Do data from children with specific language impairment support the agreement/tense omission model?

Area of Science:

• Linguistics
• Developmental Psychology
• Child Language Acquisition

Background:

• The agreement/tense omission model (ATOM) is influential in explaining pronoun case-marking errors in children's speech.
• ATOM predicts that non-nominative subjects with agreeing verbs are rare and can be disregarded as noise.

Purpose of the Study:

• To test the ATOM prediction regarding the frequency of non-nominative subjects with agreeing verbs in children with specific language impairment (SLI).
• To investigate potential explanations for observed pronoun case-marking patterns in children with SLI.

Main Methods:

• Analyzed speech data from 4 children with SLI.
• Compared the observed frequency of non-nominative subjects with agreeing verbs against chance expectations.
• Examined the role of developmental changes and unanalyzed contractions in case and agreement marking.

Main Results:

• Three out of four children with SLI used non-nominative subjects.
• Two children produced non-nominative subjects with agreeing verbs significantly less often than predicted by chance.
• Observed asymmetries were potentially explained by developmental factors and contractions.
• Non-contracted agreeing forms showed frequencies close to chance expectations.

Conclusions:

• Findings challenge the ATOM's prediction of rarity for non-nominative subjects with agreeing verbs in children with SLI.
• Developmental changes and use of contractions may explain observed pronoun case-marking patterns.
• Methodological considerations for studying pronoun case marking in children are highlighted.