Аспекты времени: обработка категории времени и вида типичными и атипичными носителями языка тема диссертации и автореферата по ВАК РФ 10.02.19, доктор наук Драгой Ольга Викторовна

• Драгой Ольга Викторовна
• доктор наук
• 2020

Оглавление диссертации доктор наук Драгой Ольга Викторовна

Table of contents

1. Introduction

2. Electrophysiology of time referencing in healthy individuals

3. Time reference and aspect impairments in individuals with aphasia

4. Tense and aspect attrition in bilingual individuals

5. Conclusions

References

Appendix A. Paper "From time to time: processing time reference violations in Dutch"

Appendix B. Paper "Time reference teased apart from tense: Thinking beyond the present"

Appendix C. Paper "Gamma Oscillations as a Neural Signature of Shifting Times in Narrative

Language"

Appendix D. Paper "Verb production and word order in Russian agrammatic speakers"

Appendix E. Paper "Aspects of time: time reference and aspect production in Russian aphasic

speakers"

Appendix F. Paper "Understanding discourse-linked elements in aphasia: a threefold study in

Russian"

Appendix G. Paper "Aspect and tense attrition in Russian-German bilingual speakers"

Введение диссертации (часть автореферата) на тему «Аспекты времени: обработка категории времени и вида типичными и атипичными носителями языка»

1. Introduction

The papers presented in this dissertation are dedicated to the experimental study of time reference across typical (non-brain-damaged monolingual) and atypical (brain-damaged or bilingual language attritor) adult language speakers. Initial cross-linguistic reports have highlighted that people with brain damage and the resulting aphasia (an acquired language impairment) experience more difficulties with past tense verb forms than with present or future tense forms (Bastiaanse, 2008; Simonsen & Lind, 2002; Yarbay Duman & Bastiaanse, 2009). Further research has revealed that those difficulties are related not to past tense, but to reference to past semantics, irrespective of the linguistic device used to convey such reference. Indeed, tense is among the most common means to express time reference. However, non-finite participles referring to the past were shown to be as impaired in aphasia, as were finite past verb forms (Bastiaanse, 2008). Similarly, perfective verb forms in general were more problematic than imperfective verb forms in Greek (Nanousi, Masterson, Druks, & Atkinson, 2006; Stavrakaki & Kouvava, 2003). These data gave rise to the PAst DIscourse LInking Hypothesis (PADILIH; Bastiaanse et al., 2011), which suggests that this specific deficit is caused by the discourse-related nature of past time reference: it requires a relation established between speech time and an earlier event - that is, discourse linking. No such linking is required for a verb form referring to the present, as the time of speech and the time of the event coincide. The PADILIH also conformed to earlier reports on a range of similar linguistic phenomena: reflexive pronouns and who-questions are better preserved in individuals with aphasia and are acquired earlier by children than anaphoric pronouns and which-questions, respectively (Avrutin, 2006; Hickok & Avrutin, 1995), because only the latter two require processing at the level of a broader context and need discourse linking.

The goal of the studies included in this dissertation was to test the universality of the PADILIH, which is still an open relevant research question. The first specific objective was to test whether, similarly to individuals with aphasia, the brains of healthy people are sensitive to the reference to different time frames and to the shifts between them. That question was addressed using electrophysiological measures of healthy Dutch speakers' reactions to verb forms referring to the past and non-past, in three experiments. The second objective was to test whether the PADILIH's specific prediction about past time reference and Avrutin's (2006) broader prediction on other similar phenomena are cross-linguistically valid. To answer this question, we performed three behavioral experiments: we tested brain-damaged individuals with aphasia who were native speakers of Russian, a language which represents a complex interaction between the verb categories of tense and aspect. Finally, the third objective was to test whether, once acquired, the means of time reference in a

language are subject to attrition under the influence of another language with a different repertoire of time reference means. That issue was studied by employing the similarities and differences in the tense-aspect systems of Russian and German, and by testing the linguistic behavior of bilingual speakers of these two languages. Taken together, the presented studies used behavioral and electrophysiological methods, characteristic properties of three languages (Dutch, Russian and German), a thorough experimental design regarding both the selected materials and the employed experimental procedures, and up-to-date statistical approaches to empirical data analysis, which constitute the novelty of the work.

In four of the papers included in this dissertation, Olga Dragoy is the first author. Her contribution to the three other co-authored papers is as follows. In the paper by Bos, Dragoy, Stowe, and Bastiaanse (2013), both Olga Dragoy and Laura Bos were involved in conducting the experiment, analyzing the data and preparing the draft of the paper. In the paper by Brederoo, Bos, Dragoy, Bastiaanse, and Baggio (2015), Olga Dragoy co-supervised the project at the stages of the study conception and design, and also considerably contributed to the data analysis. For the paper by Bos, Dragoy, Avrutin, Iskra, and Bastiaanse (2014), Olga Dragoy co-supervised the design of the project, developed the Russian materials, supervised the data collection and contributed to the draft of the paper.

The following theses are proposed for the defense:

(1) Reference to the past and to the non-past, expressed through verb forms, is processed differently in the healthy brain, irrespectively of tense: non-past is related to local morphosyntactic processing signatures, while past features discourse integration mechanisms.

(2) Shifts between linguistically expressed past and present time reference are taxing for the healthy brain and are processed in the gamma band frequency of brain oscillations.

(3) Russian speakers with post-stroke aphasia are sensitive to the grammatical information expressed in verbs and, specifically, to the prototypical and non-prototypical matches of time reference and aspectual semantics.

(4) In Russian-German bilingual speakers, Russian tense and - to a larger degree - aspect are vulnerable to attrition, with an advantage for the prototypical match of temporal and aspectual semantics.

(5) Individuals with aphasia experience consistent problems with a range of linguistic expressions requiring discourse linking.

Altogether, the obtained evidence confirms the cross-populational and cross-linguistic relevance

of the PADILIH, but highlights a strong modulation of time reference processing by language-

specific categories.

Заключение диссертации по теме «Теория языка», Драгой Ольга Викторовна

Issues for further research

Needless to say, there are many issues left unresolved or not addressed at all which are raised by the current study. The issue of sentence-final negativity remains underspecified within the Unification Model. However, Baggio (2008) argues that within his experiment this effect reflects the readjustment of verb constraints, so that the whole set of constraints becomes satisfiable. in our experiment, such a readjustment may force a historic present (for the past adverb-present verb combination) or narrative past reading (for the present adverb-past verb combination). This claim is in line with the view that sentence-final negativity might reflect the extra memory load needed to support the search for a successful referential link and/or repair of the semantic interpretation of the whole discourse model constructed on the basis of the sentence. This view is also compatible with a single perspective for time reference, pronominal reference (Osterhout & Mobley, 1995) and referential ambiguity processing (Nieuwland & van Berkum, 2008). However, it remains an issue whether this negativity reflects such a specific integration, or rather a more general effect elicited by any incoherence in the sentence as a whole regardless of whether it implicates the discourse model or not. Such a conclusion would be interesting: although in the current study no sentence-final effect was found in control sentences with semantic and morphosyntactic anomalies, there are examples of sentence-final negativities elicited by both lexical semantic and non-referential morphosyn-tactic violations (e.g. Hagoort, 2003b). There is some variation in the scalp distributions and latencies of the sentence-final negativities discussed above, and thus more in depth research using within subject comparisons is indicated before the exact status of these negativities is clear.

We have suggested that the dissociation found in aphasia (e.g. Bastiaanse et al., 2009) can be explained by the difference between local binding and discourse-linking. A related issue for further research is whether the dissociation is due to a problem with discourse-linking or with the working memory resources necessary to carry out discourse-linking. The proposed involvement of extra memory load in establishing a discourse link while processing past time reference may well be the reason for a dissociation between present and past found in individuals with

aphasia. It has been shown that linguistic and working memory deficits positively correlate in aphasia (Caspari, Parkinson, LaPointe, & Katz, 1998; Sung et al., 2009; Wright, Downey, Gravier, Love, & Shapiro, 2007). Those working memory limitations could make discourse-linked relations more vulnerable than locally bound relations. Also, effects similar to those apparent in the spontaneous speech or non-speeded behavioral experiments of individuals with aphasia become visible in healthy individuals under time pressure. In line with the working memory load interpretation of the sentence-final negativity mentioned above, in our RT experiment more errors were observed for sentences with verbs referring to the past: the speeded task imposed extra processing constraints, and healthy participants showed more difficulties in sentences for which they could otherwise successfully have determined whether a successful discourse-linking was possible. The role of working memory provides an attractive explanation of aphasic and normal error patterns, but its role in establishing discourse-linking requires additional clarification.

For the past discourse-linking theory discussed here, a necessary line for further research is the systematic comparison of ERP effects related to processing of locally bound and discourse-linked elements. This systematic examination has several facets. For one, we have appealed to a comparison between pronominal and tense processing to support the existence of a dissociation similar to that found in pronominal processing for tense processing. If we are really seeing the same effects in both modalities of reference, they should show comparable ERP responses. However, a real comparison of ERP effects is best done with the same group of subjects, since the scalp distribution of various effects can differ quite noticeably between different groups of participants.

Furthermore, we have barely scratched the surface of tense processing. The status of future tense is an interesting issue. Faroqi-Shah and Dickey's (2009) results suggest that future tense patterns with present tense: that is, it is locally bound and relatively easy. Data from aphasia collected in a language with a more extensive morphologically marked tense system, on the other hand (Yarbay Duman & Bastiaanse, 2009), suggest that like the past tense, future is more difficult than present. To add to the possibilities, according to Zagona's (2003) theoretical account, future is neither bound nor discourse-linked. Temporal relations conveyed by future time reference therefore urgently need further investigation.

Finally, the initial impulse for this line of study into time reference came from morphological processing studies. However, time can also be indicated by lexical adverbs and event sequences, which present another interesting direction for future research.

Список литературы диссертационного исследования доктор наук Драгой Ольга Викторовна, 2020 год

References

Allen, M., Badecker, W., & Osterhout, L. (2003). Morphological analysis in sentence processing: An ERP study. Language and Cognitive Processes, 18, 405-430.

Aronson, H. (1977). The interrelationships between Aspect and mood in Bulgarian. Folia Slavica, 11, 9-32.

Avrutin, S. (2000). Comprehension of discourse-linked and non-discourse-linked questions by children and Broca's aphasics. In Y. Grodzinsky, L. Shapiro, & D. Swinney (Eds.), Language and the brain: Representation and processing (pp. 295-313). San Diego: Academic Press.

Baayen, R H., Piepenbrock, R., & van Rijn, H. (1993). The CELEX lexical database. (CD-ROM). Philadelphia, PA: Linguistic Data Consortium.

Baggio, G. (2008). Processing temporal constraints: An ERP study. Language Learning, 58, 35-55.

Bastiaanse, R. (2008). Production of verbs in base position by Dutch agrammatic speakers: Inflection versus finiteness. Journal of Neurolinguistics, 21,104-119.