Examining Speech Production in 3- and 4-Year-Old Typically Developing Children: A Comparison Study for the American English Phrase Sample

The standard of care by multiple cleft teams includes utilizing controlled speech samples, such as the American English Phrase Sample (AEPS), which is controlled for each sound class in different word positions to rate cleft speech characteristics, mainly resonance, within multi-word contexts. This study aimed to provide information on traditional speech errors and speech sound accuracy in typically developing (TD) children aged three to four years on this phrase repetition task. Additionally, it compared speech sound accuracy between single-word articulation and phrases. Finally, the speech samples from a small group of non-cleft children with a speech delay were described in relation to their TD peers. Thirty typically developing children without cleft palate and seven children with speech delays, ranging in age from 3-4;11 years old, were recruited from a larger study. The Sounds-in-Words subtest of the Goldman-Fristoe Test of Articulation-3rd Edition (GFTA-3) and the AEPS were administered. The GFTA-3 and AEPS were analyzed for traditional speech errors, Percent Consonants Correct (PCC) total and PCC by manner. Additionally, phonological processes were examined using the Khan-Lewis Phonological Analysis-3rd Edition (KLPA-3). Analysis of variance (ANOVA) and effect sizes were computed for the Substitutions, Omissions, Distortions, and Additions (SODA) and PCC comparisons of the children with typical development. The data for the children with speech delays are presented descriptively due to the small numbers. Results revealed significant decreases in PCC for certain categories at the phrase level for 3-year-olds, with little variation in PCC for 4-year-olds. Children with speech delays exhibited lower PCCs for multiple manner classes compared to their TD peers. Age showed significance in increased PCC for 4-year-olds. Substitution errors were prevalent in TD children, while children with speech delays demonstrated various error types. Error reduction correlated with increased age and varied by word position. Patterns differed between TD and speech delay groups across linguistic contexts. Though originally intended to assess cleft palate speech characteristics, normative data on the AEPS helps contextualize speech characteristics observed within typical development. The current study addresses the lack of normative data on the AEPS for comparison to children with cleft palate with or without cleft lip (CP+/-L). Additionally, it provides normative data for PCC and PCC by manner at the single-word and phrase level. Overall, the results of this study support the claim that children perform similarly on the AEPS as the GFTA-3, with a few variations depending on context.