 160 seconds. Over 24 trials the time fell, with large fluctuations, to something like six or seven seconds in the later trials.
Thorndike drew a bold conclusion from the smooth, gradual slope of those time curves. He wrote that "the gradual slope of the time-curve, then, shows the absence of reasoning". The cat, he said, did not look over the situation and decide what to do. Instead, the one impulse out of many accidental ones that leads to pleasure "becomes strengthened and stamped in".
That phrase, "stamped in", is the law of effect in miniature, and it was an argument for explaining animal learning through its observable consequences rather than through assumed insight.
The two laws, and what changed
In the 1911 version Thorndike set out two laws of learning:
- The law of effect: consequences strengthen or weaken the link between a situation and a response.
- The law of exercise: a response becomes more strongly connected with a situation in proportion to how often it has been connected with it, and with the vigour and duration of the connections.
He later revised both. According to Wikipedia, by 1929 he had said his early theory was wrong and dropped the law of exercise. He also removed half of the law of effect. Reward still strengthened connections, but punishment turned out to be much less effective at weakening them than the symmetrical original had claimed. Britannica describes the same change: rewards substantially strengthened associations, whereas punishments weakened the unwanted link only slightly. The APA dictionary calls the original the "strong" law of effect and the narrowed version the "weak" law of effect.
From puzzle box to Skinner box
Thorndike's cats were a major influence on B. F. Skinner's later animal experiments. Wikipedia describes the Skinner box, or operant chamber, as an updated puzzle box, and notes that "satisfying" and "annoying" were replaced in later work by the more neutral words "reinforcing" and "punishing". Skinner's operant conditioning is the law of effect with better equipment, a cleaner vocabulary, and decades of experiments on how often and how predictably rewards arrive.
This is also what separates the law of effect from the better-known Pavlovian conditioning. In Pavlov's experiments, a stimulus such as a bell becomes linked to a response that happens automatically. In Thorndike's, the animal's own behaviour is shaped by what follows it. If you want the Pavlov story, we tell it in Pavlov's Dogs.
How the law of effect shows up in everyday life
- Habits. An action that reliably produces something pleasant, even trivially, gets repeated: reaching for the phone, the afternoon biscuit, checking a feed. The consequence does the teaching, not your intentions.
- Teaching and parenting. Attention, praise and privileges are consequences, and they shape what gets repeated, often unintentionally.
- Design of apps and games. Systems that deliver small rewards promptly are applying the same principle, whether or not their designers call it that.
- Workplaces. What gets noticed or paid is what gets repeated, which can differ from what the organisation says it values.
It also works backwards. If a behaviour keeps appearing, ask what it is getting.
What do replication and criticism say?
This is not a replication-crisis story in the usual sense. The idea has not been withdrawn after a failed study. It has been refined, narrowed and rebuilt, and the broad principle that consequences shape behaviour underpins the operant tradition that Skinner developed from it. But there are real criticisms, and the original evidence was thin by modern standards.
The original sample was small. The cat chapter lists twelve animals. That was normal for 1898, but it would not pass for a main finding today. Later work did the heavy lifting.
The "punishment" half was wrong. As above, Thorndike himself concluded in the late 1920s and 1930s that punishment did not simply mirror reward.
Circularity. If "satisfying" is defined only by the fact that it makes a response more likely, the law becomes circular. Wikipedia notes that the behaviourist Richard Herrnstein said in 1970 that the qualitative law "may be a tautology", while arguing that his own quantitative version was not.
"No reasoning" went too far. Thorndike read the gradual curves as evidence against any insight in animals. That is a conclusion about the mechanism, and it was stronger than a time curve alone can support. It is better to treat the experiments as showing that a great deal of behaviour can be explained by consequences, not that nothing else is going on.
Related concepts
- The Zeigarnik Effect Explained: another effect of unmet goals on attention.
- The Curiosity Gap Explained: why anticipated payoffs pull us forward.
- The Testing Effect Explained: how feedback on retrieval strengthens learning.
- The Ben Franklin Effect Explained: a case where behaviour changes attitude rather than the reverse.
- The wider set is collected on our psychology hub.
Frequently Asked Questions
What is the law of effect in psychology?
It is the principle that behaviours followed by satisfying consequences become more likely to be repeated in the same situation, and behaviours followed by discomfort become less likely.
Who came up with the law of effect?
Edward Thorndike, an American psychologist. He developed it from puzzle-box experiments with cats in the late 1890s and stated it in its best-known form in Animal Intelligence (1911).
What was Thorndike's puzzle box experiment?
Hungry cats were placed in boxes that opened only when they operated a mechanism such as a string loop or a button. Their escape times fell gradually across repeated trials, which Thorndike took as evidence of trial-and-error learning rather than sudden insight.
How is the law of effect different from operant conditioning?
Operant conditioning, developed by B. F. Skinner, is a more precise and experimentally rigorous successor. It replaces "satisfying" and "annoying" with "reinforcing" and "punishing" and studies how patterns of reinforcement shape behaviour.
How is the law of effect different from classical conditioning?
In classical (Pavlovian) conditioning, a neutral stimulus becomes linked to an automatic response. In the law of effect, a voluntary behaviour is strengthened or weakened by what follows it.
Did Thorndike change the law of effect?
Yes. By the late 1920s and early 1930s he dropped the law of exercise and concluded that punishment was much less effective at weakening a response than reward was at strengthening one.
Learning more
Learning theory is easier to remember as a story than as a list of laws. Chunks tells psychology's landmark experiments as short narrated stories, with a free tier and no ads.

Founder of Chunks Microlearning. Software engineer with 15 years of experience.


