Learning is colloquially described as a process by which we acquire knowledge about the world, but for experimental science, the way in which we can measure or observe this phenomena is by a behavioral change of the organism. With this in view, learning can be defined as a relatively permanent change in behavior due to experience. This definition therefore excludes behavioral changes derived from developmental maturation, or by drugs or illnesses. The behavioral change has to come from exposure to its environmental medium, that has to act on sensory systems to produce stimuli.

Some authors exclude from the definition of learning very simple experience-induced behavioral modifications, such as sensitization or habituation, although most neuroscientists identify them as nonassociative learning. Nonassociative learning is a behavioral change brought by repeated presentation of one stimulus. In the case of sensitization, a stimulus that originally elicited a weak or no response, starts evoking stronger responses after several presentations, or when the presentation of one very intense stimulus evokes stronger responses at other stimuli. For example, after a very loud crash sound, smaller noises can startle a person, which otherwise would go almost unnoticed. In the case of habituation, repeated presentation of the same stimulus produces decreasing responses to it. In the example of the loud crash, if it keeps sounding repeatedly every twenty seconds, the startle to it will decrease in further presentations.

What is commonly called learning refers to what is known in experimental science as associative learning. This name is derived from the fact that after careful and detailed study of learning, specially in animals, most authors have concluded that learning derives from a association of events. The type of these events, and the way they are associated are classified into two experimental paradigms, one called classical or pavlovian conditioning, and the second is operant or instrumental conditioning. Conditioning in this sense is not learning itself, but the means (procedure) by which learning is achieved.

In the case of classical conditioning, what is achieved is that a response that was originally elicited by one stimulus, can now be elicited by another one that originally had no effect. In the experiments made at the turn of the century by Ivan Pavlov, he proved that the effect was due to the association or pairing of two stimuli. The first stimulus was one that evoked the response to be tested, without prior experience. These types of stimulus-response pairs are known as innate reflexes, and the stimulus is named an unconditioned stimulus (US) and its response an unconditioned response (UCR). For example, the smell of food (US) elicits the salivation response (UCR) naturally without any training. For conditioning, a second stimulus is used that does not evoke any response that affects the UCR studied, and is called a conditioned stimulus (CS), in out example it could be the turning on of a light (which obviously elicits several responses by itself, but none that affect salivation). After the CS is presented repeatedly with the US, either simultaneously or with the US slightly after, the CS can produce a response that is similar, if not identical to the original UCR, and is therefore called conditioned response (CR). In our example, if a light is turned on slightly before the presentation of food, and this is done repeatedly, the turning on of the light will produce the animal to salivate. In the case of this example, the conditioning is said to be appetitive, since the US is rewarding, but if an aversive unconditioned stimulus is used, such as an electric shock, the conditioning is said to be defensive.

Once conditioning has taken place, each time the CS is presented, the animal will respond with the CR, but if this is done repeatedly, without ever pairing again with the US, the conditioned response will no longer present itself. This is called extinction conditioning. It is important no to confuse extinction with forgetting, although in both cases the conditioned response is no longer present. Extinction is an active process which depends on unpaired presentations of the CS, while forgetting does not depend on any conditioning procedure, but depends on other factors such as time passage, biological cycles, developmental stages or brain physical or chemical manipulations.

In the case of operant conditioning, what is achieved is that an animal increases the rate or probability of a particular response, innate or not, when that response is often followed by a reinforcing (rewarding) stimulus. The typical example of this type of conditioning is a rat placed in a box with a lever sticking out. If the rat presses the lever by accident, or during exploratory behavior, it is given a small food pellet. After a few of these events, the rat will start to press the lever more and more often. In this case, the great difference with classical conditioning is that the conditioned response (or, more exactly, the response to be conditioned) appears before the unconditioned stimulus, either spontaneously, or by previous learnings. Once the response appears, if it is followed shortly by a reinforcer (either an appetitive stimulus or the removal of an aversive one) on several occasions, the animal most likely will start to repeat frequently that response, even when the response is not followed by the reinforcer anymore. However, if this unpaired response occurs too often without any reinforcement, extinction will occur, and the response will disappear, or go back to the spontaneous levels it had before conditioning

Go to the list of topics