PSYCHOLOGISTS study learning in a laboratory. This means that many important variables influencing behaviour can be excluded or controlled.
Learning is defined as an adaptive change in observable behaviour. If you fall from the top of the Eiffel Tower, your behaviour when you reach the ground will have changed from your behaviour on the viewing platform, but you haven’t learned anything (well, you might have – but too late).
Learning also requires that the adaptive change is contingent upon certain stimuli in one’s environment.
I won’t go on about defining the term learning, but it’s important to agree on what we are talking about to make sure things can’t slip past unintended.
Learning falls into two categories, classical conditioning and operant conditioning.
In classical conditioning, as pioneered by Pavlov and his dogs, a stimulus (a bell) is followed by another stimulus (food). Soon, when the bell is rung, saliva begins to flow. The dog has been conditioned to associate the sound of the bell with the presentation of food. It has learned. However, you will notice that the dog has not had to do anything to achieve this learning. The dog’s autonomic nervous system has done the learning for him. This is an important observation. You can learn without wishing to learn. Circumstances can dictate.
In operant conditioning, pioneered by B F Skinner, the organism under observation is required to do something: to operate on its environment in some way. The simplest way to achieve this is to get it to press a lever. The lever can be wired to an electric switch, which can deliver a reward – a food pellet, or access to a sexually receptive mate, or a punishment such as an electric shock. This result of the initial behaviour is contingent upon it. It won’t happen unless the behaviour happens first. This is an important distinction from classical conditioning, in which the stimulus happens first, and the behaviour in response occurs afterwards.
If the rat is motivated, by being hungry for example, and presses the lever and receives a food pellet, the probability of it pressing the lever increases. After only a short time the rat will eagerly press the lever. Its behaviour has adapted. It has learned.
So far so good. We have cracked the laws that govern learning; now what?
A series of experiments by Seligman and Maier led to the coining of the term ‘learned helplessness’. In one of my psychology lectures the following experiment was described:
A large tank is half filled with water. In the centre of the tank is a raised platform atop a slim pole, which cannot be climbed up or down. At one end of the tank is a broad shelf, divided into two compartments. Each compartment is fronted with a drop-down flap, which can be locked shut. One flap is painted in black and white stripes, the other in black and white dots, to be perceptually distinguishable.
A rat which has been previously conditioned to this experimental set-up is positioned on the platform. Food pellets are placed behind the striped flap. The dotted flap is locked. The rat has already been trained to leap. If it leaps at the unlocked flap, the flap drops back and the rat gains food pellets and freedom. If it leaps at the dotted flap, it smacks its nose and falls into the water. This occurrence is called a negative reinforcement. The rat quickly learns to leap only at the striped flap.
Then the locks are swapped. The striped flap is now locked and the dotted flap is unlocked. After a few unsuccessful leaps, the rat switches to the dotted flap.
For the next run of the experiment, both flaps are locked. The rat leaps at one then the other. The result is always a dunking. Eventually the rat refuses to jump and remains motionless on the platform. In order to force a leap, a blast of cold air is directed at the rat. The experiment is run until behavioural breakdown occurs. Whenever the rat is placed on the platform, it leaps off randomly in all directions, always ending in the water.
Now comes the clever part. Both flaps are removed. The food pellets and the path to escape are clearly visible. But the breakdown of the rat’s behaviour is ineradicable. When placed on the platform the rat freezes. Even the jet of air won’t budge it. If it does eventually leap, it will not be to the escape shelf. Only by lifting the rat from the platform and manually placing it on the shelf can the dysfunctional behaviour gradually be extinguished.
Conclusion: When faced by a chaotic, unpredictable, random environment, wherein the rules are constantly changing, and whatever the rat does makes no difference, the rat will be ‘broken’ and will cease to be able to behave adaptively. An unkind experiment, but revealing.