Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript presents a solid and well-founded theoretical framework, coherently integrating classic literature on multilingualism, multi-word expressions, and learning theories with recent studies on the use of AR in educational settings. However, it is recommended that this section be strengthened by more clearly distinguishing between the conceptual frameworks of educational AR and those of multilingual learning, which would allow for a more precise highlighting of the study's specific contribution. Furthermore, although the general and specific objectives are well-defined, it would be pertinent to incorporate explicit research questions derived from them, which would more directly guide the methodology and the discussion of results. It is also suggested that the identification of research gaps be explored further, emphasizing which aspects have not been addressed by previous literature and how this work expands upon or reinterprets them.

The Methodology and Design section of the manuscript presents a solid structure and a high level of detail, reflecting careful planning of the research process. However, it also reveals some aspects that could be improved to strengthen methodological coherence and clarity of exposition. The method is described as exploratory, inductive, and descriptive, framed within a qualitative paradigm with the integration of mixed methods. This characterization is consistent with the stated objectives and the type of analysis performed. However, a more detailed justification of the mixed methods employed is suggested, explaining how and why the quantitative and qualitative approaches are combined, specifying the sequence or relative weight of each component (for example, whether it is a convergent design, sequential exploratory design, etc.). Ethical treatment and obtaining informed consent are adequately mentioned and meet the minimum standards required for educational research. However, it would be advisable to integrate a brief reference to these aspects within the "Context, Materials and Methods" section, instead of relegating them solely to the final sections. This would allow for greater structural coherence and would underscore ethical responsibility from the outset of the methodological design, further elaborating on the pseudonymization of personal data.

The analysis procedures and the use of various tools are detailed, combining descriptive analysis in Excel and Google Forms with the innovative application of Large-Scale Language Models for qualitative coding and summary generation. While this technological integration brings novelty and efficiency to the processing of large volumes of data, the process description suffers from a certain lack of methodological precision regarding how the validity, reliability, and traceability of the results produced by the artificial intelligence tools were ensured. Human review is mentioned as a control mechanism, but the verification criteria, the percentage of agreement between automatic coding and manual review, and how potential biases or interpretation errors generated by the models were managed are not specified. Furthermore, the explanation of the analytical sequence between manual and automated procedures could be more systematic, showing more clearly how human decisions guided the AI ​​intervention and how consistency with the original qualitative approach was preserved.

The results are presented clearly and coherently, supported by tables, figures, and percentages that facilitate understanding and maintain a good correspondence with the study's objectives. However, this section could benefit from greater synthesis and interpretation, as the abundance of descriptive data sometimes overshadows critical analysis.
The manuscript's discussion and conclusions are generally coherent, balanced, and convincing, demonstrating a good articulation between the empirical results and the reviewed literature. The arguments clearly highlight the effectiveness of AR as an inclusive and motivating tool and link the findings to relevant previous studies, thus providing theoretical rigor to the interpretations. Likewise, the conclusions adequately synthesize the study's achievements and reflect the coherence between the objectives, the methodological design, and the results obtained. Nevertheless, from a critical perspective, the discussion tends in some sections to be more descriptive than analytical, limiting itself to confirming agreements with the literature rather than problematizing discrepancies or theoretical tensions. Further exploration of the relationship between empirical results and the conceptual frameworks of multimodal learning or digital inclusion could strengthen the interpretive argument. Regarding the conclusions, while well supported by the data, it would be advisable to more clearly define the inferences within the exploratory scope of the study, avoiding excessive generalizations.

Comments on the Quality of English Language

The English in the manuscript is functional and allows for a clear understanding of the scientific content, but it shows several signs of having been translated from Spanish and not originally written by a native speaker. These signs appear at different linguistic levels and partially affect the fluency and naturalness of the text. Syntactic borrowings from academic Spanish are observed, with lengthy sentences and structures uncharacteristic of scientific English, such as the repeated use of expressions like “taking into account that…” or “in this sense…”, which are excessively literal. Furthermore, errors or unidiomatic uses of verbs and prepositions were detected, for example, “to identify a corpus” or “it results from the combination of the physical and virtual worlds,” which replicate the grammatical logic of Spanish. Stylistically, the text exhibits lexical repetition and redundant connectors, as well as punctuation influenced by Spanish syntax, with excessively long sentences and irregular use of commas. Finally, a lack of naturalness in discourse transitions and some inconsistency in verb tenses are evident, characteristics of translated texts. For these reasons, a professional linguistic review by a native speaker of academic English is recommended, aimed at improving the manuscript's fluency, textual cohesion, and idiomatic appropriateness.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Overall, despite the exploratory intent of the study, the submission presents several substantive concerns that go beyond issues of structure, APA formatting, and the presentation of results (e.g., the extensive use of pie charts, which is not appropriate for journal publication and reduces clarity).

More critically, the study exhibits major problems related to construct validity. The instructional procedure described is complex, involving augmented reality, multilingual learning, and engagement with multiword expressions; however, the methods and analytical procedures used to evaluate these variables are relatively simple and largely descriptive. This mismatch results in a lack of alignment between the complexity of the constructs under investigation and the simplicity of the measurement and analysis techniques employed.

Consequently, the evidence presented is insufficient to support claims related to learning effectiveness or cognitive engagement, even within an exploratory research framework.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This article investigates the application of mobile Augmented Reality (AR) as a pedagogical resource for teaching and learning Multiword Expressions ((MWEs) in higher education contexts, with a special focus on promoting linguistic diversity. The study involves 220 undergraduate students in Education and combines quantitative and qualitative instruments, including questionnaires, content analysis, and support from Large-Scale Language Models (LLMs) for data coding.

The work presents three specific objectives:

• to implement a multilingual AR resource in initial teacher training;
• to evaluate the didactic and psychopedagogical impact of the resource; and
• to identify and organize a localized corpus of MWEs in the Murcia region (Spain).

The results indicate a highly positive perception on the part of the students, as well as the potential of AR to promote active, multimodal, and inclusive learning.

This study addresses a still little-explored intersection between AR, multilingualism, idiomatic expressions, and cultural diversity, representing a relevant contribution to the fields of educational technology, applied linguistics, and inclusive education.

The manuscript demonstrates scientific quality, thematic originality, and pedagogical relevance. It does, however, present some methodological and structural weaknesses (which are acknowledged by the authors). Furthermore, the article is very long and, in some places, excessively descriptive.

Although the study involves N = 220 students, the central analyses are based on small subsamples (n = 69, n = 54, n = 86). This limits the robustness of the conclusions. This limitation should be highlighted in the Results section.

There are minor grammatical and typographical errors in English. For example:

• Line 17 - “Overal” should be Overall
• Line 92 - duplication of the article "the"
• Occurs several times: tables, figures and text - “Dislake” should be Dislike
• Table A1:
  • Item 4 - “Applicartion” should be Application
  • Item 3 - “wuth simple answer” should be "with simple answer"
• In the “Abbreviations” section, PDPC is defined twice

(there are more; the authors should carefully review them9

• I think the word "Subsection" is superfluous in the title of subsection 3.4.

3.4 Subsection Identification of a corpus of located MWEs (SO3)

• A substantial number of the figures need to be improved, as their current size is too small to allow for clear visualization and accurate interpretation of the presented data.

Comments on the Quality of English Language

There are minor grammatical and typographical errors in the English text that require careful proofreading.

Author Response

Please see the attachment

Author Response File: Author Response.pdf