IALP 2025

Decompositional verb processing in children with DLD in Croatian

Lucija Milic 1 Prof. Lidija Cvikić 2
1Department of Psychology, University of Zadar, Croatia
2Faculty of Teacher Education, University of Zagreb, Croatia

Difficulties in mastering morphological rules and lexical processing deficits are well-known deficits in Developmental Language Disorder (DLD). Despite this, there is a lack of research on processing of morphologically complex words in people with DLD. According to the dual-route model (Schreuder and Baayen, 1995), morphologically complex words can be processed directly, as whole word forms and decompositionally, as morphemes. Words that are semantically transparent and have productive affixes tend to be processed decompositionally. So, this type of processing is more pronounced in morphologically rich and transparent languages and it possibly develops earlier. The goal of this study was to investigate whether decompositional processing develops in children with DLD in morphologically rich language such as Croatian. The effects of decompositional variables, i.e. prefix productivity and semantic transparency, were measured in an auditory lexical decision experiment. The stimuli included prefixed verbs, pseudoverbs with true prefix and false stem (+P-S) and pseudoverbs with false prefix and real stem (-P+S). Participants were children with DLD and typically developing (TD) children in 1st and 4th grade of elementary school. Results showed that children with DLD were less accurate in processing (U=1535.00, Z=2.189, p<.05 for real verbs, U=1201.50, Z=-3.82, p<.01 for pseudoverbs), while there was no difference in speed of processing (F (1,124)=3.09, p=.081) compared to TD children. All participants processed real verbs more efficiently compared to pseudoverbs and pseudoverbs +P-VS more efficiently compared to -P+VS stimuli (F (1.39, 172.67=15.05, p<.001). The effects of prefix productivity and semantic transparency were determined for all participants. Verbs with productive prefixes had shorter response times (F (1,122=9.59, p=.002) and higher accuracy (Z=-4.69, p=.00), while semantically transparent verbs were processed with greater speed (F (1,122=4.41, p=.038), but with no effect on accuracy (Z=-.95, p=.34). Taken together, effects of prefix productivity and semantic transparency obtained for all participants indicate that children with DLD, as well as TD children, have developed decompositional processing of morphologically complex words. However, the lower accuracy rates in lexical decisions indicate that the underlying lexical representations of existing morpheme combinations in Croatian are underspecified in children with DLD.

Schreuder, R., Baayen, R.H. (1995). Modeling morphological processing. In L.B. Feldman (Ed.) Morphological Aspects of Language Processing (pg. 131-154). Hillsdale, NJ:Erlbaum.