Why Distance Learning?

In this episode of Why Distance Learning, your hosts talk with Kristen DeBruler — Assistant Director of the Michigan Virtual Learning Research Institute — about what makes online learning work for K-12 students, and what quietly undermines it. Drawing on survey data from over a thousand virtual teachers and fresh findings on student and teacher AI perceptions, her work reveals where the conventional promises of online learning diverge from what the data actually shows. The episode centers on a tension that practitioners rarely name directly: the flexibility that makes online learning valuable for some students can work against those who are still developing the self-regulation skills to use it well.

Together, the hosts and Kristen explore how a researcher inside a state virtual learning organization sets and pursues a research agenda — including the unusual advantage of having direct access to student data. They examine common failure modes in classroom-level research, particularly how vague research questions leave teachers vulnerable to the biases they're trying to surface. Kristen walks through her team's findings on the gap between how students and teachers define acceptable AI use, and why that ambiguity is already showing up in the classroom. The conversation turns to teacher feedback as the connective tissue of asynchronous online learning — and what the data shows about what makes it work and what makes it hollow. And Kristen makes a pointed case about applying adult learning research to K-12 populations: the data doesn't transfer as cleanly as the field sometimes assumes, and the consequences land on students who are still building the executive functioning skills that adult learners already have. Her pacing research is illustrative: students who cross unit boundaries — not just move around within one — end up with final grades 9.5 points lower on average, a gap large enough that teachers should treat it as a warning signal, not background noise.

Key topics:
  • Researcher role inside a state virtual learning organization
  • Setting a research agenda: legislative directives vs. internal needs
  • Avoiding bias in classroom-level research
  • The AI acceptable-use gap between students and teachers
  • Teacher feedback as the primary relationship-building mechanism in async courses
  • What makes feedback substantive (personal, formative, actionable) vs. hollow
  • AI-generated feedback and trust erosion in online learning
  • Student pacing deviation and its effect on learning outcomes
  • Executive functioning support for K-12 online learners
  • Cautions in applying adult learning research to adolescents

Links & Resources

Guest Bio: Kristen DeBruler
Kristen DeBruler is the Assistant Director of the Michigan Virtual Learning Research Institute, where she has spent more than 14 years studying K-12 online learning — including student pacing, teacher communication, mentor support, special populations, and AI use in virtual environments. Her research is oriented toward practitioners: she publishes findings in formats designed for teachers, administrators, and program leaders to act on, not just cite. She holds a PhD in Educational Psychology and Educational Technology from Michigan State University.

About the Hosts
Seth Fleischauer is the founder of Banyan Global Learning and host of Why Distance Learning. Through Banyan, he designs live virtual programs that connect K-12 classrooms to global peers and expert facilitators — building the kind of structured, human-centered distance learning the podcast explores. See https://banyangloballearning.com/
Tami Moehring and Allyson Mitchell work with CILC, the Center for Interactive Learning and Collaboration, to help educators implement high-quality live virtual learning experiences across grade levels. Discover more at CILC.org.

Creators and Guests

Host
Allyson Mitchell
SF
Host
Seth Fleischauer
TM
Host
Tami Moehring

What is Why Distance Learning??

Why Distance Learning? is a podcast about the decisions, design choices, and assumptions that determine whether live virtual learning becomes shallow and transactional—or meaningful, relational, and effective at scale.

The show is designed for education leaders, instructional designers, and system-level practitioners responsible for adopting, scaling, and sustaining virtual, hybrid, and online learning models. Each episode examines the structural conditions under which distance learning actually works—and the predictable reasons it fails when it doesn’t.

Through conversations with researchers, experienced practitioners, and field-shaping leaders, Why Distance Learning? translates research, field evidence, and lived experience into decision-relevant insight. Episodes surface real tradeoffs, near-failures, and hard-won lessons, equipping listeners with clear framing and language they can use to explain, defend, or redesign distance learning models in real organizational contexts.

Hosted by Seth Fleischauer of Banyan Global Learning, and Allyson Mitchell and Tami Moehring of the Center for Interactive Learning and Collaboration, the podcast challenges outdated narratives about distance learning and explores what becomes possible when live virtual education is designed intentionally, human-centered, and grounded in evidence.

# Clean Transcript: Kristen DeBruler — Michigan Virtual
*Why Distance Learning*
*Edited: 2026-06-01*
*Note: All timestamps after the first cut are approximate (marked with *approx*). Adjusted for ~10s removed at 03:19, ~36s at 28:44, and minor trims at 29:19 and 30:44.*

---

**[INTRO — recorded separately]**
**[00:00:00]**

**Seth:** Hello, and welcome to Why Distance Learning, the podcast for education leaders and practitioners who are making real decisions about how virtual learning gets designed, adopted, and sustained. I'm Seth Fleischauer, founder and president of Banyan Global Learning, and my cohosts are Allyson Mitchell and Tami Moerhing of the Center for Interactive Learning and Collaboration. Every episode we try to surface at least one assumption worth questioning.

Today's guest is Kristen DeBruler, a researcher embedded inside Michigan Virtual — the state's virtual learning organization — where she's spent 14 years studying what actually happens when students and teachers navigate online learning. Her team just published findings on how students and teachers understand acceptable AI use, and what the gap between those two groups looks like in practice.

The assumption on the table today: that giving K-12 students flexibility in pacing is straightforwardly a feature. Kristen's research suggests that for adolescents still developing executive functioning and self-regulation skills, flexibility without structured support isn't a benefit — it's a risk that shows up directly in learning outcomes.

*This episode is brought to you by CILC, the Center for Interactive Learning and Collaboration, connecting students to real experts through live virtual field trips and experiences — visit cilc.org to learn more. And by Banyan Global Learning, which brings K-12 classrooms face to face with global peers and expert facilitators through live, thematic international exchange programs — find them at banyangloballearning.com.*

---

**Seth (00:00):** Kristin, welcome to the podcast.

**Kristen DeBruler (00:03):** Thank you, Seth.

**Seth (00:05):** Tammy, could you start off with our first question?

**Tami Moehring (00:08):** I'd be happy to. Kristen, you've been at Michigan Virtual for about 14 years. What does a researcher embedded inside a state virtual learning organization actually do day to day? And how is that different from being a university researcher?

**Kristen DeBruler (00:25):** Yeah, great question. So day to day, I would say my work involves a lot of different parts of Michigan Virtual. So I work with our student learning services, I work with professional learning services. We have a data insights team. We have my team, which is the research team. We have a piloting team. We have an instructional design. And so my work really works with all those different teams to meet the needs that they have and meet internal needs that we have. So my job is a lot different, I would say, from a university in that we have student data, right? We have access to student data. And so we can use that data and we can really take it, analyze it and use it to inform practice, which I think is a unique place to be. I also think my role in Michigan Virtual is really unique at a state level. There's very few other states that have funds allocated to have like a research institute. And so other virtual schools will do internal research with the capacity that they have, but we've really been lucky to be able to fund high quality research that benefits our own practice, right? Our students, our teachers, but also share that statewide as well so that other schools can benefit from the work that we do too.

**Seth (01:31):** That's a whole lot of people that you're working with, which is amazing. I've been a fan of Michigan Virtual for some time now. We've had several people on this podcast. I've had several people on my other podcasts from Michigan Virtual. I think I'm at like a half dozen of y'all now. Everyone's impressive. The fact that it exists is impressive. Who sets the agenda for the research, right? You've got all of these different people. Is it that like the student learning services, they're like, hey, we really need more data on this so that we can act on this, or —

**Kristen DeBruler (01:41):** Fuh.

**Allyson (01:41):** You —

**Kristen DeBruler (01:46):** Agree.

**Seth (02:00):** Do you like look at all the data and you're like, hey, I think this thing is something that's happening across all of these, or the state's like, hey, we need more AI — like, who sets the agenda?

**Allyson (02:10):** Mm-hmm.

**Kristen DeBruler (02:13):** So all of the people you mentioned. So at the highest level, the state sets it. The state has what are called our legislative directives and those are on our website. It's the technical state language of, you know, research, identify and share best practices for digital learning or evaluation of cyber schools, right? So there's these sort of really high level things that the state asks us to do with the funds that they allocate towards us. But within there, there's a ton of flexibility, right? Identifying best practices can mean a bunch of different things. And so each year what we'll do is set our research agenda based on — like you said — conversations with, let's say, PLS. So, you know, one of the things that we did years ago was they were thinking about how people are using their courses. Are they staying because they want the full content, or are they coming for resources, right? So like we can design a research study to answer questions that they might have. And then we can also look at our own internal data and say, this is interesting, this is strange, this is — let's look more into this, something like that — and then design a research study around that.

**Allyson (03:19 *(approx)*):** Yeah, I was just processing everything — all the different people you work with, the ecosystem that you have. It's exciting to have an opportunity to have lots of people, but I could see that being a little overwhelming as well. But for research purposes, yay, that you get to contribute so much. And having been able prior to the podcast to dive into some of the work that you have done and published, one thing that stuck out was that you like to do the classroom-level research where it's kind of try-and-see — you don't have to wait for the peer-reviewed study, which is a great form, right? Because sometimes that can be a long time. But I wonder just in your experience, have you ever seen any of the practitioners have a little bit of bias, or you feel like it's something they believe before? Just wondering — what are the most common mistakes you see when people are studying their own practice with those that you've worked with?

**Kristen DeBruler (04:13 *(approx)*):** Yeah, so I am a huge advocate for teachers doing classroom-level, small-scale research. I think it can be really informative. It can be really helpful. And I think they run into the same problems that even researchers do — like, you said bias, right? I can have bias going into a research study. The best way to counteract that, and the thing I advocate for like from the start, from the get-go, first meeting is — what questions do you want to ask, and what are you trying to understand here, right? And being —

**Seth (04:25 *(approx)*):** You —

**Kristen DeBruler (04:40 *(approx)*):** — extremely specific with that. Because the more specific you can be, the better measures you can do, the better data you can collect, and the more concrete your answers will be. If your questions are too broad, then yes, you're susceptible to bias. If your question is, does this improve learning? Well, how do you measure learning? Is it a test? Is it a quiz? Is it any number of things? And what do you mean by measure? Is it a standardized test? Do you want to see time on task? Do you want to see all these other different metrics? Right. So being hyper-specific with that, I think, can counteract biases that you have.

**Seth (04:56 *(approx)*):** You —

**Kristen DeBruler (05:09 *(approx)*):** And maybe even just acknowledging your biases straight up too, right? Like, I think this tool is better than this one. So when I'm collecting data, I need to make sure I'm collecting fair, unbiased data. I'm collecting the same thing from each of them, right? So these are just simple ways that, you know, a teacher or even a researcher can keep their work focused on having usable outcomes at the end of whatever project they're working on.

**Allyson (05:31 *(approx)*):** Yeah, I like that idea. It's like the idea of asking — is the question answerable, how? Being able to really have those sources out and almost like flip what you might think, or have a preconceived — you know, your bias going in. In some cases, have you ever seen anybody kind of flip it and make it the question? Does that — you know, does technology belong in museums? For example, if we from here at CILC think it does — is that something that you see as a helpful practice in that kind of being mindful?

**Kristen DeBruler (06:02 *(approx)*):** Yeah, it certainly could be, right? So you're acknowledging whatever bias you might have. And then as long as it's something that you can measure and you can collect data on and has like a measurable outcome, then absolutely, I think that's a great way to structure it.

**Seth (06:17 *(approx)*):** Sorry, it's 2026 in America. We don't acknowledge our biases. That's, that's so 2012. In 2024, speaking of dates, you synthesized — you did a synthesis and it surfaced the fact that only 30% of district administrators report having any sort of formal AI policy while 80% of educators expect AI to matter significantly within the next like five years. What is the cost of that gap for leaders? And what does a useful first version of district AI policy actually look like?

**Kristen DeBruler (06:21 *(approx)*):** Yeah.

**Kristen DeBruler (06:25 *(approx)*):** You —

**Kristen DeBruler (06:30 *(approx)*):** Yeah.

**Kristen DeBruler (07:01 *(approx)*):** That's, yeah, so there's two parts there. I think the cost of that gap is something that we're seeing in other research that we've done. So we just published last week or two weeks ago, a report that looked at our Michigan Virtual students and teachers, and it looked at their perceptions of AI, their knowledge of AI, and how they understand like acceptable use. And what we saw is that while there was pretty good alignment on their knowledge of AI and what they considered to be like unacceptable uses — like, you know, ChatGPT writes my whole report for me — on acceptable use, there was a lot less clarity about what's acceptable use. So can — is it okay if ChatGPT writes an outline for me and then I write it and fill it in? Is that okay? Students said, no. Teachers said, sometimes, maybe. Yeah. Right. So we have these other examples too. And I think that speaks to that — like you said — the harm potentially in not having a clear vision or policy is that there's not a unified message around AI. There's not a uniform understanding around when they can use AI and when they can't use AI. And I think that could lead to unintentional misuse on a student's part, unintentional misuse on a teacher's part, right? I think having that gap leads into gray areas that can be really uncomfortable for educators and students and parents and teachers — for everybody, basically. As for first steps in district policy, if you're going to write a district policy, talk to Carly Dallow — she's who I would go to on our team to talk about it. From my perspective, I think it goes back to that same being hyper-specific. So what's the vision for your school or district? And how does AI fit into that vision? And then kind of crafting it around that. Don't start with AI. Don't start with AI and whatever tool it's going to be. Start with the vision for your school or district and how AI fits into that. And then what that looks like in terms of use for your school, and be specific about that. I don't know if that has to go in your overarching mission, vision, policy, but teachers and students want those kind of resources about how to use AI appropriately in their classes.

**Allyson (09:07 *(approx)*):** That's so interesting, the idea of like not having that directive as the educator and then being able to say, students, let's use this. It kind of creates — like you said — really a bit of a challenge because if they don't know how the district wants to use it, how can they help the students and then the wider family group or care team for the student use it. So I do wonder — because of, in your research — do you have anything where you've seen teachers apply ways to test AI? And could you tell us a little bit about like, what would that do in helping a student, especially online, learn not just what they like, but how to actually use the tool? Like what would they measure, and what would you want to tell them to ignore so that they don't get off focus, to your point?

**Kristen DeBruler (09:56 *(approx)*):** Yeah, we've done a few pilots with different AI tools at Michigan Virtual. So we've looked at tools that are student-focused, tools that are teacher-focused, and we've measured different things. So the hardest thing to measure is always going to be learning outcomes, right? Because it's hard to get sort of a learning outcome measure that actually measures what you want it to, that is timely, right? Is that the same time as sort of this AI intervention that you're doing? And the one that you can control, so there's not other outside factors impacting it. So that's really hard to get like a clean learning outcome measure, but that's like the gold standard, right? Because if you want to see if an AI tool works, you would want to know if it improves learning. But we've also measured things like time on task. Does a student — if it's a game, or if it's some sort of interactive item — do they stay there longer, right? Do they play it longer? Do they report enjoyment playing it? And is that — with the assumption that staying there longer improves some sort of whatever the learning outcomes are for that game. You can also look at things like, does it increase teacher self-efficacy or wellbeing if it's decreasing their workload in some administrative way, right? So we've measured it in different ways because again, learning outcomes is hard and it's not always the only thing you want to look at, but there are other things that you can look at in terms of what the tool is purported to do and how you intend to use it, sort of working backwards and then figuring out how you might measure that in a concrete way.

**Seth (10:57 *(approx)*):** Hmm.

**Kristen DeBruler (11:18 *(approx)*):** What the tool is purported to do and how you intend to use it, sort of working backwards and then figuring out how you might measure that in a concrete way.

**Seth (11:27 *(approx)*):** I think it speaks to how difficult it is to do any educational research. Right? Like, controls are really hard to establish. They're incredibly dynamic systems. What's happening even just within the school, let alone what's happening at home in their personal lives. I think it's really clever to measure the impact in those other ways — time on task, teacher self-efficacy, learning outcomes — you know, even just how to assess whether or not students have reached the learning outcomes. That's a whole quick area of practice that isn't always super clear. So I guess thank you for your contribution to the very frustrating world of educational research. One of the things you found though in terms of online learning is that teacher communication is the connective tissue of online learning. And I am curious, even within all those dynamic systems, like how you gained the confidence to say that that is the case, right? Like, how were you able to isolate that and say, yes, this is definitely having the impact we think it's having. But you also said something specific — that the relationship was built primarily on substantive feedback. So what counts as substantive feedback, and what is the non-substantive — that's a hard word to say — feedback? What does that look like, right? So what does good feedback look like? What does bad feedback look like? And sorry, I layered in a third question there of how do you know it's true? Okay.

**Allyson (12:58 *(approx)*):** Hehehehehe

**Kristen DeBruler (12:59 *(approx)*):** Yes.

**Allyson (13:08 *(approx)*):** Hehehehehe

**Kristen DeBruler (13:13 *(approx)*):** Yeah, okay. So start — I'll start with question one. How do we know? So I think this work is pulled from some work that Chris Harrington and I did several years ago. And so for that —

**Seth (13:22 *(approx)*):** Friend of the podcast.

**Allyson (13:24 *(approx)*):** Yes.

**Kristen DeBruler (13:43 *(approx)*):** — he and I surveyed Virtual Learning Leadership Alliance teachers. So the VLLA — Michigan Virtual is part of the VLLA, which is a group of state virtual schools who sort of share resources, information, they help each other. It's a network of virtual learning programs across the country. And so we talked to our teachers, we talked to their teachers, we surveyed them and their administrators as well. And we asked them, what does your job look like? What do you do day to day? What's important when you're trying to connect with a student? What behaviors do you do when you're trying to connect with a student? What's the biggest challenge in your classroom that you're having? So we asked these teachers all these things. And one of the things that came up is relationship building is really hard. That was number one. And then number two for them was — but the way that we do it is primarily through feedback — because they don't, they're not checking in with students every day. They're not doing Zoom calls with students every day. These are primarily asynchronous courses. And so giving students feedback was really the primary communication method for most of these teachers. And so that's where we got that information, was from — I think over a thousand full-time, well-established, highly skilled virtual teachers. So I feel good about those findings. I feel good about what came out of that. I think there's a couple of reports and maybe a book chapter out of that one, which was really interesting and well received, I think, by the field as well. And so, yeah, we asked them, what does good feedback look like? And two of the things that I can remember pretty well was it's personal. And it's really basic stuff — like use the student's name, say that like — check in with something about the student that you remember them telling you. Like, hey, John, what a great Lions game last night. So you just kind of pull in something that you know about the student, keep it so that they actually know that you're not AI, you're not just writing something on the back end — you're an actual person who knows them. So keep it personal. That tended to be really highly ranked by teachers. And also formative feedback. And we know that from — this isn't new, this is education research from decades ago — formative feedback tends to help students. So keep it stuff that they can work on, give them tools and feedback that they can use to move forward with their work, right? And maybe revise, or things like that. So those two things rated pretty highly from the teachers that we worked with. I think I forgot question number three — good feedback versus bad feedback, was that it?

**Allyson (15:57 *(approx)*):** Yeah.

**Seth (15:58 *(approx)*):** Yeah, like — you said what ranked highly, but like, what are some of the pitfalls that people fall into with feedback where they're like, I think I'm doing this right, but it's not actually building that connective relationship tissue?

**Kristen DeBruler (16:12 *(approx)*):** Yeah, I would say if it's not actionable for the student, it may not be as impactful for them. If it's short and if it doesn't really speak to the actual work — like if it's impersonal, if it's short, if it's not actionable — I can imagine that that wouldn't be very well received by a student who may be looking for connection. Not all students are, but a lot of them are, right? And I think given that it's the primary method of communication, it's an opportunity for online teachers to connect with students that they may not otherwise have.

**Seth (16:45 *(approx)*):** And I think maybe another bad example — aside for a second — I met a friend of mine, her daughter, met a 12 year old. We went on a walk and Prospect Park. I love talking to kids about AI because I just want to catch the vibe. And I've heard this from a number of adolescents where they are like, my teacher's telling me not to use AI and then my teacher's using AI.

**Allyson (16:49 *(approx)*):** I —

**Allyson (16:57 *(approx)*):** Hahaha

**Allyson (17:11 *(approx)*):** Yeah —

**Seth (17:11 *(approx)*):** And they're giving me feedback and I can tell that it's not them. And I'm like, well, how can you tell? And they're like, well, I just know. And you know, that might be true. The fact that that's even out there is of course creating — I think — breaking this idea of trust between the student and the teacher. Combining these last two things that we talked about here — you work on AI and this teacher feedback. If feedback is the primary relationship-building mechanism and AI is increasingly being used to generate or grade it, that's creating this tension. And your own data found that students who used AI as a facilitator of learning outperformed both non-users and students who used AI only as a task completion tool. So what does that distinction imply about where AI should sit in a teacher's workflow and where it shouldn't?

**Kristen DeBruler (18:09 *(approx)*):** Yes, great question. And I've heard similar things from students as well — the resistance and the messaging is very unclear, right? As far as teacher use of AI, I don't want to prescribe when anyone can or can't — but I think there's a real space for reducing administrative tasks with AI that teachers could leverage, right? So simple things, maybe not student-facing to start until you're maybe more comfortable, if you're interested — that's a good place to start. I think there is a tension though, like you mentioned — students: don't use AI, it's bad, it's cheating, it's whatever, right? It's sort of the messaging around that. And then teachers using it is muddying up that water a little bit, is creating some mixed messages for students. And I can see how receiving AI-generated feedback or responses from a teacher would sort of completely undermine — maybe in an online or virtual course — completely undermine that relationship-building aspect of the feedback, right? Or reaching out to the student, or preparing it in such a way that the student can tell it's AI, would definitely undermine that message from the teacher.

**Seth (19:25 *(approx)*):** Yeah.

**Allyson (19:27 *(approx)*):** Yeah, it's interesting too, because it kind of just made me think about what you were saying earlier — the idea of personalizing. Even if you, as the educator, are using AI as a tool, it's a nice reminder. And for the students, even if you get the base summary, how do you personalize that messaging? How does the student still make it — how does it work as a tool to building that relationship if there's already a potential idea that there might be mistrust in how that feedback is being provided? It's interesting.

**Allyson (20:03 *(approx)*):** And especially in an online context, just wonder — it feels like that's even more important, that you don't want that trust to break. You don't want to feel like you're talking to a robot.

**Kristen DeBruler (20:19 *(approx)*):** Yeah, I mean, so we've had what we call auto-graded assignments in Michigan Virtual courses for years. That's not new, right? Like a quiz that's graded automatically by Blackboard — that's not new technology. But I think maybe the difference is that students are aware that that's auto-graded, right? They know that this is graded automatically. The teacher's not here going through it question by question and giving me a score. It's graded right there by your LMS. So that's not new for students, but maybe it's confusing when it's expected to be a teacher kind of relationship-building moment. And then it turns out to be sort of this auto-graded thing. And I can see where that would create a lot of the confusion for the students. Clear expectations going in would alleviate a lot of that confusion, right? Like — this is an AI tool or assignment or feedback mechanism, and this is your teacher, right? Keeping those expectations clear, I think, would go a really long way.

**Allyson (21:18 *(approx)*):** Yeah, it's always interesting to think about the online learning environment. I think a lot about that balance of the asynchronous and the synchronous experience. Where does that blend come in? And then when you put in new tools that you're using as a tool, not a trend, how does that balance for each — how the courses look? And I know that in your research, you talked a little bit about pacing, like how people pace through a course, and that you also found that there were students who would sometimes deviate from how the course was supposed to be going, even if it was just once. But they ended up with a final grade — I think it was 9.5 points lower — than students who just stuck with the methods, stuck with the guide. So I wondered — in thinking about that idea of deviating, what does that mean in your research? How does that look, even with the one time? What kind of happens there? What does the gap suggest about how we structure the opening of online courses, or even just the — in my mind — that idea of the balance of the course?

**Kristen DeBruler (22:22 *(approx)*):** Yeah, so deviating from like the pacing guide is very normal behavior. I'll say that to start, right? So I think like 99% of students deviated at least once from the pacing. So that is very normal. We're going to see that. That happens. That is fine. Where we saw the biggest issues was when the students were deviating what we would consider like outside of the unit. So if a student is moving around within a unit — that tended to not be like a huge issue, right? So if you're going from like 1.1 to 1.3, not huge, right? If you're going from like 1.4 to 2.6, that's where we saw it started to create problems for students and it started to impact their learning outcomes in really significant ways, which makes sense, right? You're not scaffolding the information in a way that it's meant to be scaffolded. And it also suggests that the student is sort of progressing through the course in a chaotic or unorganized way, which —

**Allyson (22:56 *(approx)*):** Yeah.

**Kristen DeBruler (22:50 *(approx)*):** — isn't great for learning outcomes either. And so again, the messaging that we sort of concluded and what we've shared out with our mentor advisory network and with our own teachers is — it's okay, don't be alarmed if students are moving a little bit within the unit. But if they start to move outside the unit, like check in with them, make sure everything is okay, help them get back on track if they're sort of confused or lost or not sure where they should be, because that's where we see the largest detriment to their work.

**Seth (23:48 *(approx)*):** For those students who are deviating, you've connected these pacing problems to self-regulated learning and executive functioning skills that K-12 students really are still developing. And so if that's true, like — are we setting those students up to fail by offering them flexibility that they're not ready for?

**Kristen DeBruler (24:15 *(approx)*):** Yeah, this is a question I think about a lot as well. This is one that I really do think about, especially lately — we've been thinking about this a lot — this idea of anytime, any place, any pace. So we are working on sort of like a research project around that. But conceptually thinking about it, I think online learning can serve students really well. I mean, of course my own bias for Michigan Virtual, right?

**Allyson (24:17 *(approx)*):** You —

**Seth (24:39 *(approx)*):** You —

**Allyson (24:39 *(approx)*):** Hahaha

**Kristen DeBruler (24:41 *(approx)*):** But I do see a real opportunity for students that they wouldn't otherwise have without online learning. So students for medical reasons or safety reasons, or you live in a rural school and you don't have any world languages at your school but you can take them with us, right? So I think there's definitely a space there. But I do think that when we're working with adolescents who are still very much developing executive functioning and self-regulation skills — planning, goal setting, time tracking, all of these things that we do day to day to get our work done — they need a lot of support. And to me, the caution becomes applying adult-level research to the K-12 population. So adult-level research is easier to obtain. You can use undergraduate students to get it, right? So it's easier than trying to do research at a K-12 school. But you're not looking at the same population in a lot of different ways. So caution when applying adult-level research to a K-12 program. But also what supports are our programs providing to students becomes really important. So I love Jared Borup's work, his A's for Community Framework —

**Seth (25:48 *(approx)*):** Friend of the pod.

**Allyson (25:49 *(approx)*):** Hehehehehe!

**Seth (25:51 *(approx)*):** He —

**Kristen DeBruler (26:10 *(approx)*):** — I really do think that that provides a nice perspective on the multi-tier levels of support that students need, right? And it can feel overwhelming and staggering — like, I just want the students to take this course, but that student needs a lot of support on the backend to be successful. Again, not all students — some students can go in, progress through their course, finish it, and they're great, they're fine, they're done. But a lot of students really do need that onsite mentor involvement, parent involvement, community involvement, right? There needs to be sort of an acknowledgement that even though a student isn't maybe physically present in a school, they still need physically present support for them to go through their course.

**Allyson (26:30 *(approx)*):** I've been thinking a lot about that idea of head, heart, and that idea of learning, getting everyone involved and the support mechanism. And I just wonder — have you ever seen a case of AI being used as a tool for students or the teacher family network as a tool to set up like what those relationship-building things would be? Like what are those routines that people are able to stay in touch with? How does it help with that kind of executive-level thinking and functioning that you were talking about, but also across — has that been something that you've seen explored before? Because just knowing that there's not a specific policy all the time, do teachers sometimes say, hey, this is how you can use AI as a way to help support you in being able to feel successful?

**Kristen DeBruler (27:29 *(approx)*):** You know, I haven't seen any research on it and I don't have any specific examples, but I do know that there are some AI tutoring kind of technology tools that have come up in the market in recent years that could fill in sort of that just-in-time support, right? So if a student is working on their course off hours of when the teacher may be available, that could be sort of a first contact support for them so that they stay in the course, they stay working on it, but they still get some of the instructional support they might need. I haven't seen it work in terms of sort of like the more executive functioning stuff, but it certainly could, right? It could help students plan. It could say like, what do you want to get done this week? Or how many times have you signed into your class this week? Or even track it for them in a real concrete way — oh, well, you still have 60% of the course to go, but only 40% of the time. So maybe this week, can you do two extra assignments on top of it? So there is definitely an opportunity there with the amount of data that AI can process and display. I think there'd be an opportunity there to build something within the course that could support that student progress more explicitly for them.

**Allyson (28:28 *(approx)*):** Mm-hmm.

**Seth (28:44 *(approx)*):** One question that we ask everyone who's on the podcast — it's the title of the podcast. From your vantage point as a researcher who identifies all these different threads of what works and what doesn't within online learning, Kristen DeBruler — why distance learning?

**Kristen DeBruler (29:10 *(approx)*):** Distance learning is a great opportunity for students. And I think the work that Michigan Virtual and I do through them as sort of a partner to public schools and a support for public schools really does provide students with an incredible opportunity to supplement their learning. It provides them with opportunities they wouldn't otherwise have, or allows them to pursue other outside opportunities that a public school structure wouldn't allow. I think distance learning — yeah, for me it means opportunities for students. And I think we have a real obligation in that way then to maximize those opportunities so that students get the best education possible in this sort of other format.

**Seth (29:56 *(approx)*):** Hear, hear. Tammy, Allyson, any last thoughts before we let Kristen go?

**Allyson (30:01 *(approx)*):** I am so grateful for all of the research that you put out and I can't wait to continue to learn from you and I really appreciate you taking time for the conversation today.

**Tami Moehring (30:10 *(approx)*):** Yes, thank you for being on. Everything's spinning and I'm thinking all about questions I have to ask my kids now to do my own little research on all this. So thank you. Yes.

**Allyson (30:19 *(approx)*):** Ha ha!

**Seth (30:22 *(approx)*):** Local science. Kristen, thanks so much for being here. Is there anywhere where our listeners can find your work? Where would you like to send them?

**Kristen DeBruler (30:31 *(approx)*):** MichiganVirtual.org — we put all of our reports up there and we also have what's called the Digital Backpack, which is our blog. And so you'll get our work, PLS's work, innovation leadership work — it's all there. So MichiganVirtual.org.

**Seth (30:44 *(approx)*):** And we will link all the previous episodes with Dr. Tova Sheldon. I have an episode in my other podcast, both with Carly Delo and Aaron Boffman. You mentioned Chris Harrington. Is there any other Michigan Virtual folks? Probably seven or eight. Kristen, thank you so much for being here.

**Allyson (30:51 *(approx)*):** You —

**Allyson (31:06 *(approx)*):** Thank you!

**Kristen DeBruler (31:06 *(approx)*):** Thank you all.

---

**[OUTRO — recorded separately]**

**Seth:** That's a wrap on our conversation with Kristen DeBruler.

Teacher feedback is not as a nice-to-have but the primary mechanism through which an online student knows a real person is paying attention. In an async course, feedback isn't just instructional. It's the thread. When that gets automated or feels generic, there's nothing else there to replace it.

The pacing data adds a practical signal on top of that: a student skipping across units — not just moving around within one — is a student who may be lost or checked out, and that's exactly the moment a teacher check-in makes a real difference.

Kristen's published research and Michigan Virtual's Digital Backpack blog are both at michiganvirtual.org — link in the show notes.

Thanks for listening to Why Distance Learning, and we'll see you next time.