Drive reduction theory: what Hull proposed, and the experiments that challenged it

Drive reduction theory is Clark Hull's theory of how needs drive behaviour. A biological need, such as a lack of food or water, produces a drive, an aroused internal state. Any behaviour that reduces the drive is reinforced, so it becomes more likely the next time the need returns. Hull set it out in his 1943 book Principles of Behavior. The AP Psychology course framework uses a short version: drive reduction theory explains how certain behaviours help the body keep homeostasis, its internal balance.

• Need, drive and reinforcement are three separate terms. The need is the physical deficit, the drive is the state it produces, and reinforcement is what happens to the behaviour that ends the drive.
• Primary drives, such as hunger and thirst, are innate. Secondary drives, such as fear or the wish for money, are learned through association with primary drives.
• Hull wrote a theory of learning first. His central claim was a law of reinforcement, and textbooks now teach it under motivation.
• The theory lost ground because animals learned when no need was reduced. Monkeys solved puzzles for nothing, and rats worked for a sweet taste with no calories. Other rats pressed a lever thousands of times an hour for brain stimulation.
• On exam questions, drive reduction theory fits a scenario about restoring a physical balance, not about seeking excitement or a reward.

Drive reduction theory in one table

The theory is a chain of four terms. Each one has a precise meaning, and exam questions often test whether you can keep them apart.

The APA Dictionary of Psychology gives a short definition. Drive reduction theory is "a theory of learning in which the goal of motivated behavior is a reduction of a drive state". The theory assumes that all motivated behaviour comes from drives, and that drives come from a disruption in homeostasis. Responses that reduce those drives are strengthened.

What Hull actually wrote in 1943

Most summaries give Hull's theory in one sentence. His book is more careful than that summary, and four points in it often go missing.

The core of the theory is a law of learning

Hull's central statement is what he called the law of primary reinforcement. It starts with a response that occurs together with a stimulus. Hull then wrote (p. 80): "and this conjunction is closely associated in time with the diminution in the receptor discharge characteristic of a need, there will result an increment to the tendency for that stimulus on subsequent occasions to evoke that reaction." Put simply, a response that is followed by a drop in a need gets stronger.

Hull used this one law to explain both trial-and-error learning and the conditioned reflexes that Pavlov studied. That is why the theory belongs as much to the history of classical and operant conditioning as to the history of motivation.

Behaviour starts with a need and ends with its reduction

Hull described behaviour as a series of cycles. In his words (pp. 24–25): "The rise of either a primary or a secondary need normally marks the beginning of a behavior cycle, and the abolition, or substantial reduction of that need marks its end." A need arises, in his account, when one of the conditions the body needs for survival "deviates much from the optimum" (p. 29).

The homeostasis link comes from Cannon

Hull built on the work of the physiologist Walter Cannon. The APA Dictionary credits Cannon with the first description of homeostasis. Hull noted that the body corrects some deviations by itself, for example by keeping the blood "at a practically constant state". He cited Cannon's name for this automatic process, "the wisdom of the body" (p. 68). Other needs are "remediable only by movement". Hull wrote: "This kind of activity we call behavior." Drive reduction theory is about this second kind of need only.

Drive multiplies habit, but zero drive is not zero behaviour

Hull separated what an animal has learned (habit strength) from how strongly it is driven at the moment (drive, written D). His seventh postulate (p. 253) says that the tendency to respond is a product. It is an increasing function of habit strength multiplied by an increasing function of drive. All the primary drives active at a given time add to D, so drive gives energy to whatever habits the animal has. His sixth postulate adds that each drive also has its own drive stimulus, a signal whose intensity rises with the drive.

Some summaries conclude that with zero drive there is no behaviour at all. Hull's own corollary says something else. When habit strength is zero, the tendency to respond is zero. When primary drive is zero, the tendency "has an appreciable but relatively low positive value" that grows with practice (p. 247). He wrote this to fit the data from Perin's rat experiments.

Hull did not know exactly what does the reinforcing

Hull saw two possible candidates: the reduction of the need itself, or the reduction of the nerve signals that the need produces. He chose the second "in the interest of definiteness" but admitted, "The writer is aware of no critical evidence of this kind. Until such becomes available the issue must remain uncertain" (p. 81). The APA Dictionary's entry on the drive stimulus keeps his choice: in Hull's theories, behaviours that reduce the drive-produced nerve impulses are the ones reinforced.

Primary and secondary drives

A primary drive is innate. The APA Dictionary says it comes from the deprivation of a needed substance, such as food, or from the need to do a specific activity. Hunger, thirst and the drive to avoid pain are the standard examples.

A secondary drive is learned through association with a primary drive. The APA Dictionary's example is fear of an electric shock in a rat that has learned to escape from one compartment into another.

A classic study of this kind is Neal Miller's 1948 study of fear as an acquirable drive. Miller gave rats electric shocks in a white compartment, and they escaped through an open door into a black one. Then he closed the door and stopped the shocks. The rats still learned to turn a small wheel that opened the door. When Miller changed the setup so that only a bar opened it, wheel turning stopped and the rats learned to press the bar. Control experiments showed that the new habits depended on the moderately strong shocks of the first stage. Miller's conclusion was that fear itself acted as a drive, and that a drop in fear acted as reinforcement.

Miller's study extended the theory rather than refuting it. It showed how a theory built on hunger and pain could explain learned motives. Miller's opening paragraph names fear and the wish for money, approval or status.

Examples of drive reduction theory

The clearest examples involve a physical balance that the person restores by acting:

• A student who skipped breakfast feels hungry by eleven and buys a sandwich. Eating reduces the hunger drive, so buying food at that time becomes a habit.
• A runner feels thirsty after a race and drinks water. Drinking reduces the thirst drive.
• A person who feels cold puts on a coat. The action brings body temperature back towards its optimum.
• A student who feels anxious before an exam starts to revise, and the anxiety drops. In drive reduction terms, anxiety is a secondary drive and revising is the behaviour that reduces it.

Some behaviours do not fit the theory. People seek out puzzles, horror films and roller coasters when no need is pressing. These behaviours are the main reason psychologists looked for other theories, and the experiments below show the problem in the laboratory.

Drive reduction is not the same as negative reinforcement

Google's search suggestions include "drive reduction theory negative reinforcement", and the two ideas do overlap. They are still different kinds of term. Negative reinforcement is a procedure: a response removes, prevents or postpones an aversive stimulus, and the response becomes more likely. Drive reduction is Hull's proposed mechanism for every kind of reinforcement.

The difference shows in the hunger example. For Hull, a hungry rat that gets food for a lever press is reinforced by drive reduction. In operant terms, the same event is positive reinforcement, because the procedure adds a stimulus, food. Escape from a shock is both: it reduces a drive, and it removes an aversive stimulus. Our guide to negative reinforcement and punishment explains the operant definitions in detail.

Drive reduction vs arousal, incentive and instinct theory

AP Psychology's course framework lists drive reduction theory next to other motivation theories in Topic 4.6. It pairs drive reduction with arousal theory, because both deal with physical needs. Drive reduction explains how behaviour keeps homeostasis, and arousal theory explains how people look for an optimal level of arousal. The table gives each theory and the cue that usually points to it in a scenario question.

Incentive theory is the closest rival. The APA Dictionary defines it as the theory that motivation depends on two things together. One is the incentives in the environment, and the other is the organism's state, such as a drive state. Drive reduction pushes behaviour from inside, and incentives pull it from outside. Maslow's model is another need-based theory, which our article on Maslow's hierarchy of needs covers. It ranks needs, but it does not explain how a need reinforces a response.

The experiments that challenged drive reduction theory

The challenge to the theory came from experiments in which animals learned, or kept working, when no physical need was reduced.

Monkeys who solved puzzles for nothing

In 1950, Harry Harlow, Margaret Kuenne Harlow and Donald Meyer gave rhesus monkeys mechanical puzzles in their home cages. Each puzzle had a pin, a hook and a hasp that had to be opened in order. Their study used eight monkeys in two matched groups of four. One group had the assembled puzzle on the perch for 12 days, with no food or other reward. Each group then had 40 tests on days 13 and 14. The experienced group solved the puzzle 31 times, and the control group solved it 4 times. More than half of the experienced group's solutions took 20 seconds or less.

Then the researchers put a raisin under the hasp. In three of the four experienced monkeys, the food "seriously disrupted" their skill. These three monkeys made 39 errors in three tests, against 17 in the ten earlier tests. Harlow and his colleagues agreed that hunger, thirst or other drives might have been present. But solving the puzzle did not lead to food, water or sex. It also gave no more relief from heat or cold than random movement. So these drives could not explain why the monkeys worked on the puzzle. The authors proposed that "the performance of the task provided intrinsic reward". The same idea is now taught under the name intrinsic motivation.

A sweet taste with no nutrition

In 1950, Frederick Sheffield and his co-author Roby showed that animals learn to work for saccharin, a sweetener with no calories. A 1965 paper by Elliot Valenstein and Margaret Weber summarises the argument. Saccharin is non-nutritive and passes through the body unchanged. Even so, animals prefer it to water, and it works as a reward. Sheffield and Roby described their own studies this way: "essentially sham-feeding experiments in which the animal was innately stimulated to ingest a substance which did not change his state of need."

Valenstein and Weber added a caution. In their experiment, rats that drank saccharin after a dose of insulin were more likely to die than rats that drank water. Their conclusion: "It may be an error to assume that saccharin is physiologically inert." The saccharin result still shows that a reward can work without reducing a need. It does not show that saccharin has no effect on the body.

Sheffield later proposed an alternative, drive-induction theory: the excitement of eating, drinking or mating is what reinforces behaviour, not the reduction of the drive.

Rats that worked for brain stimulation

James Olds described the next challenge in a report on NASA's Technical Reports Server. Work on positive reinforcement by brain stimulation "began with an accidental observation made in Hebb's laboratory late in 1953". Olds expected the electric stimulus to act as a punishment. Instead, the rat "returned to that corner over and over again". Olds wrote that it was "a brain stimulus with all the characteristics of a primary reward." Olds and Milner published the finding in 1954.

The numbers that followed were large. According to the same report, rats would press a lever 10,000 times an hour to stimulate the lateral hypothalamus. For stimulation of olfactory areas of the cortex, the rate was about 500 times an hour. No tissue need was reduced. The stimulation itself acted as the reward.

Rats that learned without a reward

A further line of evidence was older. In the latent learning experiments of 1929 and 1930, rats ran a maze for days with no food at the end. When food was added, their errors dropped sharply within a day or two, so they had learned the route without a reward. Our article on the latent learning experiments of Blodgett and Tolman explains how these results challenged reinforcement theorists such as Hull. Harlow and his colleagues compared the design of their 1950 study with the same experiments.

Harlow's verdict

By 1953, Harlow was ready to reject the theory. His paper was called "Mice, monkeys, men, and motives". In it he wrote: "It is my belief that the theory which describes learning as dependent upon drive reduction is false, that internal drive as such is a variable of little importance to learning". One reason he gave was that internal drives are cyclical. They are active for only a brief part of an animal's waking life, so they give little opportunity for the learning the theory needs. His "primary quarrel", in his words, was that drive reduction theory "tends to focus more and more attention on problems of less and less importance."

What is left of drive reduction theory

The theory did not disappear. Two parts of it are still in the textbooks.

The first is homeostasis. Hunger is still taught as a regulated system. The AP course framework names the hormones ghrelin and leptin as controls of hunger and satiety, and the hypothalamus regulates them. Drive reduction theory remains a good description of why a hungry person eats.

The second is the vocabulary. Need, drive, primary drive, secondary drive and drive stimulus are still standard terms, and drive reduction theory is part of AP Psychology Topic 4.6. The part that did not survive the experiments above is the claim that every reward works by reducing a drive. Curiosity, sweet tastes and brain stimulation showed that behaviour can be reinforced when no need is reduced. For those cases, the AP course framework teaches other theories next to drive reduction: arousal theory, incentive theory and self-determination theory, which covers intrinsic motivation.

Learning the vocabulary

Drive reduction theory comes with a small set of terms that look alike. They are need and drive, primary and secondary drive, drive stimulus, homeostasis, arousal and incentive. Terms this close together are easy to confuse when you reread them in a list. That is why our guide to memorising psychology terminology recommends testing yourself on them as a set. In Psychology World you can save these terms to a named list. You can then drill the list with flashcards or with typing, which tests recall most directly. Multiple-choice questions work once the list has four or more words.