10 Learning through conversations

Purpose and important information

Simply being given the answer to a problem or question is not an effective way to learn. When individuals actively participate in the process and are challenged to come up with answers themselves, they gain a more comprehensive understanding of what they are studying.

One teaching method based on this premise is the Socratic method. Through conversation, the teacher elicits the learner’s knowledge and uses questions to encourage exploration and linking between concepts.

You can use a language-based generative AI tool to take the place of the teacher and it will facilitate this conversation. You can prompt the AI tool to lead a dialogue that probes your knowledge on your chosen topics, while also encouraging you to make connections between concepts within each topic.

Important note: As you engage with the AI tool, be mindful that the outputs can contain incorrect information, and that current information may not be included. Check the information that is being generated with trusted sources from your course materials.

Prompt template

Note: The following prompt type is most suitable for language-based generative AI. In some generative AI tools, you will obtain higher-quality responses by using a series of prompts, rather than one longer prompt.

For this type of prompt, we recommend giving the AI tool a role which mimics a person who would teach the topics you wish to study (e.g. a university tutor in education). In the prompt, specify the task (e.g. ask me questions) along with any specific requirements or parameters (e.g. to test my understanding of important concepts …).

You can provide instructions so that the AI tool knows how you’d like to interact with it moving forward, e.g. (ask me questions to help me arrive at the answer myself).

You are <a role>. I want to study <topics>. Ask me questions to test my understanding of these topics. After each question, wait for my response. Do not tell me the answer directly. Instead, ask me questions using the Socratic method to help me arrive at the answer myself.

Example prompts and outputs

The following example was tested in RMIT’s Val.

Screenshot of Val prompts and outputs, please see link to an appendix with an accessible text version in the reference below the screenshot.

RMIT University. (2024). Val [Large language model]. https://val.rmit.edu.au. See Appendix 3 for prompts used and text generated.

Licence

Icon for the Creative Commons Attribution-NonCommercial 4.0 International License

Generative AI at RMIT Copyright © by RMIT University Library is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, except where otherwise noted.

Share This Book