Effects of frequency and morphosyntactic structure on error detection, correction, and repetition in Swedish-speaking children

Abstract

Grammatical error detection and correction are often used to test explicit language knowledge. This study investigated effects of token frequency and error type in error detection, correction, and repetition, and performance on the three tasks were compared and related to models of metalinguistic awareness and development. Thirty Swedish-speaking 10-year-olds with typical language development participated in the study, which focused on four morphosyntactic errors: the infinitive instead of past tense for regular and irregular verbs, and the omission of the obligatory indefinite article in common and neuter gender noun phrases. Target verbs and nouns were of high or low frequency. Results showed significant effects of verb frequency in all tasks, and effects of noun gender for error detection, but not for correction and repetition. Children detected significantly more past-tense errors than they accurately corrected, but the opposite result was seen for noun phrase errors. The patterns of results both within and across tasks imply that implicit language knowledge affects performance, and that lexical frequency, even of familiar words, needs to be controlled when designing tasks for measuring grammatical knowledge. The particular challenge of the Swedish neuter noun phrase in language development and language processing needs to be further investigated.

Can we utilize Large Language Models (LLMs) to generate useful linguistic corpora? A case study of the word frequency effect in young German readers

“…Word frequency is an essential and well-researched parameter in word recognition research (Brysbaert, Mandera, & Keuleers, 2018;Brysbaert, Stevens, Mandera, & Keuleers, 2016;Gregorova, Turini, Gagl, & Võ, 2023). One well-replicated finding is that words that often occur (i.e., high-frequency words) are recognized faster and more accurately compared to less common words (i.e., low-frequency words; Adelman, Brown, & Quesada, 2006;Baayen, 2010;Brysbaert et al, 2016;Gregorova et al, 2023;Hallin & Reuterskiöld, 2018;Lieven, 2010;S. A. McDonald & Shillcock, 2001;Stokes, 2010).…”