Review Reports

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Comments and Suggestions for Authors

Thank you for taking the time to address the changes suggested. One can see you've devoted time and effort to the paper and it has greatly improved. Please find below some comments regarding the new version, based on your report about the changes:

- The sentence or prompt (not "phrase", as written by the authors) "You are a nice English teacher who helps the client to practice 151 the conversations about ordering food in a restaurant" (lines 151-152) should be delimited with inverted commas, since it is a direct quote.

- Comment 12: the comment in the authors' response says "Sessions lasted 45 minutes. The experimental group used the chatbot for 15 minutes of guided conversation". In the text of the paper, it says "The experimental group 193 students were put in front of the GPT-driven chatbot in a laboratory setting in a 30-minute 194 supervised session, three times a week. During this period, an average of 14.8 minutes 195 during each session was spent communicating with the chatbot". Therefore, it's not clear whether the sessions lasted 30 or 45 minutes.

- Comment 13 has not been completely addressed. The title above the graph is still in the image of the figure. The figure has already a caption. There is no need to have a double title above, which actually includes a misspelling (testS).

- Comment 14 has not been totally addressed. The mentioned Figure 2 has not been found.

Author Response

Response to Reviewer 1

1. Direct quote (lines 151–152):
  Done. The sentence "You are a nice English teacher who helps the client to practice the conversations about ordering food in a restaurant" has been placed in inverted commas as a direct quote.

2. Clarification of session duration (Comment 12):
  Done. The text has been revised to ensure consistency: all references to session duration now clearly indicate supervised sessions, with approximately 14.8 minutes of chatbot interaction. Any conflicting mentions of “186 minutes” have been corrected.

3. Graph title above the figure (Comment 13):
  Done. The redundant title within the figure has been removed, leaving only the properly formatted caption. The misspelling ("testS") has also been corrected.

4. Figure 2 missing (Comment 14):
 Done. Figure 2 has now been included and correctly referenced in the text.

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

The version I received is probably a draft version, as there is evidence of track changes in the document, with suggested revisions and omissions. This needs to be addressed.

In any case, the article presents an interesting and well-carried out study with clear implications for EFL pedagogy and AI-assisted learning. To improve the paper, I would advise the authors to revise for language clarity, fixing several instances of awkward or imprecise phrasing, especially in the abstract and introduction. The referencing format could also do with some attention, placeholder names should be replaced, and inconsistencies in citation style corrected. In addition, I would encourage the authors to engage more with their key sources by comparing findings or highlighting theoretical contrasts, rather than simply using them to support their claims. Including an explicit hypothesis would further clarify the study's direction. These revisions will improve the article’s readability, give it academic weight, and strengthen its overall impact.

Comments on the Quality of English Language

While the manuscript is generally readable, the quality of English requires improvement to ensure clarity and precision throughout. Several sentences, especially in the abstract, introduction, and discussion, contain awkward phrasing, redundancy, or grammatical errors that may obscure the intended meaning. There are also instances of inconsistent article use, imprecise vocabulary, and overly long or convoluted sentence structures. The manuscript would benefit from careful language editing by a proficient academic English speaker to enhance fluency, cohesion, and professional tone. Improving the expression will ensure that the strength of the study’s content is more effectively communicated to an international academic audience.

Author Response

Response to Reviewer 2

1. Track changes / draft version issue:
 Done. A clean version of the manuscript has been prepared, with all track changes fully accepted. Only highlighted revisions remain for clarity, as per editorial guidelines.

2. Language clarity and English editing:
  Done. The manuscript has been thoroughly edited by a proficient academic English editor. Awkward phrasing, redundancy, inconsistent article usage, and long sentence structures have been corrected, particularly in the abstract, introduction, and discussion.

3. Referencing format and consistency:
 Done. All references have been revised for accuracy and consistency. Placeholder names have been replaced, and citation formatting has been standardized according to the journal’s style.

4. Engagement with key sources:
  Done. The revised manuscript engages more critically with cited sources, comparing findings and highlighting theoretical contrasts instead of simply listing supportive references.

5. Explicit hypothesis:
  Done. An explicit hypothesis has been clearly stated in the methodology section to better guide the study’s direction.

Round 2

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

Generally speaking the article is fine, though there are some minor inconsistencies in formatting, as well as the sections that are in different colours, which should be corrected before publication. Also the visible track changes need to be taken care of.

Author Response

Dear Reviewer,

We appreciate your valuable comments. All formatting inconsistencies have been addressed, the text colour has been standardized, and the visible track changes have been carefully reviewed and accepted. The manuscript has now been revised accordingly.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper explores the impact of a GPT-powered chatbot on improving English conversation skills among Saudi secondary students in a low-resource region. The topic is timely, aligning with the growing interest in AI in education and the national goals of Saudi Arabia’s Vision 2030.

Using a quasi-experimental design, the study compares chatbot-based learning with traditional instruction, supported by test scores, student surveys, and teacher observations. The results suggest notable improvements in speaking skills and learner engagement. Overall, the study addresses a relevant educational gap and offers useful insights into the potential of AI in EFL classrooms.

Specific comments about formal issues are presented below:

- The phrase “the attitude towards the students” in the abstract should be corrected.

- The parenthesis in line 31should be removed.

- “allowed” is used in lines 100 and 102. One of them should be paraphrased to avoid repetition.

- “Comprised of” is incorrect.

- The sentence “Research…settings” (l. 78-81) seems not to fit there. The gap is mentioned above, and the study is also justified above.

- The objective should be clearly stated only once. It should also appear before the research questions. Therefore, there correct location for the objective is line 73. However, the real objective appears in line 89. Authors should merge the two mentions and state the objective in the correct place.

- The heading 2.1. Research Design could be removed. There shouldn’t be two headings together (2 and 2.1.) and the information presented in 2.1. is a valid introduction for section 2.

- Sentence in lines 184-186 should be reformulated.

- Groups should be consistently called “experimental group” and “control group”. Please correct this to avoid other terms like “treatment group” and the like.

Specific comments about content issues are presented below:

- The pre and post tests should be described in more detailed. What did students have to do? A presentation? A role-play? A spontaneous conversation whose topic was given by the teacher? Also, the test rubric should be provided.

- A chatbot is said to be “built”. This should be explained in more detailed. To understand the impact of the chatbot, the characteristics of the chatbot must be clearly presented. Authors must remember that enough information must be given in a methodology section so as to allow other researchers to replicate the study.

- (Classroom) Sessions should be described in more detail. How long is an ordinary session? How much of a session is devoted to using the chatbot? What is that time devoted to in a “traditional” session? Actually, what does a traditional session look like? Because if students use the chatbot to talk while the control group does not speak at all, it’s not the chatbot per se, but merely devoting time to practising the speaking skill, no matter how.

- Figure 1 should be renamed as “Figure 1: Pre- and post-test scores (experimental group)”. Also, the title above the graph (“Pre- and Post-testS”) should be removed, since we already have the caption below.

- After Figure 1 and Table 1, which show the progress of the experimental group, a similar table and figure would be expected to see the progress of the control group. Was the improvement similar when following the traditional teaching method? What was the purpose of the control group if this is not it?

- Table 2 presents the results for the questionnaire administered to students. The questions in this questionnaire are never shown, since the table does not include the questions, but the aspect measured through the questions. This should be clarified, and the reader should have a clear idea of what the questionnaire looks like, ideally by seeing the questions.

- When describing the results in 3.2., since the information is obtained through students’ opinion, the authors cannot state that “the findings indicate that the chatbot promoted skills development…”, but “the findings indicate that the students perceive…”. What students think and reality may or may not be the same. Information must be accurately expressed.

- In 3.3., why were only 10 students rated? How were they chosen? How can representativeness be ensured? The fact that, unexpectedly, not all students were assessed is concerning in terms of reliability. Additionally, there is no pre-and post-test comparison regarding these results. How can authors state that “teachers universally reported high improvement rates in all areas measured” (also, reconsider the use of the term “universally” in this sentence)? What is the difference in, for example, engagement before and after using the chatbot? Since the research is over, this can’t be fixed now. Therefore, authors should reconsider including this section as part of the study. In future research, they should reconsider the procedure.

 

In general terms, there is an important lack of information about certain factors in the methodology section, and the methodology followed in the study, based on the information presented in the text, raises important reliability concerns regarding the validity of the study and its results.

Comments on the Quality of English Language

Minor corrections should be made regarding English and language (some of them have already been indicated, but the whole text should be revised).

Reviewer 2 Report

Comments and Suggestions for Authors

p.1.10 I am a bit surprised by the fast-food interactions. It is common that individuals with ASD have food allergies (or are at least are sensitive to food) and are on some sort of diet (or have at least restrictions). The choice for fast food doesn't seem logical to me in that case.

If you are following APA 7 guidelines, make sure you place the p in italics.

p.1.34 I would urge to nuance this more. Thus: "anxiety, and/or increased". 

p.1.35 The introduction of VR is great. You explain clearly what it offers; however, I am concerned about how much it costs. It is not easy accessible due to how expensive it is.

p.2.43 How do the cultural norms affect the use of VR? In addition, you are mentioning the limited digital readiness. I am wondering if it would be better to invest in alternatives, to allow for implementation in public schools. 

p.2.85-87 This also implies that you are asking a lot from educators. On top of needing to implement a highly individualized learning trajectory without VR, educators need to apply this with VR as well. As a result, you are asking quite a lot of them (this is something to keep in mind; I expect that something similar or related will be mentioned in the discussion). 

p.3.99 You refer to accessible tools/systems. For now, VR is still too expensive and requires specific knowledge to use. I feel like you are "down talking" your research here.

Table 1 is presented in a different font. Please revise. 

I do not understand the first paragraphs in the section "Materials and Methods". It seems copied and pasted. Please check and revise accordingly. 

p.8.258 The summary of performance: for whom? Is this something the teachers use, or should the students use that? Moreover, competitive feedback isn't always the best choice. If you are offering an individualized approach--due to the heterogeneity of the target group--this might not be the best type of feedback to go for. In addition, I--as a reader--need more information about how this feedback is displayed and/or communicated to students (or teachers). It is a crucial aspects of your VR research but you do not present any information. 

In your other section (p.9.337) you refer to performance feedback (as compared to competitive feedback). Keep it consistent or clarify the difference. 

In figure 5 the font isn't the same for every box. Please revise accordingly. 

For figure 7: make sure the distinction between the two colours is sufficient (I think the contrast isn't suffice). Again, make sure the text in this figure is presented in the same font.