All of these laws have been called into question by recent investigations. It has been shown that the alleged influence of repetition does not always operate and that mental and psycho-motor functions sometimes continue to improve after practice ceases. As for the law of effect, it is difficult to understand how satisfaction which follows a successful response could operate retroactively to strengthen the bonds which have already acted. The law of trial and error, which alleges that learning to solve problems is typically accomplished by the gradual elimination of unsuccessful responses, at least in the case of animals and young children, has also been called into question on the basis of experimental data.

Though the laws of learning which once seemed to have been so definitely established are in dispute, learning nevertheless takes place. The mistake was probably in assuming that the laws were more absolute than they actually are, whereas a given law operates only at times and under certain conditions. As originally formulated and as usually understood, they make learning a more mechanical process than it in fact is, and postulate a neural mechanism far too simple. The investigations of Lashley and Spearman point to the conclusion that in learning the brain does not operate merely as a very complex set of bonds or connections, each serving a specific purpose, but that large areas of the cortex act to a greater or less extent in a unitary manner. The center of attention is shifting from the mechanical aspects of learning to the role played by in sight. In the newer terminology, learning takes place by the formation of new Gestalten, or meaning complexes, not by the mere addition and subtraction of bonds. It is more a rational and less a mechanical process than was formerly supposed.

The preceding statements will doubtless seem very vague to the reader who is not acquainted with the recent literature of Gestalt psychology. However, the applications of the newer theories to education are fairly clear. The psychologist is coming to view the educative process less and less as mere habit formation, and to lay greater stress upon the part played by insight, attitudes, and interests. The more recent textbooks on educational psychology give less space to economic methods of memorizing and more to explanations of how meanings develop. The effect upon school work should be to reduce the emphasis upon routine drill and to give larger place to activities of the kind that stimulate the child to set problems for himself and to seek rational methods for their solution.

Because of limitations of space I shall omit entirely from this discussion consideration of certain issues which have figured so largely in texts in the psychology of learning: the form of the learning curve and of the curve of forgetting, the relative advantages of learning by parts or by wholes, and the economic distribution of practice periods. These issues have their interest, but they have to do chiefly with habit formation, which is the least important kind of learning.

A very controversial problem in connection with learning is the extent to which the effects of training are capable of transfer to other functions. This is the so-called theory of formal discipline. Until a half century or so ago it was almost universally believed that extensive transfer takes place, and that one of the main purposes of education should be to train such hypothetical faculties as perception, reasoning, imagination, etc. This view was vigorously combated by Herbart on theoretical grounds, and later by James and Thorndike on the basis of experimental findings. The influence of these psychologists brought an extreme reaction to the older views and gave prevalence to the dogma that transfer is entirely too negligible to merit being taken into account in the choice of subject matter for the school curriculum. The weight of opinion of to-day favors an intermediate position. It is certain that the earlier educational theorists laid entirely too much stress upon mind training. It is hardly less certain that Thorndike and his followers have overestimated the specificity of training. The latter viewpoint favors selection of curriculum material almost entirely upon the basis of so-called "practical" values, to the neglect of values that have to do with the cultivation of intellectual attitudes.