Thorndike's Cat

Thorndike, E. L. (1898)

The cat is put into a box that can only be opened if the outside latch is opened (by putting a paw through the gap from inside) and the bolt is pulled back (by pressing on a pedal). At first, the cats move around in a 'random' fashion because they don't know how to get out. Eventually by accident they will press the pedal and push the latch, and can escape the box to eat the tasty food outside. By associating those two actions with a positive outcome, the behaviour is reinforced: push pedal + open latch = food. Next time the cat is put in the box (this happened a lot) it is more likely to do perform these actions again because of the association with food. The cat gets quicker at opening the box as it 'learns' (the association between behaviour and food becomes stronger). This produces a 'learning curve', and the trail-and-error approach is known as 'instrumental conditioning'.

(image from www.psywww.com)

Thorndike claimed the cat had no idea about the connection between its actions and possible consequences, because when the cat first goes in the box its behaviour is largely erratic. If it knew about a link, Thorndike suggested it would show some kind of systematic approach to getting out of the box, such as repeatedly attacking the door. The cats never demonstrated an obvious strategy or sudden flash of realisation, therefore his conclusions were that:

a) there is no 'mind' to house mental processes that control behaviour

b) patterns of behaviour where only engaged in because of their association with positive outcomes.

The first point can be thought of as a premature conclusion for a relatively new science, like the 'world is flat' theory. Early psychologists denied the existence of a thinking 'mind', instead claiming that practically all behaviour was the result of learning to associate certain situations with negative outcomes (and avoiding them, e.g. the Little Albert experiment) and other situations with positive outcomes (such as in Pavlov's dogs experiment). The mind was largely ignored in early psychology, being thought of more as a 'black box'. The second point is one of the fundamental principles of psychology, as it is something that is relatively easy to observe and manipulate in a wide variety of situations.

Thorndike trialled his puzzle box on a number of animals, including dogs, monkeys, chickens and, strangely, fish, finding that each produced the same kind of S-shaped learning curve. Although each species took a different number of trials to 'solve' the box, because the shape of the graph was the same it was suggested as evidence that different species learn the same way.