In today’s episode, we'll review strategies for integrating artificial intelligence (AI) into the reading process. Visit AVID Open Access to learn more.
In today’s episode, we'll review strategies for integrating artificial intelligence (AI) into the reading process. Visit AVID Open Access to learn more.
#269 — WICOR and AI: Reading
Keywords
ai, students, reading, text, lottery, strategies, story, tools, engage, process, learning, response, readers, prompt, questions, generative, content, context, provide, comprehension
Speakers
Paul (98%), Transition (2%)
Paul Beckermann 0:01
Welcome to Tech Talk for Teachers. I'm your host, Paul Beckermann.
Transition Music 0:06
Check it out. Check it out. Check it out. What's in the toolkit? What is in the toolkit? So, what's in the toolkit? Check it out.
Paul Beckermann 0:17
The topic of today's episode is WICOR and AI: Reading. This week, I'll be exploring how AI can be used to help students with the fifth and final component of AVID's WICOR model, reading. To clarify, I'm not going to be sharing ideas for helping students learn to read. While some of the strategies will likely help struggling readers, the focus will be on strengthening students' ability to learn from the text that they are reading. In other words, this will be less about learning to read and more about reading to learn. I'll be approaching this through the lens that students gain meaning, understanding, and knowledge through the process of reading. And part of that process is making connections between texts, self, and the world. In this episode, I'm also going to focus mostly on the use of generative AI tools like ChatGPT, Google Bard, Microsoft Co-Pilot and Anthropic's Claude. You can choose whichever one you're most comfortable with. For younger students who aren't allowed to use these tools, the teacher can model and lead the full class or a small group of students through the experiences soliciting questions and ideas from them and entering those ideas into the AI tool for them. In this way, the students can participate by providing the input, observing their responses, and engaging in the discussions about those results. They can even prompt follow-up questions. They just won't be the one actually doing the typing. If you're teaching older students who can use these tools, you might still consider modeling at first, but eventually, students can engage with the AI themselves. So let's get started.
Transition Music 1:56
What are some cool tech tips I can use? instructional technology.
Paul Beckermann 2:04
Number one, the pre-reading portion of the reading process. Generative AI can be used to prepare students and set them up for successful reading experiences. Ideally, the students will have access to a digital version of the text they're reading. This will allow them to copy and paste it into the AI, allowing the AI tool to know more specifically what is being referenced and discussed. If it's a famous piece of writing, like Martin Luther King Junior's I Have a Dream speech or a play by Shakespeare, the AI will likely be able to gather quite a bit of context from databases. But for less well-known texts, it will be important to be able to feed it into the AI. For my example, I'm going to use the short story The Lottery by Shirley Jackson.
One way I could use AI in getting ready to read the story would be to gain some historical context. That type of context will often help me to better understand what I'm reading. In my example, I asked ChatGPT: What historical context should I be aware of before reading The Lottery? ChatGPT returned a list of seven helpful contextual points that a student would likely not be aware of. Some of these included a description of the time period that the story was written, the political and cultural climate at the time, and a brief note about the controversy that surrounded the story when it was released. These points piqued my curiosity and were helpful in setting up a better context of the story that I was going to read. Another pre-reading strategy I could apply as a student would be to identify key vocabulary that might be challenging yet critical to understanding the text. For my example, I posted in the text of the story, and asked ChatGPT: "What are some key vocabulary words from this excerpt of Shirley Jackson's story that a reader must understand?" The AI returned a list of 11 terms that were key to the text, as well as brief definitions of the terms in the context of the story. I didn't mind getting definitions provided to me, but if you didn't want your students to be given those and you wanted them to find them themselves, you could instruct AI to simply provide the terms and not the definitions. I thought the AI did a really good job with the words it chose from the passage. It defined lottery in the context of the story, and words like civic and tradition, which are central to the story's theme.
One more pre-reading strategy that you could use, especially with a nonfiction text, is to paste in the text from the article or document and then ask ChatGPT, or your generative AI tool of choice, to provide a brief summary. You can even tell it to limit the summary to three sentences to make sure that it's short. This initial summary can provide a valuable overview and preview of the content to be read and improve comprehension. You're priming the pump. It's a little bit like scanning an article ahead of time and getting that general idea of what is to come, which can be really helpful, especially if the text is complex or technical. If you're reading as part of a research process, it can even inform whether or not you should continue to read that article. The brief summary might quickly let you know if it's even relevant to your research.
Number two, during reading. So, readers can also use generative AI is a reading partner during the actual reading of the text. This might take a little bit of practice as well as some trial and error in order to get the most beneficial results. But here's some ways that I used AI to help me interact with The Lottery. One basic way to use AI to coach me through the reading process was to ask it for a list of questions related to the content of the text. Now, I want to say right away that I'm a big fan of having students generate their own questions, but there is a place for having the AI assist with this as well, especially as a scaffold for students who are struggling in this area, or as a way to save time. You might even have students paste a list of their own questions in first, and then ask the AI to identify key questions that they might have missed. For instance, here's a prompt that they might consider: "These are the questions that I have about the text. What am I missing? What else should I be asking in order to better understand this text?" This process allows a student to generate their own questions first, and then prompt the AI to provide feedback on their work. Students could even ask the AI to evaluate the quality of the questions they submitted and perhaps identify the best ones. In a similar fashion, students could submit their thoughts on some aspects of the text and ask the AI for feedback on that. For instance, they could submit their understanding of the main point being made in the text and then ask the AI: "Do you agree with me? Why or why not?" Or again, "What am I missing?" Or, "What are my blind spots?" Or, "What is the perspective somebody else might have?"
One more strategy that I really like is to offer AI my own thoughts or interpretations about the text, and then direct the AI to ask me a follow-up question to continue the conversation and help me probe deeper into the meaning of the text or into my own thoughts. In this way, the AI is not doing all the work for me. Instead, it's helping me think more deeply and process the text. In my conversation about The Lottery, here's what I entered: "I wonder if the students know why they are gathering stones. Since children are often berated for throwing stones, it might feel ominous that they're gathering them. What do you think? Please ask me a follow-up question to keep the conversation going." So this was in response to the first scene in The Lottery, where students are piling up stones. As a response, ChatGPT returned, "That's an insightful observation about the students gathering stones in The Lottery. It indeed presents an ominous undercurrent, especially given the normal rules of behavior of children. Considering this, how do you interpret the adults' reaction, or lack thereof, to this activity? Does their response or absence of one give you any clues about the nature of the lottery, and the community's attitude toward the tradition?" These are great questions. And this engagement with the AI could lead to some really valuable deep thinking and processing on my part as the reader.
Number three, let's move on to post-reading. So once the student has reached the end of the text, some of the same strategies that were used during the reading can also be applied here. You could ask the AI for a list of probing questions to help you consider the deeper meaning of the text. Then, one at a time, you could submit a response to the question and either ask the AI to pose a follow-up question, or identify potential gaps or flaws in your logic. For some students, it might actually be safer to begin this type of interaction with a computer algorithm rather than a peer. Because the computer might feel less judgmental or personal to them. It's still really important that students graduate to peer-to-peer interactions at some point, but AI could be used as a scaffold.
Another strategy that can be really engaging is to prompt the AI to have a debate with you about a key point or controversial aspect of the text. To do this, it's important to tell the AI that you wish to engage in a friendly debate about the text. You should tell the AI which position it will take, what your position is, and then instruct it to end each response by asking a follow-up question to help continue the conversation. Students can engage in a similar give-and-take process with educational platforms like Khanmigo would use generative AI to allow students to have virtual conversations with authors and historical figures. This can be a really great way for students to internalize the context they've read and take it to the next level. And it's fun. All right, fourth, a couple bonus strategies that can improve accessibility. For readers who struggle decoding words, text-to-speech tools like Speechify can unlock the text and allow students to access the content. And this can be an important life skill for struggling readers, as well. Generative AI can often translate content into a student's primary language. If the content, and not the act of reading and English is the goal, this can help students get to the content and not be left behind in the learning process. Finally, AI can rewrite text at different reading levels. I've regularly seen adults take a complex piece of text and ask AI to explain it to them as if they were a fifth grader. So if it's good for adults, it should be good for students too, right? This is one way to make sure all students have access to the academic concepts in the text, and help them engage in the learning and conversations around that content.
As with anything AI, the landscape and capabilities are changing rapidly. By the time this podcast episode releases, there will very likely be new tools and capabilities available. That's okay. And it's a good reminder that we can never stop learning and growing as educators. It's also a reminder that we need to continually engage with the evolving world of AI and ask ourselves if there are ways to leverage these tools to help our students engage meaningfully in the learning process, and in the case of today's topic, reading. If we use it well, generative AI tools can become reading partners or comprehension coaches. Part of the key will be to help students generate the right prompts. And that's an important life skill in itself. Of course, these strategies are not the end-all-be-all to reading comprehension and learning. I like to think of them as additional strategies in our teaching toolkit. The more tools we have, the more likely we'll be able to find the right approach for each of our students and their unique learning needs.
To learn more about today's topic and explore other free resources, visit AvidOpenAccess.org. Specifically, I encourage you to check out the collection of articles about AI. You can find it by going to AvidOpenAccess.org and searching for AI in the K-12 classroom. And of course, be sure to join Rena, Winston, and me every Wednesday for our full-length podcast, Unpacking Education, where we're joined by exceptional guests and explore education topics that are important to you. Thanks for listening. Take care, and thanks for all you do. You make a difference.