A strict test of the phonological loop hypothesis with Libyan data.

In Arabic, there are two ways, differing in length, of pronouncing each digit. This feature of word pairs that are conceptually identical but of different length allows for a stricter test of the phonological loop hypothesis than has been reported previously. Libyan schoolchildren, both boys and girls, of two grades took part in a quasi-experimental study in which shorter and longer digits and shorter and longer words were presented in recall and pronunciation tasks. The results confirmed the validity of the phonological loop model in that memory span was longer for shorter stimuli. Analysis of the data with structural equation modeling showed that only 17% of the individual-level variation in memory span could be explained by pronunciation speed. In an analysis of covariance, it was determined that pronunciation speed could account for a substantial proportion of the variation found across age (8-year-olds vs. 10-year-olds), stimulus modality (digits vs. words), and stimulus length (short vs. long), but not for gender differences.