Background: Phonological representations are important for speech and literacy development. Mispronunciation detection tasks have been proposed as an appropriate measure of phonological representations for children with speech disorder. There has been limited analysis, however, of the developmental complexity of task stimuli. Further, the tasks have not been used widely with typical populations.Aims: The study aimed to examine the developmental progression of children's performance on a mispronunciation detection task, to evaluate the complexity of the task's stimuli, and to analyse the association between task performance and other skills important for literacy success.Methods & Procedures: A cross-sectional design was used to compare the mispronunciation detection performance of monolingual children within mainstream classrooms aged 4;04;5 (n = 26), 4;65;0 (n =49), and 5;15;6 (n = 64). Comparison of the complexity of vowel mismatches, single consonant mismatches, consonant deletion mismatches, and consonant transposition mismatches was conducted.Outcomes & Results: Children in the youngest age group were outperformed by the older age groups in the phonological representation task. Fine-grained mismatches were more difficult to detect than coarse-grained mismatches. There was a trend for vowel substitutions to be the most difficult mismatch to detect. Performance in the mispronunciation detection task was moderately correlated with the children's receptive vocabulary, speech, phonological awareness, and letter knowledge skills.Conclusions & Implications: Mismatch type must be taken into account when designing mispronunciation detection tasks. Access to segmental phonological representations is related to speech, language, and letter knowledge in children from mainstream classrooms.