Skip to Main Content

EDUC 379A: Educational Psychology

Instructor: Diana Fenton

Using AI for Research

Use genAI to brainstorm and explore your topic further. You can turn on "learning" or "research" modes of various AI chat tools to deepen your understanding of an overall topic and discover avenues for your own research.

Tools

Microsoft Copilot: Our campus preferred chatbot. Copilot asks follow up questions to learn more about what you want to learn more about!

Google Gemini: Deep Research mode

ChatGPT: Study mode

When you've got a pretty good idea of your topic and some areas you'd like to explore more, you can use AI chatbots to help you develop search terms to use in library databases.

Use any of the tools on the previous tab (Copilot, Gemini, ChatGPT) and follow this prompt example:

"My research question is "what is the correlation between a snow-less winter and climate change?". Give me some search terms I can use in library databases"

The chatbot will likely give you information about how to search in library databases, as well as provide recommendations for search strings you could try using.

While you can certainly use genAI chatbots to continue developing your research, there are specific tools out there that can be used as a research assistant. A lot of them, however, do have membership costs.

NOTE: These tools do not have access to articles that are behind paywalls. In order to gain access to anything you find using these tools, you will need to find the article in the library catalog.

  • ResearchRabbit: A free citation-based literature mapping tool, showing links and relationships between authors and sources. Allows you create collections, categorize resources, and share collections.

While you can use the other genAI chatbots for understanding particular articles, I recommend creating a study bot with Google's NotebookLM.

Unlike large language models (LLMs) like Copilot, Gemini and ChatGPT, NotebookLM is a small language model, meaning it's source inputs are more closed off than LLMs which use a huge set of data to create outputs.

With NotebookLM, you can upload your own source material (e.g. PDFs, links to online content) and then ask the bot questions that help you understand the content better. It will draw upon the sources you gave it and make connections between the content based on the questions you ask.
Example questions:

  • "What is the main argument in article xyz?"
  • "How does concept a connect to concept b in articles 1,2&3?"

Prompting Tips

Follow RTF (Role, Task, Format)

Role: Tell the AI who it should act as

  • "You are a high school principal..."

Task: Clearly state want done

  • "...justify xyz decision"

Format: Specify how the output should look

  • "...provide a bulleted list with short explanations and citations for justifications"

Voice and Context

  • AI outputs are often generic and may not match your personal tone.
  • Without guidance, responses can sound too formal, too casual, or inconsistent.
  • GenAI does not know you or the exact parameters and context for your work.
Always evaluate and edit outputs, and never use the word-for-word output in your final product.