Access to the full content is only available to members of institutions that have purchased access. If you belong to such an institution, please log in or find out more about how to order.



DOI: 10.4324/9780415249126-W021-1
Version: v1,  Published online: 1998
Retrieved March 17, 2018, from

Article Summary

Learning is the acquisition of some true belief or skill through experience. Rationalist/idealist philosophers held that the very constitution of thought guarantees that fundamental laws hold of the world we experience, and that our understanding of these laws was therefore innate, not learned. The empiricist tradition, doubtful of these Rationalist claims, denied that much was innate, and held that learning occurred through associations of mental representations. This view was lent support by the nineteenth-century development of physiological psychology, which led to a view of learning as a system of adjustments in a network without any intervening representations, a perspective that led in turn, in the twentieth century, to behaviourist studies of stimulus–response associations, and eventually to contemporary neural net computational models.

Empiricism, however, had also invited, especially with Hume, doubts that the correspondence between mental representations and the world could be known. Hume believed people learn, or at least form new habits, but he did not think there could be any normative theory of learning – any way of making it ‘rational’. His scepticism led to the development by Bayes and other statisticians of formal theories of how learning from evidence ought to be done. However, the standards that developed in the form of the theory of subjective probability proved impossible to apply until very fast digital computers became available.

The digital computer in turn prompted both novel normative theories of learning not considered by the statistical tradition, and also attempts to describe human learning by computational procedures. At the same time, a revolution in linguistics held that humans have an innate, specialized algorithm for learning language. Applications of computation theory to learning led to an understanding of what computational systems – possibly including people – can and cannot reliably learn. Major issues remain concerning how people acquire the system of distinctions they use to describe the world, and how – and how well – they learn the causal structure of the everyday world.

Citing this article:
Glymour, Clark. Learning, 1998, doi:10.4324/9780415249126-W021-1. Routledge Encyclopedia of Philosophy, Taylor and Francis,
Copyright © 1998-2018 Routledge.

Related Searches


Related Articles