Measuring memory.

Three broad approaches to the measurement of memory functioning will be described. The first of these involves using memory as a general indicator of any dysfunction in the central nervous system. This approach will be illustrated using Sternberg's short-term memory scanning paradigm. Its strengths are that such tests are often very sensitive, but they are often very difficult to interpret both theoretically and in practical terms. A second approach is to use a range of tasks selected so as to tap different aspects of human memory. Such an approach is of considerably more theoretical interest, and is discussed in more detail by Eysenck (this volume). Its weaknesses are that theories of memory are still changing relatively quickly, and that mapping such results onto memory outside the laboratory is often complex. A third approach is to attempt a more direct measure of everyday memory. The use of questionnaires for this purpose will be critically discussed, and a new test of everyday memory will be described. This test, the Rivermead Behavioural Memory Test, correlates well with observations of memory lapses in patients, and appears to offer a promising new line of development.