A Corpus-Based Analysis of Frequency Effects in English Pluralization

Abstract

When it comes to how the human mind processes linguistic information, frequency effects have been considered as supporting evidence for the dual mechanism approach. This approach proposes that only irregulars are stored in the associative memory, while regulars rely on rule-based processing. Accordingly, followers of this approach predict that irregular forms may exhibit responsiveness to frequency due to their storage nature, whereas regular forms may show a lack of sensitivity to frequency. The study aims to explore the influence of frequency on regulars and irregulars in the English language, with a specific focus on the noun domain. We investigate whether irregular nouns are more frequent than regular nouns and whether frequency effects differ between the two frequency-based groups. The study depends on data from the Corpus of Contemporary American English. The top regular and irregular nouns, as well as their low-frequency counterparts, were selected for analysis. The study then compares the word frequency distributions of these nouns, both in singular and plural forms, and employs descriptive and statistical analyses to assess the significance of variation in frequency effects between regulars and irregulars. Contrary to the predictions of the dual mechanism approach, the findings indicate that regular nouns generally occur more frequently than irregular nouns in Contemporary English, both in singular and plural forms, challenging the dual mechanism's expectations. These findings have significant implications and consequences for our understanding of language processing mechanisms and indicate the requirement for more elaborate models.