Developing Metacognitive Writers through SLS

Dept.: CLC

Leader: Qi Yanping

Members: Yeo Hwee Yanne, Sun Heyu

What was the current need/gap that you were addressing?

Sec 4 students often approached stimulus-based argumentative questions（材料议论文）by breaking the stimulus into separate writing points. While they understood each part, they did not always recognise the logical relationships connecting them. As a result, students could address the different parts of the question but still produce arguments that lacked depth and coherence. They needed to move beyond knowing what to write towards understanding how the logic of the question should shape their argument.

How had it been experimented and enacted?

The Sec 4 AI in NLC team developed an SLS package, “From Linear Segmentation to Deep Logical Construction”（从线性切分到深度逻辑建构）, to strengthen students’ question analysis and argumentation. Students examined past examination questions and learnt to identify six common logical relationships. They then learnt how each relationship required a different way of reasoning. For example, causal arguments (因果关系） required students to explain how a cause leads to a result, while conditional arguments （条件关系）required them to explain why a condition is necessary. Through model essays, PEEL analysis, targeted practice and SLS-supported learning, students applied these strategies to their own writing. They ended by reflecting on how their thinking had changed and why their writing had improved.

What was the positive impact?

For students: Deeper question analysis: Students became more aware of the logical relationships within a stimulus instead of simply treating each part as a separate writing point. Stronger argumentation: Students learnt to develop their evidence and reasoning according to the specific logical relationship in the question. Greater metacognitive awareness: Students reflected on how their thinking and writing had changed and identified strategies that helped them construct stronger arguments. More critical use of AI: Students were encouraged to examine AI feedback and use it to identify gaps in their reasoning, rather than simply accepting AI-generated answers. For teachers: The package provided a common approach for teaching stimulus-based argumentative writing. It was refined based on student feedback, shared with cluster schools through PLC sessions, and uploaded to the SLS Community Gallery for teachers in other schools to adapt.

What is a future need that this IdEas@work could meet?

The next step is to help students apply these logical reasoning strategies independently to unfamiliar questions. AI can also be further used as a reflective partner to help students identify gaps in their reasoning, compare different arguments and decide whether to accept, modify or reject AI feedback. This can develop more independent and discerning writers who use AI to support, rather than replace, their own thinking.