Time reference, morphology and prototypicality: tense production in stroke aphasia and semantic dementia in Greek

ABSTRACT

The present study aims at investigating verb inflection in aphasia and semantic dementia. In particular, it addresses the contribution of time reference and morphological complexity. Moreover, it investigates whether the lexical properties of the verb, such as argument structure and lexical aspect interact with the production of tense. Ten individuals with (different types of) stroke aphasia and five individuals with semantic dementia and their respective control groups conducted a sentence completion task. Three tenses were tested: past perfective, past imperfective and present. All tenses had to be produced with three different verb classes, which differed with respect to syntactic and semantic properties: unergative, unaccusative and transitive verbs. The findings imply problems with marking aspect and an interaction between inflection and lexical aspect but no effect of morphological complexity or across the board difficulties with reference to the past in aphasia. Moreover, the results suggest problems with inflection in semantic dementia, an interaction between inflection and lexical aspect and a selective difficulty with imperfective tenses. The study contributes to a better understanding of inflection problems in aphasia and it provides evidence for inflection problems in semantic dementia.

KEYWORDS:

• Aphasia
• semantic dementia
• inflection
• time reference
• morphology
• prototypicality

Acknowledgments

We are deeply grateful to the individuals with aphasia and individuals with semantic dementia as well as their families. We also thank warmly the speech therapists Thalia Zolota, Georgia Kolintza, Flora Koutsi and Erifyli Pouliou for helping us with the recruitment of the individuals with aphasia.

Data availability statement

The data that support the findings of this study are available from the corresponding author, V.K., upon reasonable request.

Disclosure of interest

The authors declare no conflict of interest.

Notes

¹ Note that Smith (1997) uses the terms situation and viewpoint aspect for lexical and grammatical aspect, respectively. We use the terms lexical and grammatical for the sake of simplicity. For this reason, we also leave aside other situation types, such as semelfactives (e.g. tap, knock) and other classification criteria, such as dynamism and durativity.

² The performance in future has not been found consistent across languages and we will not consider it in this study.

³ For the frequency comparisons, we used the log percentages and the frequencies were obtained from the Hellenic National Corpus (http://hnc.ilsp.gr) on the 12^th June 2007.

⁴ For convenience, the levels of the factor tense will be written as following: Present for the production of present tense, PastImperf for the production of past imperfective, PastPerf/Pres for the production of past perfective from present and PastPerf/Imperf for the production of past perfective from past imperfective.

⁵ We defined Helmert contrasts for the factor tense, so that the two conditions of past perfective production are compared to each other, past imperfective is compared to the mean of the two conditions of past perfective production and present is compared to the mean of all past tenses.

⁶ We defined Helmert contrasts for the factor Tense (see above) and treatment contrasts for the factor Verb Class, so that unaccusative and transitive verbs are compared to unergative verbs. Unergative verbs were the baseline because they are the simplest class in terms of argument structure (they take only one argument and the subject is an Agent), according to previous studies.

⁷ In general, the error pattern for transitive verbs is not clear, as in some conditions they pattern together with unergatives and in other conditions with unaccusatives. We take it to be a consequence of their aspectual “instability”, as their aspectual properties depend both on the lexical aspect of the verb as well as on the referential properties of their arguments.

⁸ Our data resemble but are not identical to those of Bastiaanse and Platonov (2015), who found the interaction between verb class and tense within past time frame. We did not find any dissociation among verb classes in past imperfective. This could be due to methodological differences, as Bastiaanse and Platonov (2015) did not use any cue phrases, rather they elicited the target form cued by the verb of a preceding sentence. Moreover, the cueing phrase for the elicitation of past imperfective in our material was not homogeneous: depending on the aspectual properties of the target verb we used phrases which denoted either iteration or progress. This is a methodological fault of our study which might have confounded the interaction of lexical and grammatical aspects within the past time frame. However, even if we used only cueing phrases which denote continuation like Fyndanis and colleagues did, this does not assure a homogeneous interpretation of the adverbial as a continuation cue. Adverbials like “for an hour” with telic predicates might be interpreted as iterations of a completed action.

Additional information

Funding

This work was supported by the State Scholarships Foundation (IKY) with a grant to the first author for doctoral research in Language Therapy;Greek State Scholarship Foundation.