Integrated narrative analysis in multilingual aphasia: The relationship among narrative structure, grammaticality, and fluency

Abstract

Background: Amid robust evidence for the efficacy of language treatment in aphasia, equivocal results have been reported for the generalisation of treatment effects to items and tasks not practised during therapy. Moreover, measuring generalisation using functional language production has proven challenging, especially in the context of multilingual aphasia.

Aim: In this paper we analysed four domains—discourse structure, sentence structure, instances of linguistic dysfluencies, and lexical choice—in narratives produced following treatment to assess aphasia treatment generalisation within and across languages.

Methods & Procedures: Two treatment phases were administered to a trilingual speaker with mild, chronic, non-fluent aphasia, targeting language production. Multiple baseline testing was administered prior to and following each treatment phase and included elicited narrative production in three languages: Hebrew, English, and French.

Outcomes & Results: The results demonstrated treatment-related changes in the treated language as well as in the non-treated languages. Positive change was found in narrative structure and in sentence grammaticality. A trade-off pattern between grammaticality and fluency was evident, with higher proportions of grammatical sentences associated with increased instances of linguistic dysfluencies. Within-language and between-language generalisation was observed, especially following the first treatment phase which targeted morphosyntactic structures.

Conclusions: We argue for the importance of an integrated analysis of connected language production in the assessment of aphasia treatment efficacy and in multilingual aphasia.

Keywords:

• Aphasia
• Treatment
• Generalisation
• Narrative
• Grammaticality
• Dysfluency
• Trilingual
• Integrated narrative analysis

Acknowledgments

We thank the participant, and Rebecca Kastl, Lauren Liria, Monica Levy, Valerie Levy, Loraine K. Obler, Gloria Olness, Carol Park, Yana Pugach and Joel Walters. This paper was supported, in part, by NIH grant DC-009792 (PI: Goral).

Notes

¹Due to logistic constraints a total of 9 hours of therapy was administered in each block. Whereas several treatment studies have also reported short-term treatment (e.g., 10–20 hours in Kiran & Thompson, 2003; 8–10 in Edmonds et al., 2009; 10 in Rodriquez et al., 2006; 14–16 hours in Peach & Reuter, 2010), others, including the current authors, have reported on studies comprising 30 hours of treatment or more (24–35 in Marini et al., 2011; 30 in Kempler & Goral., 2011).