arxiv: 2604.27344 · v1 · submitted 2026-04-30 · 💻 cs.CY

Recognition: unknown

Bibliometric Mapping of AI-Supported Social Presence in Online Learning Environments: Trends, Collaboration, and Thematic Directions

Agnes R. Regala, Aileen P. De Leon, Almer B. Gamboa, Erika M. Pineda, Jan Henry B. Sunga, John Paul P. Miranda, Rhiziel P. Manalese, Roque Francis B. Dianelo, Vernon Grace M. Maniago

Authors on Pith no claims yet

Pith reviewed 2026-05-07 09:26 UTC · model grok-4.3

classification 💻 cs.CY

keywords AI-supported social presenceonline learning environmentsbibliometric analysispublication trendsinternational collaborationethical issues in AI educationinstructional designScopus data

0 comments

The pith

Research on AI-supported social presence in online learning has risen since 2020, led by the US and Brazil, yet international collaboration stays limited and ethical questions on trust remain underexplored.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This bibliometric review charts the growth and structure of studies that use AI to foster social presence in virtual classrooms. It draws on 59 open-access empirical papers indexed in Scopus and applies citation counts, co-authorship maps, institutional rankings, and keyword clusters to reveal patterns. Publications accelerate after 2020, with core themes around learner engagement, AI tools, instructional design, and emerging ethics. Leading output comes from the United States and Brazil, but cross-border co-authorship is sparse. A sympathetic reader would care because online education continues to expand and these maps point to where effort and gaps now sit.

Core claim

Analysis of 59 Scopus-indexed open-access studies shows an upward publication trend beginning in 2020. Dominant research foci include engagement, AI tools, instructional design, and ethical issues. The United States and Brazil contribute the most papers, while international collaboration remains limited. Ethical concerns around trust and fairness appear as emerging topics but receive little deep treatment. The study concludes that future work should emphasize ethical integration, wider interdisciplinary ties, and learner-centered AI designs.

What carries the argument

Bibliometric mapping performed with Python tools on citation networks, co-authorship graphs, institutional output, and keyword co-occurrence clusters drawn from 59 Scopus papers.

If this is right

Future studies will likely prioritize ethical safeguards for trust and fairness when deploying AI in online settings.
Greater international co-authorship is needed to move the field beyond national clusters.
Instructional designers should integrate learner-centered AI features more systematically to sustain engagement.
Policy and funding bodies can target the currently thin ethical and cross-border research areas.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If open-access restrictions shape the sample, including paywalled papers might alter the picture of collaboration strength.
The identified themes suggest room for new empirical work that directly tests AI interventions for social presence rather than mapping them.
Limited collaboration could be addressed by creating shared datasets or joint projects across the leading countries.
As online learning scales, the underexplored ethics topics may become bottlenecks for widespread adoption.

Load-bearing premise

The 59 open-access empirical studies retrieved from Scopus are sufficient to represent the overall development, influence, and collaboration patterns across the entire field of AI-supported social presence research.

What would settle it

Locating a substantial body of relevant studies from other databases or published before 2020 that show different leading countries, higher international collaboration rates, or mature ethical analysis would undermine the reported trends and gaps.

Figures

Figures reproduced from arXiv: 2604.27344 by Agnes R. Regala, Aileen P. De Leon, Almer B. Gamboa, Erika M. Pineda, Jan Henry B. Sunga, John Paul P. Miranda, Rhiziel P. Manalese, Roque Francis B. Dianelo, Vernon Grace M. Maniago.

**Figure 1.** Figure 1: Top co-authorship network view at source ↗

**Figure 2.** Figure 2: Keywords co-occurrence network 225 view at source ↗

read the original abstract

This study examines the development, influence, and collaboration patterns in AI-supported social presence research within online learning environments. Utilizing 59 open-access empirical studies from Scopus, the study applies citation analysis, co-authorship mapping, institutional analysis, and keyword clustering using Python-based bibliometric tools. Findings reveal an upward trend in publications since 2020, with research focusing on engagement, AI tools, instructional design, and ethical issues. While countries such as the United States and Brazil are leading contributors, international collaboration remains limited. Ethical concerns related to trust and fairness are emerging but underexplored. The study highlights the importance of ethical integration, interdisciplinary collaboration, and learner-centered AI applications in education.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This is a routine bibliometric snapshot of AI-supported social presence in online learning that adds little beyond standard trend counts from a thin 59-paper slice.

read the letter

The paper runs citation analysis, co-authorship maps, and keyword clusters on 59 open-access empirical studies pulled from Scopus. It reports an uptick in publications after 2020, emphasis on engagement and AI tools, US and Brazil as top contributors, and limited cross-border work, plus some mention of emerging ethics questions around trust and fairness. That is the core output, delivered through Python bibliometric scripts. The approach follows the usual playbook for this kind of mapping exercise, and the descriptive summaries are straightforward to follow. What it does cleanly is pull together publication counts and basic network views for this narrow intersection of topics. Similar exercises already exist for broader AI-in-education or online-learning corpora, so the incremental value sits mainly in the specific focus rather than any new method or dataset. The soft spot is the data foundation. Restricting to open-access Scopus items without a reported search string, explicit inclusion criteria, or any check against paywalled or non-Scopus sources leaves the reported trends and country rankings open to selection effects. No sensitivity test appears for that filter, and the abstract gives no sense of how many papers were screened out. That makes the central claims about growth, collaboration patterns, and thematic directions rest on an untested sample. The work is aimed at educational-technology researchers who want a quick overview of activity in this sub-area rather than a theoretical advance or new measurement tool. A reader already familiar with bibliometric reviews will find the execution competent but not surprising. It deserves a serious referee round because the topic is timely and the basic mapping could be made more reliable with clearer methods and broader coverage; the current version is too thin on justification to stand as is.

Referee Report

3 major / 2 minor

Summary. The paper conducts a bibliometric analysis of 59 open-access empirical studies retrieved from Scopus on AI-supported social presence in online learning environments. It applies citation analysis, co-authorship mapping, institutional analysis, and keyword clustering via Python tools to identify trends, collaboration patterns, and themes. Key claims include an upward publication trend since 2020, research emphasis on engagement, AI tools, instructional design, and ethics, leadership by the United States and Brazil, limited international collaboration, and emerging but underexplored ethical concerns around trust and fairness.

Significance. If the sample proves representative, the mapping could usefully highlight growth areas and gaps for researchers in AI-enhanced education, particularly the call for ethical integration and interdisciplinary work. The application of standard bibliometric techniques on a focused corpus is a modest contribution, but the absence of reproducibility details and sensitivity checks limits its reliability as a field overview.

major comments (3)

[Methods] The methods description provides no explicit Scopus search string, date range, inclusion/exclusion criteria, or justification for restricting to open-access empirical studies only. This omission directly undermines replicability and the validity of all downstream claims about publication trends, country leadership, and thematic clusters derived from the 59 papers.
[Results and Discussion] The central claims of upward trend since 2020, US/Brazil dominance, limited international collaboration, and underexplored ethics rest on this single-database, open-access filter without any sensitivity analysis or comparison to the full Scopus corpus. The skeptic note correctly identifies this representativeness assumption as load-bearing and untested, risking systematic bias in observed patterns.
[Thematic Analysis] Keyword clustering and thematic interpretation (engagement, AI tools, ethics) are presented without validation steps such as inter-rater checks or cross-referencing against non-open-access literature, weakening the assertion that ethical issues are 'emerging but underexplored.'

minor comments (2)

[Methods] The abstract and results would benefit from clearer reporting of exact Python libraries and parameters used for co-authorship and keyword networks to aid reproducibility.
[Figures] Figure captions for bibliometric maps should explicitly state the number of nodes/edges and clustering algorithm to improve interpretability.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed comments, which have prompted important improvements to the transparency and robustness of our bibliometric analysis. We have revised the manuscript to address the concerns about replicability, representativeness, and validation of themes. Our responses to each major comment are provided below.

read point-by-point responses

Referee: [Methods] The methods description provides no explicit Scopus search string, date range, inclusion/exclusion criteria, or justification for restricting to open-access empirical studies only. This omission directly undermines replicability and the validity of all downstream claims about publication trends, country leadership, and thematic clusters derived from the 59 papers.

Authors: We agree that the original Methods section was insufficiently detailed for replicability. In the revised manuscript we have inserted the exact Scopus search string, the date range of the search, the full list of inclusion criteria (open-access empirical studies focused on AI-supported social presence in online learning environments), and the exclusion criteria (non-empirical works, non-English papers, and studies outside the defined scope). We also added a justification for the open-access restriction: it ensures all source material is publicly verifiable and aligns with the goal of mapping accessible research for the education community. A PRISMA-style flow diagram documenting the retrieval and screening process has been included. revision: yes
Referee: [Results and Discussion] The central claims of upward trend since 2020, US/Brazil dominance, limited international collaboration, and underexplored ethics rest on this single-database, open-access filter without any sensitivity analysis or comparison to the full Scopus corpus. The skeptic note correctly identifies this representativeness assumption as load-bearing and untested, risking systematic bias in observed patterns.

Authors: We acknowledge that restricting the corpus to open-access records from a single database introduces potential bias and that the representativeness of the 59-paper sample was not previously tested. In the revision we have added an expanded Limitations subsection that explicitly discusses these constraints and their possible effects on the reported trends, country rankings, and collaboration patterns. Although a complete sensitivity analysis against the full (including paywalled) Scopus corpus is not feasible within the present study, we have contextualized our findings against two recent broader reviews of AI in education. The original claims are now qualified with appropriate caveats regarding the open-access focus. revision: partial
Referee: [Thematic Analysis] Keyword clustering and thematic interpretation (engagement, AI tools, ethics) are presented without validation steps such as inter-rater checks or cross-referencing against non-open-access literature, weakening the assertion that ethical issues are 'emerging but underexplored.'

Authors: We have strengthened the Thematic Analysis section by documenting the exact parameters of the Python-based keyword clustering procedure and by adding an inter-rater validation step: two authors independently reviewed and labeled the resulting clusters, resolving disagreements through discussion. We also performed a limited cross-check by examining abstracts of non-open-access papers returned by the same Scopus query. These steps corroborate that ethical dimensions such as trust and fairness receive comparatively little attention. The manuscript language has been revised from “emerging but underexplored” to “emerging and still limited,” with an explicit call for future work. revision: yes

Circularity Check

0 steps flagged

No circularity in descriptive bibliometric analysis

full rationale

The paper conducts a standard bibliometric analysis by retrieving 59 open-access empirical studies from the external Scopus database and applying citation analysis, co-authorship mapping, institutional analysis, and keyword clustering via Python tools. All reported findings (upward trend since 2020, focus areas, country leadership, limited collaboration, emerging ethics themes) are direct descriptive summaries of this external dataset with no equations, derivations, predictions, or first-principles results. No steps reduce by construction to the paper's own inputs, self-definitions, fitted parameters renamed as predictions, or load-bearing self-citations; the work is self-contained as an empirical description of the chosen sample.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central descriptive claims rest on the assumption that Scopus open-access records form a representative sample of the field and that standard bibliometric metrics faithfully capture influence and collaboration.

axioms (1)

domain assumption Scopus database plus open-access filter yields a sufficiently complete and unbiased set of empirical studies on AI-supported social presence
Invoked when selecting the 59 studies; no discussion of coverage gaps or language bias appears in the abstract.

pith-pipeline@v0.9.0 · 5467 in / 1306 out tokens · 72653 ms · 2026-05-07T09:26:38.655432+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

15 extracted references · 12 canonical work pages

[1]

Aljawharah M Aldosari, Saad M Alramthi, and Hala F Eid. 2022. Improving social presence in online higher education: Using live virtual classroom to con- front learning challenges during COVID-19 pandemic. Front. Psychol. 13, (2022). Retrieved from 10.3389/fpsyg.2022.994403

work page doi:10.3389/fpsyg.2022.994403 2022
[2]

Rex Bringula, Roman Paulo Baet, Ralph Lawrence Garcia, Franchesca Mari, Jan Carlo Ramos, Hanna Sophia Sarmiento, and Edmon Torres. 2024. Factors Contributing to the Negative Online Learning Academic Self-Concept of College Students. In Proceedings of the 32ndInternational Conference on Computers in Education, 2024. 1–6. Retrieved from https://library.apsc...

2024
[3]

Manuel B Garcia. 2023. Fostering an Innovation Culture in the Education Sector: A Scoping Review and Bibliometric Analysis of Hackathon Research. Innov. High. Educ. (2023). https://doi.org/10.1007/s10755-023-09651-y

work page doi:10.1007/s10755-023-09651-y 2023
[4]

Garcia, Clievenze Karl Quejado, Clark Raven B

Manuel B. Garcia, Clievenze Karl Quejado, Clark Raven B. Maranan, Owen N. Ualat, Rossana T. Adao, Ari Happonen, Ramazan Yilmaz, and Aras Bozkurt. 2024. Social Relationship Development in the Metaverse: The Roles of Embodiment, Immersion, and the Moderating Effect of Copresence. In TENCON 2024 - 2024 IEEE Region 10 Conference (TENCON) , 2024. Singapore, 15...

work page doi:10.1109/tencon61640.2024.10902707 2024
[5]

Philip Hallinger and Jasna Kovačević. 2023. Applying bibliometric review meth- ods in education: rationale, definitions, analytical techniques, and illustrations. In Robert J Tierney, Fazal Rizvi and Kadriye B T - International Encyclope- dia of Education (Fourth Edition) Ercikan (eds.). Elsevier, Oxford, 546–556. https://doi.org/10.1016/B978-0-12-818630-...

work page doi:10.1016/b978-0-12-818630-5.05070-3 2023
[6]

Yungwei Hao. 2016. Chapter 12 - Investigating Students’ Feelings and Their Perspectives Toward Web 2.0 Technologies in a Teacher Education Course. In Emotions and Technology, Sharon Y Tettegah and Martin B T - Emotions Gartmeier Technology, Design, and Learning (eds.). Academic Press, San Diego, 239–271. https://doi.org/10.1016/B978-0-12-801856-9.00012-8

work page doi:10.1016/b978-0-12-801856-9.00012-8 2016
[7]

Carol Johnson. 2017. Teaching music online: Changing pedagogical approach when moving to the online environment. London Rev. Educ. 15, 3 (2017), 439–456. Retrieved from https://files.eric.ed.gov/fulltext/EJ1163219.pdf

2017
[8]

A Kaur, A Mustafa, L Mehta, and A Dhall. 2018. Prediction and Localization of Student Engagement in the Wild. In 2018 Digital Image Computing: Techniques and Applications (DICTA), 2018. 1–8. https://doi.org/10.1109/DICTA.2018.8615851

work page doi:10.1109/dicta.2018.8615851 2018
[9]

Jia Miao and Li Ma. 2022. Students’ online interaction, self-regulation, and learn- ing engagement in higher education: The importance of social presence to online learning. Front. Psychol. 13, (2022). Retrieved from 10.3389/fpsyg.2022.815220

work page doi:10.3389/fpsyg.2022.815220 2022
[10]

Lynette Nagel and Theuns G Kotzé. 2010. Supersizing e-learning: What a CoI survey reveals about teaching presence in a large online class. Internet High. Educ. 13, 1 (2010), 45–51. https://doi.org/10.1016/j.iheduc.2009.12.001

work page doi:10.1016/j.iheduc.2009.12.001 2010
[11]

Elise Schurink. 2019. The role of perceived social presence in online shopping: The effects of chatbot appearance on perceived social presence, satisfaction and purchase intention. University of Twente. Retrieved from https://essay.utwente. nl/77361/1/Schurink_MA_BMS.pdf

2019
[12]

Emma J Stodel, Terrie Lynn Thompson, and Colla J MacDonald. 2006. Learners’ Perspectives on what is Missing from Online Learning: Interpretations through the Community of Inquiry Framework. Int. Rev. Res. Open Distrib. Learn. 7, 3 SE-Research Articles (December 2006). https://doi.org/10.19173/irrodl.v7i3.325

work page doi:10.19173/irrodl.v7i3.325 2006
[13]

Murat Turk, Benjamin C Heddy, and Robert W Danielson. 2022. Teaching and social presences supporting basic needs satisfaction in online learning environ- ments: How can presences and basic needs happily meet online? Comput. Educ. 180, (2022), 104432. https://doi.org/10.1016/j.compedu.2022.104432

work page doi:10.1016/j.compedu.2022.104432 2022
[14]

Qiaosi Wang, Ida Camacho, Shan Jing, and Ashok K Goel. 2022. Understanding the Design Space of AI-Mediated Social Interaction in Online Learning: Challenges and Opportunities. Proc. ACM Hum.-Comput. Interact. 6, CSCW1 (April 2022). https://doi.org/10.1145/3512977

work page doi:10.1145/3512977 2022
[15]

Okan Yeti̇şensoy and Hıdır Karaduman. 2024. The effect of AI-powered chatbots in social studies education. Educ. Inf. Technol. 29, 13 (2024), 17035–17069. https: //doi.org/10.1007/s10639-024-12485-6 227

work page doi:10.1007/s10639-024-12485-6 2024