Context effects on verb production in specific language impairment (SLI): Confrontation naming versus connected speech

Abstract

A handful of studies have shown that verbs are more vulnerable than nouns to retrieval deficits on picture-based naming tasks for children with specific language impairment (SLI). The aim of this study was to examine if the disproportionate verb as opposed to noun production deficit reported for naming is also found in connected speech. Sixteen children participated in the study: eight children diagnosed with SLI (mean age: 6:3 years) and eight typically language developing (TLD, mean age: 5:9 years) controls. Verb and noun production was measured in connected speech and compared to picture confrontation naming. Both groups of children showed a significant difficulty naming verbs compared to nouns. In contrast, they did not differ on the total number of both verb tokens and verb types produced in connected speech. The findings indicate that the previously reported verb retrieval difficulties in SLI are a product of the confrontation naming task demands rather than a true verb deficit.

• SLI
• verb production
• confrontation naming
• connected speech

Notes

ⁱNote that 30 out of the 33 children with SLI reported in the published studies showed a significant impairment for verbs. The remaining three showed equal difficulties retrieving verbs and nouns. The study by Kambanaros et al. (2010) was not included in this analysis as the children were significantly younger (preschoolers) and the verb-noun difference failed to reach significance.

ⁱⁱThis claim was made by Bastiaanse & Jonkers (1998) for all patients with aphasia.

ⁱⁱⁱEight of the 11 children with verb impairment in naming are reported in the present investigation, from which narrative samples were elicited.

^ivThis measure was used to match the language-impaired and non-impaired groups on narrative ability.

^vThe methodology is adapted from Pashek and Tompkins (2002).

^viA stricter criterion of 0.05 divided by 4 (i.e., 0.0125) was used because four tests were run simultaneously.

^viiFor example, all of the following were judged to be acceptable from the narrative re-tell samples: “The driver (repaired) (mended) (fixed) (helped) (got) the bus”.

^viiiThe difference between the two groups across the different verb types was significant for the SLI group (Kambanaros, 2013, for details).