Non-treated languages in aphasia therapy of polyglots benefit from improvement in the treated language

https://doi.org/10.1016/j.jneuroling.2008.07.003Get rights and content

Abstract

The present case study aims to investigate whether the treatment of a third language (L3 French) in a trilingual chronic Wernicke-aphasic patient leads to the parallel improvement of the first (L1 German) and second (L2 English) languages. After a linguistic assessment in each language by means of the Bilingual Aphasia Test (BAT) [Paradis, M. (1987). The assessment of bilingual aphasia. Hillsdale: Lawrence Erlbaum], the patient was intensively treated in his L3 twice a day for 45 min for three and a half weeks. The treatment focussed on lexical–semantic deficits. Subsequent to the treatment, the patient was tested again in the three languages. The resulting recovery pattern shows that both L2 and L3 significantly improved after treatment with a larger effect for the treated language (L3) than for the L2. In contrast, L1 (German) did not show significant improvement, probably because the test results were already above-average for the first testing (T1). The present findings lend support to cognitive models of bilingual word recognition postulating that a bilingual's two languages share a common semantic–conceptual memory system.

Introduction

In the last decade, several psycholinguistic studies of bilingual language processing have addressed the question of whether the language system of polyglots relies on a common semantic–conceptual memory system shared by the different languages or on separate memory systems. Combined behavioral (Francis, 1999, Francis, 2005) and neuroimaging studies (for a review see Illes et al., 1999) support the idea of a common semantic–conceptual system which might be accessible through the different mental lexicons of a bilingual individual. Different psycholinguistic models of bilingual word recognition follow different hypotheses with respect to the internal organisation of the components (i.e. lexical and conceptual) constituting the language processing system. One class of models consisting of “hierarchical models” postulates two separate lexical stores (one for each language) and one common conceptual store (French & Jacquet, 2004 for a review). These models differ regarding the type of links they hypothesize between the L1 (first language) and L2 (second language) lexical nodes and the conceptual node. One of the most important existing models of bilingual word recognition is the Revised Hierarchical (RH) model put forth by Kroll and Stewart (1994) (see Fig. 1).

The core idea of the RH model is that bilinguals initially map a newly learned L2 word onto its L1 translation, with conceptual access based on the L1 linkage between the lexical representation and its conceptual representation (see Fig. 1). With increasing proficiency, bilinguals can instead directly map a word in L2 to a conceptual representation without the need for in a study which resembles our work in some central aspects, L1 mediation. In this way, the RH model postulates a stronger connection between the L1 lexicon (L1) and the conceptual system (C) than between the L2 lexicon (L2) and the conceptual system. Furthermore, a strong link exists going from the L2 to the L1 lexicon, while the reverse connection is weaker.

In the present study, we aimed to examine whether the architecture of the language system in multilinguals has an impact on the recovery of non-treated languages in polyglot aphasics. Our rationale was that if a common system containing semantic–conceptual information for the different languages of a multilingual exists, a training stimulating this system using one language should also affect the other two languages that were not directly treated. In contrast, if the different languages are represented separately at the conceptual level, such transfer effects between the languages should not be observed. In order to approach this issue we adopted a neuropsychological viewpoint by treating a trilingual Wernicke-aphasic patient (L1 German, L2 English and L3 French) in his L3. All three languages were assessed by means of the standardized Bilingual Aphasia Test (BAT) (Paradis, 1987) before and after training.

Section snippets

Aphasia in polyglots

Case studies on aphasia in polyglots go back to the middle of the 19th century. Albert and Obler (1978) and Paradis (1983) collected case studies of multilingual aphasics and evaluated them according to different criteria. One central criterion was the recovery pattern the different languages of the polyglot subject underwent, i.e. whether each of the languages of the aphasic patient was affected in a similar manner following a specific cerebral lesion. Paradis (1983) found a parallel recovery

Language background

BL,2 a highly educated (Ph.D.) right-handed man, was 48 years old when he suffered a left-hemisphere ischemic cerebrovascular stroke. BL was raised in southern Germany with German as his L1 and both parents he being native speakers of German. At the age of 10, BL started to learn English in school and, at the age of 13, he began to learn French. After having completed his Ph.D. in Germany, BL began to work as a scientist at an

Predictions

  • i.

    If an intensive aphasia therapy in the chronic stage is effective, then an improvement in the language of treatment, French, should be observed.

  • ii.

    According to psycholinguistic models of bilingual word recognition postulating a common conceptual system in multilingual speakers, an improvement in the languages that were not treated, i.e. English (L2) and German (L1), should also be expected. According to the RH model (Kroll & Stewart, 1994), a larger improvement in the L1 of the patient should

Language performance at T1 (before the training)

Before the training, language performance of BL was tested by means of the BAT in French (L3), German (L1) and English (L2). Fig. 3 shows the total score for each language at T1 as well as the norm score.4

Paradis (1987, p. 24) assigns each subtest of the BAT to one or several linguistic levels so that we were also able to measure

Linguistic performance at T1

The patient's performance before training is represented in Fig. 3, Fig. 4. It is clearly visible that the patient performs far better in his L1 than in his L2 and L3. As mentioned before BL was treated in his L1 German for 2½ years post-onset. BL's performance advantage in his L1 could be ascribed to this long-term treatment which obviously did not fully transfer to L2 or L3. This observation is in line with the hypothesis that non-native languages are sustained by declarative memory to a

Conclusion

The aim of the present study was to explore whether the lexical–semantic treatment of only one language (L3) in a multilingual chronic Wernicke-aphasic patient leads to a parallel improvement of all languages of the subject. Our results show that apart from the language of therapy, French, one non-treated language, namely the L2 English, improved significantly after therapy. L1 German did not improve – a result that could be explained by the patient's already strong performance in this language

Acknowledgements

The data presented in this paper are part of the master thesis of the first author (B.M.). First and foremost, we would like to thank BL for taking part in the study. We are also grateful to Nils Skotara for his valuable statistical suggestions and to one anonymous reviewer for very helpful comments on the text.

References (34)

  • M.L. Albert et al.

    The bilingual brain. Neuropsychological and neurolinguistic aspects of bilingualism

    (1978)
  • S.K. Bhogal et al.

    Intensity of aphasia therapy: impact on recovery

    Stroke

    (2003)
  • M.W. Chee et al.

    Left insula activation: a marker for language attainment in bilinguals

    Proceedings of the National Academy of Sciences of the United States of America

    (2004)
  • L.A. Edmonds et al.

    Effect of semantic naming treatment on crosslinguistic generalization in bilingual aphasia. Journal of Speech

    Language and Hearing Research

    (2006)
  • F. Fabbro et al.

    Differential impairments in four multilingual patients with subcortical lesions

  • D. Filiputti et al.

    Nonparallel recovery of languages in a quadrilingual aphasic patient

    The International Journal of Bilingualism

    (2002)
  • W.S. Francis

    Cognitive integration of language and memory in bilinguals: semantic representation

    Psychological Bulletin

    (1999)
  • Cited by (49)

    • Bilingual aphasia: Explanations in population encoding

      2019, Journal of Neurolinguistics
      Citation Excerpt :

      However, one can speculate as to whether the intense focus on grammatic morphology could have achieved generalization via an extrinsic effect — providing this brilliant and highly motivated patient a technique that he could use outside therapy to improve grammatic morphologic function. The results of some multi-lingual aphasia treatment studies seem to be best interpreted as reflecting the successful engagement of a fundamentally non-linguistic mechanism, as in CILT, e.g. Miertsch, Meisel, and Isel (2009). That is, formal treatment and/or ambient treatment (McClung, Rothi, & Nadeau, 2010) led to development of a new intentional bias — a language set — that favored increased use of language in general or of non-preferred language(s).

    • A case study about the interplay between language control and cognitive abilities in bilingual differential aphasia: Behavioral and brain correlates

      2018, Journal of Neurolinguistics
      Citation Excerpt :

      There is a lot of debate whether linguistic treatment should be given in one or in both languages of a bilingual patient. Some researchers found that the effects of language therapy in one language generalize to the other, untrained, language (Edmonds & Kiran, 2006; Filiputti, Tavano, Vorano, de Luca, & Fabbro, 2002; Kiran & Edmonds, 2004; Marangolo, Rizzi, Peran, Piras, & Sabatini, 2009; Miertsch, Meisel, & Isel, 2009), while others were not able to observe such generalization (Abutalebi et al., 2009; Galvez & Hinckley, 2003; Meinzer, Obleser, Flaisch, Eulitz, & Rockstroh, 2007). In recent years, the importance of training control abilities for cross-linguistic recovery of aphasia becomes more and more recognized (e.g., Abutalebi, et al., 2009; Ansaldo & Saidi, 2014; Brownsett et al., 2014; Kohnert, 2004; Radman et al., 2016), although control abilities are still rarely trained in aphasic patients.

    View all citing articles on Scopus
    View full text