Verb morphology impairment in a bilingual speaker with non-fluent aphasia

• ¹ Lehman College, CUNY, Speech-Language-Hearing Sciences, United States

Introduction Current literature on bilingual aphasia offers little evidence regarding the manifestation of verb morphology impairment in two languages. De Diego Balaguer et al. (2004) report parallel impairment in the two languages of early, highly proficient Catalan-Spanish bilinguals with non-fluent aphasia, both demonstrating greater difficulty conjugating irregular than regular verbs on a sentence completion task. Further, greater difficulty with tense selection in verb production has been found in structured tests as compared to semi-spontaneous connected speech production (Faroqi-Shah & Dickey, 2009). The present study examined tense selection and verb conjugation in two languages of a bilingual on different tests of language production, to address the following question: What aspects of verb morphology should be considered to determine parallel or non-parallel impairment in the two languages of bilinguals with aphasia? Methods The participant was a 73-year-old Hebrew-English speaker with moderate non-fluent chronic aphasia. He has been exposed to English in an immersion context for 35 years and used both languages at the time of the CVA. The Western Aphasia Battery (WAB, Kertesz, 1982) administered in English revealed an aphasia quotient of 71. He underwent a comprehensive battery of language tests in both his languages in an alternating, counter-balanced order. The battery included a structured test, a modified version of the Verb Inflection Test (Faroqi-Shah, unpublished), and a semi-spontaneous picture sequence description task (based on the Narrative Story Cards, Helm-Estabrooks & Nicholas, 2003). Results In the Verb Inflection Test, the overall performance in Hebrew (L1) and English (L2) was comparable, but the error patterns varied. In Hebrew, 37% of verbs (10 out of 27) were produced in the appropriate tense and with the correct conjugation. The participant’s most frequent error on the remaining verbs (9 out of 27, 33%) was producing the verb in an inappropriate tense (e.g., “Tomorrow, girl painted pictures”). None of the errors in Hebrew were verb conjugation errors. In English, he provided correct responses for 33% of verbs. Errors in tense selection occurred as often as in Hebrew (33% of verbs); however, in English verb conjugation errors (e.g., “Yesterday, girl weared the skates) were also frequent. In the picture sequence description task, the participant was less impaired in Hebrew than in English. Specifically, out of 19 opportunities in Hebrew, he produced 16 verbs (84%), all of which were correctly conjugated. In English, he produced verbs in 8 out of 22 opportunities (36%). Out of these, 3 were correctly conjugated (38%). Conclusion Whereas accuracy levels on the structured task were comparable in the participant’s two languages, he was more accurate in L1 than in L2 on the semi-spontaneous task. Further, production in L1 revealed only tense-selection errors evident only in the structured task; production in L2 included both tense-selection and conjugation errors, and those were found on both tasks. Thus, what may appear as comparable impairment in producing conjugated verbs on a structured test could rather be construed as non-parallel levels of morphological deficit. These findings highlight the importance of considering both tense selection and conjugation across different tasks.