Exercise Effects on Detection Dog Acuity with Liza Rothkoff

Hey guys, this is the editor, Sarah Fangton. I just wanted to hop on here to let you know that we unfortunately did have some issues with one of the microphones during the episode. It was just a bit on the sensitive side. Specifically, when Liza speaks, you're going to hear probably a lot of echo and banging and wrestling and just overall background noise. I tried my best to get rid of as much as possible, but unfortunately, I'm not able to get rid of those sounds while the person is speaking. But we wanted to make sure this episode got out to you guys. That's a really cool information in this episode and enjoy. Hello and welcome to the Canine Conservationists podcast where we are positively obsessed with conservation detection dogs. Join us every other Thursday to talk about detection training, Canine welfare, conservation biology and everything in between. I'm Kayla Fraet, one of the three co-founders of Canine Conservationists. We trained dogs to detect data for researchers, conservation organizations and government agencies. Today is April 14th and you may notice, so I'm reporting three podcast episodes today, so they're going to be three episodes with April 14th and the date. Hopefully I still have a voice about the time we talked to our last guest. And I'm talking to Liza Rothkopf about exercise effects on the old factory abilities of detection dogs. I'm super duper excited about this episode. It's really, really cool stuff and it builds off of a science highlight from a previous episode. But before we get into this episode, which, you know, basically the whole thing is a science highlight, we're also going to go into our brief science highlight, which was prepared where I are wonderful volunteer Eloise Hedge Cot. A group of researchers in Finland investigated whether dogs could be trained to detect a rust fungus called Cronardium Pine I. So this is a pathogenic rust fungus that infects and kills scots pine trees, leading to significant losses in managed forests. So it is a heteroescious life cycle that can infect other unrelated hosts. So as well as infections passing from pine tree to pine tree, it can also spread via 50 or more other plant species. And as the icing on the cake, it can remain latent lying in weight in young trees for a long time before any symptoms of disease appear, which makes it difficult to detect early and remove infected trees and prevent its spread. Dogs have not yet been trained to detect rust fungi. So the authors aim to, quote, "clarify if dogs can be trained to detect the scent produced by sea pine I, in trees and plants to improve early detection and therefore forest health." So for the methods, they had volunteer dog handler teams that were that embarked and Eloise says part of the pun. On an extended one-year training process supervised by professional trainers, the dogs were trained using lots of different parts and forms of the fungus, including spores, fruiting bodies, Aisha, which are the reproductive structures, Mycelia, which are the rootlets, structures, ranch lesions, and infected leaves from different host species. Many biological and non-biological controls were used, including samples of healthy host plants and other rust fungi. Five dog handler teams took part in a final indoor scent discrimination test, and four of those went on to take part in field studies. So as far as our key results, the dogs are able to detect the rust fungus on Scott's Pine and alternate hosts. They were able to detect various different life cycle forms, and performance varied in the indoor tests. The authors note as it occurred six to eight months before the field test that the lower success rate in the indoor stage, which is pretty surprising that it would be lower success in the indoor stage might be because the dogs were at an earlier stage in their training. So remember if this whole thing took a year, the indoor training took place about half or maybe three quarters of the way through that. However, quote, in the field, dogs identified C-Pyronis spores, branches with Aisha, fresh and old lesions, and even latent infections without disease symptoms and branches and stems from Mycelia inside woods and standing trees. So therefore trained dogs could be used to detect C-Pyronis in the field to offer a faster alternative to existing time-consuming detection methods, such as genetic testing. However, many factors are likely to affect detection efficacy and efficiency, including for example, environmental conditions, forest size and tree density. Therefore, in Eloise's opinion, for the research and larger sample sizes would be beneficial to test the viability of this method in real-world forestry context. And for example, again, I so Eloise prepared this, which we're very thankful for. I'm not sure what the actual setup was of this forest test. You know, was this kind of a realistic size? How closely were the dogs actually being guided to investigate the trees? And, you know, definitely sounds like useful as a first step, but it'll be interesting to see how that translates into actual field work. So, okay, Liza, welcome to the podcast. Do you want to give our listeners a brief introduction to you and how you ended up getting involved in detection dog research? Sure. I am currently a third year PhD student at Texas Tech University in the Canine Ulfaxima, the Nathan Hall. And I originally started doing detection work when I was in college. I must have been 20 or 21. It was summer before my senior year of college. And I interned at Penn vet record dog center, helping them out with their research. And so mainly that was the air and cancer detection at the time. And I loved it at that time. I had no idea that I always loved dogs, and I had no idea that this was even a career that people had. I thought, okay, like dogs either event or just a dog trainer. And I realized that there was this whole world of research as well that wasn't just like more medical based. So, from there, I decided that I wanted to work with dogs in a research capacity. And then I got after I graduated college, I got a master's at Hunter College, and kind of pivoted a bit. And I did work, cognition work, and I got my master's in animal behavior and conservation. And then for my PhD, I'm now back to Elvactuar. And I've been in Texas for three years, working with the wonderful dogs in our lab. And they're all very special in their own ways. And, it's definitely a great experience. And I'm very excited to continue learning more about the field, and basically groaning within this field as well. Yeah, really cool. Yeah, I didn't realize that you were another Penn vet alumni. So yeah, we've had several Penn vet alumni on the podcast, obviously. And then what if you from Dr. Hall's lab as well? So we've really got that pedigree going for you. And I'm also, I'm a third year PhD student as well. So I think we're in probably a similar think of it phase going at this point. It's really not. Congratulations on already having a paper out because I don't. I will. Yeah, I did. I better. I don't want to make PhD student money for the rest of my life. So why don't we talk a little bit about how this study came to be? So as for listeners who maybe it's been it's been eight minutes already since this episode started, we're talking about exercise effects on olfactory acuity. I suppose you'll suburrient a better than I do. How did this actually come to be your project and who's who's asking it? Why are we interested in this question? So this project when I started my PhD, there were two projects that I could have started with. And this one was really just out of convenience. I was around during the time, during the summer when this project was starting. So that's really why I was put on it. And this project is based off of where it's a continuation of a paper that my lab published. I think at this point, it was in 2024 where we did a similar paradigm where dogs were exercising, but we wanted to see one of the things that we were interested in seeing was how concentration and duration affect their performance. So duration of exercise, concentration of odor, and we did find that concentration of odor and duration of exercise do affect the dog's performance, but especially when they are doing a more high intensity exercise. So we wanted to build on that and we wanted to add in more physiological measures. So we wanted to know, okay, yes, we know this, but what are there some predictors? Would we find out whether if the dog's queried by temperature is a certain degree? Does that affect how well they detect the odor? Or their heart rate is high. Is that a predictor? Okay, my dog's heart rate is above a certain amount. Does that mean that the dog probably will not be able to detect the odor? So we basically wanted to do a similar thing to the prior study, but we wanted to add in these predictors. And then for this study, we kind of tweaked it a little bit. We wanted instead of booking out the different concentrations. We already knew the concentration did affect their performance, so we wanted to look at just threshold concentration. So we wanted to see when these dogs are really pushed to the limit in their defection abilities, how does it affect them, and what is physiological what is it? geologically is going on during that. - Yeah, definitely. And what are you kind of, you and the rest of the lab envision as the application for this? Like, are we hoping that we can find a physiological measure or duration of search where maybe then a recommendation to handlers could be given about how to address that is that kind of what we're thinking? - Yeah, so if everything was in science as perfect, there would be okay, certain dog temperatures out of certain temperature, okay, then the dog should not be working anymore. We did find out, I'll talk more about study, but we do find out the study, it's not that simple. - Um, I know, right? - Yeah, so, but we were like, that would be great, we wanted to see if, because he knows, we get, we have to cries every day by what we find. But yeah, that was not necessarily the answer. We want to basically see why are there certain predictors and can we tell handlers, okay, your dog and has been working for this long, has been exercising a certain pace, something like that to see, okay, this is, like you said, the recommendation of a pig, your dog really is not fit to work right now. Your dog, like, we will not be detecting the target and depending on what that is, whether it's explosive detection, conservation, narcotics, matter what it is, that can't have detrimental effects. We want to basically give some kind of recommendation to handlers where we can tell them, okay, your dog needs a little bit of a breath. - Yeah, wonderful. So we're trying to figure out where exercise and action kind of come into conflict with each other. And obviously part of our problem here is that when our dogs are searching, almost always they're moving while they're doing that. They're trotting, they're, you know, in the case of some of our dogs, they might be moving a little bit faster than a trot. And they're probably covering quite a bit of ground, but they need to be doing that while sniffing, which involves sniffing and respirating, kind of coming to conflict, eventually as your dog is trying to both keep, their muscles fully oxygenated and actually bring odor molecules through their nose. So yeah, what is the actual setup here? Because I imagine if you want to get these like precise measurements of various things, you're not actually just having a dog run around a field. It's something a little bit more controlled, right? - Yes, so the point of this study in particular is to create a laboratory model where we can measure all of these very, very precisely. Whereas, yes, it'd be great if we can do something more uncontrolled and more what you would see in the field. But in order to figure this stuff out, we really did need, we do need more of a controlled environment where we can control as much as possible. So for this, we recorded all of our sessions. We had two cameras. We had one facing the dog. And then one, that got a profile view of the dog. So then we also, the our main measurements that we used for this were, we created a Ulfat doing treadmill, which was used in our prior paper from 2024 where it's just normal dog treadmill, but we mounted an Ulfat Tabitor. And then we trained the dogs to both walk and trot on the treadmill for 100 trials. And they would simultaneously, order would be delivered and they would sign, simultaneously poke their nose into the port if they detected the odor. - Sure, yeah. - That was how we controlled the odor portion. - Yeah. - And then for body temperature and heart rate, we used these really cool pills. Now, are their polymetric pills that the dogs ingest and they take the temperature of the GI tract? So those were really cool. And so that's how we got precise temperature measurements. And then we put a polar heart rate monitor on the dogs to get heart rate. And then we also used Pays, so we kept the pace very consistent. So walk and trot were always the same pace. It was at 2.5 miles per hour for the walk and by-moves per hour for the trot. - Every dog of course has a different date when walk and trotting in a different natural pace. That is what was recommended to us to do. So we did choose those and we kept it pretty consistent for all the dogs. - Yeah, that makes sense. I mean, this is one of the things that I think is both interesting and really challenging about doing science with that involves dog training is, at some point you kind of need to decide whether you're going to say, okay, we're going to set the treadmill at the speed that helps each dog trot. And then, okay, how do you define a trot and how do you make sure that that's equivalent effort for each dog? Or we're just going to say, okay, five miles an hour that should get all of the dogs trotting more or less. And maybe our short-legged guy is working a little bit harder than our longest-legged guy. But, you know, yeah, you kind of do have to make those calls at some point because you have to go one way or the other. - Yeah, and it was definitely for, we had a mix of a bunch of different freights on this study. And definitely for our, we had two sprinter spaniels. And it was definitely, they had to maybe run a bit faster to keep up, but they are pretty quick at the big, pretty fast anyway. So, both their little legs, which is kind of trot very quickly. - Okay, so I'm imagining, and I've actually seen video of this before, I think, for one of Dr. Hall's presentations, where, so we're imagining a dog's trotting on a treadmill. We've got an olfactoryometer in the front. When the dog smells the odor, that target odor, they do a nose poke. And I think I remember in the paper that there was different durations of nose poke for the different dogs, but they each had like a duration that they were trained to. And then kind of the idea is, okay, as we exercise them more or bring the odor concentration down, which is the amazing thing that we can do with an olfactoryometer. At what point do we know that odor is being presented, but the dog is not successfully nose poking. That's basically our setup, yeah? Cool, yeah, and we've got video of it. And I know in the paper, and again, if anybody gets the chance to see anybody from the lab talking about this, there's some cool visualizations where you're actually were able to put some sort of visual something out. I don't remember what it was that let you kind of see what that odor looked like over the treadmill. Yeah, what is it? We have a powder that we put into the olfactoryometer. And when the odor was released, it showed basically the powder came out of olfactoryometer and showed the odor. Cool, yeah, it's a really, really interesting visual. So how did you end up with the dogs for this project? Was there any specific recruitment who ended up being on your team of subjects? So this project had four dogs. We were hoping to build on the last study that we did about this was kind of a continuation of how three dogs were. Like we need more, then we ended up with four. We, as you can imagine, getting dogs to go on a treadmill is not always the easiest, and we wanted to make sure that the dogs enjoyed it. So we probably tried out 10 to 15 dogs and only four were what we ended up with. Yeah, because we had a bunch of dogs that would do it. And then once we'd make it a little harder, and they would start going a little bit faster, they would just check out and they could, OK, I'm good and wanted to jump off. Yeah, we wanted dogs that loved walking. Yeah, and felt comfortable on that treadmill. Yeah, it's loud too. So it is. I've been teaching. I got a dog treadmill off Facebook marketplace last fall. And I've been teaching my dogs to use it. And one of them picked it up basically right away. And if flare my younger border collies, just like, oh, I get to go fast on this thing without you taking anywhere. Great. I'll do like he immediately. I mean, and not quite immediately, but very, very quickly was comfortable on it versus barley. It has taken him months and months to get to the point where now he will like he can walk for about a minute on it without showing signs of distress, like, whiskers flared, pupils dilating. And we're not asking dogs to do anything under big amounts of duress. But there was free to leave. I know I saw that in your paper. But even so, it can take a long time of positive reinforcement to try to get over that. And I understand why that wouldn't make sense for this study. You really want to dog who takes to it naturally. Yeah. And the training is what takes so long. So if it's also part of it, we have death lines. And we have a dog that is progressing slowly. We can't always include them on the study because we need to make certain death lines. But so for this study, we had four dogs. One was a mixed breed. And we have two springer spaniels and a malinois. So the mixed breed is our current postdocs hares dog. He's on basically every study that we do. He's the dog that we always call in. And when we need a dog, we know no's odor. And he also just loves the trust. mill so much. So that was never a worry with him. You're like, okay, you know we have Charlie. And then we did, where our studies be collaborated with Paul Bonker, where our working dogs and he recruits or procures working dogs for us. Most of the times they're rejects from different programs like TSA, DOD. So we got from that, we got two Springer Spaniels and an owl law. And so we kind of had a mix of different breeds and sizes. Our Malamaw was very tiny Malamaw, one of our Springer's Tuma. He was obviously very tiny and then I think he sits around like 37 pounds. Yeah, he's small. Yeah, and then Boomer was 45 ish pounds and then Chelten's Charlie's about 45 pounds. So we had a mix of a bunch of different dogs. So once we were happy with the four that we had, I mean, yes, we would have loved to have more. We didn't want to keep on pushing it. We were like, okay, we need to have your four because also it's a lot of just effort to train the dogs. Training the dogs takes so much longer than the actual study. Of course, yeah. So, trying to get a lot of the dogs really take to it very easily and just as long as you're giving them food, they will happily do whatever you want on the treadmill. Some the minute that the starts moving, they jump off and you get a slowly rick into that. So with all of the four dogs that we had on the study, they definitely showed the least amount of any kind of fear out of what we saw from the dog. Some of the dogs we that we tested to see if they would be candidates for the study, they showed a little bit of fear and we try to work through it, but for most of the dogs, we were just okay, you don't seem happy on this. You want to jump off. So even though we have a small sample size, it is we're very happy with it considering how long the training does take. Sure. I mean, for in like animal behavior studies that require this amount of work is not crazy. You know, it's really hard to get much more than that and we do see it. We definitely do. There are labs that are really good at getting together bigger studies than that, but yeah, or it's pretty good. So let's talk a little bit. We've talked a little bit about the treadmill training. Did you introduce the old-factorometer and the nose poke to the dogs separately? And then kind of bring it back together to the treadmill and how did that go? So pretty much all of the dogs on the study, I think actually I think all four were already bullfactorometer trained. So they already understood the old-factorometer just not in the treadmill setting, but they already knew our said well, the old-factorometer is now going on to it. So they did a lot of them on that we generalized to I see a nose poor I put my nose in. All of our dogs had already known our target owner except were boomer and we did slowly teach it to him on the treadmill and then so he kind of did the opposite of the other dogs were healing it all the treadmill first and did other studies with that. Whereas most of our dogs will do our said well first or like the panels first and then move on to something like the treadmill if they're going to be one of these studies. So mostly the old-factorometer training was separate, but not necessarily because it had to be just because most of these dogs were on other studies that they were already trained on the old-factorometer. So what we would do to start is we would bring the dog onto the treadmill and just no movement and then try to get the dog to nose poke and with the odor we turn the odor on turn the odor off and once they seem to get that then we'd start slowly increasing the pace of the treadmill and once the dog successfully does the walk pace then we'd slowly increase the trough pace and then alternate and see make sure that they could do both paces and depending on some dogs catch on fairly quickly, but for some mistakes the dogs. Sure, yeah that makes sense. What was the actual setup of your trialing then with these dogs as far as you know the physiological measures? Did you have the dogs working at a given duration? How did you experiment with that intensity and what was the setup there? Yeah so we had two experiments as part of this. So our first experiment it was just one pace the dogs either walked or trotted on the treadmill and they did this for 100 trials. It was a fix duration but it was 33 minutes and 30 seconds and so the duration was pretty standard so when the odor was on it would be on for up to 10 seconds. So if the dog alerted then it immediately would go to the inner trial interval. If the dog did not alert then the odor would be on for all 10 seconds. So each trial was 20 seconds so whatever that additional remaining time was that is what the ITI would be and then did this 100 times for the first experiment. Okay so let me make sure I'm getting this right. So if the dog alerted within one second of the odor being presented then there would be like 19 seconds of blank and then the next odor is presented versus if the dog alerted on second eight then there would be 12 seconds of inner trial interval that ITI and if the dog didn't alert at all then the odor would be presented for 10 seconds, 10 seconds of ITI inner trial interval just kind of blank nothing odor and then the next one would start. Am I getting that right? Exactly. Okay great thank you. Great so that's that's what's going on and then sometimes they're walking sometimes they're trotting we've got their heart rate monitors on we know their internal body temperature from that like swallowable capsule. Are we experimenting or like playing with duration of trotting, duration of walking, the order, am I getting my papers confused? No you have your papers so for our first experiment we only did one case and then we found for that experiment that temperature was the number one predictor of perform. Temperature is a covalent with duration because the longer that the dog is trotting the harder they're going to get. So we wanted to try to separate that and we wanted to do a mid-session case reversal so then for experiment two we did 140 trials so for that the dog was trotting for 70s trials and walking for 70s trials. So I'm obsessed with dogs to start with the walk some sessions with the dogs with the cart of the trot and so we wanted to kind of separate the temperature from the pace so if the dog is going longer but they are out of walk and but they were originally out of trot then how does that affect the results? Right yeah a lot of the dogs were pretty confused with that because from them it was okay once the dog is starting with a trot and moving to the walk the dog was new from the prior experiment that okay I know once the terminal slows down that the session's over and they were all very confused. Oh yes so they come like wanting to hop off. Yeah yeah they're like okay so it's done now and you know you were actually just changing the speed I see. But for that it was 45-ish minutes. For 140 trials so it was still the fix duration but because 140 trials it was longer. I mean it chose 70 trials per pace because you want it to be a long enough duration at each pace where we might be able to see effects of the pace. Yeah that makes sense. So yeah what did you end up finding? What did we actually see then as far as our our results for these dogs? Yeah so what we found was that it was not necessarily temperature so we found that when dogs started off at a slower pace and then went to the faster pace their performance their sexual performance immediately collapsed. But starting off at the faster pace and then switching to the walking and we're slower pace then their performance stayed fairly steady throughout session. Yeah so when dogs started the truck and then moved to the walk their pace stayed pretty steady throughout. It was not as high as when they started up the walk so during the first half of when they were walking it was of course not as high but it was pretty steady and they were still getting a decent amount correct. We basically found that it was not temperature it was more so a pace and the duration that really did affect the dog's attraction performance. Gotcha so the longer they were on it if they were trotting more they were less able to detect that odor when it was presented to them and remind me I can't remember is the odor always being presented in the same way or was there shifts in like that parts per million of the odor. So it was consistent during each experiment the virtual experiments with during each experiment we did keep it consistent so what we think found was their second experiment with the paste reversal. We did find that the temperature was not necessarily the predictive factor, but it was mostly the exercise intensity and the exercise intensity and duration interaction that mostly affected their performance. So when dogs are walking pretty good performance trotting performance is not great, but that's also duration. So when dogs are walking the duration doesn't not affect them as much. When they're walking they have pretty steady performance throughout the session, but when the dogs are trotting a very slowly threat the session, but mostly by around the 21 minute mark they tend to have a steeper decline. Interesting. Okay. So yeah, we kind of see that. Yeah, they can walk for longer and maintain odor like detection performance versus with trotting. There's kind of a steady decline until about 2021 minutes and then we see a bigger drop. Interesting. So yeah, what do you think are kind of your next steps for this sort of research? What do we think are how do we implement this? Because I know I'm thinking about God, like the some of the surveys I do, we're out there for like two or three hours and the dog is moving the whole time, probably trotting most of that time. You know, they're allowed to speed up and slow down as they so choose, but most of my dogs work at a trot or again, in the case of some of the spaniels a little bit faster even. And you know, we're still seeing good performance out of them. So I've been trying to figure out like how do we actually implement this and what are our next steps? What do we, what can we do with this information so far? So with the experiment too, when we found that more moderate and intensity trot into the walk. So start off the trot when moving into the walk. Had a better performance outcome at the end of the session, then start off the walk and move into the trot. So what is meant to suggest is that potentially it is best when you are working a dog to maybe start your dog off with the more intense task and then switch to a task that lower intensity. Interesting. Yeah, yeah, I wonder, yeah, that makes sense and that does feel like a possibly actionable part of this. Yeah, and what are what are you looking at next? Are you going to continue working within this exercise realm or what else are you hoping to work through as part of your PhD? Yeah, so next steps are to look at this more operationally. So when dogs are able to regulate their own pace, how does that affect their performance? So this is very much just a laboratory setting, but we also are very interested in something more applied. We got a lot of people telling us that, okay, things that results are really cool, but I don't think my working balance would necessarily do this and show a decrement of performance. So we also want to look at breed and whether working breed versus say, like a pet dog has any effect on that. So there are a bunch of different things we can work look at. One of the main thing really is the operational lookability of the study and bringing it out of this laboratory setting because it can do so much for us, but for handlers, the goal is to have it out to them and understandable to them. So when it's just a laboratory model, it's hard to tell handlers, okay, this is what our results show. You really want to show it in a more operational way. Yeah, yeah, definitely. No one I'm excited to hear that yeah, there's so many different ways that you can take this and you're already thinking about a lot of the same ones that I'm sure a lot of our audience members are thinking about as far as gosh, how does this vary with breed? How does it vary when the dog is able to control their own pace? Yeah, that makes a lot of sense. So anything else you're like really excited about, really wondering about like what what else kind of is tickling your mind with this study or future studies that you're excited about? I think one of the big things for me is the operational lookability of this and how we can get this out to handlers. I really, something that's important to me is for open communication between handlers and academics and I think things are changing, but historically it's very okay, we do the lab work, you do the dog work, we don't talk and we don't necessarily bear with each other. So I think that is something that hopefully many studies in this study included help bridge that gap how to apply scientific principles to more handler based roles and how that can help the dogs and also I think research in general should maybe like yes, the lab, laboratory portions are very important for understanding concepts, but with the dog research, maybe we get more operational and applied maybe we can bridge that out. Yeah, absolutely. No, and I'm excited to hear more about that and more about kind of how to continue connecting these in the future and I know I think I've really started coming around on these these foundational studies where it's like man, how do we take something that involves a treadmill and an olfactometer and apply it to what you do in the field for real with your dog, but like this is where you have to start to figure out what these baselines are and figure out what it looks like when it's under its most controlled and I'm so glad that labs like Dr. Halls exist where they're able to start looking at those questions because when I'm looking at like you know, I might try to run a model on my detection dogs in the field in Alaska seeing if like scats found per our decrease or per minute say decrease per duration of the search to as kind of a weird proxy for this, but there are so many reasons that we could find an effect or not find an effect that has nothing to do with exercise intensity and we would have no idea whether it was pacing or temperature or duration, any number of things. So I I'm so excited that there is work like this that comes out of the lab that yes, it's really limited in what we can actually apply with it, but it gives us an idea of what might be happening out in the field where we have so few variables under control and you know, we really need both. Yeah, and I'm also very interested in learning more about maybe humidity and outdoor temperature and how that would affect a search or especially conservation dogs and maybe military working dogs for working outside and there are probably our other factors that we can't account for, but that might be very important. So trying to think of all the possible doctors that we can put into models even if we can't experiment like for them, I think for us. Yeah, absolutely. No, I'll be really curious to hear about humidity and temperature as well because I I feel like one of my dogs performance decreases when he's really humid. I think I feel like he doesn't cool off as well, but I can't prove that. I know it's yeah, don't I and even when you're just trying to do field trials with a dog over the course of like day after day after day and you keep track of all your weather variables, it's still really hard to pull up. It's so much harder to pull apart those different relationships when you're not able to just control things. So well, Liza, thank you so much. Yeah, thank you. This was great. I had a lot of fun. Yeah, excellent. Well, I hope that everybody, you know, keeps up with the Cannot All Faction Lab and, you know, I'm excited to see where you and the rest of your career goes. Do you have social media or like a research gate or anything that people should follow or just keep up with Dr. Hall's lab? I do have a research gate and the Google Scholar page channel and that and then I'm also on our website. Perfect. We'll make sure to link that in the show notes. Yeah, thank you so much for coming on. This was a lot of fun. It's exciting to get to learn a little bit more about some of Tumos past work and, you know, all of the other dogs as well. And for those of you at home, thank you so much for listening. I hope you learned a lot and you're feeling inspired to get outside and be a canine conservationist in whatever way suits your passions and skill set. Our podcast is edited by Sarah Fankton, our science communication intern is Evelyn Combs. We've got a whole bunch of volunteers who help get this podcast polished and published. Brooks Shroeder is on guest logistics. Meg DeBray is on the website. Madison Davis organizes our coaching calls and you at home listener can find all of the resources you need from canine conservationists at you guessed it canineconservationists.org. Until next time, bye.