ABSTRACT
A person’s memory span is measured by the number of items (usually digits) that can be reproduced in the correct order immediately following stimulus presentation. Performance in this task is subject to strict limitations—for adults it is defined by the Miller parameter of 7±2 (Miller, 1956). Memory span is seen as a traditional measure of short-term-memory or working-memory capacity, and it is also used as a subtest in intelligence test routines (i.e., Wechsler, 1956, 1958). In stark contrast to the fact that this parameter is so commonly used as a means of defining short-term-memory performance, our knowledge over the mechanisms that ultimately limit memory span is, as yet, highly incomplete. This has led to an intense discussion in recent years (i.e., Cowan et al, 1998; Dosher & Ma, 1998; Hulme, Newton, Cowan, Stuart, and Brown, 1999). With the present study, we aim to contribute to this discussion by reexamining data we have published since 1997 and analyzing data gathered specifically for the purpose of furthering the quest for the mechanisms that limit memory span.
