Concepts can be learned without the use of language. Rats can learn the concept of triangularity- by being rewarded for selecting triangles of various shapes and sizes and not rewarded for responding to other geometrical forms, they can learn to respond consistently to triangles. Since the triangles varying shape and size they are not responding to a specific object but to the concept of triangularity. Monkeys can learn the concept of “oddity”.

They can learn to select the odd stimulus object from a set of three objects, two of which are identical. The stimuli vary from trial to trial (for example, two circles and a square on one trial and two squares and a triangle on the next) so that the animal is not responding to a specific object but is learning to abstract the common property oddity

Generalization and Discrimination in Concept Learning:

Concept learning utilizes the psychological processes of generalization and discrimination. In learning the concept of triangularity, a rat generalizes the response initially to other geometrical forms, but since these responses go unrewarded they are extinguished and it eventually narrows the discrimination to triangles.

ADVERTISEMENTS:

A child learning the concept dog may generalize the term initially to include all small animals. But from parental corrections and personal observations the child learns to make finer discriminations until his concept further and distinguish between friendly dogs, whose wagging tails indicate that approach is safe, and unfriendly dogs, whose growls signify that avoidance is the best response. Eventually the child will learn to distinguish among various breeds.

Drawings similar to these were shown one at a time, starting on each trial from the top of the column and proceeding to the bottom.

As each picture was shown, the subject tried to anticipate the nonsense word paired with it. After the subject responded the experimenter called out the correct nonsense word, which is listed below each picture. On the first trial the subject had no way of knowing what word was paired with a picture, but over the course of several trials he gradually learned to anticipate correctly on all pictures.

Note that none of the pictures is repeated, so it is not learning a specific response to a specific stimulus, but rather learning a response to a concept. In this example the concept of face = RELK, building = LETH, tree = MULP, circle = FRD, the number two = LING, and the number five = DILT. (After Heidbreder, 1947)

ADVERTISEMENTS:

Human ability of concept formation:

Because of their language ability, humans are able to deal with all sorts of concepts, from fairly concrete ones, such as dog, to highly abstract ones, such as gravity, justice, and God. Concrete characteristics are generally easier to conceive than the more abstract relationships of form and number.

These studies used a paired-associate technique with the type of material illustrated in Figure. On each trial several pictures were presented one at a time, and the subject was required to anticipate the response paired with each one. The experimenter arbitrarily assigned a different nonsense word as the response for each concept.

Thus MULP might refer to the object concept tree, FARD to the spatial concept of circular patterns, and DILT to the number concept of five objects. On the second trial new pictures representing the same concepts were paired with the original responses. For example the Figure shows six pictures that might be presented on trial 1, a new set of six pictures for trial 2, and so on. The experiment continued until the subject responded correctly to all stimuli on a trial.

ADVERTISEMENTS:

A series of such experiments showed that object concepts (shoe, book, bird) were the easiest to learn, spatial forms the next easiest, and then numbers. Our thinking apparently tends to run to objects rather than to abstractions.

Piaget’s stages of intellectual development indicate that the child first learns object concepts and develops more abstract concepts only as he grows older. Interestingly enough, with certain types of brain damage an individual may lose the ability to deal with abstract concepts and respond only in terms of concrete ideas.

For example, he may be able to throw balls accurately into three boxes located at different distances from him but not be able to state which box is nearest and which farthest or explain his procedure in aiming. This inability to think abstractly is so marked in certain types of brain damage that performance on concept-learning tasks is sometimes used as a basis for diagnosis.