Science of Reading: The Podcast

S10 E12: Filling the gaps with inferences, with Kristen McMaster, Ph.D.

• Amplify Education • Season 10 • Episode 12

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 48:32

In this episode of Science of Reading: The Podcast, Susan Lambert is joined by Kristen McMaster, Ph.D., Guy Bond Chair in Reading and professor of special education in the Department of Educational Psychology at University of Minnesota. Together, they explore how reading comprehension isn't just about what's on the page—it's also about what's not there—and share practical insights on how to support students in developing inference skills. Susan and Kristen also discuss the dual processes of activation and integration when making inferences; the distinction between teaching students to process text actively versus teaching students to apply comprehension strategies; and different types of inferences, including causal, bridging, and elaborative.

Show notes:

Quotes:

"Inferencing is really central to comprehension. We wouldn't comprehend if we didn't make inferences." —Kristen McMaster

"I would encourage teachers not to underestimate the importance of supporting even the inferences that might seem obvious to us." —Kristen McMaster

"Good comprehenders are often making very automatic inferences that they don't even realize." —Kristen McMaster

"It helps to explicitly teach what an inference is in language that students will understand." —Kristen McMaster

Timestamps*:
00:00 Introduction: Filling in the gaps with inferences, with Kristen McMaster, Ph.D.
05:00 Comprehension is how we make sense of the world around us
09:00 The types of inferences: Causal, bridging, elaborative, and theory of mind
17:00 How teachers can help students develop inference skills
22:00 Creating an effective questioning strategy
27:00 How teachers can preview a text and think about the inferences that might need to be made
31:00 Supporting students who process texts in different ways
37:00 The timing of comprehension questions
40:00 The connection between oral language comprehension and text comprehension
45:00 Final thought: Teacher's shouldn't underestimate the importance of inferences that might seem obvious.
*Timestamps are approximate, rounded to nearest minute


[00:00:00] Kristen McMaster: I would just encourage teachers not to underestimate the importance of supporting even the inferences that might seem obvious to us.

[00:00:14] Susan Lambert: This is Susan Lambert, and welcome to Science of Reading: The Podcast from Amplify. We are now in the homestretch of this 10th season on comprehension. And as we head toward the end, today's episode is all about research-backed methods for improving students' comprehension abilities. My guest is Dr. Kristen McMaster, Guy Bond Chair in Reading and Professor of Special Education in the Department of Educational Psychology at University of Minnesota. She's also a former special education teacher. On this episode, she helps us learn how to support students through assessment and intervention. She breaks down different types of inferences, such as bridging inferences and elaborative inferences. She shares the kinds of questions educators can ask to support students' comprehension. Let's get right to it. Well, I'm so excited we have with us today Dr. Kristen McMaster. Kristen, thank you so much for joining us on today's episode! So excited about the content and the topic that we're going to dig into. But before we do that, I would love if you could take a little bit of time to introduce yourself to our audience, um, and tell us a little bit about yourself, maybe even how you got into literacy.

[00:01:33] Kristen McMaster: Sure, thank you so much for having me! I'm of course very excited to talk about this topic as well. Um, so I am, as you said, Kristen McMaster. I am a professor at the University of Minnesota in the Department of Educational Psychology, and my specific expertise is in special education and literacy, specifically supporting students through assessment and intervention and also their teachers in learning how they can support their students. I've been doing this work for many years. Couple of decades. Um, but before that I was a special education teacher, and I guess, like, how I became interested in literacy is I've always been an avid reader and writer myself, since I was a child. And, I think for a long time I took for granted the many ways that avid reading and writing contributed to my quality of life. But when I started in the field of special education, even when I was learning to be a teacher and doing my student teaching, and then of course when I was an actual teacher, I saw that many of my students really struggled, and had significant and persistent difficulties with reading and writing, and those difficulties had a huge impact on so many aspects of their lives. But I also saw, firsthand, that they could learn, and that that was a really powerful thing. The vast majority of students can learn, especially when they're taught using scientifically based instructional approaches that have been tested and we know can be effective. And then, when they do learn it opens so many doors for them for their success in school, and in later life, and their wellbeing. Just their enjoyment of things. And their ability to fully participate and contribute to society. So I just feel like supporting and advancing literacy, especially for kids who experience significant difficulties, is really an educational imperative. And that's what kind of motivates me, I guess.

[00:03:42] Susan Lambert: Motivates you every day, 'cause you're still in the research doing that work. And I love that you are interested in both. Helping students, but also helping teachers help students.

[00:03:54] Kristen McMaster: Mm-hmm.

[00:03:54] Susan Lambert: So that you have, sort of, that dual role. I'm sure that's been just very rewarding in terms of the work that you're doing.

[00:04:02] Kristen McMaster: Yeah, and I think that comes from I know what it was like to be a teacher, and to be sitting in front of a student, or a group of students, who are really having difficulty, and sometimes not knowing quite what to do. And so, it's really important to me to be able to provide teachers with tools that can help them get beyond that terrible feeling. So, yeah.

[00:04:26] Susan Lambert: Yeah. And we'll talk a little bit more about that, but, um, we brought you on because you're particularly interested in the area of comprehension, and this season is all about comprehension. And we've been asking our guests if they could provide their own definition of comprehension. So, I would love it if you could give us a definition of comprehension, sort of, from your point of view.

[00:04:50] Kristen McMaster: Well, so, very broadly speaking, comprehension is how we make sense of the world around us. So, we're constantly taking in information. We see things. We watch things. We hear things. We read things. And as that information comes in, we are constantly integrating it with what we already know and creating this sort of, in cognitive science we say coherent mental representation. Um, but it just basically means, sort of, like, this image or this picture in our minds of what we're learning, or what that information is adding to what we know. Um, and that just becomes a clearer and more meaningful representation of the world around us. And helps us understand. So, that's, kind of, a broad definition of comprehension overall.

[00:05:39] Susan Lambert: I like that, because it's interesting. I think this is the first time that anybody's answered that question with the broad view. It seems like we're always narrowed either into language comprehension or reading comprehension, but this really broad view I think is super helpful to remind us that, just like you said, we take in information from all different ways.

[00:06:00] Kristen McMaster: Mm-hmm.

[00:06:00] Susan Lambert: Sort of, to help us. And then we use that information in lots of different ways. So, that's great. I love that. One thing we do know, and we're gonna jump right into it, because we wanna talk a little bit about inferencing and the work that you've done around there. We know inferencing is really important to comprehension. We haven't talked a lot about inferencing on this podcast. Maybe some references to it here and there, but we've never really taken a deep dive into the concept of inference. So I wonder if you can, one more time, give us a definition.

[00:06:32] Kristen McMaster: Mm-hmm.

[00:06:33] Susan Lambert: Help our listeners really understand when we're talking about inferencing, what are we talking about?

[00:06:39] Kristen McMaster: Sure, yeah. And I'm glad you asked that question, because inferencing is really central to comprehension. We wouldn't comprehend if we didn't make inferences, and that's because a lot of the information that is coming at us all the time doesn't necessarily explicitly tell us everything. There's always some gaps that we need to fill in or, kind of, read between the lines, so to speak.

[00:07:03] Susan Lambert: Mm-hmm. Yeah.

[00:07:03] Kristen McMaster: And so, in order to fill in that missing information, we make inferences. And what that involves is both activating existing knowledge, or tapping into things we already know that's needed to fill in those gaps, but then also integrating it with that new information, to help us build that coherent mental picture that's gonna help us understand the information. And so it's both of those things, activating and integrating information together.

[00:07:36] Susan Lambert: Is there an example that you can provide of when we might need to make an inference? Putting you on the spot, I know.

[00:07:45] Kristen McMaster: No, um, no. So I'm just gonna use an example about plants. I'm not really a plant expert, but for some reason that comes to mind. So let's say you go to the garden store, and you know you have house plants, and so you know something about plants. Like, you should water them every day. So you're shopping in the garden store, and you see this tool that is a moisture tester for plants.

[00:08:14] Susan Lambert: Mm-hmm.

[00:08:15] Kristen McMaster: And you might not have read the fine print yet, but you might make the inference that, "Oh, this is a cool tool for my plants, because I can see how much moisture is in the soil and that will help me know how much to water it." So that, kind of adds, you already have this mental picture of plants needing water. And you know when they wilt, you need to add water. And now, you're adding this new thing like, "Oh, there's a tool that's gonna help me." So that just, kind of, expands your existing mental representation, and you made that inference that that's what that tool is for.

[00:08:50] Susan Lambert: Yeah. I love that. That's a really great example, because, again, sometimes when we talk about inferences, we go right to making inferences in print.

[00:08:59] Kristen McMaster: Mm-hmm.

[00:08:59] Susan Lambert: But it's a more general ... inferencing is more general than just doing it when we're reading text and in print.

[00:09:06] Kristen McMaster: Yeah. Exactly! Maybe we'll talk about this later, but that's really critical when we're thinking about young children, because they might not necessarily be fluently reading texts just yet, but they can still make inferences. They do make inferences, and they can learn to make them even before they're reading text fluently. So, that's why I have that broad approach, just because we know it's critical for reading, but we know we can start developing before the reading is actually happening.

[00:09:37] Susan Lambert: So, it's critical for thinking.

[00:09:39] Kristen McMaster: Yeah.

[00:09:39] Susan Lambert: Maybe that's a good way to say it, right?

[00:09:41] Kristen McMaster: Yeah. For thinking and for understanding.

[00:09:43] Susan Lambert: Yeah. Thinking and understanding. I love that.

[00:09:44] Kristen McMaster: Mm-hmm.

[00:09:45] Susan Lambert: Are there different types of inferences?

[00:09:49] Kristen McMaster: Yes, there are several different types. So, we make a lot of different kinds of inferences in a lot of different kinds of media that we encounter. So, if we're listening to a news story, watching a show, reading book, a lot of the kinds of inferences that we make are causal inferences. So, there are inferences that help us understand why. So, for example, I might say, "Sam really wanted a new bike. He saved up his money for a year." And if I just told you that, you probably made the inference that Sam is saving his money so he can buy a bike.

[00:10:27] Susan Lambert: Mm-hmm.

[00:10:28] Kristen McMaster: And so that's a, like, why did he save his money? So he could buy the bike. So, that's an example of a causal inference. That's a really simple one. Like, you probably didn't even realize you were making that inference.

[00:10:39] Susan Lambert: Yeah.

[00:10:39] Kristen McMaster: But sometimes —

[00:10:39] Susan Lambert: Kind of automatic, yeah?

[00:10:40] Kristen McMaster: Yeah. The inferences we have to make can be very simple like that, or they can be very complex, but it's often about like why something is happening.

[00:10:50] Susan Lambert: OK.

[00:10:50] Kristen McMaster: And getting back to the definition, I mean, so when I said that you were probably activating information that you already know, like bikes cost money.

[00:11:01] Susan Lambert: Yeah.

[00:11:01] Kristen McMaster: And then you were integrating it with what I told you about Sam, that he wants a bike and he's saving up. You integrated all of that to have this image in your mind of Sam.

[00:11:11] Susan Lambert: Yeah.

[00:11:12] Kristen McMaster: Saving up to buy his new bike.

[00:11:14] Susan Lambert: I like those two words. We're gonna maybe keep those, uh, the activation and integration. I know you've mentioned them two times. So, I love that, uh, you connected that example. That's great.

[00:11:24] Kristen McMaster: Yeah.

[00:11:25] Susan Lambert: So, that's a causal inference.

[00:11:26] Kristen McMaster: Yeah, so that's a causal inference, and that can happen in different ways. So sometimes we make inferences by connecting information that's all there in the text. We call those bridging inferences, because there might be one idea in the text that we have to connect with another idea in the text to understand what the text is saying. There might be a little bit of background knowledge that comes into that, but it's mostly based on what we see in the text. Other times we have to really rely on our background knowledge. So, I'm going to go back to my plant example. So, what if I said, "Sasha carefully watered her cactus every day. Sadly, her cactus died." Now, for that to make sense, I mean we just talked about how plants need water and we can check the soil moisture and so forth, but for that to make sense, you need to know something specific about cacti and that in dry conditions they tend to survive. But if we water them a lot, that's probably not gonna be a helpful thing. So, we call that an elaborative inference. Because we're connecting beyond the text to our own background knowledge that's gonna help us understand that specific information. And then there's other kinds of inferences, like something we call referential inferences where we're connecting pronouns to names or objects that come up. Whether it's in a text or other media. So, when I talked about Sam and his bike, I think I said, "He bought a new bike" instead of "Sam bought a new bike," 'cause it would be really boring if I just kept repeating the names over and over again. So, oftentimes, we use pronouns, but then we have to make an inference about who is he? He is Sam. And connect back to those individuals who are introduced. But that can be really challenging. You might have read books where there's a lot of different characters to keep track of, or there's different objects and things. And when the author is saying he or she or they or it or whatever, we have to sometimes very intentionally look back and be like, OK, who is he again?

[00:13:41] Susan Lambert: I just experienced that last night while I was reading a book. I literally had to trace back to figure out, "Wait a minute. Who is the "she" that they are talking about here?"

[00:13:49] Kristen McMaster: Yeah.

[00:13:49] Susan Lambert: So yeah, I can relate to that.

[00:13:50] Kristen McMaster: Right. There's also an a really interesting kind of inference that's related to Theory of Mind, which is basically where you're making an inference about what you think the character's beliefs, or their thoughts, or their desires are in order to understand an action that they do.

[00:14:12] Susan Lambert: Mm-hmm.

[00:14:13] Kristen McMaster: So, we were talking about Sasha and her cactus. If I said, "Sasha decided to get a pet rock," you might infer that she now believes she's not a very good plant caretaker based on her cactus experience. She desires to care for something. So, maybe a rock. A pet rock is a better idea. So there you're, sort of, putting yourself in Sasha's shoes. Thinking she probably feels bummed about her cactus. She still wants something to take care of. So maybe a rock is a good idea.

[00:14:45] Susan Lambert: That's a lot of different kinds or types of inferencing. And some of these are, like I said when you were using the plant example, some of them is, uh, automatic. We automatically infer that.

[00:14:57] Kristen McMaster: Mm-hmm.

[00:14:58] Susan Lambert: And sometimes it's not as automatic. Is that right?

[00:15:01] Kristen McMaster: Right. Sometimes we need to be more strategic. And so, just first of all, because there's both the activation and the integration components of inferencing, I think to be a good inferencer, you need to be aware that, maybe not using those exact words, but you need to be aware that you've gotta, sort of, tap into the relevant background knowledge, and then you need to connect it to the new information that's coming in. So, good comprehenders they're often making those very automatic inferences. Where they don't even realize.

[00:15:38] Susan Lambert: Yeah.

[00:15:38] Kristen McMaster: But they also, kind of, need to monitor their own comprehension. And when they realize that something's breaking down, kind of like when you were reading and it said, "she," and you had to look back and figure out who "she" was, you were monitoring your comprehension, and realizing, "Oh, I don't remember who that was. So I need to look back and make that connection." So, having that very deliberate strategic approach to helping yourself make the connection and integrate it with what you know about whoever "she" was that likely really supported your comprehension. Whereas if you were just like, "I don't know who "she" is. I'm just gonna keep reading," then you might not have gotten as much out of that text.

[00:16:18] Susan Lambert: Yeah, yeah, for sure. I noticed that I couldn't comprehend it, because I couldn't figure out who the "she" was in that whole situation. So, knowing this, is there a way that teachers can help students develop some of these inferencing skills?

[00:16:33] Kristen McMaster: Yeah. So, teachers can support inference making in lots of ways. First of all, it just helps to explicitly teach what an inference is in language that students will understand. So, we work with kindergartners a lot in our work. So, we very simply say an inference is when you connect something you see, hear, or read to something you already know to make a brand new idea.

[00:16:58] Susan Lambert: Oh, I love that. So connecting something you see. What did you say?

[00:17:02] Kristen McMaster: Something you see, hear, or read.

[00:17:05] Susan Lambert: Mm-hmm.

[00:17:05] Kristen McMaster: To something you already know to make a brand new idea.

[00:17:08] Susan Lambert: Oh, that's great.

[00:17:09] Kristen McMaster: Yeah, and kids get that.

[00:17:12] Susan Lambert: Yeah.

[00:17:12] Kristen McMaster: So teaching what it is, and then modeling it. And there's lots of opportunities to model and practice with kids, especially, you know, their teachers are often doing read alouds or showing videos to teach various kinds of content. So, if they're doing a shared book reading activity, or other kind of lesson, they can stop and ask questions that help activate and integrate the knowledge that's needed to make the inferences. And they can do that very deliberately, like, by thinking aloud, "Oh, you know, I need to make an inference here. What do I already know? How does that connect to what we just learned in the book? And how does that help us make a new idea?" And then another really powerful thing that teachers can do is ask questions that help activate and integrate the information. And oftentimes those are why questions or how questions, like, why didn't Sasha's cactus survive? Just getting kids to think about, "OK, the cactus died even though she gave it a lot of water. So it must be that water doesn't help the cactus. Well, what do I know about cacti?" So asking those questions. And then when a student doesn't quite get it, then providing some scaffolding and feedback to help them get there. So, they can scaffold by maybe pointing to information that needs to be connected to background knowledge. Or giving them background knowledge. Sometimes kids don't have the background knowledge, so that needs to be taught as well. But, we can support that with questioning and then, kind of, hearing where kids are coming from and giving them feedback and supporting that.

[00:19:01] Susan Lambert: We're gonna dive into this questioning thing just in a minute here, but what about assessing students inferencing ability? Is it possible to do that outside of, you know, some of the examples that you've already provided?

[00:19:15] Kristen McMaster: Yeah, I mean, so we have an inferencing assessment that is basically, because we work with young children, they're watching videos. And, periodically, where an inference is needed, we ask a causal question, usually one that is based on all the information that's within the video. So they need to make connections between pieces of information presented there. And if they can answer that, and this happens to be a multiple-choice approach, so they select an answer, but it could also be an open-ended questioning approach. So that can help us assess inferencing to see if they are actually making those connections. We can also do things like think alouds, where as students read we can ask them to say what they're thinking as they're reading.

[00:20:06] Susan Lambert: Oh yeah, yeah.

[00:20:07] Kristen McMaster: Um, because that can give insights into whether they're just, sort of, parroting back what they read or they're making connections to things they know or to the text. We can have them recall what they read. Because sometimes, when they recall what they read or what they heard read aloud to them, they will fill in gaps as they do the recall. But that can be a little challenging, because then kids are also needing to articulate what they —

[00:20:37] Susan Lambert: Right. Yeah.

[00:20:37] Kristen McMaster: Heard or read. And it can also be pretty time consuming, 'cause then you need to interpret what they said. So sometimes the more, sort of, multiple-choice type thing can be a little bit more efficient. There's also tests or assessments that ask kids to infer meanings of vocabulary. So, they might read a text that has unfamiliar words embedded in it, and then they are asked what those words mean and they have to use the context to infer the word meaning.

[00:21:08] Susan Lambert: I never thought of that being an inference.

[00:21:11] Kristen McMaster: Mm-hmm.

[00:21:11] Susan Lambert: I just learned something new. This makes me think of when I was in the classroom. So, I taught third grade for a long time. And there was, you know, focus on quality questions.

[00:21:24] Kristen McMaster: Mm-hmm.

[00:21:24] Susan Lambert: And I feel like teachers often use questions, right? Like, that's a great strategy. Use questions to determine how well students are comprehending. But, in focusing on thinking about the quality of the questioning, it's a lot harder to create effective questions or have an effective questioning strategy.

[00:21:46] Kristen McMaster: Mm-hmm.

[00:21:46] Susan Lambert: Then one might think. So —

[00:21:48] Kristen McMaster: Right.

[00:21:48] Susan Lambert: I remember, right, like, think about and plan. Look at your text or whatever. Think about and plan the kind of questions you might want to ask. And it's like, "Oh, OK, well yeah, maybe you should plan 'em ahead of time." But trying to think of those questions on the fly is difficult, isn't it?

[00:22:05] Kristen McMaster: Yes. Yes, it can be. For sure.

[00:22:08] Susan Lambert: So, what are some ways, and, first of all, what have you learned, because I know you've done a lot of work here, what have you learned about effective and maybe ineffective questioning strategies?

[00:22:19] Kristen McMaster: Yeah, I think when questioning is not planful, like you were saying, then it can be less effective. And I think a lot of times, we talked about how a lot of inferences we kind of make automatically, and I think, sometimes, if teachers haven't been planful, they don't necessarily think about the inferences that are needed that seem really obvious to us but that might be really challenging for a student who's experiencing difficulty. So, I think, first of all, just, sort of, having teachers understand that some inferences are really tiny, little inferences that we don't really think of as inferences.

[00:23:03] Susan Lambert: Right, yeah.

[00:23:03] Kristen McMaster: But some kids might not be making those. And so just, kind of, raising awareness and sensitivity to that can be helpful. And then knowing those different types of inferences that we talked about. So, like, the pronoun-type inferences that kids need to make. The more teachers are aware of those, they can be more sensitive as they're going through a text. And maybe, if it's on the fly realizing, "Oh, there were just a lot of "he's," "she's," and "it's," let's make sure everyone knows and is tracking who these different people are." Or if they have time to go through the text ahead of time, it's certainly helpful to do that. Because then you can identify those places where there might be gaps that need some support in filling them. So, both the sort of planful approach, where we're, kind of, looking through a text and identifying places where information isn't explicitly stated. Especially, like, sometimes you can read a text and not make every single inference that's possible and still get the gist. So, it's also helpful to, kind of, know what that main thread is, that's, kind of, the storyline. That there's lots of causal connections to that main thing. And then there's some, sort of, tangential things that maybe aren't as important. And really focusing on the things that are related to the main theme, or the thing that we want students to take away from that text. But I think just asking questions along the way. Making them those kind of why questions or how questions, that get at the causal nature of the text. Those are probably gonna be good, high-quality questions.

[00:24:49] Susan Lambert: Yeah. It reminds me of, I heard an example, and I can't remember the exact two sentences. I use it sometimes in presentations, and you'd think I'd remember it, but basically the gist of it was Carla forgot her umbrella, and when she got to school, she was wet. Right?

[00:25:07] Kristen McMaster: Mm-hmm.

[00:25:07] Susan Lambert: And so, well, why did Carla get wet, right? And there's sort of two levels of that. Well, because she forgot her umbrella. Which is right there in the text. Or, well, because it was raining, right? So that's the sort of deeper inference there. And when somebody presented that to me as an example of how you have to fill in the gaps, or how you have to make an inference, this was post being a third grade teacher, somebody provided me that example. I'm like, "Oh man, I don't think as a third grade teacher I would've understood that that took an inference, because for me the inference was automatic." And I knew what an umbrella was. And I knew like when you're walking and you get wet and you don't have an umbrella, it's because it's raining. But there are kids that may be English language learners that don't know what an umbrella is, or for other reasons they don't make that connection. That's a really simple example, and, to me, that example was a huge "Aha!" in terms of the importance of simple, automatic inferences that we make as adults that children might need scaffolding and supporting in.

[00:26:22] Kristen McMaster: Exactly.

[00:26:23] Susan Lambert: Which is a really long way of me saying, how in the world as a teacher can I go about looking at a text and thinking about this inferencing that might need to happen? I mean, are there ways that you can tell teachers, "Here's some things that you can do when you're previewing a text to help you develop these really great questions?"

[00:26:45] Kristen McMaster: Well, the wonderful thing about teachers is they also know their students a lot, you know, a lot of the time. And so, they might have some insight into what might be challenging for students, and where they might need to provide extra support. But, if they're new to their students, they can still think through, like, what might be not immediately obvious to a group of third graders, especially if I know I have multilingual learners in my classroom, some of whom are just learning to speak English or maybe have a different fund of knowledge depending on their cultural background. And so, what are those things that they might not know? And sometimes it's thinking about vocabulary. So, because vocabulary, the meaning of words, is so central to us then inferencing, like you use the umbrella example, so identifying a handful of words, you probably wanna keep it —

[00:27:51] Susan Lambert: Limited, yeah.

[00:27:52] Kristen McMaster: Yeah. But three or four words that are really central to the main ideas of the text that students will need to understand to be able to comprehend the text. So pre-teaching those can be really helpful, because that's gonna create space to ask questions that will make sense to students. Because they'll understand what the words are. And so, that's one thing teachers can do is just identify what are those words that might be challenging? But then, also, what are the main ideas of the text? What are the sentences, or the parts of the text, that really align closely to those main ideas? Because that's where you're probably gonna wanna focus your questions. I mean, you might want to also make sure kids are picking up on the details. To make sure they're getting, sort of, the causal chain of the text, right? Or like that thread that's gonna be really important for comprehension. Thinking about that and, kind of, going through and seeing where those gaps are. And that might be where there's really a gap asking a question. I like simple examples. And one that I often use with teachers, 'cause they're familiar if they're early elementary, is "If you give a pig a pancake, she'll want some syrup to go with it."

[00:29:15] Susan Lambert: Right. I love that book!

[00:29:17] Kristen McMaster: I know, me too! It's one of my favorites. So, why would she want some syrup to go with it? It seems really obvious, right?

[00:29:23] Susan Lambert: Right.

[00:29:24] Kristen McMaster: But if I have students in my class who come from cultures where they eat pancakes with savory meat and vegetables, that might not make as much sense. So, that might be a place where if the teacher thinks that's important to their comprehension of that text, we might wanna ask a little question about that.

[00:29:42] Susan Lambert: Yeah.

[00:29:42] Kristen McMaster: Sometimes we put syrup on pancakes, because it makes 'em taste yummy. And it's a sweet treat that we often enjoy. But then, you know, of course there are much more complex examples and just being really attuned to that and identifying those gaps ahead of time.

[00:30:00] Susan Lambert: Yeah.

[00:30:00] Kristen McMaster: And being prepared to ask questions. The other thing is, so, kids aren't all the same. Um, we know that students process texts in different ways. So, sometimes we've seen students who tend to just, kind of, paraphrase what they've read. So, if you stop to ask them what they're thinking about as they're reading, or you ask them a question, they might just repeat the last sentence that they read. Or just give a paraphrase that doesn't really extend beyond what the text is saying. So, they don't really make the inference. So, in that case, they might really need some prompting to make a connection. Not just to restate the text, but connect it to something they already know. Other students might be connecting to what they already know, but what they already know might not necessarily be completely relevant. As a third grade teacher, you're probably familiar with —

[00:30:58] Susan Lambert: Oh yes.

[00:30:58] Kristen McMaster: The student who reads the text and then tells you a little story about something.

[00:31:03] Susan Lambert: Yes.

[00:31:03] Kristen McMaster: Rather unrelated.

[00:31:05] Susan Lambert: That's right.

[00:31:05] Kristen McMaster: So in that case, you might need to bring them back with a very specific question. Why did this happen in this text? What do we know in this text about whatever we just read that can help us understand, or help us make an inference? So, kind of, hearing what students say and then tailoring our response to the way that they're thinking about the text is also quite important.

[00:31:31] Susan Lambert: Yeah, that makes sense. So, would you say that our ultimate goal then is to create causal comprehenders?

[00:31:40] Kristen McMaster: I think that's an excellent goal. Again, there are different kinds of comprehension questions. You know, we also wanna create critical thinkers.

[00:31:51] Susan Lambert: Yeah.

[00:31:51] Kristen McMaster: So, that's where maybe some of, well, the causal questions definitely help with that, because we don't necessarily just wanna take what the text says at face value and not go beyond that. But also the Theory of Mind pieces, especially as kids get older. Kind of, understanding what's motivating characters, and why that would lead to certain behaviors. Those kinds of inferences are also important. But causal questions, if I was going to focus on any particular kind of inference, I think those are essential.

[00:32:27] Susan Lambert: You know, for our listeners, just an aside here, is that one thing that I like to do in our pre-call with our guests, so I did this with you, Kristen, is I say, "Please send me anything you would like me to read to prepare for our conversation." And it's my favorite part is to get introduced to new ideas, and do this reading. And so, I did my homework and I did some reading when you sent it to me.

[00:32:50] Kristen McMaster: Mm-hmm.

[00:32:50] Susan Lambert: And something jumped out at me when I was reading one of the articles, which is that there is some evidence that helping students with text processing and, help me if I'm getting this right, is actually more effective than comprehension strategies.

[00:33:13] Kristen McMaster: Yeah, I think it's an important thing to keep in mind. I mean, there are definitely strategies that are also helpful. Sometimes strategies can get a little bit in the way, because if we're so focused on a strategy, it can, kind of, distract from the actual content.

[00:33:33] Susan Lambert: Oh, from the content of the text?

[00:33:35] Kristen McMaster: Yeah.

[00:33:35] Susan Lambert: Yeah, yeah.

[00:33:36] Kristen McMaster: I mean, there's a lot of great strategies that can be really helpful, but especially if I'm not really fluent to using those strategies, I might get, sort of, bogged down and, like, "What were the steps I was supposed to do?"

[00:33:48] Susan Lambert: Right. Right. Yeah.

[00:33:49] Kristen McMaster: And also, I mean, when we're reading skillfully, sometimes we're using strategies and, sometimes we're, most of the time we're just processing the text. We use the strategies when that processing kind of breaks down, and we notice.

[00:34:05] Susan Lambert: Yeah.

[00:34:05] Kristen McMaster: And so I think both approaches are needed. And sometimes, like teaching kids by using the process, like inferencing, to start with can be helpful. And then once they've been doing it for a while, and we don't want them to rely on the teacher all the time to be asking questions and so forth, we can make it a strategy. So, after doing this for a while, we could be like, "Did you notice," like whenever we pause and ask a question, like a why question, you're making connections between what you're reading and what you know. And you can do that anytime. Not just when I, the teacher, asks you questions, or you're prompted in your app to —

[00:34:51] Susan Lambert: Mm-hmm.

[00:34:51] Kristen McMaster: You know, answer a question or something. But also, when you're reading, if you find that you're not comprehending things, you can ask yourself a question. Why did X happen? Or who is she? Those kinds of questions that help you understand. So, it can become a strategy that then students can use when they notice that they're not understanding. So, I think it's a both-and kind of thing.

[00:35:18] Susan Lambert: Yeah, that makes sense. And just to be clear, like text processing, if we were gonna define it really quickly, that would, sort of, be these things that are pointing students back to the text when you're asking why questions. Or when we're trying to connect pronouns between, right, like, we're actually trying to process this text. And we're doing these things to help us make sense.

[00:35:40] Kristen McMaster: Mm-hmm.

[00:35:41] Susan Lambert: Of what we're reading, right?

[00:35:41] Kristen McMaster: Exactly. Yeah.

[00:35:42] Susan Lambert: Yeah. That's really interesting. I never thought about that idea that these text-processing ideas, or asking questions about the text when you employ them, they're actually the strategy. And the strategy is really a questioning strategy, right?

[00:35:57] Kristen McMaster: Mm-hmm. Mm-hmm.

[00:35:58] Susan Lambert: So all of this is starting to come together. That's great. I did wanna ask you about the timing of questioning. Does that impact comprehension? And by timing, I mean, you know, often we talk about we do these things before we start to read. And then during reading we're doing this. And then after reading, we're doing this.

[00:36:16] Kristen McMaster: Well, it depends on the goal. And so, theoretically, and we have empirical evidence to support this, it can be much more helpful to ask questions DURING text if you want to influence that mental picture that the child is building. Because if you wait until after they've read the text, they've already built that mental picture representation. And it may or may not be quite what you were hoping they would build. And it's harder to change once they have that. Or they might just not have anything there. But if you ask questions DURING reading, that should be helping them build that mental picture along the way. And they, kind of, need to be doing that as they read to make sense of the text as a whole.

[00:37:07] Susan Lambert: Yeah.

[00:37:07] Kristen McMaster: And so, asking questions during reading can be really helpful and supportive. Especially if, as that's happening, you're giving feedback and scaffolding when it's needed to support coming to the right answers. And sometimes there's more than one right way to understand a text. But, successful inferences. I think that the danger with that is either if the questions aren't high quality, so making sure those are good questions that are along that, sort of, causal thread that we want students to be following, and also there's probably a point of either diminishing returns or even distracting if we're asking so many questions. And it's sort of disrupting that comprehension process. So, I think sometimes we just need to be sensitive to not asking so many questions that there's more questions than actual text.

[00:38:06] Susan Lambert: Reading happening.

[00:38:07] Kristen McMaster: But, other times, we might ask questions afterwards if we want to know what the child took away. Or if you want to dive more deeply into themes, or think critically about the text, or relate it to other texts that they've read. Sometimes, then, waiting until after reading is gonna be more appropriate. So, now they've built their mental representation and are going to dive in even more deeply to think about the concepts that they were reading about.

[00:38:39] Susan Lambert: That makes me think of a lot of the guests that we've had on this season have talked about the differences in the comprehension process, or how we're comprehending while we're reading, which is, I think what you're saying. That those questioning that's happening during that unfolding process of comprehension can help support the product of comprehension, which is that mental model, right?

[00:39:03] Kristen McMaster: Right. Exactly.

[00:39:04] Susan Lambert: Yeah. And a really "Aha!" we had over the course of this season, too, was that constructing a poor mental model, or a weak mental model, or a mental model that is wrong, a misconception, is more difficult to correct. So that —

[00:39:21] Kristen McMaster: Yes.

[00:39:21] Susan Lambert: Like this just tracks with some of the things we've been talking about this season. That's great.

[00:39:26] Kristen McMaster: Mm-hmm. Yeah.

[00:39:27] Susan Lambert: What about the connection. Now, I know you're more in the early literacy space, but this probably is relevant all the way up to even me. We know there's a connection between oral language comprehension and text comprehension. How does that relate to what we've been talking about?

[00:39:47] Kristen McMaster: Yeah. Well, and I think because I've kind of taken this broad approach. Um.

[00:39:53] Susan Lambert: Yeah.

[00:39:53] Kristen McMaster: It's all about language. For me, it's a little hard to separate, but oral language, I mean, a lot of what goes into that is understanding the meaning of words. You know, even if you're not reading, we're talking with each other. We need to understand the words. Or if you're watching a show, or even, you know, in sign language, you need to understand the language, the words, the sentence structure, like the syntax, the semantics. Also, understanding genre, different kinds of text, or discourse more broadly speaking. All of those things are language comprehension, and we need to be able to comprehend in a variety of different media, but certainly text. We need to do all of those things to comprehend text. So, language comprehension is necessary for text comprehension.

[00:40:46] Susan Lambert: Yeah.

[00:40:46] Kristen McMaster: It's not sufficient though, because text comprehension involves reading words, and we need to be able to decode those words and recognize them, eventually automatically, so we're not stopping at every word, which is also taking up cognitive resources. And so we need both. Sort of, the word recognition, which, if you unpack that, there's a lot of components there. We need the language comprehension, which, also, if you unpack that, there's a lot of components. But together, those two things are needed for reading comprehension to happen. Without one or the other, you wouldn't comprehend what you read.

[00:41:25] Susan Lambert: What a great review of the Simple View of Reading.

[00:41:27] Kristen McMaster: That is the Simple View of Reading. Yep. Which is a little too simple, a lot of people might say, but kind of at least it gives some building blocks to think about.

[00:41:37] Susan Lambert: Yeah, yeah, for sure. We had Dr. Wesley Hoover on and he explained it that yes, it's a Simple View of Reading, but all we've done is capture the complexity into two simple concepts.

[00:41:48] Kristen McMaster: Exactly. Yeah.

[00:41:48] Susan Lambert: We definitely know the complexity. And, especially as we start unpacking this whole broad view of comprehension. Um, it's just amazing to me how much I missed when I was teaching third grade. And how much better a teacher I could have been had I understood more of these concepts. So we appreciate this. I wanna go back before we close up and just highlight these two words, which I think I'm going to remember forever about inferencing, this idea of activation and integration. And would you say that those are two words that we can think about when we're thinking about this process of inferencing?

[00:42:30] Kristen McMaster: Yes, absolutely. The other piece is of course, having the background knowledge to activate and integrate. That is essential, but the process of using that information, I think for me, that idea of those two components makes a lot of sense, and is what's happening.

[00:42:50] Susan Lambert: You know, we've already said we know literacy is very complex, but, sort of, distilling it down in ways that you can, sort of, hold and for teachers ways that you can hold that and apply it over and over and over again. What am I doing to help kids both activate and integrate? And just like the Simple View of Reading, you can't have good inferencing without both of those things.

[00:43:15] Kristen McMaster: Mm-hmm. Yeah, I would agree. And I would also say, you know, sometimes in activating, you also need to help build that background knowledge. But yes, yeah. And I would credit, I mean, I was a teacher who at the time didn't know these things. So like you, you know, if I could go back, I feel like I would do a better job, or I hope I would. And I've learned a lot of that from my colleague Pani Kendeou, who's a cognitive scientist, and who I work with very closely, and we've, kind of, brought together her theoretical background and her deep knowledge about just how people learn to read with my experience in developing instruction and interventions. And I think that's been what's made them work well, um, is that they do have that really strong theoretical underpinning. And then we've just tried really hard to translate it into things teachers can do.

[00:44:12] Susan Lambert: Hmm. Yeah. That's awesome. So, as we wrap up, do you have any closing thoughts or advice for our listeners, particularly those teachers out there that are trying to get a handle on helping their students be better at inferencing?

[00:44:29] Kristen McMaster: Yeah, I mean, I guess I would just encourage teachers not to underestimate, you know, the importance of supporting even the inferences that might seem obvious to us as good readers. That we have to, sort of, meet students where they are and listen really closely to how they're responding to text. Um, whether it's through answering questions or other ways, and making sure that we are supporting those really basic pieces. And we can do that in explicit ways. We can do that with modeling, and practice, and lots of feedback, and support. And kids will really benefit from that.

[00:45:12] Susan Lambert: Well, thank you so much for joining us. This is such an important topic. Thank you too for the work that you continue to do, um, in this world, both understanding this and then helping to support teachers. So, it's been such a pleasure to have you on today, Kristin. Thank you so much.

[00:45:26] Kristen McMaster: Well, thank you so much! And thank you for doing this work. And it's wonderful that you're diving so deeply into comprehension and sharing that with your listeners.

[00:45:35] Susan Lambert: I learn every single day, and I know our listeners do too, so thank you again for joining. That was Dr. Kristen McMaster, Guy Bond Chair in Reading and Professor of Special Education in the Department of Educational Psychology at the University of Minnesota. Check out the show notes for some links to her website and some of the research we discussed. Next time, we're doing something very special. I'm going to be sharing some of my biggest learnings from this comprehension-focused season of the podcast. And then, we'll be closing out Season 10 two weeks after that with a listener mailbag episode.

[00:46:16] Ben Rogers: Hi Susan. My name is Ben Rogers. I'm a physics teacher from the UK. I've got a question for you on comprehension. When my students are reading a text ...

[00:46:25] Susan Lambert: That's all coming up on Science of Reading: The Podcast. Meanwhile, our sister podcast, Beyond My Years just wrapped up its second season with a conversation about the research on efficient and effective change.

[00:46:39] Mike Flynn: The thing is that leaders, when I say leaders, I don't mean principals, I'm talking about anyone that has influence. One thing that you can do is orchestrate experiences to disrupt unproductive beliefs.

[00:46:51] Susan Lambert: Listen today on Beyond My Years, available wherever you get this show. There's also a link in the show notes. Science of Reading: The Podcast is brought to you by Amplify. I'm Susan Lambert. Thank you so much for listening.