Theory that a belief can be called justified if it is formed by a process that is reliable, that is normally produces true beliefs.
This is an externalist account of justification if it is not insisted that the believer be aware of the method's reliability.
This appeal to reliability may also contribute to an analysis of knowledge, though the questions of when a belief is justified and when it amounts to knowledge are different.
This is because to have knowledge we may need more than justified belief (for example the belief must at least be true, and even a method that normally produces true beliefs might on some occasion produce a belief that was indeed true but only by accident: would that still amount to knowledge?); also some knowledge, for example of some of our inner states, may not need justification.
A I Goldman, 'What is Justified Belief?', Justification and Knowledge, G S Pappas, ed. (1979)