Autodidactic Learning with GPT-4

Autodidacticism with GPT-4:

A Theoretical Study of Potentials and Pitfalls of AI-Driven Education in under-resourced context

MA DTCE

26 June 2024

Presenter: Muhammad Zubair Madni

Autodidacticism with GPT-4:

Research Question?

What specific opportunities and challenges does ChatGPT present in supporting autodidacts in under-resourced contexts?

A Theoretical Study of Potentials and Pitfalls of AI-Driven Education in under-resourced context

Research Methodology?

- Systematic Literature Review- Autoethnography

PART 1

Terms widely used in literature for Autodidacticism

Etymology

The term 'autodidacticism' comes from two Greek words: 'autos' meaning 'self' and 'didaktikos' meaning 'teaching'

Definition

Autodidacts are defined as “who teaches themself”. Solomon, J. (2003).

Definitions:

Knowles (1975) defined it as a process wherein “individuals take the initiative, with or without the help of others, in diagnosing their learning needs, formulating learning goals, identifying human and material resources for learning, choosing and implementing appropriate learning strategies, and evaluating learning outcomes”

Autodidactic learning, or self-directed learning, refers to the process of learning independently, without the guidance or support of a teacher or instructor (Candy, 199).

Zimmerman (2022) introduced and placed the focus on the metacognitive monitoring and control of the learning process.

Most researchers agree that the essential abilities of a self-directed learner are:

• The ability to assess learning needs and to establish learning goals,
• The ability to select learning strategies
• The ability to identify and efficiently use learning resources.
• The ability to monitor the progress.
• The ability to adjust the behaviour towards the desired learning outcome.

Perspective on Autodidacticism

When Initial SDLmodels were developed, face-to-face instruction was the predominant mode in higher education. Almost a decade after the last model was developed (cf., Garrison, 1997), higher education is occurring in a variety of contexts, ranging from face-to-face classrooms to virtual classrooms (Song and Hill, 2007).

Autodidacticism with Context

There was a need for new perspectives on how context influences autodidacticism. The following model introduces a conceptual model for understanding SDL in a context. The model incorporates SDL as a personal attribute and a learning process as pointed out by most scholars in the literature of SDL. Then Song and Hill (2007) added a third dimension --the learning context -- to indicate the impact of environmental factors on SDL.

Case Study (Context) 1

• 12 million girls are out of school
• Only 13% of girls advance to grade 9
• In urban areas was recorded as 58.3%
• while only 12% among rural women.
• 21% of girls get married before turning 18.
• Role of patriarchy and imposed religious sentiments.

Children who have never attended school by sex (%)

Case Study (Context) 3

Case Study (Context) 2

• Nomads or seasonal migrants in Deserts of Pakistan.
• Poverty, vulnerability, lack of services, spatial isolation.
• Radio, Television, MOOCs, Distance Education
• Mobile Schools, Boat Schools, Seasonal Schools
• Semi-permanent Schools, Online Tutoring
• Estimates number vary between 2-3 million

• Among Pakistan’s prime teaching systems.
• Known as Islamic seminaries religious institute.
• Theology and Quranic Studies.
• Estimates number vary between 32,000 and 40,000
• Enrolled number of students estimated is 4 million
• No source for scientific or contemporary education

PART 2

ChatGPT

a conversational chatbot equipped with artificial intelligence, capable to understand and generate natural human language in at least 95 languages

ChatGPT was launched in November 2022 as a conversational AI developed by OpenAI for providing human-like responses for any natural language prompt As with other generative AI, balancing innovation possibilities with ethical precautions around bias, accuracy, and academic integrity will be crucial.

SWOT Analysisof ChatGPT

The SWOT analysis of ChatGPT by Farrokhnia et al. (2024) reveals that this new AI technology has various potential applications for education, but also comes with certain challenges, as shown in figure below.

SDL Conceptual Framework

Based on the features of SDL, Kim et al. (2014) developed an SDL conceptual framework for Generative AI that consists of five main aspects of the SDL process

Locating/Accessing Resources

Establishing Learning Goals

• Help learners proactively find resources essential for their learning.
• ChatGPT can provide external and relevant resources aligned with personalized learning plans.
• Recommend various formats like articles, videos, books, and social media groups.
• Suggest interactive and peer-learning opportunities to facilitate the exchange of ideas.

• Encourage adult learners to set SMART goals (Specific, Measurable, Achievable, Relevant, Time-bound) to stay focused.
• ChatGPT can help learners identify specific learning goals based on their interests, motivations, and current knowledge.
• Recommend methods like quizzes and reflection papers to track progress.

• Relate learning goals to personal and professional contexts to enhance relevance.

Monitoring and Evaluating Performance

Adopting and Executing Learning Activities

• Assist learners in developing a specific plan of action that aligns with their goals.
• Tailor plans based on learners' preferred styles and available time.
• Recommend time-management techniques and learning strategies suitable for different learning styles.
• Provide resources in various formats (multimedia, podcasts, articles) based on learning preferences.

• Provide timely and constructive feedback on learners’ progress.
• Analyze interactions and offer suggestions for improvement.
• Share progress data with instructors for targeted feedback and support.
• Ensure learners receive both general and personalized feedback to enhance learning.
• Help learners and instructors make necessary adjustments to achieve learning goals.

Reassessing Learning Strategies Progress

• Generate quizzes and other tools to evaluate learning progress.
• Offer suggestions to adjust learning strategies based on self-assessment results.
• Encourage self-reflection to develop metacognitive skills.
• Help learners set new goals and adjust strategies based on reflections and feedback.
• Ensure learners can think critically about their learning experiences to continuously improve.

Limitations and Challenges of Using ChatGPT for Autodidacts

• The effectiveness of ChatGPT's recommendations heavily depends on the quality and specificity of the input provided by learners.
• Over-reliance on ChatGPT might hinder learners' ability to critically engage with the material and develop independent thinking skills.
• Heavy dependence on ChatGPT could reduce interactions with instructors and peers, diminishing the benefits of collaborative learning experiences.

Limitations and Challenges of Using ChatGPT for Autodidacts

• Ensuring the privacy and security of user data is crucial, and there may be challenges in protecting sensitive information when using AI tools like ChatGPT.
• ChatGPT may provide incorrect or irrelevant resources, leading to confusion and misunderstandings among learners.

Conclusion

In conclusion, several critical questions arise regarding the use of ChatGPT in asynchronous online learning environments: Will it reduce interaction between learners and human instructors? Can it match the engagement and social interaction provided by human instructors? What are its limitations in offering personalized feedback and support? While AI systems like ChatGPT have made significant advancements, they cannot fully replace human instructors. Human instructors offer unique abilities in providing personalized feedback, emotional support, and fostering higher-order thinking and community building, which AI cannot replicate. Relying solely on ChatGPT may lead to a lack of critical thinking, social interaction, and targeted guidance. Therefore, it is essential to balance the use of ChatGPT with human instructor engagement to ensure optimal learning outcomes.

Refrences:

Shah, S., & Shah, U. (2012). Girl education in rural Pakistan. Revista Internacional de Sociología de la Educación, 1(2), 180-207. Suleman, Q., Aslam, H. D., Habib, M. B., Yasmeen, K., Jalalian, M., Akhtar, Z., & Akhtar, B. (2015). Exploring Factors Affecting Girls' Education at Secondary Level: A Case of Karak District, Pakistan. Journal of Education and practice, 6(19), 95-109. Baughn, B. (2018). Nomadic education in Northern Pakistan: The Bakarwal case. UMT Education Review, 1(1), 1-22. Siddique, A., Ahmad, S., & Chaudhary, A. H. (2017). Exploring nomadic parents’ attitudes towards education of children in district Kasur. Journal of Secondary Education and Research, 1(1), 45-55. Suri, K. (2014). Education, conflict and development: A case study of mobile schools for pastoralists in Jammu and Kashmir. IOSR Journal of Research and methods in Education, 4(1), 12-9. Ahmed, Z. S. (2009). Madrasa education in the Pakistani context: Challenges, reforms and future directions. Peace Prints: South Asian Journal of Peace building, 2(1), 1-13. Sajjad, F. (2013). Reforming Madrasa Education in Pakistan; Post 9/11 Perspectives. Journal of Islamic Thought and Civilization, 3(1), 104-121. Firat, M. (2023). How chat GPT can transform autodidactic experiences and open education?.

Kim R., Olfman L., Ryan T., Eryilmaz E. (2014). Leveraging a personalized system to improve self-directed learning in online educational environments. Computers and Education, 70(2014), 150–160. Crossref Farrokhnia, M., Banihashem, S. K., Noroozi, O., & Wals, A. (2024). A SWOT analysis of ChatGPT: Implications for educational practice and research. Innovations in education and teaching international, 61(3), 460-474. https://seo.ai/blog/how-many-languages-does-chatgpt-support (accessed April 20, 2024) Song, L., & Hill, J. R. (2007). A conceptual model for understanding self-directed learning in online environments. Journal of interactive online learning, 6(1), 27-42. Solomon, J. (2003). Theories of learning and the range of autodidactism. In The Passion to Learn (pp. 15-34). Routledge. B. J. Zimmerman (2022) Becoming a self-regulated learner: An overview, Theory into practice, vol. 41, no. 2. M. S. Knowles (1975) Self-directed learning: A guide for learners and teachers. ERIC. Candy, P. C. (1991). Self-direction for lifelong learning: A comprehensive guide to theory and practice. San Francisco, CA: Jossey-Bass. Ahmad, I., Said, H., Hussain, A., & Khan, S. (2014). BARRIERS TO CO-EDUCATION IN PAKISTAN AND ITS IMPLICATIONS ON GIRLS EDUCATION: CRITICAL REVIEW OF LITERATURE. Science International, 26(1).