In this episode of the Social Learning Amplified podcast, host Eric Mazur engages with guests Tricia Bertram Gallant, UC San Diego, and David Rettinger, University of Tulsa, to explore the concept of cheating in education, its implications, and the evolving landscape of academic integrity in the age of generative AI. They discuss the importance of redefining cheating as a lack of learning, the reasons behind student cheating, and the role of assessment in fostering or deterring academic misconduct. The conversation also highlights practical strategies for educators to promote integrity and adapt to the challenges posed by AI technologies.

Eric Mazur
Thank you for joining us today for this episode on cheating, or actually the opposite of cheating, in the Social Learning Amplified podcast series. I'm your host, Eric Mazur, and our guests on the episode today are Tricia Bertram Gallant and David Rettinger. Thank you both for joining us.

Tricia Bertram Gallant
Thanks for having us.

David Rettinger
Thank you for having us.

Eric Mazur‍
Tricia Bertram-Gallant is the Director of Academic Integrity Office and Triton Testing Center at the University of California, San Diego, Emeritus Board Member of the International Center for Academic Integrity and former lecturer for both UCSD and the University of San Diego. Tricia has authored or coauthored and edited numerous articles, blogs, guides, book chapters, and books on academic integrity, artificial intelligence, and ethical decision-making. Following her 2022 co-edited book with David Rettinger, Cheating Academic Integrity, Lessons from 30 Years of Research, published by Josie Bass in 2022, Tricia and David are about to release a new book. That book, The Opposite of Cheating, Teaching for Integrity in the Age of AI, will be released by the University of Oklahoma Press in March 2025.

Tricia regularly consults with and trains faculty, staff, and students around the world on academic integrity, artificial intelligence, and ethical decision making. David is a polite professor and director of the undergraduate program in psychology at the University of Tulsa. He's taught at the college level for over 20 years, including 15 at the University of Mary Washington, where he holds the title of Professor Emeritus.

During that time, he served as procedural advisor to UMW's student-run honor system and has published frequently on the psychology of academic integrity. With a PhD in cognitive psychology, he's also well-versed in the basic principles of learning science that underlie excellent teaching. Rettinger is president emeritus of the International Center for Academic Integrity. In addition, he leads the organization's effort in assessment and survey research, continuing the McCabe Academic Integrity Survey. In this podcast today, I'd like to focus on your upcoming book. Its title, The Opposite of Cheating, resonated very strongly with me for the following reason. About a decade or so ago, I eliminated all high-stakes testing from my teaching as this form of assessment, I think, tends to incentivize bad study habits and also cutting corners. Maybe we can talk a bit more about assessment and grades later on, but I'd like to start with a very simple question. What is the opposite of cheating?

David Rettinger
Well, I think for me, the opposite of cheating is learning. It's not not cheating. It's the thing that we're all trying to do. We're trying to get students to interpret our material, to grow as people and to become better learners. And all of those things are what is being circumvented when we when students commit academic misconduct. So the opposite of cheating is learning.

Eric Mazur
What about you, Tricia?

Tricia Bertram Gallant
And for me,

yeah, for me from the teacher's perspective, since David did it from the learner perspective, I would say the opposite of cheating is teaching, even though it's really an anagram, right, of cheating. It's not really the opposite. But for me, that's really important because when I started in this work 18 and a half years ago, I spent a lot of my time, and I still do, trying to refocus faculty away from focusing on cheating.

Why do students cheat and how can I stop them in my class? And focusing to David's point on how do I best facilitate and assess learning in the 21st century given the realities that exist like contract cheating, artificial intelligence, and so on. So for faculty and students, it's both about the opposite is learning, but on the faculty side, it's how can I best facilitate and assess that?

Eric Mazur
So you sort of restrict cheating to the context of learning and teaching. think of course teaching is a much broader concept, right? I was thinking more maybe the opposite of cheating is adhering to the highest integrity, standards of integrity. But I see now what you mean in the context of simply learning.

So Tricia, you mentioned the reason for cheating. And so that's a great segue, in a sense, to my next question. Why do students, or maybe I should say people in general, why does cheating occur?

Tricia Bertram Gallant (05:23)
David, I'm gonna take an overview and I'll let you get into the details. So I like to say that students cheat because they're human. And our colleague Jason Stevens wrote about this, did a great article in Change Higher Ed Magazine a few years ago and talked about this. All species cheat, humans cheat, we do it for survival, it's just a learned behavior. so...

Ultimately, when faculty asked me, why are my students doing this? I said, because they're human. And beyond being human, they're 17 to 21 year olds, a large part of them, right, if we're talking about undergrads. And so, you know, they're still developing some knowledge that, to your point, Eric, about making better ethical decisions and making better decisions under stress and pressure. So that's kind of the overview. And David can provide more details on specific powerful whys.

Eric Mazur
But maybe before we get to David, let me interject this. You said humans cheat, which is true. People tend to, especially the more high stakes things are, money, you name it, people tend to cheat, take a shortcut to obtain a goal that they need to obtain.

But the interesting thing that in the context of learning, a cheating student is really cheating themselves because they're robbing themselves of an opportunity to learn.

David Rettinger (06:51)
This is true, but to the extent that we've commodified higher education or education in general, they are cheating the medical board or they're cheating the employer who is relying on us as higher ed institutions to certify that students can know things and can do things. I you don't want to go to a doctor who's cheated on their medical boards, right? And that's so if they're...

Tricia Bertram Gallant (06:52)
Maybe.

David Rettinger (07:19)
Cheating is very much not a victimless crime, even at the undergraduate level, because as I say, if you leave the University of Tulsa not knowing what you claim to know, every time someone walks into a job interview with a TU diploma, the employer is going to think about that time where you showed up and you didn't know what you were doing. So cheating is very much not just about cheating yourself, although it is about cheating yourself.

Tricia Bertram Gallant (07:45)
And to tie that back into your earlier questions, one of the biggest influencers of student cheating behavior is their perception or their understanding of what other students are doing. If other students are cheating in the class, they're more likely to cheat. So it's obviously harming the class culture, right? Students are saying, well, they're cheating, they're getting away with it, therefore I might as well cheat too, or I have to cheat to keep up. And it also hurts the instructor-student relationship.

the more cheating that happens, or even sometimes one incident of cheating, faculty now say, well, I can't trust students anymore. I can't do this assignment anymore because students are going to cheat on it. So it harms that relationship. And Eric, I would say, we all cheat even on little things. It's not just about high stakes. Think about a white lie. That's dishonest. And it's not big stakes most of the time. That's why it's called a white lie. But yet we do it all the time. And so it's.

It's more insidious than a cost benefit, not a cost benefit, but a cause, high stakes and result cheating. It's more of a insidious, intrinsic thing that we do in response to our environment.

Eric Mazur
So.

David Rettinger
In fact, a lot of students are happier to cheat on something small than something big because it's just a little thing. It's just a small assignment. The biggest irony, and I bet Tricia will tell you the same thing, is when we get a student through an academic misconduct hearing who is being punished at a semester level for something that was worth 2 % of their grade, they would have been far better off not to turn it in. But whether it's perfectionism or whether it's a perception of competition in their class,‍ that led them to a place where they felt like that violating the rules, pulling something off the internet, sticking it into chat GPT was a better solution than just asking for an extension or even taking a zero. low stakes helps a lot because it changes students' sense of why they're doing what they're doing and it gives them maybe a little bit more sense of agency that they can actually be successful on this assessment.

But it's never going to remove that motivation to cheat. And moreover, it's going to give them more opportunities to do it and to feel like it's not such a big deal.

Eric Mazur
Yeah. Yeah. I'd like to maybe briefly challenge the notion that, you know, the students are cheating future employers or medical board or hospitals where they're being employed. I think that everybody who hires people knows the transcript tell, you know, nearly nothing about a person's ability in terms of, you know, its relevance to future job, I was recently talking to a number of doctors and they all admitted that if they were to take the, what is it called again, MCAT, exactly, they would probably fail it, yet they are successful doctors. So the question is, know, what does that really tell us and have we constructed a system that students --

Tricia Bertram Gallant
MCAT.

Eric Mazur
but that is actually not even that relevant for society as a whole. mean, just one experiment I would love to run on my campus. I would love my colleagues to take their colleagues' I think it would be very revealing. I don't think I'll ever manage to get such a test done, but it would be very revealing. What are your thoughts on that?

David Rettinger
Ha.

think you're talking about assessment. And that is something that's at the heart of higher education, all education. And for me, it's the biggest challenge that we face. Medical school is a great example. The MCATs are not at all predictive of what sort of family physician you're going to be or what sort of surgeon you're going to be. Even remotely, they are pretty good predictors of how good of medical student you're going to be, but that's a different point. And so...

when you bring it back to an individual course, to what extent is 60 multiple choice questions in December a good assessment of what you know at that moment, let alone what you will know in six months? Any cognitive psychologist will tell you that even if you've learned it well enough to know it on December 12th, there's no reason to think that that's going to last even through the break. And so once you set up this assessment regime,

that isn't about learning, it's about getting the scantrons read so you can go ski, then students pick up on that right away and they understand that the assessment is neither authentic nor actually that predictive of the things we care about. And that in turn undermines all of the reasons that they might choose not to cheat. And so better assessment is not going to eliminate cheating, but bad assessment is almost begging students to commit misconduct.

Kath Ellis talks about that all the time. The way you assess is you can build in cheating, but you can't build it out.

Eric Mazur
So the web page for your book states that your upcoming book provides practical suggestions to help faculty revise the conversation around integrity. What are some of the things instructors can do to curb cheating right away, even before your book comes out, given that the semester will start before your book comes out?

Tricia Bertram Gallant (13:19)
Yeah, well, so I'm getting, you know, we both get asked this a lot because the Gen AI age is very overwhelming

faculty, right? Like, what am I supposed to do? Where am I supposed to start? And a very clear kind of natural place to start is communicating with your students. And notice I say with and not at your students, right? So we give some suggestions in there and, you know, one of them is ask them how they're using

if they're using the tools, how they're using them, how they think they would be helpful, even going through your syllabus together and looking at the assessments that are upcoming and asking the students, you think you'd be able to you know, a ChatGPT on this assessment, ethically, responsibly, or do you think it would undermine the integrity of this assessment? And having students talk with you and with each other can be very powerful.

And then I would say after you, and by talk with, don't mean you necessarily have to a class discussion, maybe it's surveys, maybe it's some other asynchronous method. And then after you have that discussion and hopefully maybe after you've played with the tools, with your assessments a little bit, you then come up with your policy for the class and you clearly articulate the policy to the students. So I have met with over 200 students now who've been reported for GEN.AI misuse.

And the most common thing I hear from them is, I just wish the professors were more clear about whether it's allowed, when it's allowed, and how I can use it, or if I can't use it. And so of course, that doesn't eliminate all cheating. There are people who are to cheat because you told them the rules, what the rules were. But it will eliminate the cheating of the students sleepwalking through, well, I used it on this assignment, so I thought I could use it on this one, or I used it in this class, so I thought I could use it in this one. So communicating with your students about it. Both in terms of taking information from them and also being very clear with them about what the rules are for that class and the why. And the why should be connected back to the learning outcomes. You can use it on this assignment because of these two learning outcomes or you can't use it on it because of these learning outcomes. The why is extremely important to us as humans, right? We know why we're being told to do something the way we're being told to do it.

Eric Mazur
David, any other tips you want to highlight?

David Rettinger
Tricia's little codicil at the end is exactly what I wanted to talk about. So the place to start with any change you want to make is your learning objectives in the class. And you want to ask yourself, are the things that I'm assessing the things that are actually in those learning objectives? And so, for example, I love having students give oral presentations. But in a lot of those courses, that's not a learning objective of the class. And so when a student says to me, I have social anxiety, or I hate giving presentations,

It really begs the question, can I give them an accommodation? And if it's not a learning objective, then it changes, it should change my approach relative to a class where it is a learning objective. So once you've given some thought to what the real objectives, learning objectives are of your course, like Tricia said, now is how you connect those assessments to those learning objectives explicitly for your students. They love doing things that mean something to them and they don't like busy work.

my students say to me, one of the favorite things you ever say is there's no busy work in this class. Now frankly, I don't think most of my colleagues are assigning busy work, but they're not explaining why what they're doing isn't busy work. So once you've set up learning objectives and you've given some thought to your current assessments, the next thing I'd suggest is giving some thought to some authentic assessments. And I know that's a buzzword and I know not everybody loves the notion.

But what I really mean is just things that are meaningful to the students, whether that be in their profession, whether that be in their lives as humans, as citizens, as members of a family, give some thought to what the assignment is doing for assessment and what it's doing for them. And if you do those things, then you're going to get some of those persuadable students on board with what you're doing because they know why and they think it's worth it to them. And again, the non-persuadable students are not persuaded by that.

You can move some folks along for sure.

Eric Mazur
Can you give an example of what you consider an authentic assessment?

David Rettinger
Well, sure, in a lot of cases what it is giving students the choice of, a deliverable. So in a lot of my classes, they do what I call an application project, say in cognitive psychology or decision science. And I say to them, your audience is, you can pick your audience, but it's not an academic audience. And you can give me anything you want as a deliverable for this. You just have to write a proposal at the beginning of the term telling me what your topic is and what the deliverable is.

And so of course I do get a lot of research papers, right? Some students really are most comfortable in that space and you don't want to push them out of it if that's not a learning objective. But I mean, I've gotten all of it. I've gotten art projects. My absolute all time favorite was a student who was the head of the peer tutoring program at the University of Mary Washington. And she wrote a teaching and learning guide for peer tutors, how to help students learn how to study on the basis of cognitive psychology that couldn't have been more authentic and I could have never assigned that to her, only she knowing herself and what's authentic to her could make meaning from that assignment. And so working with your students collaboratively to create an assignment that works for them is a fantastic way to do it. Now, the grading rubric for that assignment is really complicated. You can really get anything from interpretive dance to a research paper. And so you have to be smart about that, but it took me a few iterations, but it worked. It won't work for every discipline, but in those classes it did.

Eric Mazur
So you talked briefly about the rubrics there. Do you do any self and peer grading?

David Rettinger
I do some. We tend to, I do a lot of drafting in my classes. That's another way to help students avoid misconduct is giving them a sense that they have control over their situation. And one of my favorite ways to do that is to just give them more than one shot at something, which sometimes means a traditional draft and revise. Sometimes it means I give them a couple of hall passes to redo an assignment that they're not satisfied with. And sometimes it means having a writing workshop day where they bring in the particularly results section. APA statistics can be really tough for some students and really easy for others. And it's a great opportunity to give them a chance to be social, collaborative. And if they're going to work together when they're not supposed to, why not give them the opportunity to learn from this behavior and take it from being academic misconduct to leveraging their natural behavior to being something that's useful for them.

Eric Mazur
Now, Tricia mentioned Gen. AI a moment ago, and I am surprised that the rate at which things are progressing in Gen. AI, what was true at the beginning of the semester is no longer true at the end of the semester. And being in semester, this is actually a deliverable that Gen. AI really can produce. And then a few weeks later, it just does it, right?

I require my students to hand in handwritten problem solutions broken down in a particular way, four steps, you know, getting started, developing a plan, executing plan and evaluating the answer. And typically if you give the problem to Gen. AI, it just boom gives the solution without any of those four steps that, you know, take the force, the problem solver to think through those steps.

But now I can just go to even to chat GPT and say, please solve this problem using Eric Mazur's four step problem solving procedure. I'm not kidding you. And it's piss it out. mean, more often than not, there are pretty big errors in it. And it has a lot of trouble with evaluating. It simply regurgitates things, which is what students often do too. So it becomes harder and harder to, you know, take into account the role of Gen AI. I'd hope that COVID would be an enormous disruptor, bigger than any other disruptor in education, only to discover that people went straight back to what they'd done before COVID, at the end of COVID. So the big difference, of course, between Gen AI and COVID, the genie is out of the bottle. It won't go away. COVID, you know, we're all back on campus now. So to what extent do you think that we've reached an actual turning point that we can maybe 10, 20, 30 years from now look back at and say, this is the moment that education really changed.

Tricia Bertram Gallant
Okay, so you're asking me on a day I'm optimistic or pessimistic? So to take us back in time, in 2008 when I wrote my first book, I said we're still teaching as if it's the 20th century and the internet doesn't exist. Largely...

Eric Mazur

You're being very generous here because, you know, many people are still teaching as if Gutenberg had never existed and the printing press was never invented.

Tricia Bertram Gallant
True, true. That's true. We had an instructor here who is still using his handwritten lecture notes from 25 years prior. So do I feel like this time will be the disruptor? Internet wasn't the disruptor, right? Printing press wasn't the disruptor for some. Will this be the disruptor? I don't know. So in my optimistic times, see a lot of pockets of faculty saying everything's gotta change. But I'm not seeing a lot of change at institutional levels. I'm not seeing a lot of change in the rhetoric. The rhetoric remains the same. David mentioned the, I call it the industrialization of higher education, but the language remains the same about reducing time to degree, retention, not that those things don't matter, grades, GPAs.

But they're false idols, right? They're not really getting at the heart of the matter. What I hope to see is that we recognize students can learn 24-7 outside of the university. They don't need us for learning. And they haven't used us for a while, necessarily, right? Learning content, learning fundamentals, perhaps there's a way they can learn that, especially if we end up with personalized AI tutors at some point.

What they need us for is human-to-human interaction. And I know, Eric, you do a lot of active, engaged learning, right? So problem-based learning, team-based learning, the flipped classroom model. That's what I hope. That's what I hope we look at when we see this new era and we say, look, this is finally the time where we have to say it's not worth it for students or teachers to come to class, you know, to get in the car, to commute, to find parking, to walk to class, to find a seat, all of that logistical stuff, it's not worth it for them if they're just gonna sit there in a lecture class. That's gonna be podcasted and made available outside of class later. They want, they need to come to us for human-to-human interaction, to develop those human durable skills that they can't necessarily develop with AI. And so I really do hope that this is fulcrum of change.

I hope but I'm not optimistic about it because higher ed is quite known as you pointed out, at just resisting change. The one thing I think is different, at least in America right now, is we are losing trust of society as being relevant. Like you mentioned the doctor conversation. And if we continue down this road, we're gonna continue to lose our relevancy, right? We both have to be relevant as certifiers of developing knowledge and abilities and then certifying that knowledge and abilities and that knowledge and abilities has to be relevant to what people need to be citizens and professionals and so on. so we may, that might be end up being the fulcrum that makes us change more so than this. But I don't know, I'd like to hear what David thinks if he's more optimistic than I am.

Eric Mazur
let me add another dimension to this question, because I think that Generative AI will not simply impact education because it can do the work that we expect the students to do. I think that probably the biggest impact of GenAI on education will be indirect. In the sense that it will affect the jobs that our graduates are going to take. And this is something, of course, that we don't exactly know, but you definitely see it in computer science. mean, there's a precipitous decline in the need for programmers because, you know, generative AI can do programming quite well. And so it completely changes the need for, you know, programmers in the workplace and therefore the number of students who are going to take computer science as a major. So I think one of the impacts is still completely unknown. How is it going to change the workplace? How it's going to change the needs of our graduates?

David Rettinger
Absolutely. When I'm feeling optimistic, so I agree that I think that the change in the workplace and the loss of trust in our society are going to conspire to send fewer people to university and college. And hopefully those folks will be able to have a productive place in the workplace, not unlike the 1950s, say, where you could build cars and make a great living and send your kids to college if they chose to or take a union job themselves. If that's not if and if that's the case, that'll be great. What I think higher education will become will come back to is. Both a luxury good, which I'm not so thrilled about, but also getting back to what it's really for, which is the humanities, which is making life worth living, not necessarily building that life. I don't think engineering is going away anytime soon as a profession, but as you said, we're not going to need the sheer volume of it. So how does how does Gen AI affect that? Well, who's to say, but my best guess if it goes well, is that it will be, it'll give students the ability to focus on their process of learning as opposed to the product of learning. Our assessments have been since before Gen AI and since before contract cheating, were tended to be fully born products, whether that be a physics proof, whether that be a research paper, whether it be an art project, whatever it might be. And we are reasonably able to infer process from product. But writing is the best example of this. The late, great Frank Yates used to tell us, writing is nature's way of telling you how unclear your thinking is. And it's also how to tell you whether you're a professor, whether you know what you're talking about.

Nowadays, in the case of writing, knowing something and writing about it well are decoupled with the age of gen AI. I can give you some half-baked thoughts and give you a C or even a B paper in a lot of contexts. So then if we are going to continue to both assess learning and ensure learning, we're going to have to think about what does that look like as a learning process.

Perusall is an absolutely fantastic example of a way you can take something like writing an article summary, which is a product, and turning it into a social, either real-time or asynchronous process, and have insight into the student's process. That's just one of many changes that I think we'll have to see as the result of Gen. AI if we do well. If we don't do well, we're just going to continue down the same road of allowing students to turn in work that didn't have much intellectual contribution from them, which is not a great road to be on.

Eric Mazur
Wonderful. Unfortunately, the time always goes faster than we want. I want to thank you for an absolutely thought provoking discussion. And I very much look forward to the event in March and the publication of your book. I would like to conclude by thanking our audience for listening and inviting everyone to return for our next episode on behalf of

All our listeners, Tricia and David, thank you again for joining us today.

Tricia Bertram Gallant
Thank you.

David Rettinger
Thanks for having us.

Eric Mazur
Their book, The Opposite of Cheating, Teaching for Integrity in the Age of AI, will be available in March 2025. And I'm excited to announce, I already alluded to this, that starting on March 11, 2025, we will have a Perusall Engage event with Tricia and David for their new book. For those of you who have not yet joined such an event, Perusall Engage events are author-facilitated events where for a nominal fee, can access

the book and you can read and interact with the authors and other like-minded educators like me, for example. And all of this is done on the Perusall interface. I hope you will join the brainstorming on how to restore integrity and well, do the opposite of cheating in the classroom. To learn more about this event and reserve a spot, go to perusall.com slash engage. Finally,

You can find our social learning amplified podcast and more at perusall.com slash social learning amplified all in one word. Subscribe to find out about other episodes. I hope to welcome you back on a future episode.