Connectivism and Networked Learning in the Age of Artificial Intelligence

Abstract

The rapid and revolutionary appearance of AI technologies has been changing many aspects of our modern lives. Among these, AI advancement is transforming the educational landscape and practice. Yet, integrating such technologies into educational practices necessitates re-examining theories related to learning and education. Connectivism, a more recent learning theory introduced by Siemens (2005), provides a valuable framework for this matter. This theory describes learning as a process that happens within networks connecting knowledge, people, and technology.  This paper aims to examine how AI supports connectivist principles as a node connecting learners to relevant information, through building networks based on each learner’s unique characteristics and preferences. However, challenges such as algorithmic bias, over-reliance on AI, data privacy concerns, and unequal access to technology must be addressed when introducing AI technologies to learners. Hence, the primary purpose of this paper is to address this question: How can AI support and enhance the principles of connectivism in educational environments, and what ethical considerations must be addressed to ensure AI’s fair and beneficial impact on learning?

Introduction

The rapid development of artificial intelligence (AI) in recent years, has significantly influenced the educational landscape and traditional practices. The interaction between the learners and AI has added another layer to the learners’ relationship with information and learning (Slimi, 2023). Consequently, these changes raise the need to re-examine learning through updated approaches. Connectivism, a modern learning theory introduced by Siemens (2005), can help us to further investigate this new phenomenon in the world of education. This theory provides a valuable framework for understanding how learners navigate and interact with different technological systems as it defines learning as a process of connecting with knowledge, people, and technology in a networked environment. Currently, AI-driven learning tools, are the new technology widely used in educational settings, yet there are few studies, investigating their impact on the way education is both delivered and experienced (Almasri, 2024).

 Connectivism, as introduced by Siemens (2005), represents a shift in learning theory by emphasizing the role of networks in the knowledge-building process. Unlike traditional theories that focus on individual knowledge acquisition, connectivism sees learning as a distributed process across various information nodes, including digital and technological platforms. This theory aligns well with recent applications of artificial intelligence (AI) in education. In this theory, AI functions as a node that can facilitate access to information, provide feedback, and support personalized learning experiences. AI’s capacity to act as a connector in a learner’s network underlines its relevance to the principles of connectivism, where knowledge and resources are accessed through interconnected nodes in a digital space (Glassner & Back, 2020).

 Current applications of AI in education illustrate this alignment with connectivist principles. Adaptive learning is one of the emerging concepts of using AI in education in which, an AI algorithm is used to provide customized content, based on learners’ individual needs, preferences, and progress. This feature can enhance tailored educational experience that reflects the core connectivist idea of personal learning (Baker & Hawn, 2022; Sajja et al., 2023). Additionally, intelligent tutoring platforms extend this personalization by providing guided instruction and real-time feedback, mirroring the ways connectivism emphasizes active engagement within a network. Moreover, AI-powered content curation technologies further support this model by helping learners access timely and relevant information, thereby streamlining the learning process and reinforcing the connections that form the basis of connectivism learning (Zheng et al., 2024).

Research question

Given the rapid integration of AI in educational contexts, this paper addresses the following research question: How can AI support and enhance the principles of connectivism in educational environments, and what ethical considerations must be addressed to ensure AI’s fair and beneficial impact on learning?

This question explores AI’s role in providing personalized learning paths, acting as a knowledge node, and fostering collaboration, while also considering ethical challenges such as algorithmic bias, over-reliance on technology, data privacy, and issues of equity, diversity, and inclusion (EDI).

Literature review

a)    Defining Connectivism and Networked Learning

Connectivism is a theory developed by George Siemens. It describes learning as a process that unfolds through recognizing and navigating networks to access distributed knowledge. Traditional learning theories emphasize learning as an individual knowledge acquisition or overlook the role of recent technologies in this process. Connectivism argues that learning happens by navigating through networks of knowledge that are spread out across various sources. This process involves recognizing these networks such as online resources, communities, or databases, and navigating through them whether by searching for information, joining discussions, or using tools to process the information (Siemens, 2005). In this framework, knowledge does not solely reside within the learner but is spread across a network of nodes that include people, digital resources, and technological systems. Based on this theory, each of the components of this network contributes to the learning process (Glassner & Back, 2020).

Another foundational element in this theory is decision-making. It reflects how learners actively engage in selecting relevant knowledge nodes and determining which information aligns best with their learning goals (Bates A.W., 2019).In this context, learning is seen as the process of forming new connections, navigating through connected nodes, and making decisions to strengthen them.

Today’s digital society is changing rapidly, and connectivism provides a way to understand how learning happens in this environment. It explains learning as a process where information is part of a network that adapts and grows with the flow of new information. This approach views learners as active participants in a larger knowledge network, where the key to learning is the ability to access, connect, and combine information effectively (Klara et al., 2024).

b)    AI and Learning Systems

Artificial Intelligence in education refers to the application of AI technologies to enhance teaching and learning processes (Chen et al., 2022). These days, AI-powered technologies are playing a revolutionary role in education. They provide technologies like adaptive learning systems, intelligent tutoring platforms, and content curation tools. Adaptive learning systems, powered by AI, tailor educational experiences to each learner’s needs, preferences, and progress. By analyzing individual behaviors and interactions, these systems adjust content delivery, pacing, and feedback, to provide a personalized learning pathway (Baker & Hawn, 2022; Sajja et al., 2023).  

Intelligent tutoring platforms are another example of AI’s capacity to support connectivist learning. These platforms offer real-time, targeted guidance, simulating the role of a personal tutor who adapts instruction based on the learner’s responses and performance. This functionality immediacy reinforces the connectivist principle of learning through active engagement within a network. This way, learners can receive instant feedback and resources that encourage deeper understanding and networked exploration(Zheng et al., 2024).

Additionally, AI-driven content curation technologies, organize and deliver relevant information to learners. This feature helps the learners navigate through vast digital resources to find materials that support their learning goals (Gligorea et al., 2023). Facilitating the accessibility to relevant knowledge is also one of the other features relating AI to connectivism theory.

Ultimately, AI’s role in adaptive learning, intelligent tutoring, and content curation technologies are some of the few examples explaining how this technology can function as a critical facilitator within a connectivist network of learning.

AI’s Role in Enhancing Connectivist Learning

a)    Providing Personalized Learning Paths by Analyzing Learners’ Individual Needs, Preferences, and Behavior

AI can potentially play a transformative role in creating personalized learning experiences. It can analyze learners’ performance such as needs, behaviors, and preferences based on the collected data. It can also assess each learner’s unique interactions and adjust content, pacing, and complexity, accordingly. With the collected data, it can then craft a customized learning path for each learner. This approach aligns with the connectivist principle of responsive, learner-centered practices (Laak & Aru, 2024; Sajja et al., 2023). The unique way of personalization provided by AI tools, not only enhances learner engagement but also enables learners to build more meaningful connections within their knowledge networks. AI helps students navigate information in ways that align with their specific goals and preferences, which is fundamental to a connectivist approach.

b)    AI as a Knowledge Node: The Role of AI in Providing Faster Access to Information and Real-Time Feedback

In the connectivist framework, AI acts as a dynamic knowledge node that offers learners rapid access to vast amounts of information and immediate feedback. AI Serves as a bridge between students and a wide array of digital resources. AI tools such as intelligent tutoring systems and knowledge retrieval platforms allow learners to search for more specific information, receive instant responses, and explore diverse perspectives (Hu & Wang, 2024). This real-time interaction aligns with the connectivist notion that learning is distributed across networks, where AI enables learners to locate and connect with relevant information more efficiently. Moreover, the real-time feedback provided by AI systems facilitates immediate reflection and deeper understanding of the subject which allows learners to build a stronger network of knowledge. Moreover, by serving as a resourceful and accessible node, AI strengthens the learner’s ability to engage with a broader network of information and knowledge. It can connect learners to more relevant sources of information and reduce the time searching for required information (Afzaal et al., 2024).

c)     Collaborative Learning Provided by AI: Facilitating Peer Collaboration and Networked Learning

Peer-to-peer interaction within digital and online learning communities is a core component of connectivist learning. AI supports collaborative learning by acting as a smart fascinator among learners. Through AI-enhanced discussion forums, peer-matching algorithms, and virtual learning environments, learners can connect, exchange ideas, and build shared knowledge networks (Joseph et al., 2024; Zheng et al., 2024). AI-driven collaborative platforms often recommend peers with similar learning goals or complementary strengths. This technology provides a unique opportunity for learners to build effective study groups in which they can engage in cooperative problem-solving procedures with peers across the globe. This interaction promotes a sense of community and encourages learners to construct knowledge in a shared, universal networked environment. Additionally, the use of AI in moderating discussions and providing resources enhances these connections, allowing learners to draw from collective insights and expertise, which is a key aspect of networked learning in connectivism (Archibald et al., 2023).

AI enhances the connectivist learning experience, through all the mentioned functionalities. It facilitates access to tailored content, can provide immediate personalized feedback, and fosters community interactions. Additionally, it helps learners engage more deeply with their networks, fulfilling the goals of connectivism and adapting education to meet the needs of the digital age.

Challenges and Ethical Considerations

a)    Algorithm Bias: Potential Biases in AI Systems and Their Impact on Education

Algorithmic bias is one of the most significant concerns in the deployment of AI technologies in education. Machine learning is the core of all AI models’ functionality. Such systems learn from large databases of information that often include existing societal biases. Consequently, implementing AI in students’ learning paths can inadvertently influence educational outcomes. For instance, predictive analytics which is used to evaluate student performance can unintentionally disadvantage certain demographic groups, reinforcing existing inequalities instead of promoting equal learning opportunities (Baker & Hawn, 2022; Idowu, 2024). This bias can potentially affect everything from grading to resource recommendations, resulting in an uneven learning environment. As a result, it is essential to implement strategies to identify and reduce such biases in educational AI systems to ensure equity for learners from diverse backgrounds.

b)    Over-Reliance on Technology: Impact on Learners’ Critical Thinking and Decision-Making

The integration of AI in education brings the risk of learners becoming overly reliant on technology for guidance and problem-solving. It can potentially undermine learners’ critical thinking and decision-making abilities. When students rely heavily on AI tools for feedback, answers, or direction, they may miss out on developing independent problem-solving skills essential for real-world scenarios (Miller, 2023). Learners’ overreliance on AI can lead to a reduction in cognitive engagement, as they might passively follow AI recommendations without questioning or analyzing underlying concepts. To address this issue, it is important to create a balance between AI assistance and active learning practices to ensure that learners maintain their ability to think critically and make informed decisions independently.

c)     Data Privacy: Ethical Concerns of Constant Surveillance and Data Collection by AI

As mentioned before, AI-driven educational systems often require extensive data collection to personalize learning experiences. This act leads to concerns about data privacy and the over-surveillance of learners by third-party applications offering AI technologies (Stewart, 2023). To provide personalized learning experiences, AI tools collect information on student behaviors, preferences, performance, and even location. This constant data collection raises ethical questions about how this data is managed, stored, and used. This continuous surveillance can violate students’ privacy rights, making them subjects of data analysis beyond their awareness (Hooshyar et al., 2023). To manage this situation, there is a need for transparent data policies, secure data handling, and informed consent practices to protect student privacy in AI-enabled educational environments (Chang, 2021).

d) Issues with Equity, Diversity, and Inclusion: Accessibility of AI for All Learners

Equity, diversity, and inclusion (EDI) considerations are crucial in evaluating the accessibility of AI in education. Not all students have equal access to the technology, devices, and internet connectivity needed to benefit from AI-enhanced learning systems. This accessibility gap can lead to inequalities in educational outcomes, where only students with sufficient resources can fully engage with AI-based tools (Shelton Ken, 2024). As a result, ensuring that AI systems are accessible and inclusive for all students is necessary to prevent further inequalities and to support a diverse and equitable learning environment (Reiss, 2021).

Conclusion

This paper has explored how AI, aligned with the principles of connectivism, enhances learning practices. This technology serves as a powerful facilitator in adaptive learning systems, intelligent tutoring, and content curation. It can also act as a central knowledge node, and foster collaboration in digital learning communities. Incorporating AI technologies in education aligns with the distributed, networked approach to knowledge sharing that connectivism advocates (Glassner & Back, 2020; Siemens, 2005). AI’s presence in education demonstrates how technology can create dynamic, interconnected learning environments that provide learners with access to timely information, personalized pathways, and collaborative opportunities (Gligorea et al., 2023; Sajja et al., 2023)

However, as AI becomes more integrated into educational systems, ethical considerations, and challenges must be addressed to ensure its responsible application. Algorithmic bias, over-reliance on technology, data privacy, and EDI issues are some of these obstacles. Biases embedded in AI can spread inequities, impact student outcomes, and limit fair access to educational resources (Baker & Hawn, 2022; Idowu, 2024). Additionally, over-reliance on AI can mitigate critical thinking skills in the users of such technologies. On the other hand, the extensive data collection necessary for personalization raises concerns about privacy and surveillance (Hooshyar et al., 2023; Miller, 2023). Lastly, ensuring that AI is accessible to all students, regardless of socioeconomic background, is also essential to prevent a digital divide that could aggravate existing educational inequalities (Shelton Ken, 2024).

Moving forward, educators and policymakers must develop frameworks that promote equitable and ethical AI use in educational contexts. Addressing these challenges will be crucial to harnessing AI’s potential in ways that uphold the principles of connectivism and ensure fair, beneficial learning experiences for all students.

Future research should continue exploring ways to reduce biases, protect users’ privacy, and broaden accessibility to establish a responsible foundation for AI’s role in education.

References:

Afzaal, M., Zia, A., Nouri, J., & Fors, U. (2024). Informative Feedback and Explainable AI-Based Recommendations to Support Students’ Self-regulation. Technology, Knowledge and Learning, 29(1), 331–354. https://doi.org/10.1007/s10758-023-09650-0

Almasri, F. (2024). Exploring the Impact of Artificial Intelligence in Teaching and Learning of Science: A Systematic Review of Empirical Research. Research in Science Education, 54(5), 977–997. https://doi.org/10.1007/s11165-024-10176-3

Archibald, A., Hudson, C., Heap, T., Thompson, R. “Rudi,” Lin, L., DeMeritt, J., & Lucke, H. (2023). A Validation of AI-Enabled Discussion Platform Metrics and Relationships to Student Efforts. TechTrends, 67(2), 285–293. https://doi.org/10.1007/s11528-022-00825-7

A.W. (Tony) Bates and Bates, A. W. (2019). The nature of knowledge and the implications for teaching. In Teaching in a Digital Age (second). Tony Bates Associates Ltd.

Baker, R. S., & Hawn, A. (2022). Algorithmic Bias in Education. International Journal of Artificial Intelligence in Education, 32(4), 1052–1092. https://doi.org/10.1007/s40593-021-00285-9

Chang, B. (2021). Privacy Issues in Online Learning Environment. https://doi.org/https://doi.org/10.1080/01587919.2020.1869527

Chen, X., Zou, D., Xie, H., Cheng, G., & Liu, C. (2022). International Forum of Educational Technology & Society Two Decades of Artificial Intelligence in Education. Technology & Society, 25(1), 28–47. https://doi.org/10.2307/48647028

Glassner, A., & Back, S. (2020). Connectivism: Networks, Knowledge, and Learning. In Exploring Heutagogy in Higher Education (pp. 39–47). Springer Singapore. https://doi.org/10.1007/978-981-15-4144-5_3

Gligorea, I., Cioca, M., Oancea, R., Gorski, A. T., Gorski, H., & Tudorache, P. (2023). Adaptive Learning Using Artificial Intelligence in e-Learning: A Literature Review. In Education Sciences (Vol. 13, Issue 12). Multidisciplinary Digital Publishing Institute (MDPI). https://doi.org/10.3390/educsci13121216

Hooshyar, D., Azevedo, R., & Yang, Y. (2023). Augmenting deep neural networks with symbolic knowledge: Towards trustworthy and interpretable AI for education. http://arxiv.org/abs/2311.00393

Hu, S., & Wang, X. (2024). FOKE: A Personalized and Explainable Education Framework Integrating Foundation Models, Knowledge Graphs, and Prompt Engineering. http://arxiv.org/abs/2405.03734

Idowu, J. A. (2024). Debiasing Education Algorithms. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-023-00389-4

Joseph, G. V., Athira, P., Anit Thomas, M., Jose, D., Roy, T. V., & Prasad, M. (2024). Impact of Digital Literacy, Use of AI tools and Peer Collaboration on AI Assisted Learning: Perceptions of the University Students. Digital Education Review, 45, 43–49. https://doi.org/10.1344/der.2024.45.43-49

Klara, I., Gyonyoru, K., & Ida, K. (2024). The Role of AI-based Adaptive Learning Systems in Digital Education Journal of Applied Technical and Educational Sciences jATES The Role of AI-based Adaptive Learning Systems in Digital Education. 14(2), 1–12. https://doi.org/10.24368/jates380

Laak, K.-J., & Aru, J. (2024). AI and personalized learning: bridging the gap with modern educational goals. http://arxiv.org/abs/2404.02798

Miller, K. (2023, March). AI Overreliance Is a Problem. Are Explanations a Solution? Standfor University . https://hai.stanford.edu/news/ai-overreliance-problem-are-explanations-solution

Reiss, M. J. (2021). The use of AI in education: Practicalities and ethical considerations. London Review of Education, 19(1). https://doi.org/10.14324/LRE.19.1.05

Sajja, R., Sermet, Y., Cikmaz, M., Cwiertny, D., & Demir, I. (2023a). Artificial Intelligence-Enabled Intelligent Assistant for Personalized and Adaptive Learning in Higher Education.

Sajja, R., Sermet, Y., Cikmaz, M., Cwiertny, D., & Demir, I. (2023b). Artificial Intelligence-Enabled Intelligent Assistant for Personalized and Adaptive Learning in Higher Education.

Shelton Ken, L. D. (2024, August). Thinking About Equity and Bias in AI. Edutopia. https://www.edutopia.org/article/equity-bias-ai-what-educators-should-know

Siemens, G. (2005). connectivism. https://jotamac.typepad.com/jotamacs_weblog/files/connectivism.pdf

Slimi, Z. (2023). The Impact of Artificial Intelligence on Higher Education: An Empirical Study. European Journal of Educational Sciences, 10(1). https://doi.org/10.19044/ejes.v10no1a17

Stewart, B. E. (2023). The problem of the web: Can we prioritize both participatory practices and privacy? Contemporary Educational Technology, 15(1). https://doi.org/10.30935/cedtech/12668

Zheng, L., Fan, Y., Chen, B., Huang, Z., LeiGao, & Long, M. (2024). An AI-enabled feedback-feedforward approach to promoting online collaborative learning. Education and Information Technologies, 29(9), 11385–11406. https://doi.org/10.1007/s10639-023-12292-5