Review Reports - Main Concepts in the Spoken Discourse of Persons with Aphasia: Analysis on a Propositional and Linguistic Level

Round 1

Reviewer 1 Report

Dear authors,

All suggestions to improve your submission are stated in the attached document. Please take time to revise your paper considering these suggestions and comments, because qualitative discourse / corpora analyses as you do are very precious for aphasia research and speakers ability assessment, and your contribution will be very valuable with a partial re-analysis of data (for instance : calculate the verb ratio that is to variable from one to the other corpora in raw numbers, then do again your statistical analyses).

Comments for author File: Comments.pdf

Author Response

Dear reviewer,

we appreciate your valuable comments! All have been addressed and we feel our manuscript had indeed been improved, thanks to Your expertise. We hope you will find our manuscript a valuable contribution. Please note that all the changes are marked in red, and please find all the responses to the suggestions here.

Kind regards,

The authors

Author Response File: Author Response.pdf

Reviewer 2 Report

In this manuscript, the authors carry out an investigation of the production of discourse in PwA. In particular, they check their ability to identify and express main concepts, and the linguistic accuracy and completeness they show when describing events. The authors find PwA to express fewer concepts than healthy speakers, and to often produce linguistically inaccurate discourse, particularly omitting objects and verbs. The study competently contributes with data on Croatian on knowledge on the discourse abilities of PwA. The manuscript requires more referencing and clearer definitions of the concepts employed, but it is overall well written. Some suggestions of edits are given below.

lines 28-78 - Introduction - Please insert references on "difficulties in forming discourse in PwA", "the peculiarities of aphasia" (please elaborate), and integration of "different types of knowledge" when creating informative discourse.

line 29 - "discourse genres" - please elaborate on what is meant by "genre" in this context

line 56 - pragmatic, propositional, etc. meaning

line 63 - "reduced language knowledge" -reduced structural/syntactic abilities?

line 66 - "the functioning" - please rephrase

line 71 - these structural properties of Croatian are not mentioned further as the authors do not utilise them to interpret results (apart from inflection, but not related to the specificities of Croatian inflection). Therefore, I do not see the point of mentioning them here in passing as the authors have done.

line 80 - Bryant et al.

line 82 - substitute "including" with "namely"

line 131 (and elsewhere) - As they are concepts which are very relevant to the study, the authors should define the measures of "accuracy" and "completeness" when they first appear.

lines 191 - 195 - I feel like Rossi and Bastiaanse's reference is relevant for verb omission and verb inflection mistakes, while mention of the differences in verb types can be omitted

line 213 - I think it is worth mentioning that use of subordinate clauses does not necessarily mean use of more complex syntax, as it depends on what types of clauses these are. For example, PwA have been found to struggle with "complex" syntactic constructions such as object relative clauses (Martini, Belletti, Centorrino & Garraffa 2020 a.o.).

line 223 - change "was" to "is"

lines 223 to 226 - elsewhere, these aims are divided as two separate aims (for example in lines 744-5). I think it best to keep them divided in this subsection too.

line 226 - please substitute "approach"

lines 260-267 This paragraph repeats the contents of the Table and is therefore not needed.

lines 276-277 I am confused by this. How is this a reason for inclusion, when you previously state that an inclusion criterion was linked to performance to expressive language tasks? Please clarify

Lines 281-283 - this division is not explored in the data analysis, so it is incorrect to state that PwA were divided in two groups based on this criterion. I would prefer this to be another level of Table 1 with the descriptives of the groups.

line 318 - can the authors include the selected picture?

line 408 - remove space before first word

lines 419 - 20 - Tables should be referenced for the first time in the paragraph where they are described. Please remove reference from here and add in relevant sentences.

line 423 - either remove generic p value or state "alpha value set at 0.05"

line 424 - "including" > "specifically"

line 428 - (Table 3) at end of sentence

line 431 - please explain what "range of results" refers to. SD?

line 433 - repetition of "the difference"

line 434 - remove *

line 437 - (Table 4) at end of sentence

line 447 and elsewhere - I am usure what is meant by "marking a concept linguistically". What linguistic structures are used? Could you please clarify somewhere in the manuscript.

line 468 - produces > produce

line 489 - in discourse in aphasia

line 511 - it makes sense that this criterion should be applied to the speech of healthy individuals, so I would avoid the conditional. The result is still interesting.

line 529 -520 - this inference feels a bit stretched. I would avoid it or rephrase it accordingly.

lines 530ss - why the italics?

line 565 - the omission of the object is not necessarily a problem of word retrieval, but it could also be a (syntactic) pronoun omission. I would mention this.

line 660 - please add references to literature on aphasia and tense marking (Friedmann and Grodzinsky 1997, etc.)

line 663 - precisely less frequently

line 701-708 - the clinical implications, namely generalisations on discourse abilities in PwA (which are relevant for speech therapy) and the use of this task for assessment could be stated more clearly/overtly if this is to be a separate heading.

line 756 - I am confused by this sentence, as it seemed to me that the main concepts are the ones that are recognised the most by PwA, and rather what they are not producing are the concepts which are less relevant to the story (descriptions, etc.). Please rephrase.

Author Response

Dear reviewer,

Hereby we are re-submitting our original scientific paper entitled Main Concepts in the Spoken Discourse of Persons with Aphasia: Analysis on a Propositional and Linguistic Level.

We would like to thank you for very valuable comments and suggestions that, in our humble opinion, substantially improved our work. We addressed all the issues raised, and we marked all changes in red. We replied to every comment and we provide the responses in the uploaded document.

We appreciate your suggestions and we hope you will find our work suitable for this special issue.

Kind regards,

The authors

Author Response File: Author Response.pdf

Reviewer 3 Report

This study aims to investigate the impact of post-stroke aphasia on specific quantitative measures of connected speech production, such as main concepts and secondly to identify possible relationships among the aforementioned variable and other oral language production indices. The study presents a detailed quantitative analysis of connected speech samples derived from a picture description task. Although interesting, since not many studies address this question in patients with post-stroke aphasia, there are several issues, which I address in detail below:

Abstract

· Line 9: “Individuals use different discourse genres as part of their daily functioning”.

This introductory sentence appears misleading, as the reader may expect that current study includes more than one speech genre in participants’ assessment.

Introduction

In general, the introduction is well written, yet quite extended; consequently, it is not easy for the reader to follow the rationale of presenting all these studies. I suggest the introduction to be made shorter and more concise, coherent and focused in order for the flow of arguments to be smoother and to highlight the issue at hand. More importantly, the authors do not discuss at all previous evidence with regard to the phase of stroke.

Aim of the study

· Lines 223-224: “The main goal of this study was to examine the two categories of discourse production – propositional and linguistic”. As previously mentioned for other parts of the manuscript, this sentence appears misleading, as it implies that the authors used more than one speech genre to assess their participants. Please rephrase appropriately.

· The authors do not mention that they assessed patients in acute and in chronic phase of stroke.

Methods

· Table 2 is not very informative. I believe a more detailed table presenting individual information for lesion location would be much more informative for the reader.

· The picture from CAT-HR used for participants’ assessment could be included in Appendix.

Results

· It is not clear why the authors chose t-tests instead of equivalent non-parametric for the analysis presented in table 5, since they have previously mentioned that all their measurements deviated from normality (lines: 419-421).

· More importantly, the authors have conducted several comparisons, using either parametric or non-parametric statistical tests, without correcting p values for multiple comparisons. A Bonferroni or at least an FDR correction should be applied especially when parametric statistical tests were implemented.

· Error bar charts should be included for all comparisons, at lease as Supplementary material.

· Scatter plots should be included for significant correlations.

My major concern refers to the fact that the authors do not account for whether the patients are in acute or chronic phase of stroke in their analyses, ignoring the issue of spontaneous recovery (see for example: Laska et al., 2001; Lazar et al., 2010; Lazar et al., 2008). The authors should include this kind of information in their analyses (perhaps by changing their analyses to one-way ANCOVA) and more importantly they should report whether patients in acute phase present different patterns in all these speech output indices compared to patients in chronic phase of stroke.

General comments:

I appreciate and I acknowledge the amount of work necessary for the accomplishment of the present study as well as the importance of presenting evidence from connected speech analysis. My major concern is related to the inclusion of patients in both acute and chronic phase of stroke in their analyses. Based on that, the authors should strengthen the rationale of their study to investigate both patients in acute and chronic. Investigation of specific associations between indices of connected speech in either acute or chronic phase of stroke could be much more informative.

References:

Laska, A. C., Hellblom, A., Murray, V., Kahan, T., & Von Arbin, M. (2001). Aphasia in acute stroke and relation to outcome. Journal of internal medicine, 249(5), 413-422.

Lazar, R. M., Minzer, B., Antoniello, D., Festa, J. R., Krakauer, J. W., & Marshall, R. S. (2010). Improvement in aphasia scores after stroke is well predicted by initial severity. Stroke,41(7), 1485-1488.

Lazar, R. M., Speizer, A. E., Festa, J. R., Krakauer, J. W., & Marshall, R. S. (2008). Variability in language recovery after first-time stroke. Journal of Neurology, Neurosurgery & Psychiatry,79(5), 530-534.

Author Response

Dear reviewer,

Hereby we are re-submitting our original scientific paper entitled Main Concepts in the Spoken Discourse of Persons with Aphasia: Analysis on a Propositional and Linguistic Level.

We would like to thank you for very valuable comments and suggestions that reflect your expertise in the field. In our humble opinion, they contributed to the improvement of our work that we invested a lot in. We addressed all the issues raised, and we marked all changes in red. We replied to every comment and we provide the responses in the uploaded document.

We appreciate your suggestions and we hope you will find our work suitable for this special issue.

Kind regards,

The authors

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Current form of the manuscript is significantly improved. However, major issues still exist. My major concern is related with the fact that the authors insist to include patients of both acute and chronic phase in the same analysis. This is not methodologically correct or informative.

Please see some more detailed issues below:

- Results are not clearly presented in the Abstract. Differences between performance in patients of acute and chronic phase should be highlighted.

- The authors recruited patients with mild and moderate aphasia. However, they did not report based on which criteria they characterized aphasia severity. Moreover, they claimed “We included mild and moderate aphasia for two reasons: 1) during the acute phase of a stroke patients are medically unstable, so it is difficult to test them, especially those with severe aphasia”, which could not be considered as a valid argument for patients’ selection. If all patients are considered to be “unstable”, why to exclude only those with severe aphasia? I would assume that speech output deficits of patients with severe aphasia would be a major issue.

- “post-acute phase” is not very informative. “Chronic phase” would be more appropriate since time post-onset is between 5 and 79 months.

- Comparisons between patients (as a whole group) and healthy speakers is not informative, as previously mentioned. Same for correlation analyses, which are only being reported by treating patients of acute and chronic phase as a whole group. Scatter plots should also be included for correlation analyses, at least in supplementary material.

- The authors should clearly present the difference between content units and main concepts and further discuss this in “Discussion” section.

- Differences between performance in patients of acute and chronic phase should be further discussed in “Discussion” section.

As previously mentioned, I appreciate, and I acknowledge the amount of work necessary for the accomplishment of the present study and the revised manuscript, as well as the importance of presenting evidence from connected speech analysis. However, I insist that the authors should only present results from distinct comparisons of acute phase and chronic phase patients. Moreover, the authors should clearly state that different patients exist in each group, and this is not a longitudinal study. Investigation of specific associations between indices of connected speech in either acute or chronic phase of stroke could be much more informative.

Author Response

Dear Reviewer,

thank you for your valuable comments. We provide responses below (marked in red):

Current form of the manuscript is significantly improved. However, major issues still exist. My major concern is related to the fact that the authors insist to include patients of both acute and chronic phase in the same analysis. This is not methodologically correct or informative.

We appreciate your sincere and professional inputs. We made all the changes and we provide the responses point per point.

Please see some more detailed issues below:

Results are not clearly presented in the Abstract. Differences between performance in patients of acute and chronic phase should be highlighted.

The results are better elaborated in the Abstract.

The authors recruited patients with mild and moderate aphasia. However, they did not report based on which criteria they characterized aphasia severity. Moreover, they claimed “We included mild and moderate aphasia for two reasons: 1) during the acute phase of a stroke patients are medically unstable, so it is difficult to test them, especially those with severe aphasia”, which could not be considered as a valid argument for patients’ selection. If all patients are considered to be “unstable”, why to exclude only those with severe aphasia? I would assume that speech output deficits of patients with severe aphasia would be a major issue.

We added information on how we defined the level of severity.

“post-acute phase” is not very informative. “Chronic phase” would be more appropriate since time post-onset is between 5 and 79 months.

We systematically changed post-acute to chronic throughout the paper.

Comparisons between patients (as a whole group) and healthy speakers is not informative, as previously mentioned. Same for correlation analyses, which are only being reported by treating patients of acute and chronic phase as a whole group. Scatter plots should also be included for correlation analyses, at least in supplementary material.

We performed additional analyses for acute and chronic phases and presented them separately. Scatter plots for significant correlations are added in Supplementary material.

The authors should clearly present the difference between content units and main concepts and further discuss this in “Discussion” section.

The link between these two measures was explained in the first lines of the Discussion section.

Differences between performance in patients of acute and chronic phase should be further discussed in “Discussion” section.

Section Discussion has been revised according to new data obtained in the analyses between PwA in the acute and chronic phase.

The suggestion has been taken into account, and the manuscript and the analyses were changed accordingly. We provide the data on PwA and HS, as well as on PwA in the acute and PwA in the chronic phase of recovery.

We hope you will find our paper clearer and the analyses more suitable. We as authors feel that your suggestions increased the theoretical and methodological value of our work.

Kind regards,

Authors