Jonathan Y. H. SIM1 and LEE Li Neng2
1AI Centre for Educational Technologies (AICET) and
Department of Philosophy, Faculty of Arts and Social Sciences (FASS), NUS
2Centre for Teaching, Learning and Technology (CTLT), and
Department of Psychology, FASS, NUS
1jyhsim@nus.edu.sg
2psylln@nus.edu.sg
Sim, J. Y. H., & Lee, L. N. (2024). From solitary to social: How GenAI can enhance social learning processes with peers in reading [Paper presentation] In Higher Education Conference in Singapore (HECS) 2024, 3 December, National University of Singapore. https://blog.nus.edu.sg/hecs/hecs2024-jyhsim-leeln/
SUB-THEME
Opportunities from Generative AI
KEYWORDS
Reading, interpretation, social learning, peer teaching, peer learning
CATEGORY
Paper Presentation
EXTENDED ABSTRACT
Historically, reading has been a communal activity, where people came together to read aloud and engage in discussions and debates (Fischer, 2003; Sedo, 2011), enabling readers to collectively interpret and make sense of readings in their contexts.
Today, however, reading is typically perceived as a solitary activity.
Interestingly, Generative AI (GenAI) has begun to revive communal aspects of reading. Instead of engaging with other humans, students now interact with ChatGPT to discuss their readings. GenAI provides patient and non-judgemental responses, making these tools an essential learning partner for many students who now claim they “cannot imagine being a student without ChatGPT.”
Can interactions with GenAI chatbots elicit the same benefits as human social learning?
Social learning can activate “higher forms of human mental activity” (Lantolf & Thorne, 2006). Collaborative processes can help learners move from an understanding with the help of others (social learning), to understanding independently (individual learning) to perform tasks that previously were harder than what one can usually handle (Vygotsky, 1978; Commander & De Guerrero, 2013). Mastery of reading and sense-making does not happen in isolation; instead, it is achieved through social collaboration, leading to a “co-construction of meaning” (Sweet & Snow, 2002).
However, present-day GenAI has limitations. GenAI’s responses are influenced by the user’s knowledge and constrained by the user’s ignorance. For example, if I know nothing about periodontitis, I can only ask basic questions and cannot probe deeply. GenAI, at this point in time, cannot facilitate my discovery the way a human instructor can. As such, a student’s ability to interpret and make meaning of what she reads is likewise constrained by her existing knowledge, even with GenAI assistance.
One way to address such limitations is to introduce students to communal reading activities that incorporate GenAI, teaching students to consult these tools as part of social learning processes with peers. The more we practice this in the classroom, the more students will discover the benefits of social learning augmented by GenAI. This will allow us to effectively condition better GenAI usage habits in our learners over time.
As a start, I have begun to condition learners in social learning processes with GenAI through the course“PH2301 Classical Chinese Philosophy I” (taught during Semester 1, AY2023/24). In each tutorial, students received handouts containing curated passages from an ancient Chinese thinker. They discussed their interpretations in groups before consulting ChatGPT (role-playing as an ancient philosopher). After being made aware of ChatGPT’s potential inaccuracies and hallucinations, students were tasked to evaluate ChatGPT’s interpretation and consider incorporating its ideas if they found any merit.
Engaging in communal reading practices enhances students’ individual cognition through shared understanding and collaborative sense-making (Lantolf & Thorne, 2006). In an end-of-course survey to the entire class (26 students) about these activities, 72% (18 of 25) of respondents appreciated the social learning opportunities. In the open-ended remarks, respondents highlighted how it helped them put their “thoughts into words,” and deepened their understanding. They enjoyed learning from each other, “build[ing] upon each other’s ideas,” and discovering their own contradictions and insights.
The use of ChatGPT in communal reading activities catalysed richer collaborations. In the same survey, students reported positive learning experience with ChatGPT. Although some remarked that it gave “textbook responses,” many valued it “as a springboard” to “gather their thoughts” and engage in deeper discussions with their peers.
Students typically worry about causing others to “lose face” or appear hostile if they voice their criticisms of their peer’s ideas (Sim, 2019) Discussing ChatGPT’s answers gave them opportunities to practice evaluating and voicing critical comments whilst overcoming such concerns. One student mentioned how this helped him to be more thoughtful “and critical about the answers it produces.”
A handful of students (mainly the stronger ones) felt that ChatGPT “lowered the quality of discussion,” this challenge mirrors the dynamics of groups with weaker students. For the next semester, I hope to provide more facilitation guides to ensure that all students, regardless of their calibre, can benefit from this activity. Nonetheless, this teaching and learning experience demonstrates the potential for GenAI to enrich social learning processes in various applications, from communal reading to peer learning.
REFERENCES
Lantolf, J. P., & Thorne, S. L. (2006). Sociocultural theory and the genesis of second language development. Oxford University Press.
Fischer, S. R. (2003). A History of Reading. Reaktion Books Ltd.
Commander, M., & de Guerrero, M. C. M. (2013). Reading as a social interactive process: The impact of shadow-reading in L2 classrooms. Reading in a Foreign Language 25(2), 170–191. https://dx.doi.org/10125/66865
DeNel Rehberg, S. (2011). An introduction to reading communities: Processes and formations. In Reading Communities from Salons to Cyberspace. Palgrave Macmillan.
Sim, J. Y. H. (2019). The ‘face’ barriers to partnership. Teaching and Learning Together in Higher Education 1(27),1-4. Retrieved from https://scholarbank.nus.edu.sg/rest/bitstreams/1373572/retrieve
Vygotsky, L. S. (1978). Mind in society. The development of higher psychological processes. Harvard University Press.