Pause Length and Differences in Cognitive State Attribution in Native and Non-Native Speakers
Round 1
Reviewer 1 Report
General comments:
1. Could the authors speculate on some other factors related to non-native speech potentially influencing the variables? E.g. how do disfluencies, speech rate, and prosody of Polish trivia sentences recorded by Chinese native speakers impact the ratings? This would leave a reader with impression that pauses are one of many factors that influence the ratings.
2. In the Introduction, authors provide the examples from Conversation Analysis. This section would benefit from including some other frameworks, perhaps more suitable for a quantitative study. The CA is not applied in the analyses which makes giving it a lot of space in the Introduction difficult to justify.
3. Can we assume that content of the trivia questions can have an impact on the ratings of speakers' knowledge? Since the sentences were not delexicalized, could the meaning of applied stimuli (for example - no. 32 about vegetables in space) impact the ratings? Some further explanation accounting for the semantics of the recorded sentences would be appreciated.
4. Can willingness to grant the request be inferred from the content? The effort required to handle the requests differs across the examples (compare the requests no. 14 and 25). Perhaps authors should elaborate on the meaning of the requests or even try to analyze whether willingness markings reached ceiling for the easy requests compared to the demanding ones.
5. What is a possible influence of non-native, Chinese-accented speech on the ratings? The non-native speakers show noticeable differences in fluency. One female speaker is clearly more fluent than the others. Also, the answers given by the non-native speakers were close to read speech - in contrast to the recordings of Polish native speakers. Could this effect be treated as a confound in the analyses? This could have been a distractor influencing the ratings.
6. In the Polish instructions (cf. json file), the relation between the speakers was defined. Could this possibly prime the subjects and impact their ratings of willingness?
7. In native-speakers' condition, the element which ratings could be based upon was 'myślę' [I think]. The pronunciation of this predicate by a female native speaker varied across the samples on both segmental and suprasegmental levels. This onset could have been perceived as a hesitation marker and impact the ratings to a higher extend than various pause lengths. This could have been a cue for knowledge ratings too. Maybe even more obvious than the pause length.
Specific comments:
p. 2, line 48: please explain 'second pair parts SPP'. Similar to 'first pair part' on page 3, line 92.
p. 2, line 52: the symbols used in the examples are not quite clear.
p. 3, lines 95-96: page 5617.
p. 3, lines 107-109: consider rephrasing.
p. 5, line 171: please explain 'Floor Transfer Offsets'.
p. 5, lines 182-183: 'b) how confident (for questions) ... the request (for requests)' - this explanation would read well earlier in the text.
p. 8, lines 271-274: except examples no. 5, 13, 14, 17, 24, 25, 29... where the answers do not repeat the questions. See also request no. 5 and 17 where the response is logical but different from the question.
p. 8, lines 288-290: is to possible to provide more detailed profiles of linguistic competence of the recorded speakers (e.g., in CEFR scale)?
p. 8, line 293: sampling rate, bit depth, and file format would be more informative than a manufacturer's name.
p. 10, line 341: I am not sure if 'preregistered' fits here, since the analyses were already conducted.
p. 16, lines 444-449: this conclusion was based on a dataset that is missing in the repository.
p. 17, line 423: 'cognitive states' were briefly mentioned on page 2, lines 41-42 as 'production difficulty, honesty, and being comfortable with the conversation topic to pauses.' The ratings though concerned knowledge, confidence, and willingness? Are they also cognitive states? If so, this would require either rephrasing or providing a definition earlier in the text. A cue appears as late as p. 19, lines 560-561.
p. 17, line 492: consider rephrasing.
p. 18, lines 502-503: can this statement be supported with speech rate data from this study?
p. 18, lines 505-506: is it a conclusion coming from this study? If so, where do we find the fluency ratings?
p. 18, lines 527-537: this paragraph does not discuss the study findings and would fit better in the Introduction. Maybe somewhere around par. 6, where pause lengths are explained.
p. 21, line 602: revise the reference.
p. 21, lines 613, 616, 618, 621, 630, 634, 637, 642, 646, 652: revise the references.
p. 22, lines 658, 660, 668, 675, 677, 681: revise the references.
p. 23, lines 697, 699, 707, 716, 721: revise the references.
p. 26, Question 10: is: 'zeby chronic sie' - should be: 'żeby chronić się'.
p. 27, Question 23: transcript: 'Peterborougs' - recording: 'Petersburgu'?
p. 32, Request 27: '?' missing.
p. 32, Request 27: is 'ja' should be 'jÄ…'.
Author Response
Thanks for the insightful review. Please find our response in the attached pdf file.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Please see the attached pdf.
Comments for author File: Comments.pdf
Author Response
Thanks for the insightful review. Please find our response in the attached pdf file.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Thank you. The previous comments and suggestions were sufficiently addressed. Just a couple of typos in the revised text:
p. 4, line 113: space missing
p. 5, line 155: space missing
p. 6, line 175: typo
p. 7, line 208: typo
p. 7, line 214: space missing
p. 10, line 296: space missing
Reviewer 2 Report
The authors have answered my comments and suggestions thoroughly and revised the manuscript accordingly. In my opinion, the paper is ready for publishing (it just needs a final proof-reading, as there were some typos still).
