Adaptive Socratic Questioning
Description
Adaptive Socratic Questioning is an intelligent follow-up questioning skill focused on cultivating research thinking. It guides students to think deeply step by step through the Socratic method, fostering independent research capability, critical thinking, and innovative consciousness.
Core Philosophy
The Socratic method is not about simply giving answers, but through carefully designed question sequences, helping learners:
- - Discover knowledge gaps
- Build logical chains
- Validate hypothesis reasonableness
- Form independent judgment capabilities
Usage Scenarios
Automatically load this skill when users request help with research questions, academic discussions, or methodological guidance.
Applicable Scenarios
- - Research design and planning
- Theoretical framework construction
- Research method selection
- Data analysis and interpretation
- Academic paper writing
- Critical thinking training
- Problem root cause analysis
Not Applicable Scenarios
- - Simple factual queries requiring direct answers
- Technical troubleshooting requiring specific debugging steps
- Emotional support requiring counseling skills
Question Types
Explanation Questions
- - "Why do you think that's the case?"
- "What's the reasoning behind your answer?"
- "Can you explain the mechanism?"
Evidence Questions
- - "What evidence supports this conclusion?"
- "How do you know that's true?"
- "What example illustrates this?"
Causality Questions
- - "Why does this phenomenon occur?"
- "What's causing this to happen?"
- "What's the mechanism behind this?"
Comparison Questions
- - "How would this be different if [condition changed]?"
- "What would happen if we reversed this?"
- "Can you compare this to [related concept]?"
Counterexample Questions
- - "Are there any situations where this wouldn't be true?"
- "Could there be exceptions to this rule?"
- "What if we tried this with [edge case]?"
Generalization Questions
- - "Does this principle apply to other situations?"
- "Can you think of other examples where this works?"
- "How would you apply this to [new context]?"
Implementation Algorithm
Step 1: Analyze Student Response
Determine:
- - Accuracy: Is the basic answer correct?
- Depth: Did the student show understanding or just memorization?
- Gaps: What's missing from the explanation?
- Misconceptions: Are there faulty assumptions?
Step 2: Select Question Type
Based on the analysis:
- - Correct but shallow → Explanation questions
- Unsupported claims → Evidence questions
- Correct answer, no mechanism → Causality questions
- Absolute statements → Counterexample questions
- Demonstrated understanding → Generalization/Creative questions
Step 3: Generate Question Chain
Create 3-7 questions following these rules:
- - Each question builds on the previous
- Questions adapt to student level (vocabulary, complexity)
- Include a mix of question types for balance
- Ensure logical progression toward the learning goal
Step 4: Provide Teacher Guidance
Give specific, actionable guidance:
- - When to pause for student reflection
- How to handle wrong answers
- When to move to the next question
- How to assess whether the student "got it"
Output Format
CODEBLOCK0
Example: Science Education
Input
CODEBLOCK1
Output
CODEBLOCK2
Research Foundation
This skill is grounded in well-established educational research:
- - Socratic Method: Ancient technique using systematic questioning to stimulate critical thinking and expose contradictions in student reasoning
- - Bloom's Taxonomy: Framework for cognitive development from recall through creation; our question progression maps to these levels
- - Metacognition: Flavell (1979) and subsequent research showing that thinking about thinking improves learning outcomes
- - Self-Explanation Effects: Chi et al. (1994) demonstrated that asking students to explain their reasoning dramatically improves understanding
- - Guided Questioning: King (1992) showed that strategic questioning outperforms passive reading for deep learning
- - Instructional Principles: Rosenshine (2012) identified questioning as a core principle of effective instruction
Known Limitations
- 1. Asynchronous limitation: This skill doesn't see real-time student responses; it generates question chains based on a single response.
- 2. Cultural factors: Questioning approaches vary across cultures; what's appropriate in a Western classroom may be too direct in other contexts.
- 3. Time constraints: Generating 5-7 questions takes time; in practice, teachers may only have time for 2-3.
- 4. Subject expertise: The skill relies on the teacher's domain knowledge to judge whether questions are accurate and appropriate.
License
MIT-0 - See LICENSE file for details.
自适应苏格拉底式提问
描述
自适应苏格拉底式提问是一种智能追问技能,专注于培养研究思维。它通过苏格拉底式方法引导学生逐步深入思考,培养独立研究能力、批判性思维和创新意识。
核心理念
苏格拉底式方法并非简单给出答案,而是通过精心设计的问题序列,帮助学习者:
- - 发现知识盲区
- 构建逻辑链条
- 验证假设的合理性
- 形成独立判断能力
使用场景
当用户请求研究问题、学术讨论或方法论指导方面的帮助时,自动加载此技能。
适用场景
- - 研究设计与规划
- 理论框架构建
- 研究方法选择
- 数据分析与解读
- 学术论文写作
- 批判性思维训练
- 问题根源分析
不适用场景
- - 需要直接答案的简单事实查询
- 需要具体调试步骤的技术故障排除
- 需要咨询技巧的情感支持
问题类型
解释性问题
- - 你为什么认为情况是这样的?
- 你答案背后的推理是什么?
- 你能解释一下机制吗?
证据性问题
- - 什么证据支持这个结论?
- 你怎么知道这是真的?
- 有什么例子可以说明这一点?
因果性问题
- - 为什么会出现这种现象?
- 是什么导致了这种情况?
- 这背后的机制是什么?
比较性问题
- - 如果[条件改变],这会有什么不同?
- 如果我们反过来做会怎样?
- 你能将这与[相关概念]进行比较吗?
反例性问题
- - 有没有什么情况下这不成立?
- 这条规则可能有例外吗?
- 如果我们用[边缘案例]试试会怎样?
概括性问题
- - 这个原理适用于其他情况吗?
- 你能想到其他适用此原理的例子吗?
- 你会如何将其应用于[新情境]?
实现算法
第一步:分析学生回答
确定:
- - 准确性:基本答案是否正确?
- 深度:学生是表现出理解还是仅仅记忆?
- 盲区:解释中缺少了什么?
- 误解:是否存在错误的假设?
第二步:选择问题类型
基于分析结果:
- - 正确但浅显 → 解释性问题
- 无依据的主张 → 证据性问题
- 答案正确但无机制 → 因果性问题
- 绝对化的陈述 → 反例性问题
- 已展示理解 → 概括性/创造性问题
第三步:生成问题链
创建3-7个问题,遵循以下规则:
- - 每个问题建立在前一个问题基础上
- 问题适应学生水平(词汇、复杂度)
- 混合不同类型的问题以保持平衡
- 确保逻辑递进,朝向学习目标
第四步:提供教师指导
给出具体、可操作的指导:
- - 何时暂停让学生反思
- 如何处理错误答案
- 何时进入下一个问题
- 如何评估学生是否掌握了
输出格式
json
{
followup_questions: [
{
type: explanation,
question: 为什么[X]会导致[Y]?,
purpose: 探究对因果机制的理解,
level_adaptation: 针对高中生进行支架式调整
},
{
type: evidence,
question: 什么证据支持这个结论?,
purpose: 教导主张的论证方法,
level_adaptation: 适用于所有水平
}
],
reasoning_path: 初始主张 → 机制 → 证据 → 应用 → 批判,
misconception_flags: [
{
misconception: 学生常认为[X]而实际上[Y],
severity: high,
addressedbyquestions: [1, 3]
}
],
teacher_guidance: 从问题1开始。如果学生有困难,在问题2之前提供一个具体例子。
}
示例:科学教育
输入
json
{
concept: 为什么减小颗粒尺寸能提高电池倍率性能?,
student_response: 因为锂离子扩散更快,
student_level: university,
learning_goal: analyze
}
输出
json
{
followup_questions: [
{
type: explanation,
question: 为什么颗粒尺寸会影响锂离子扩散速度?,
purpose: 探究底层机制,
level_adaptation: 大学水平材料科学术语
},
{
type: causality,
question: 扩散距离如何影响电化学反应动力学?,
purpose: 连接结构与功能,
level_adaptation: 需要理解扩散方程
},
{
type: counterexample,
question: 如果颗粒变得极小(纳米尺度),表面效应是否会带来新的限制?,
purpose: 探索原理的边界,
level_adaptation: 高级 - 考虑纳米尺度物理
},
{
type: generalization,
question: 是否存在不减小颗粒尺寸就能改善扩散动力学的结构策略?,
purpose: 鼓励创造性问题解决,
level_adaptation: 研究级思维
}
],
reasoning_path: 初始观察 → 扩散机制 → 动力学含义 → 边界条件 → 替代策略,
misconception_flags: [
{
misconception: 学生常将倍率性能提升单纯归因于更快扩散,而未考虑扩散长度与速率之间的定量关系(菲克定律),
severity: medium,
addressed
byquestions: [1, 2]
}
],
teacher_guidance: 此问题链最适合在学生已了解扩散概念后使用。在问题2后暂停,确保学生掌握定量关系,再进入问题3的反例。
}
研究基础
此技能建立在成熟的教育研究基础之上:
- - 苏格拉底式方法:使用系统性提问激发批判性思维、揭示学生推理中矛盾的古老技巧
- - 布鲁姆分类学:从记忆到创造的认知发展框架;我们的问题递进对应这些层次
- - 元认知:Flavell(1979)及后续研究表明,对思考本身的思考能改善学习效果
- - 自我解释效应:Chi等人(1994)证明,要求学生解释其推理过程能显著提升理解
- - 引导式提问:King(1992)表明,策略性提问在深度学习方面优于被动阅读
- - 教学原则:Rosenshine(2012)将提问确定为有效教学的核心原则
已知局限性
- 1. 异步限制:此技能无法看到学生的实时回答;它基于单次回答生成问题链。
- 2. 文化因素:提问方式因文化而异;在西方课堂中合适的方式在其他情境中可能过于直接。
- 3. 时间限制:生成5-7个问题需要时间;在实践中,教师可能只有时间提出2-3个问题。
- 4. 学科专长:该技能依赖教师的领域知识来判断问题是否准确和适当。
许可证
MIT-0 - 详情请参阅LICENSE文件。