Episode#18 Fear Conditioning – Dr. Ian Jacobs

Dr. Barkha Yadav-Samudrala: 0:06

Hello, everyone to another episode of NerdRX podcast and I'm your host, Barkha. Today we have a behavior technique called fear conditioning. And we have an expert who has done a lot of fear conditioning in his graduate work, and he is my lab mate and also a good friend, Ian Jacobs. Welcome Ian to the show.

Dr. Ian Jacobs: 0:31

Hi, Barkha thanks for having me.

Dr. Barkha Yadav-Samudrala: 0:33

It's my pleasure. Thank you so much for giving us your time en. So before we jump into fear, conditioning, dreaded fear conditioning, why don't you introduce yourself to our listeners?

Dr. Ian Jacobs: 0:47

Yeah, so my name is Ian Jacobs. I have a master's degree in experimental psychology. And I'm working on my doctorate right now and behavioral integrative neuroscience. I'm a fifth year, sixth year now graduate student at the University of North Carolina. And so I hope to graduate will be Dr. Jacobs here. And I just

Dr. Barkha Yadav-Samudrala: 1:06

that has a nice ring to it, Dr. Jacobs.

Dr. Ian Jacobs: 1:11

I've been doing fear conditioning for all of my graduate career, I started out in a lab at Appalachian State University where I got my master's, where we were studying more like learning theory type of approach to fear conditioning. And then I moved on to UNC Chapel Hill, where we're looking more at how your condition can be used in the examination of brain circuitry.

Dr. Barkha Yadav-Samudrala: 1:34

Yeah, awesome. So my first question to you would be, what is fear conditioning, and I have seen you do this in the lab. And I honestly don't know how you do it. What got you interested in?

Dr. Ian Jacobs: 1:52

Sure. So with fear conditioning, it's a form of classical conditioning. And so with classical conditioning, you have a few pieces that are all in play here, you have first something called an unconditioned stimulus, which is just some sort of event in the environment that occurs, or that is detected by the animal, which results in some untrained reflex from that animal. So since we're talking about fear conditioning, right, it might be a you know, loud noise in a movie or something like that, that causes you to start All right, so when the ghost jumps out or whatever, right, so you start with that unconditioned stimulus. And then what you can do is you can take some stimulus that's neutral, and does not provoke any sort of response, in the case of fear in this case, right. So like relative to fear, the stimulus doesn't produce a response, like something like a puppy dog, which hopefully none of us are terribly afraid of. And so then, by pairing, the neutral stimulus with the unconditioned stimulus, the puppy dog with the startling ghost, right over and over, and over and over again, the idea is that a conditioned response comes to be learned about that neutral stimulus, where now they will produce a fear response to the neutral stimulus. And now we call it a conditioned stimulus, where it's a conditioned response. And so that's kind of the basics of fear conditioning acquisition, where you're actually acquiring those fears is, you know, you take these neutral stimuli new parent with unconditioned stimuli, it can be as few as just one trial is necessary to do it. But typically, in experimental settings, they tend to do three to four presentations of the stimulus pairing before they actually move the animal on to future phases, assessments. And so how did I get into it? So, I first heard about this in a learning class in at Appalachian. And I just really enjoyed, you know, at the time in my naive understanding of it, how clean it looked, right? So you just pair these two stimuli together, and you watch the conditioned response measurement go up, right? It's very straightforward and that sense. And then, you know, I gotten to the lab with the the professor for that course, Dr. James Dennison. And he took me on and showed me all of the more intricacies of classical conditioning and how you can change all these different experimental parameters to adjust the conditioned responses. And I realized that, you know, viewing it as simple was, you know, of course, like I said, naive, and I came to appreciate the vast complexities of fear conditioning and just exploring all those different little cracks and crevices of it. So fun.

Dr. Barkha Yadav-Samudrala: 4:31

Wow, your idea of fun is different.

Dr. Ian Jacobs: 4:36

Maybe fun isn't the best word. It's interesting.

Dr. Barkha Yadav-Samudrala: 4:39

Yeah, for sure. It is definitely interesting to see that. So why would you use fear conditioning in rodents? Like what are the applications of it?

Dr. Ian Jacobs: 4:50

Sure. So at its core, right, because you're pairing that neutral stimulus with an unconditioned stimulus. What you're actually looking at with classical conditioning is a learning process. And so you can, anytime you want to assess how an animal is learning, right, you can use any number of things, you can use operant conditioning, where you're training them to receive some reward. Or you can use, you know, classical conditioning where you're just trying to train a particular type of reflex. And now the benefits of classical conditioning is that it doesn't rely on what's called motivating operations, for the most part, to actually produce that response. And so what a motivating operation is, if we roll it back for a second, you know, if you were to present somebody with, you know, a piece of cake after dinner, right, and, you know, Parker, what's your favorite flavor of cake?

Barkha Yadav-Samudrala: 5:47

Anything with fruit,

Dr. Ian Jacobs: 5:49

anything with fruit? Okay, so we have a nice, I'm not gonna say fruitcake, because it has, you know, holiday implications, but a cake with nice fruit on it, maybe like a strawberry shortcake, right. And I feed that to you, and you eat it and okay, you loved it. And then I feed you it again. And again, and again. And again. And the fervor with which you consume the strawberry shortcake at the beginning, is much stronger than you would buy your sixth piece of cake, you might even straight up refuse the piece of cake, despite the fact that you love strawberry shortcake so much, right. And so that's a motivating operation, where by satiating the person on strawberry shortcake, we've now changed the value that strawberry shortcake has to the individual. Right. So now if we roll this back to an experimental setting, if you're talking about how motivating operations can change how rewards function for an animal or a subject, than any changes in the animal's brain, due to disease, or any sort of transgenic manipulation of the organism, that changes how they process rewards, or how they process motivation to get rewards, is now going to be affecting your measurements of learning. Okay, so to get around that, what we do is we say, okay, let's use fear conditioning, fear conditioning, and essentially involves a survival reflex, right? In the wild, if an animal if a mouse is out there in the field, and it sees something that it should be afraid of, it produces a freezing response, where it's going to try to preserve its own life by limiting, its, you know, I guess, detection by other creatures by predators, right. So it's a survival reflex, when it sees something fearful to do that, and survival. I'm sure maybe some people might argue this, but I think that survival is not necessarily a subject to motivating operations in the same way, or at least to the same degree that reward based learning is, you know, you're always going to want to survive. So then, since we're separating the learning from reward systems, we can then study these different brain regions and these structures that are involved in associative learning, with a greater fidelity than we could if we use something that was more vulnerable to environmental manipulations. So that's why you would use fear conditioning over some other type of procedure to measure learning. A second reason to use it is just kind of how clean it can be right? Like, you're not, when you have something like an operant conditioning paradigm, or, you know, an appetitive classical conditioning paradigm, you're again relying on the subject to actually experience and consume that reward or reinforcer. Whereas in fear conditioning, you're not really giving them an option, right, they receive the fearful stimulus, whether or not they want to or not. And so you can kind of really force your experimental conditions on these animals without needing to do much intervention, really. So yeah, that's why I would use fear conditioning specifically over other types of, you know, methods for learning.

Dr. Barkha Yadav-Samudrala: 9:01

Okay, yeah. Thanks. That was a very nice and easy to follow explanation.

Dr. Ian Jacobs: 9:07

I felt like I was rambling there for a bit. No, no, thank

Dr. Barkha Yadav-Samudrala: 9:10

you so much for that. So would you mind taking us through the steps involved here? Like, how would you go on to set up this experiment? And how long does it usually take for you to run this entire experiment? Sure.

Dr. Ian Jacobs: 9:26

So with fear conditioning, the setup you need to do is you need to have I'm going to talk specifically about fear conditioning involving foot shocks. Okay. So there are other methods, which I think I'll talk about later, that you can use to assess fear learning, but my main experience is using foot shocks and operant conditioning chambers. So I'm going to come at it from that perspective. So all you need first is you need a operant conditioning box. You know, you can the original ones were built by BF dinner. But you know, you can just buy it for any any, you know, retailer that would sell these, and you have a electric grid floor that is down in the bottom of it. And so what this grid floor does is it passes an electric current, a very mild and you know, not painful current through the floor to the animal. And that requires animal of course to be on the floor, which you know, for rodents is not always a given, but for the most part, they tend to sit on the floor. And so when I say it's a, it's delivering a mild foot shock, I don't know if Parker you've ever had like those joke pins that, you know, oh, yeah, press the button, and it shocks you a little bit. And it kind of just feels like a vibration. So that same sort of feeling is what we're going for when we try to put these electric shocks to the ground. Right? Not a painful one. But just something you'd rather not press that pen cap again, right? I just want to experience that again. Yep. So you have these these animals in the box, and you have them with the electric shocks going on. And then you get to choose how you want to present stimuli to them. Right? What neutral stimulus Do you want to present to these animals? One way to do it, is to simply use the place that the animals in as the neutral stimulus, right, this operant conditioning box, maybe produces some hesitation at first while they explore it. But there's nothing terribly frightening about being in a box. In fact, most rodents tend to enjoy being in dark boxes, it's pretty safe for them. And so you then would give them the foot shock in this setting. And they learn now that the neutral set, stimulus of the context is now being paired with this foot shock. And so now you have contextual fear conditioning. And so that's going to engage a particular set of brain regions, if you use that one, what you can also do is something called cued fear conditioning, where you're going to present what's called a discrete cube, a cube that has a beginning and an end, something like a 62nd tone, or flashing light, for example. But you can turn on and off, you then present that that stimulus before the actual foot shock happens. And then the idea is that you will then pair that discrete cue with the foot shock causing cued fear conditioning to occur. And the conditioned response would happen to the, you know, now the discrete cue. And again, that's going to involve a different subset of associated learning circuit circuitry. So you can kind of use these different methods to parse that out. Then what you can do right is I mentioned that there's different modalities of these these stimulus stimuli that you can use, you can use visual auditory, something that's kind of difficult to do well, but would honestly be best for rodents, in particular, is to use their primary sense, and to try to use olfactory cues, right odors, to as the neutral stimuli in order to, you know, make sure that they detect and are able to see those experiences stimuli. And so then what you can do from there, right is that's just the process to acquire the fear, what you would do is you would then pair those, again, it depends on your experiment that you're trying to run, you could pair them one time or multiple times. And then you would say that you have acquired a conditioned response. So in this case, now, what you're doing is you're gonna then extinguish that conditioned response if you wanted to take this a little bit further. So whereas in what we call acquisition of that conditioned response, you are pairing the neutral stimulus with the unconditioned stimulus in extinction, you're just presenting that neutral stimulus, or what's now a conditioned stimulus alone by itself without any unconditioned stimuli present. And so the idea then, is that over time, they would learn, it's a form of new learning, right? It's not just an erasure of old learning, but it's a new learning that that stimulus, which they thought was a good predictor of this unconditioned stimulus happening is now no longer a good predictor. And so it can be more or less ignored, and the conditioned response to that conditioned stimulus will decrease over time. And that process is called extinction. Okay. So now you have a fear that you've acquired and treated. And so now the question comes, what else can you do with this? Well, one of the my favorite activities to then do is to study relapse after this. And this has a lot of implications for drug relapse, relapse and drug addiction. And what you do is you can then, you know, manipulate the different context, what sort of stimuli, the subject experiences, and you can get different types of relapse. I can talk about those a bit if you want, but I think we're kind of getting into a longer subject, but briefly, okay, so the first type of relapse is just called RE acquisition, right? This one's kind of unsurprising and boring. Re acquisition is just where you after you provide an extinction treatment and the animals treated for their fear. You then present the neutral stimulus or the conditioned stimulus us together with the unconditioned stimulus again. And then you test them to see, if they're afraid of the conditioned stimulus spoilers, they will be. Then you have spontaneous recovery, which is just the idea that over time, the animal will spontaneously recover the conditioned response. And this happens over the course of just 24 hours, right? So between extinction sessions, you can actually view the spontaneous recovery occurring in your animals. It's just where they're conditioned response will be higher at the very beginning of the session than it was on the last trial of the session before that spontaneous recovery. You have reinstatement, which is kind of like re acquisition, where you're just you're just going to present the unconditioned stimulus again, by itself, right. So you're not pairing the neutral stimulus with the conditioned stimulus. With this unconditioned stimulus, you're just presenting the unconditioned stimulus alone. And the idea is that then when you present the animal with the conditioned stimulus, that test following that reinstatement treatment, they should now have the conditioned response to the conditioned stimulus return, despite never having been trained with that, again, like they weren't in reacquisition. And finally, you have renewal, which is a contextual form of relapse. And so this one kind of requires us to back up a little bit. So this is why this one got to be a little bit long one in acquisition, right, it's known that the first learned information tends to generalize the best. What I mean by that is that if you learn that fire is hot, and it will burn you, if you touch it, right, you're going to apply that learning to everywhere that you see fire, right, every situation you see fire, it's safe to assume it's going to burn you. And that makes real, real good sense to us. Now, cases where it doesn't make quite as much sense, as let's say, you know, a neighbor, dog, you know, bit you when you were a kid, and now you're afraid of all dogs. Yeah, that's a little reasonable, but maybe a little bit of overkill as far as your responses are concerned. So that's what we mean by the first one information tend to generalize the most, right. So then when you provide an extinction treatment, right, so you provide this acquisition treatment, or this acquisition process in one context, right? If you then switch context and put the animal in a different place, and you provide them treatment for that fear in that new place, the learning about that arrangement, right about that stimulate that extinction becomes specific to that place. So then when relapse occurs, if you simply move the animal outside of that extinction context, that place where they received extinction, you will see a return of the conditioned response. You know, you haven't presented them with the fearful unconditioned stimulus, again, you haven't paired anything else, again, you simply move them outside of the extinction context. And because information learned in acquisition, that first learned information generalizes, the most readily, whatever context you put them in, they're going to pretend like they're afraid or I mean, actually be afraid right of that stimulus. And so that's how relapse renewal works. It's just that change outside of the extinction context is going to cause renewal. Now, what's kind of interesting about that, just to kind of continue this little spitballing, yeah, it doesn't just have to be physical places, right? You can also use what are called interoceptive contexts. So a place that you're in right, the room that you're in right now is something called an extended receptive context, it is external to the individual interoceptive contexts are things that are internally to the individual, whether or not the animal is under the influence of any drugs, right? Have you hopped that mouse up on on cocaine or not? Is it hungry or not? Right? Like, how have you fed it in the past 24 hours? Or is it starving, and now that's a different interoceptive state than when it was hungry. So if you were to train a fear when they were hungry, treat them when they were satiated on food when their bellies were full, and then return them to a hungry state. They should show renewal. So it's all kind of interesting like that, that you can do both physical places as well as states of the animal acting as relapse triggers.

Dr. Barkha Yadav-Samudrala: 19:21

There are so many variables to this. I can imagine so many things going wrong.

Dr. Ian Jacobs: 19:27

Yes. The kind of crucial thing about all of this is to do everything the same for every animal all the time forever. And that's a big ask for humans. Yeah, for sure. So one last thing, you asked me how long it would take? Yeah, so for acquisition, that typically takes a day, right? You just gotta provide that way. Even if you're doing multiple treatments or assign multiple trials, or just a single trial. It's only one day extinction typically takes longer. In rats. We got good extinction. After a week, when we used for foot shocks, the length of extinction is going to depend on a number of different variables, including, you know how intense your unconditioned stimuli is, how intense your neutral stimuli are, you know, the biological disposition of the animal, tons of different stuff are going to affect the extinction. And then yeah, and then so extinction takes you roughly a week, we'll say. And then the relapse test is just a day after extinction. So used up prior to reinstatement treatment, right, you would just spend one, day two or reinstatement or re acquisition, you would spend one day to present them with that either neutral unconditioned stimulus or just the unconditioned stimulus alone. And then the next day, you would test them, or in the case of something like renewal or spontaneous recovery, you just simply have to place them back in the, in the box the next day after extinction. And you should see the relapse occur there.

Dr. Barkha Yadav-Samudrala: 21:01

Okay, I didn't know you work with rats as well. So in your experience, are rats better? Are my

Dr. Ian Jacobs: 21:08

rats, 100%? Rats? Mice, I actually learned about recently in my, my dissertation, so mice largely come in inbred strains, right? Yeah. And depending on which strain you use, you may or may not see all of these different things that I've been talking about. I'll talk about a little bit later, but kind of, you know, before I read these experiments, I kind of thought this was a more universal feature, I guess of life. But you know, it does, because because, you know, the the mechanisms on the the biological level and the cellular level are preserved everywhere from a pleasee up to humans, which is what, you know, Candell won the Nobel Prize for back in the day. And so to hear that, oh, it just depends on the particular mouse strain that you're using, whether or not you'll actually get, for instance, renewal effects, was a bit surprising to me, in general, to rats just tend to connect things in their environment more readily than mice. Mice tend to take longer and need more obvious stimuli than rats. So yeah, I would, I would use rats whenever you can. But, you know, a lot of the neuroscientists out there are, you know, shaking their heads, because they understand that, you know, a lot of their transgenic strains are only available in mice,

Dr. Barkha Yadav-Samudrala: 22:34

in mice. So there is a high chance that you plan and do everything, you set this experiment, and it's not going to work because of your mouse

Dr. Ian Jacobs: 22:43

could be. And you know, what's interesting about it, too, is that unless you actually do the experiment, you won't know because they'll do, for instance, DBA, mice will show renewal. So they acquire the same, they extinguish the same as sort of the lab standard mouse to see 57, Black six J mouse, right. They'll perform the same as those mice. And then when you get to the renewal test that just met, and they just don't, they just don't do it. So last thing that I thought was more ubiquitous than it truly seems to be.

Dr. Barkha Yadav-Samudrala: 23:20

Wow. Yeah, that is one. I would say disadvantages of behavioral techniques in general, as compared to, you know, all the biochemistry techniques we do, it's like, it's gonna work, you know it, but with animal it's always so. So thank you so much for giving us that overview of all the steps. What, are there any alternative techniques to fear conditioning, in order to assess the same things you're going for in fear, right?

Dr. Ian Jacobs: 23:57

So I'm gonna take this as saying like alternatives within fear conditioning. First, like, so I talked about my main method is to use the foot shocks, right? That's kind of what I was trained on and everything. But one of the modern sort of approaches to this fear conditioning stuff, is actually to use a something called the looming predator task. And so what this does is rather than relying on foot shocks to actually provide the unconditioned stimuli, it sort of tricks the mice and the rats to into thinking there is a predator looming above their heads. And so what they do is they have a transparent box that they put these animals in, and then in one corner of the box is a, like a little hut that they can go into, it's really cute. But they have like this little tent basically, that they can escape into, and get out of the way. Right. And so, what they do is they have a screen positioned on the roof of this transparent box. So from the Animals perspective, if it were to look up, it would see the screen. And they take a stimulus a little circle. And they can sweep that circle like a black circle across the screen. Or they can make the circle go to the center of the screen and enlarge to fill the entire screen. And the idea is that this is supposed to mimic a swoop by an approaching predator. Yeah. So yeah, like the birds in the sky, right? It's circling around the field that the mouse is in the mouse is freezing at this point, to not be detected, right, because the birds vision is based on movement, as we learned from like Jurassic Park, but so it freezes in that circumstance. But when it actually, when they have the the looming predator where it's going to enlarge that circle, and that shadow basically, is going to get bigger and bigger and bigger, it's supposed to be like the predators coming right at them. So the idea is that they'll perform a different type of fear response, a flight response, where they're going to flee to safety. And so this is very interesting, because, you know, first of all, you're not actually doing anything to the mouse. So, in terms of, you know, whether or not the ethics of you know, providing foot shocks to animals, you know, you can just kind of skirt around all those because you're literally not doing anything to them. So it's kind of nice in that way. You're right, it's good to reduce the pain and the discomfort, the animals experience. But what it's also revealed, is that there are sex differences in how these animals are going to perform their fear responses. I think for the most part, males tend to do more freezing behaviors, and I have this backwards, but I think males tend to do more freezing behaviors. And females tend to do more flight behaviors. And so when you're looking at something like using foot shocks, and you're just measuring freezing behaviors, you might not be appreciating the true complexity of sex differences that are there for you. Another task they use is something called the robo Gator task. And this one's interesting, just because it's I don't know if you've ever watched videos on it, it's pretty funny. It's kind of a corny procedure. But it's nonetheless pretty cool. So what they do in this one is they have a basically a little robotic car on one side of a hallway. And they put the mouse on the other side of the hallway. And the idea is the mouse, you know, ventures out into the hallway, and it sees that there's this thing on the other end of the, the chamber and it should be a little bit wary of it. But they bait food right in front of the robot Gator. And so they're, you know, oh, this mouse is hungry, and there's some food right there. But I have to go next to the monster. And so they'll they'll have the mouse go out and approach it. And then as the mouse gets past a certain threshold, they'll have the gator, the robo Gator drive forward and use like these lightly grasping claws, basically, too. Yeah, gotcha. It's a little interesting, because, you know, you can do things that would affect fear conditioning, circuitry, like lesions to the amygdala, the hippocampus, and they'll affect how the animal will approach these, these, this sort of conundrum. Where, you know, if you were to leash in the amygdala, for instance, those animals just do not care. They will waltz right up to that Robo Gator, take the chomp and carry on with their day. And so it's kind of wild to see how they would interact in a more realistic predator situation. Because as I mentioned the beginning, right, the whole point of using fear conditioning in the first place is to tap into this survival reflex to get the animal to do defensive things to protect its own life. And so if you're using things that are more similar to the way the animal would behave in a natural environment, using these looming predators, at least physical predators, I feel like that's a bit more ecologically valid than saying, you know, I'm just gonna use foot shocks

Dr. Barkha Yadav-Samudrala: 29:05

when you're explaining the struggle get at the only image in front of my eyes was from the shining,

Dr. Ian Jacobs: 29:12

the shining like from the statues?

Dr. Barkha Yadav-Samudrala: 29:16

The hallway with the twins, and yes,

Dr. Ian Jacobs: 29:19

it's exactly like that, dude. Yes, that is perfect. Oh, man. Yeah. Winds are just sitting down there, the end of the hall, and yet there's a nice tasty burger. Yes, this is decisions. Oh my god. So but outside of fear conditioning, right? If you want to study associative learning circuitry. If you want to study the specific search circuitry that's involved in what's called conditioned aversive emotional responses, you would need to study via conditioning. There's not really a lot ethically more that you could do to the animal beyond that. That you would really want to but if you're just interested in reflexes, I mean, there's tons of reflexes, whole classical conditioning got its start in salivation, right Pavlov and his dogs and he rang the bell or it's actually a metronome fun fact. But he played this metronome and the dog salivate. And bam, reflex, classical conditioning. So you could study that circuitry. But what you're going to be involving, then if you're adding any sort of appetitive stimulus, is you're going to start involving things like the nucleus accumbens with a lateral habenula, to, you know, do these sort of feeding behaviors and to monitor rewards like that. And so that gets when we talked about the motivating operations a little bit more tricky to do like that. But, you know, if you're trying to study reflexes and don't want to use fear, that's one way to do it. Another way is to use drugs. Right? So I think, at our, at our University, Dr. Don Lyle, he does experiments where he conditions immune responses, that result from cocaine, heroin treatments, so he gives the animal hair when in a particular context, the animal then, you know, of course, has its immune response, which is to depress the immune system, when the heroine is on board, and then he, when he puts them back in that context later, right, he sees that despite the fact that they haven't received hair, when they will see this depressed immune response in their animals, and so you can study reflexes in that way as well. And that also is a type of learning that is worth studying. So there's other ways that you can get around.

Dr. Barkha Yadav-Samudrala: 31:37

Okay. So what would you say is the most difficult step, I mean, the step that involves a lot of troubleshooting in this entire experiment?

Dr. Ian Jacobs: 31:52

Well, alright, so if you're just trying to acquire the fear, right, just trying to see if there's acquisition differences between your groups, the hardest part is the initial setup, where you are trying to titrate the exact shock that you need to use. Okay, so one thing I've learned over my many years of doing this, and my many purchases from, you know, supply companies is that the equipment's not always the same, there can be a surprising amount of variation in in the actual output of some of these machines. So just to get technical for a moment, the way that the shock is produced is it has a, a box on the outside of the enclosures that produces the shock and runs up to the the floors, and you can adjust the level of that shock, the intensity of that shock. And what I'm saying is that when that unit reads that this is a, for instance, point seven milliamp shock, right? Is it actually point seven milliamps, maybe, sometimes it's pretty good. But there's always a few of them whenever you do a purchase that need a little bit more, right, maybe just set at 2.73, to actually get that point seven milliamp shot. And why that's important is because as we mentioned, the intensity of your stimuli, determine the speed and the the robustness of the learning that occurs. So if you have differences in that, and then you see differences in your groups, well, is it just due to the fact that you shocked one a lot harder than you shopped the other one? Then, you know, what I found was with, you know, we're using a transgenic line in our lab. Depending on the particular mouse strain that you use, you'll have to titrate different levels of shock, to try to produce a condition response that is comparable to your other studies, or to controls if you're using wild types, or if there are, you know, if you're, if you're actually trying to measure if there are differences in acquisition, right? You didn't know how much those types of intensities of unconditioned stimuli are playing into that. Beyond that, if you're trying to do extinction, I think the hardest part is just well, determining how long to do the extinction treatment for. So if you have a system like, like in rats, one of the benefits of using rats is that you can use something called suppression ratios to determine the animal's performance. So you train the animal to perform some operant task. The idea is that when the animal is afraid, it's going to freeze. And so when it's freezing, it is not performing the operant task. So you're asking, right, how much did the presentation of this feel fearful stimulus, suppress the opera activity, and that gives you an idea of the degree of fear that the animals getting and so that's really, really easy to calculate, right? If you doing something like lever presses, you can calculate that on the spot, you know, in the actual program itself, and so you get an idea of exactly how well your animals are doing. In our lab. We didn't do that because again, mice are really, really dumb. I cannot stress that enough. Help. Get dumb mouse's, we thought that the length of time it would take to make them perform an operant response. And in our tap mice, a basic nose poke for food response took our animals 10 months to train. So just to put in perspective of how bad they the stuff, that's how long it took to just train a simple behavior. So we weren't about to do that, for this study. So what we did was we used a visual identifier for when they were freezing of like when they were ceasing all their movement, right. And so for that one, let's say you have 32 subjects, and you put them in a box for 60 minutes, you have 32 hours of footage to go through in a day to determine whether or not your animals are at a level of extinction that you can then provide a relapse test to them. Right. So that's obviously not possible unless you have like a team of full time undergraduates. I mean, like eight hours a day, going through these videos. Yeah. So in that part, it's a little bit hard to figure out what length of time you need for your subjects, because they're actually extinguished. Because unless you're using something like a suppression ratio, or they do have alternatives, where they have computers that can code videos for you. They require very specific equipment. But determining that length of time can be very difficult. The final thing that I think is the most difficult is determining your, like getting your actual experiment room correct. And appropriate. So, you know, as you've mentioned, there are a lot of things that are going on in this experiment. There's a lot of different variables that can impact whether or not an animal freezes or not. You know, anyone who's had a pet knows that, especially like a cat, right? Like, any sort of unexpected stimulus, like a cucumber on the ground, or a loud noise in the corner, right is going to make them freak out and perform this this fear response to something that's completely irrelevant. Well, what if that happens while you're in the middle of your experiment, right? What if you're trying to code how afraid this animal is? And maybe a someone drops something outside, it makes up like a beaker crashes down and breaks, right? Like, okay, well, now you've just ruined an entire batch of animals because they experienced this extraneous variables. So limiting those factors, as much as possible having white noise on in your rooms, to make sure that the outside is more or less quiet, you know, controlling the environment outside, you know, for our labs at UNC, we sadly have our doors to our laboratories, access accessible to the undergraduate population, which you know, they're cool and all, but they also are noisy. In our lab, in particular, we have a classroom that lets out every hour is right next to our labs. So if you're doing this experiment, and you know, some undergraduate comes along, talking about, you know, their their night real loud, and they laugh real loud, the animal is going to freeze up to that. Yeah, and you don't know if it's due to your procedure, or if it's due to the actual extraneous variable there. So I think controlling your environment is extremely difficult with this.

Dr. Barkha Yadav-Samudrala: 38:15

Yeah, now I have started putting up a sign whenever I run behaviors outside the door, that please do not enter. Just keep quiet.

Dr. Ian Jacobs: 38:24

And I think when people read it, they think like, oh, I'm being quiet. No, no, no, no, we mean, absolutely. Yeah. Yeah. Yeah. So some, some laboratories are better about that. I know, at Wilmington, UNC Wilmington, they have a whole wing of their psychology building dedicated to the animals. And wow, yeah, so they got a real nice over them.

Dr. Barkha Yadav-Samudrala: 38:48

That's great. So my next question is, is this easy to learn? Like in terms of a scientist like me, if I want to learn fear conditioning? Is it easy, or it has a huge learning curve?

Dr. Ian Jacobs: 39:03

So I think performing the experiments is exceptionally easy. Okay, um, you know, it's a really great project for undergraduates to do, because first of all, undergraduates are very motivated to at least you know, if you choose them, right, are very motivated to perform well. And so they're the kind of uncomfortable with with laboratory work, and they're always worried that they're doing something wrong. Yeah. And so they're very careful about how they do things and you tell them straight up, you need to do this the same every single time. And all it involves for them is picking a rat or a mouse up, putting it in a box and running a program. The rest is history, right? They leave the room. They don't have to do anything about it, the science gets done. Where it becomes difficult is in setting up the experiments if you can have somebody provide you with a program For these boxes that will, you know, present your stimuli in the way that you want to the animals, it can be pretty easy to set up on that end, writing the programs is a bit difficult, and it requires a bit of skill. troubleshooting, your data is also pretty difficult. You know, when things work, right, and as expected, it's easy, right? But when things go wrong, and you know, as I just said, like, you know, in my dissertation that I'm working on right now, my animals didn't show the renewal effect. And so I had to go looking like, why the heck aren't they showing the renewal effect and trying all this troubleshooting stuff, only to find out, you know, at the end of this whole story, that, Oh, there's just some mouse strains that just don't show it. And so it's common in inbred and transgenic mouse lines for them to just sort of lose the ability to do this. So that's, that's that, right. So troubleshooting that takes a long time. And knowing where to look for those things can be difficult. I think it's mostly, I know, I know, it sounds complicated on the on the beginning, like to learn all the different pieces. But once you have those pieces down, I think it's incredibly easy to learn. And when you do one experiment, once you've got that one experiment down, all you have to do is change one variable, and bam, you have a whole new experiment with a whole new set of data and a whole new publication. Right? I think it's a pretty robust and easy to learn.

Dr. Barkha Yadav-Samudrala: 41:17

Yeah, when you explain it that way. I'm like, Oh, wow. Yeah, it is really cool.

Dr. Ian Jacobs: 41:23

Right. But you know, for anyone doing this in a maybe non laboratory setting, like if you're doing this in an educational setting, I think, you know, if you can get past the whole giving foot shocks to animals thing, a classical conditioning experiment is quick, it's easy to perform. And it gives the students very clear data in a way that they can, you know, feel like they actually performed an animal experiment in a very short period of time. So I think it's great for education.

Dr. Barkha Yadav-Samudrala: 41:51

Right. So what are the advantages and disadvantages of fear conditioning?

Dr. Ian Jacobs: 41:57

Alright, so advantages, right, you're free of any motivating operations. I talked about that before. I also think that in terms of fear conditioning, it's kind of like there are cases where the animal cannot experience the stimuli that you're giving to them. But in the most cases, you can be pretty sure that all your animals have, you know, receive the same treatment. Right. Another advantage is just again, how quick it is. You can have an experiment done in just over a week, which is a great turnaround for experiment, especially considering that, you know, for our population of mice, at least the turnaround for an operant conditioning experiment, where we're also studying learning in that case, was 10 months. So definitely, I changed over to doing fear conditioning for my dissertation, for just that reason. I also think that the, the conclusions you draw from it, if you're careful not to ascribe, you know, too much personification to the mouse, right, if you just purely talk in the behavioral terms are hard to argue with, because you have very clear numbers that show whether or not the animal was freezing or not, in a particular circumstance. And that's just the numbers that you can't say like, oh, well, you know, were you you know, just counting something that you shouldn't have counted, right? Or do you not have a good operational definition of your, your behaviors? Well, no, we did. We was freezing. And here's, here's the video, if you want to watch for yourself, I just think it's very hard to argue with the the findings from fear conditioning. As far as disadvantages go. The first is, it's extremely awkward to talk about what people think when people hear foot shocks, they imagine like a, an execution chamber. Yeah. And it's just not that, and it's hard to get people over that hurdle. I think, too, like I mentioned, there are sex differences in at least freezing behaviors that make the inclusion of female subjects. Not gonna say troublesome, because most of the time, you'll get no differences between your sexes. But when you do you have to ask was I using the correct assessment here and not? Another disadvantage is I guess just how vulnerable it is to environmental influences. When you're talking about something like a reflex. You're not allowing the animal to, you know, sit down to personify, but think about it too much. Right, like, like they just are performing a response based on on the data they have in front of them or the stimulus that they have in front of them. And any sort of alteration to that stimulus. The way that's detected could influence their performance, or any sort of extraneous variable that is influencing whether or not they're rate on a given trial could be a wrench in the plan so to speak. You know, for instance, let's say you are in an extinction treatment, you're training them to not be afraid to atone. And then on day six, right, a beaker breaks outside. And suddenly this loud crash happens, why, just after you presented your tone, all of a sudden, your animals are afraid, again, they're freezing to the tone, but for a different reason. Right. And so you're just your experiment just kind of got ruined by one thing. That's actually the final thing I'll mention too, about the disadvantages is, these animals are kind of one trick ponies that you use. So when you do an operant conditioning study, you can train them to press a lever for sugar pellets, look, their nose for peanut butter, not poke their nose for sunflower seeds, right? Like whatever you want to do, you can try to do all these different things. And, you know, retrain them and solve kind of, okay, with a classical conditioning experiment, once you provide a stimulus arrangement, right, that first one information is going to generalize very readily to all these different contexts and procedures. So if you were to bring them into a new experiment, the learning from that old experiment is going to heavily influence the speed at which your animals learn about the new information. It will also influence if you're trying to do like, like something like renewal or something like that, and you've used those various other stimuli before, you might not actually get good acquisition to see that renewal effect later on, because you've just done all this stuff in the past. And so for that reason, it's kind of it uses a lot of animals, which I know can be distasteful to some. And like, especially if you're talking about some of these transgenic or optogenetic ready animals, it can get very expensive to run your experiment, and then have on day seven, someone break a beaker and have a hole ruined. And you can't go back. You can't you can't take that back. You can't retrain them for a few days, it's done. And so that's kind of weak part of it, too.

Dr. Barkha Yadav-Samudrala: 47:04

So you cannot even use this animal for a complete new experiment. Like, for example, teammates or a novel object?

Dr. Ian Jacobs: 47:11

Well, like, you probably could, right? You probably could use them for something like that. But if you were to try to use them for an operant conditioning experiment in the same operant chamber, so they just got the foot shock in. Yeah, you can see the problem there. Right. Like, wouldn't they be afraid? And it would slow the actual acquisition? If that's something you're measuring? If it's not something you're measuring, then maybe it'd be okay. But I don't know. I mean, I think when you provide such a strong experience where you're, you know, threatening in the animals mind the survival the animal I think that messes with a little bit. Yeah, and then you know, for something like the robo Gator task, right, like you're putting them in a maze like situation and then attacking them. So choose to do the maze in the future. Yeah.

Dr. Barkha Yadav-Samudrala: 47:59

Yeah, that's always gonna be the shining task for

Dr. Ian Jacobs: 48:04

students, the amygdala, one.

Dr. Barkha Yadav-Samudrala: 48:10

So what are the costs associated in setting fear conditioning? Is it expensive? Or any lab with some access skin said this, I think relative

Dr. Ian Jacobs: 48:22

to the rest of science, it's pretty okay. Relative to you know, everyday purchases? God, no, it's it's so expensive. I think our last quote for getting what did we get? We get 16 boxes, I think we got 16 note, eight boxes, I think eight boxes was $32,000. Why Yeah, so it was and that's what the whole shock setup and everything and that was not even kitted out, in the ways that we could have could have them out. And it's without having the equipment to do you know, computer based analysis of the videos. So it gets very expensive in that sense to do this. But once you have it set up, there's no reagents you need to buy. There's no you know, antibodies to keep track of and to store there is simply just animals to keep and if your university or your you know, business has a good breeding program. I mean, it's essentially just the cost of Chow to keep this going.

Dr. Barkha Yadav-Samudrala: 49:23

Yeah, so it's just like the first setup costs.

Dr. Ian Jacobs: 49:26

The initial setup is a little high. But you can also build it yourself. My mentor at Appalachian you know, he's I think he set up his in the I'm gonna feel like if he ever hears this, he might get mad at me. I think he's ended up in the 80s or like in the 90s but that might make him too old. I'm not sure they didn't have all this like, you know newfangled technology with the whole boxes that produce a shock now, so he set up a neon sign transformer to produce the shock for him and Like, daisy chained all of his boxes together in a way that tried to evenly distribute the shock to each of them. So I mean, you can just, if you have some engineering know how you can just make these things, no problem. It's literally just a box with a grid. So yeah, I think it can be, depending on how but okay, so actually, here's something that I should say, I don't think you should do that. I don't think you just set up your own materials. One of the things I'd like to see in science, and especially in behavioral research, is a standardization of all these methods and techniques. Like you mentioned, the teammates, and the, you know, there's also different types of arenas that people use. And a lot of the times people just build them themselves, because it's just a box and like, why would I pay $10,000 to get, you know, a box. But I think that it's worth standardizing you know, the equipment that we're using, because these environmental variables can be so impactful on the performance of our animals, that when we're trying to compare between studies, when I'm trying to compare my renewal data to your renewal data, I'd like to know that we were using the same equipment. Correct. So I think that it actually wouldn't say, I would say, you know, buy it from a reputable source and try as much as you can to standardize what you're doing.

Dr. Barkha Yadav-Samudrala: 51:16

Okay, and now we come to my favorite question, is there any fun fact about fear conduct, so I

Dr. Ian Jacobs: 51:23

actually sort of spoiled it already, but I couldn't avoid it. The fun fact I have, and, you know, maybe put an asterisk next to it, because I am learning more each day about this. But, you know, Candell, to this experiment in a pleasing back, and I think it was 2009, which won the Nobel Prize, where he discovered the circuitry within the ecclesia, for producing the fear response, you know, when you pair that neutral stimulus, and that unconditioned stimulus together. And just real quick, with the pleasee experiment, was, is pleased to have a gill withdrawal reflex. So they're an aquatic mollusk, and they have a gill on the top of their body, where if you touch the Gill, or if you touch things around the Gill, it'll retract and pull into the creature. And so what you can do then is you can pair a, you know, a touch of the called a siphon, which is a structure near the Gill, you can prepare a touch to the siphon, which, you know, won't produce a huge response with a foot or not foot shock, a tail shock towards the back of the mollusk. And over time, what you'll see is a strengthening of the gill withdrawal reflex to just the stimulation of its siphon. This is what Kendall did in his experiment. And he followed the the whole circuitry involved in these three particular neurons that were coding for this entire learning process. And then when they scaled it up, and they looked at more complex creatures, and more and more complex creatures, they found that the basic, at least more, more of them, but the basic structure of these, these neurons, which are responsible for this process, are preserved across a number of different species, which warranted, you know, the Nobel Prize, because it is quite the discovery. So when you're talking about something like fear conditioning, and you're studying something like fear conditioning, and you say, well, you're doing it in mice, or you're doing it in rats, how does this apply to humans? Bam, directly, it's the same circuitry. So I just think that's a kind of fun fact that like, you know, when you see a mollusk learning about tail shocks, it's the same thing that's going on in your brain. And this isn't there's so I think that's interesting. Oh, my God, that's what's kept me in love with this stuff, is just learning. Like, I think as I've learned more and done more animal research, I've really begun to appreciate the cognitive abilities of these animals. Not trying to personify them and say that they're, they're fully on the level of humans, but I think, you know, people going around saying, for instance, like dogs don't have memories. don't really understand what's going on in the dog's brain. Yeah, cuz yeah, definitely, definitely do, anyway.

Dr. Barkha Yadav-Samudrala: 54:20

Well, to wrap things up, my last question to you is, can you suggest any article or a protocol about fear conditioning that I can link in the description

Dr. Ian Jacobs: 54:32

I can. So it's gonna require you to have like, at least a little bit of background in the stuff, but it's a Boulton 2002 paper. Let me see if I can pull it up. It is the paper is called context, ambiguity and unlearning sources of relapse after behavioral extinction. So that doesn't sound too interesting. But this is actually a really Good seminal paper that I think introduces people to a lot of the stuff that happens after acquisition. So the acquisition part's pretty straightforward. But one of the the contemporary issues at the time this was written was whether or not the extinction that occurs, right, that weakening of the condition response during extinction is that a type of unlearning where the information that was learned in acquisition is destroyed, and no longer present in the animal's brain. And, you know, this paper here kind of goes through a series of experiments. And it's sort of a review of the experiments that tell us that extinction is not a process of unlearning. It's actually a new type of learning that occurs, and then goes through all of its evidence for why that's the case and then wraps up with just some like, you know, how would you actually use this information in real life? So I think it's a really interesting, cool article, just introduce people to these topics and sort of get an idea of how the words are used.

Dr. Barkha Yadav-Samudrala: 56:02

Well, in that was an amazing episode. I have, like, I have a new appreciation about fear conditioning after listening to you. And thank you so much for your time to explain what fear conditioning is to us.

Dr. Ian Jacobs: 56:21

Thank you so much for having me on here. This has been a lot of fun.

Dr. Barkha Yadav-Samudrala: 56:24

Yeah, it is. It is. Thank you so much, Ian and listeners. I will catch you next week on another episode. And in meanwhile, if you have any suggestions, or if you would like to come on the podcast and discuss something with me, please email me at Barkha at Nerdrx podcast.com. And remember, it's good to be a nerd. Bye.