Abstract
Background: The majority of the world's population is bilingual. Yet, therapy studies involving bilingual people with aphasia are rare and have produced conflicting results. One recent study suggested that therapy can assist word retrieval in bilingual aphasia, with effects generalising to related words in the untreated language. However, this cross-linguistic generalisation only occurred into the person's stronger language (L1). While indicative, these findings were derived from just three participants, and only one received therapy in both languages.
Aims: This study addressed the following questions. Do bilingual people with aphasia respond to naming therapy techniques developed for the monolingual population? Do languages respond differently to therapy and, if so, are gains influenced by language dominance? Does cross-linguistic generalisation occur and does this depend on the therapy approach? Is cross-linguistic generalisation more likely following treatment in L2 or L1?
Methods & Procedures: The study involved five aphasic participants who were bilingual in English and Bengali. Testing showed that their severity and dominance patterns varied, so the study adopted a case series rather than a group design. Each person received two phases of naming therapy, one in Bengali and one in English. Each phase treated two groups of words with semantic and phonological tasks, respectively. The effects of therapy were measured with a picture-naming task involving both treated and untreated (control) items. This was administered in both languages on four occasions: two pre-therapy, one immediately post-therapy and one 4 weeks after therapy had ceased. Testing and therapy in Bengali was administered by bilingual co-workers.
Outcomes & Results: Four of the five participants made significant gains from at least one episode of therapy. Benefits arose in both languages and from both semantic and phonological tasks. There were three instances of cross-linguistic generalisation, which occurred when items had been treated in the person's dominant language using semantic tasks.
Conclusions & Implications: This study suggests that ‘typical’ naming treatments can be effective for some bilingual people with aphasia, with both L1 and L2 benefiting. It offers evidence of cross-linguistic generalisation, and suggests that this is most likely to arise from semantic therapy approaches. In contrast to some results in the academic literature, the direction of generalisation was from LI to L2. The theoretical implications of these findings are considered. Finally, the results support the use of bilingual co-workers in therapy delivery.
Acknowledgements
This project was funded by St Bartholomew's and the Royal London Charitable Foundation and Barts and the London Trust. The authors are grateful to Michael Coleman, Division of Psychology and Language Sciences, University College London (UCL), for allowing access to the Soundbytes programme. The authors thank Sharifun Ahmed, who acted as a language consultant to the project, and Salma Begum, Zahir Choudhury, and Tarik Khan who were the bilingual co-workers. The authors are grateful to the aphasic and control participants who gave their time so generously to the project. Katerina Hilari was an advisor to the project; and Susanne Marsello helped to develop the proposal.
Notes
1. At the time of the study, there was only one speech and language therapist in the UK who spoke Bengali. This therapist acted as language consultant to the project.
2. All participant names are pseudonyms.
3. When the study was conducted, there were no students of speech and language therapy in London who were bilingual in English and the Sylheti dialect of Bengali.
4. Made available by Michael Coleman at University College London.