Low short-term memory capacity in individuals with dyslexia and enhanced short-term memory capacity in musicians are well documented, yet their causes are disputed. In this paper we claim that the difficulty of individuals with dyslexia stems from an inefficient learning from multiple exposures (red curve in Figure 1), and that the benefit of musicians stems from s super-efficient learning from multiple exposures (purple curve in Figure 1). Individuals with dyslexia do not benefit to the same extent from multiple encounters with stimuli, and then, when they are tested for short-term memory and frequent items are used in the test, they get especially low results compared to controls (right hand side of Figure 1). For musicians it is the other way around. So, at least to a large extent, the deficit in dyslexia and the benefit of musicians is not of short-term memory, but of the utilization of long-term memory while performing short-term memory tasks.

By contrast, the learning of a repeated sequence was similar for all groups, so it is not that every measure of short-term memory in dyslexia is reduced and every measure in musicians is enhanced, but the deficit/advantage is specific. Thus, these results suggest that the learning of a repeated sequence is based on a different mechanism than the benefit of long-term item frequency, and only the latter is impaired in individuals with dyslexia and is enhanced in musicians.