Language profile of primary progressive aphasia in a linguistically diverse cohort in India

Abstract

AbstractBackgroundThe accurate characterization of language and semantic memory deficits is crucial for diagnosis of primary progressive aphasia (PPA). In a linguistically diverse society like India, assessment with linguistically appropriate and culturally relevant tools is essential but challenging. This study aimed at developing language and culture specific tests to diagnose PPA and profile cognitive‐linguistic characteristics of a diverse cohort of PPA in India.MethodBased on typical characteristics of Indian languages: free‐word order, rich system of morphology and case‐marking that are features that differentiate Indian languages from English, a language‐specific assessment battery was developed in seven languages: Hindi, Bengali (Indo‐Aryan languages) and Kannada, Telugu, Malayalam and Tamil (Dravidian languages). In the background of the importance of case‐markers in the syntax of Indian languages, a unique syntax assessment test was specifically developed. To evaluate semantic memory, the Indian Semantic Memory battery was developed using stimuli appropriate for a pan‐Indian cultural context. All patients underwent assessment of language proficiency and comprehensive cognitive evaluation by examiners proficient in the language of testing.Result36 patients with PPA: 22 with semantic variant (svPPA), 12 nonfluent variant (nfvPPA) and 2 logopenic variant (lvPPA) participated. The cohort consisted of proficient speakers of seven Indian languages. A particular difficulty in the appropriate use of case markers on the new test of syntax was revealed in nfvPPA compared to svPPA. Further, the core diagnostic feature of ‘agrammatism’ for PPA in English‐speaking nfvPPA was not as dramatic in Indian patients owing to the morphological richness of Indian languages, leading to reduced contrast between fluent versus non‐fluent aphasia. Consistent with diagnosis, a significant reduction in fluency was demonstrated in nfvPPA (p< 0.006 on fluency tasks). Semantic memory was significantly affected in svPPA as revealed on non‐verbal tasks i.e., word‐picture matching (p<0.02)and semantic association task(p<0.01) of Indian Semantic Battery.ConclusionUsing language and culture specific assessment batteries tailored to the Indian context, a comprehensive clinical characteristics in PPA subtypes were identified. The need to re‐define agrammatism with respect to Indian languages with additional weightage to the use of case markers that assist in identifying the PPA subtype was a novel finding.