The word length effect shows a connection between word length and working memory performance. Although the relationship between digit verbal length and digit span has been investigated between languages, it has not been investigated within a language. It was hypothesized that this effect can be shown as a connection between the prevalence of digits with two syllables and digit span score. The study examined the effect of amount of syllables on Norwegian digit span test scores by altering the prevalence of two-syllable digits using three conditions in a repeated measures design ( N = 54). Results suggest that an elimination of two-syllable digits in a digit span test significantly reduced forward span test score (Cohen’s d = 0.36), but had no effect on backward span scores. These results suggest that a balanced distribution of two-syllable digits in a forward digit span tests should theoretically increase the test’s comparability and reduce language-related biases thus increasing the test’s parallel-form reliability. A peak-span model is proposed to integrate the findings into previous research on the word length effect.