 So thank you all for coming. I'm going to talk to you today about some work that really got me first interested in the neurobiology of learning and memory back in graduate school, then continued through my postdoctoral years and into my UVM years. And I hope to convince you by the end of this talk that simple is good or at least can be good and that you actually can learn about learning and memory in the blink of an eye or by studying the blink of an eye. So just to quote Jim Henson, simple is good and as I like to say sometimes it's not easy being John Green, I decided to throw that in there. So the physical basis of memories, which is really what this is all about and it's root. It's both about what the physical basis of memories are and how you actually go about studying the physical basis of memories. So one place to start is going back over a century ago and Richard Seaman, who's actually a fairly obscure figure but coined a term that we still use in the literature and that's the term n-gram. So what he wrote about back in 1904, I'll just read you the quote here, when an organism has been temporarily stimulated and has passed after the cessation of the stimulus into the condition of secondary indifference, it can be shown that such organism has been permanently affected. This I call the engraphic action of a stimulus because a permanent record has been written or engraved on the irritable substance. I use the word n-gram to denote this permanent change wrought by a stimulus. So that's a definition of n-gram as a physical memory trace. But where are those physical memory traces? Can they be localized? And if they can be, what does it mean to say there are physical changes in the substance in the brain? It means enormously complex, of course. There are well over 100 billion cells. There are trillions of connections in the human brain. So the question is, can n-grams be localized to particular brain areas? And if so, where do you even begin to look in this massive tissue? And then once you find, or if you can find them, then you can actually study the types of changes that occur in that tissue that may be the actual physical basis of a memory. So it's actually, as I said in my kind of abstract for this, it's sort of a daunting task when you step back and think about it. It's not trivial to try and figure out where to even start, where to look. So Carl Lashley was a figure in the early 20th century, a psychologist who devoted a good part of his career to looking for the physical basis of memories, to looking for the n-gram, to trying to find one or more n-grams in the brain. And he wrote in 1950 towards the end of his career about this search for the n-gram. And I have a couple of quotes up here. He said in 1950 towards the end of his career in experiments extending over the past 30 years, I've been trying to trace condition reflex paths through the brain and or to find the locus of specific memory traces. The results for different types of learning have been inconsistent and often mutually contradictory. This series of experiments has yielded a good bit of information about what and where the memory is not, has discovered nothing directly of the real nature of the n-gram. I sometimes feel in reviewing the evidence on the localization of the memory trace that the necessary conclusion is that learning just is not possible. It's difficult to conceive of a mechanism which can satisfy the condition set for it. Nevertheless, in spite of such evidence against it, learning does sometimes occur. One of my favorite quotes. So this state of affairs actually continued for quite some time beyond Lashley. I mean, you could argue it really continued into the 1970s at least. So an important figure at this point is Richard Thompson. So Richard Thompson, who just actually just passed away a couple of years ago, was very interested, devoted basically his entire career as well to the search for the n-gram. And he wrote a paper in 1976 in American Psychologist where he talked about the search for the n-gram. Remember this is now over 25 years after Lashley wrote. And Thompson wrote at that time, analysis of the neural bases of associative learning has not progressed very far. Simpler invertebrates do not, for the most part, exhibit very robust classical or instrumental conditioning. N-gram invertebrate models may have promised, and in fact, that promise was realized to some extent in later years after this paper. He also wrote, is to be hoped that analysis of such simplified neuronal models of learning will provide information about the nature of basics and apoptic processes involved. However, the ultimate goal of studies in the neurology of learning is of course an understanding of the neural mechanisms of human learning and memory. To achieve this is necessary to undertake directly the difficult task of analyzing the neurology of learning in the mammalian central nervous system. The optimal requirements have to do with control of the experimental situation, control of non-associative variables, and the feasibility of neuronal analysis. And as he wrote in 1976, we have recently adopted classical conditioning of the eye-blink response of the intact learning rabbit as a model system for analysis of brain substrates of associative learning. So to digress for just a minute, because I want to talk more about Richard Thompson's Dick Thompson's work, that then kind of sets the stage for some of the things I've done. Dick Thompson had an absolutely extraordinary career, one of the giants in the field of learning and memory and neurobiology and learning and memory. You can go online and do these academic family trees. And I pulled up here Dick Thompson's Neuro Tree, I'll find my laser pointer here, which expands out beyond what I can even fit on the screen here. And for those of you like Bill, who have been in this area for a while, you'll recognize many names, many well-known scientists that Dick Thompson had a hand in training. Steve Marenge and Scott Kim goes on and on. New scientists that he trained were also my mentors, if I can get my pointer working here. Joe Steinmetz was a postdoc with Dick Thompson back in the late 1980s, and Joe, as Bill mentioned, was my postdoctoral mentor at Indiana University. And Diana Woodruff-Pack, that Bill also mentioned my graduate mentor, was a visiting scientist in the mid-1980s in Dick Thompson's lab. So I am essentially Dick Thompson's academic grandchild. And just to prove it, you can go down one level, and there I am, and there's the University of Vermont, and someday you'll be able to expand this, and hopefully I'll have a little Neuro Tree there. So there's Diana's Neuro Tree, and there is Joe's, and there I am. So eye-blink conditioning is a model system. So eye-blink conditioning, what is it? Well, it's a type of classical or Pavlovian conditioning. And in its simplest form, which a lot of this research uses, each trial, you have trials that consist of a neutral stimulus, such as a pure tone, followed half a second later by stimulation of the eye. But surprisingly, the eye stimulation causes an eye blink. That's not really what we're interested in. And after several dozen trials of pairing tone eye stimulation, blinks begin to be made to the tone as well, in anticipation of the eye stimulation. And these tone-illicited anticipatory eye blinks are the learned response that we're interested in, or in the lingo, the conditioned response, or the CR. So a very quick little history of eye-blink conditioning, it's kind of interesting to take the slight digression. Eye-blink conditioning was originally used in humans to study associative learning long before it was adapted to non-human animals to study the neurobiology of learning and memory. And the first known publication using eye-blink conditioning actually dates from 1899. My two mentors actually co-edited a pair of books actually that came out in 2000, contained all chapters on different aspects of eye-blink conditioning. And in the introduction to one of those books, they wrote, the history of the study of classical conditioning is relatively long, beginning late in the 19th century. First in Germany and later in the United States, investigators moved from studying the slower autonomic nervous system responses, such as salivation to assessment of conditioning and the somatic nervous system using the eye-blink response. They go on to write, in the United States, Ernest Hilgarb was among the first researchers to conduct studies on human eye-blink conditioning using techniques that might not be approved by contemporary human subject institutional boards. Instead of an airpuff, Hilgarb used a paddle to smack the subject in the face. Of course, this large wooden device was effective in producing a reflexive eye-blink that could then serve as the basis for conditioning to atone. So in the early 1960s, usually a character named Isidore Gormanzano is given credit, although Alan Wagner, who many of you know may or may not share this credit, for developing a system of eye-blink conditioning in rabbits. So rabbits became the initial kind of animal model system for a variety of reasons, which I won't get into all of them. And then more recently, rats and mice have become more common non-human animal subjects. So Gormanzano in 1983 wrote a long chapter about eye-blink conditioning and he was looking back over 20 years of research. And he wrote a few things about the very earliest days in the early 1960s. Incidentally, he developed this at Indiana University as well, where I did my postdoc, so I'm really steeped in eye-blink conditioning. He wrote, in both humans and animals, the other major conditioning preparations suffered from substantial intrusions by responding not attributable to associative processes in human eye-blink conditioning. Extensive methodological work had removed many of the measurement difficulties, but the data still displayed considerable variability in the human preparation was clearly unusable for physiological interventions. Consequently, to focus attention on the objective determinants of conditioning and to provide the robust data necessary to address both physiological and theoretical questions, we sought a more suitable animal preparation. And Dick Thompson, looking back at his early history with using this preparation and adopting it in the early 1970s, wrote, Michael Patterson was a postdoctoral fellow in my laboratory, having obtained his PhD with Gormanzano. Patterson strongly proselytized Gormanzano's preparation, classical conditioning of the eye-blink response in the rabbit, as a suitable preparation for analysis of the brain substrates of associative learning and memory. So what's so great about eye-blink conditioning for studying the mammalian neurobiology of learning and memory? Well, there are a lot of aspects of eye-blink conditioning in non-human animals that make it really ideal for this sort of study. I'll outline a few here. First of all, just the fact it can be done, as I mentioned, in humans and non-human animals means that what you learn in a non-human animal can easily be applied to what's going on in the human brain. As well, the experimenter has maximum control over the training situation in terms of when stimuli are delivered and when the learned response should occur. So you know exactly where and when to look for the learned response. There's minimal or no eye-blink response to the neutral stimulus prior to conditioning. So anything that begins, any eye-blinks that begin to be expressed to most commonly atone, you can pretty much attribute to a learned response, that it really is a learned response. Furthermore, if the reflexive eye-blink is normal, the eye-blink to an air puff or some kind of eye stimulation, then the absence of a learned eye-blink is less likely to be caused by purely movement issues. In other words, if you're trying to localize the source of the learned response and you, let's say, make a tiny lesion in a particular brain area and there's no learned response, if there's still a reflexive response, you know that the animal in this case can still make an eye-blink. So you're on firmer ground thinking that you've affected a learned eye-blink rather than just the ability to blink. And then finally, because this is probably the best one, because the learned response is simple and discreet, predictive relations between neural activity and learned behavior can be established. It's very easy to map changes in neural firing, for example, on to this simple learned response in terms of the timing, amplitude, and all those sorts of things. So eye-blink conditions had made many contributions to our understanding of several things over the years. I'll just briefly kind of give you the global view here. That's the statue of Pavlov, by the way. One thing it's contributed a lot to, which I won't really talk any more about, is a much better understanding of classical or Pavlovian conditioning in general, which, as many of you know, is a very important form of associative learning. Pavlovian conditioning plays a role in the development of and treatments for many psychological disorders. This is just one example, and a good example is panic disorder. So a lot of what I've focused on and maybe made some small contributions to is eye-blink conditioning has contributed substantially to research on the neural processes underlying learning and memory, not only on the neural processes, but also what are the fruitful approaches to take, what kind of data are most convincing to the field? And, of course, alleviating disorders of learning and memory, which would be one goal of understanding mechanisms of learning and memory better, requires first understanding those mechanisms. So that's another way that eye-blink conditioning has contributed, that understanding, that basic understanding of mechanism. And then finally, eye-blink conditioning has contributed substantially to knowledge of the processing that goes on in this structure called the cerebellum, which I'll talk a little more about, which is the main brain structure supporting eye-blink conditioning. The cerebellum is really interesting for a lot of reasons, despite what Bill said. It's been known for many years to play a very important role in movements of various kinds, which in and of itself is extraordinarily important. But even beyond that, there's increasing support for the involvement of the cerebellum beyond just motor function in terms of more cognitive or non-motor function. Some examples of this from the literature of disorders is cerebellar abnormalities have been linked most recently to autism, to schizophrenia, to ADHD. Disorders that you would not normally consider motor disorders. So the cerebellum is actually quite an interesting brain structure, and I'll come back to it. So with all that groundwork laid, come back to Dick Thompson. So he and his colleagues, his students and postdocs and colleagues, published a short paper in 1981 in a fairly obscure journal called The Bulletin of the Psychonomic Society. And this was essentially sort of a three page abstract. They're not really any hard data in here, but he wanted to get the word out about what he and his colleagues were observing in the lab. These really interesting and I think surprising results that they were uncovering about the essential role of the cerebellum in this simple learn response. And they were so excited about this, they wanted to just get something published quickly before even, you know, prior to doing full-fledged publications. So they published this short little paper in 1981 that provided the first evidence, perhaps, that there was an engram that could be found. Although around the same time Bruce Cap was working here at UVM and providing some evidence that the amygdala was also the site of a different type of engram. So it was sort of in the air at the time. But Thompson and colleagues were really excited about this idea that the cerebellum might house an engram. So a little bit about the cerebellum, also known as the little brain. Here it is in red here in the human brain. Here it is in a rat brain, the brainstem and spinal cord of the rat. Here's a drawing from Ramona Kahal, one of very famous early anatomists in neuroscience who did these beautiful drawings of cerebellar cortex back in the early 1900s. Cerebellum is present in all true vertebrates. Cellular structure is virtually identical across mammals and birds. And an interesting little factoid is it's roughly half the neurons in the entire brain, although it's only about 10% of the volume of the entire brain. So two regions of the cerebellum form and store the memory of idling conditioning. So I'm summarizing a whole bunch of research here. The interpositus nucleus, which is one of the three deep cerebellar nuclei buried in the middle of the cerebellum. And the lower part of the primary fissure in the cerebellar cortex. And these are shown here in the rat. The deep cerebellar nuclei, so one of the three deep cerebellar nuclei buried. These are a bunch of neurons buried in the middle of the cerebellum. One of them holds the stores the memory of idling conditioning, along with a region around here. So I want to talk about one of Dick Thompson's studies in particular. And this will be one of the only graphs I actually show. This is one of my favorite studies that came out the year I started graduate school in 1993 from Thompson and colleagues. And was very influential. This is I think one of the most convincing demonstrations that the cerebellum is important for storing the learning that goes on in idling conditioning. So, and this type of approach has been adopted was very influential. It's been adopted since by many different types of researchers in this field. So in any particular learning circuit in the brain, some brain areas are involved only in expressing the learned response. And we might say that they're downstream of learning related structures. We're learning related brain areas actually store the memory. So how do you actually demonstrate that? Well, Thompson and colleagues and there are fancier ways to do this now. But they did a really nice convincing demonstration back in 93 of this. And what I want you to look at in this graph here, we have a measure of learning on the y-axis here. This is the percentage of trials in which in this case a rabbit blinked to a tone. So if there are 100 trials in a session, if you have a 20 here, it means on 20 of those 100 trials, the rabbit gave a learned response. So of course, as you go up here, it means there's more and more idling conditioning going on. You can see there's a control group here that learns really well. Over days of training, tone, eye stimulation, they show more and more anticipatory learned eye blinks to the tone. So that's kind of your comparison group. The two groups of interest here are what I call the brainstem and cerebellum group. So let's take the, well, we'll take them in order here. So cerebellum, what they did here is they used a pharmacological substance to temporarily silence the activity of neurons in a particular brain region. We don't have to get into details about how they did that. But they had two separate groups here. In one of those groups, they silenced an area of the cerebellum. And in the other group, they silenced an area of the brainstem, which is sort of kind of near the cerebellum but outside it. And basically what you can see here from these data is that silencing either this cerebellum region or the brainstem region, there's no learned responses or in essence, essentially no learned responses given by these rabbits. Control group learns really well. These guys look like they're not learning at all. But what's, that's not the most interesting part. What's really interesting is what happens then when the brain region is back online because these are temporary inactivations. What happens is the brainstem guys, the guys who had a little area of their brainstem inactivated immediately start showing learned responses. It's as if they were just, they were learning but they were just prevented from showing that they were learning. In contrast, the guys with a small region of cerebellum inactivated look as if they had not learned anything. In fact, they look completely naive. They look like the control guys when they first started. It's as if that inactivation prevented them from learning at all. And now they're learning for the first time that the tone predicts the eye stimulation. So this is a very convincing demonstration at the time that you have a brainstem region in this case that is important for expressing the learned response but is not the site of storage. And you have a region in the cerebellum that is the site of storage and is important for forming and storing the memory. So cerebellar memory site, number one, the interposest nucleus, as I mentioned. There's a lot of research supporting this region as one of the two regions in the cerebellum that's important for storing the learned response. I won't get into all of that. I'll mention one contribution my lab has made. This is part of a larger paper. We're not the only ones to show this, but recording studies have shown that many neurons in the interposest nucleus fire in a pattern that resembles a neural model of the learned eye blink response. It's as if in the firing of those neurons you can actually see the brain exhibiting a learned eye blink response. One way of showing this is here where on the y-axis, you basically have this is collapsed across a bunch of trials. And on the y-axis here, let me get this to work. What this is simply is a measure of the firing of neurons above a baseline. And what you can see here is this is time here and this is an example of events that are going on in a trial where you have the tone come on, you have a learned eye blink here. And you can see neurons start ramping up their activity in this brain region before and during the expression of the learned eye blink response. Notice that they don't do that on trials where there's no learned eye blink exhibited. So it's only on trials where there's the learned eye blink occurring. There's a second region, as I mentioned, in cerebellar cortex. There's a particular cell type, a neuron type in cerebellar cortex called a Purkinje cell or Purkinje neuron that is the sole output neuron of the cerebellar cortex. So everything the cerebellar cortex does flows through the Purkinje cells and out of the cerebellum to the deep nuclei. Purkinje neurons near this one region of cerebellar cortex, the primary fissure, show this interesting pattern of activity during eye blink conditioning or develops during eye blink conditioning in concert with the learned response. And what this is is a pause in their activity immediately prior to the expression of a learned eye blink response. And this was something else that we showed again, we're not unique in showing this. This is basically, this shows one trial where you have this. These are action potentials from a single Purkinje cell that we've isolated. And it's firing away merrily here. And then the tone comes on, these are well conditioned animals. The tone comes on, Purkinje cell shuts off for a short amount of time. Meanwhile, we're also measuring eye blinks. This is a conditioned eye blink to the tone. Here's where the tone's occurring. And you can see the shut off in Purkinje neurons proceeds and somewhat overlaps with the learned response. That's probably slightly confusing why a neuron would shut off and you get the expression of a learned response. Unless you know just a little bit more about the cerebellum. And that's the Purkinje neurons normally inhibit the deep cerebellar nuclei, including the interpositis. So this pause would effectively, and the basic idea in the literature is this would disinhibit the interpositis, free it from inhibition and allow a learned eye blink to be expressed. So the idea is that the cortex and the deep nuclei work together to store and control the timing and the amplitude of the learned response. So, eye blink conditioning reveals cerebellar function more generally. I mentioned that in the beginning. And folks like Mike Mock have written about this, what he wrote some years ago. Although initially viewed as a model of associative learning, it is increasingly clear that Pavlovian eyelid conditioning is an especially useful means of studying cerebellar computation and its underlying mechanisms. And Goodlitz-Stanton and my postdoc mentor, Joe Steinmetz wrote, back in 2000, eyeballing conditioning has also been used to characterize functional consequences of clinical neuropathology of the cerebellum and brainstem. So we've contributed some to this literature as well, in my lab. One of the things I did as a postdoc and I did during my early years at UVM is work with a rat model of fetal alcohol spectrum disorder. And these are rats that receive developmental ethanol exposure. And then we test them in eyeballing conditioning to try and get a sense of how functional their cerebellum is or not. And then with the idea of being able to then treat those dysfunctions, either prevent them or treat them after they occur. So this was one study we published now over 10 years ago. This was kind of a dose response study. We had different doses of ethanol. These rats showed impaired eyeballing conditioning at the highest doses, really, but not at lower doses of ethanol exposure. The dose, actually, a little more precisely, it was the blood alcohol concentration produced. And interestingly, there was also a dose response effect in the number of inter-positive nucleus neurons these guys had. I should say this was measured when they were adults, both eyeballing conditioning and neuron counts. So these are long-term effects on the cerebellum of developmental ethanol exposure, and eyeballing conditioning allows you to test the functionality of their cerebellum. We've also done work on rat models of ADHD or attention deficit hyperactivity disorder. Two different strains here, the spontaneously hypertensive rat, the WKHA strain. Both these strains interestingly show, they show good eyeballing conditioning, but the learn-eye blinks themselves are a little bit abnormal. They occur very quickly after the tone onset. In normal rats, they tend to be a little more well timed with about when the eye stimulation would occur. These occur in these strains, they occur abnormally early. This is shown for the WKHA here, which is just showing the latency of the condition eye blink after the tone starts. You can see it's short, even across training compared to a control strain, or really two control strains here. And interestingly, at least, and I should mention also, at least in the WKHA, we only saw this in males and not females. And then also interestingly in the WKHA strain at least, we also saw that they had more prokinji neurons in the cerebellar cortex. So we think that this might be possibly the cause of the abnormal learned eye blink because there's data in the literature suggesting that the form of the learned eye blink is controlled by prokinji neurons. So what about coming back to Thompson again? What about learning related cellular changes? One of the ideas, one of the reasons to develop eye blink condition is not only to be able to trace a learning circuit in the brain, but then to zero in on the areas where the learning occurs and look at the actual physical changes. The physical cellular changes that are the basis of the engram. And I haven't really told you anything about that yet. Well, there's been a variety of folks looking at this from different angles. And my lab has looked at this in collaboration with Tony Morielli for the last five or six years focused in on a voltage gated potassium channel known as KV 1.2. Now kind of very broadly and generally voltage gated ion channels contribute to neuronal activity, to neuron communication. And interestingly KV 1.2 is expressed very highly in two neuron types in the cerebellar cortex. One of these is our friend the prokinji cell or the prokinji neuron, which I mentioned provides the sole output of cerebellar cortex. The other is a little inner neuron just localized within cerebellar cortex called the basket cell. These are really interesting because they can strongly inhibit prokinji cells. So they can control the entire output of the cerebellar cortex via controlling prokinji cells. Here's a little cartoon of how these guys connect to each other as well as other cell types. So here is, I think I have my little, there we go. There's a prokinji cell or prokinji neuron. Here's it's axon going out of cerebellar cortex to the deep nuclei. And there's a little basket cell forming a connection with the prokinji cell right here. Here's from a paper we just published. This is Katabashabi's work and this just shows some KV 1.2, very densely expressed at this synapse right here. Okay, this is where basket cells would come in and synapse with and this is a prokinji cell body here and here you can kind of see outlined there. So this KV 1.2 is very densely expressed at these synapses. It's also the second place as I mentioned it's expressed is prokinji cells and specifically their dendrites up here, this big tree at the top of the prokinji cell. So my lab's contributed to this work through a variety of studies. I'm going to more or less summarize on one slide here where we've infused various things into the cerebellar cortex, very substances. We've infused them near the base of the primary fissure and then looked at the effect on eyeballing conditioning. But we haven't picked just random things to infuse. What we've looked at are things that kind of very broadly and generally have effects on KV 1.2. And the pattern we've seen overall, again to summarize a variety of studies and agents that we've infused is that infusing things in this area that tend to down-regulate KV 1.2 or decrease its function or expression have this facilitatory effect, they improve eyeballing conditioning. That's what seems to be in common. And on the flip side, things that increase KV 1.2 expression in cerebellar cortex or at least maintain it at a high level seem to impair eyeballing conditioning. So we think KV 1.2 has an important role and is one of the cellular mechanisms in cerebellar cortex controlling eyeballing conditioning. We looked at this another way as well. So those studies suggest that messing around with KV 1.2 can modulate eyeballing conditioning. We looked at this in an opposite way. Jason Fuchs, a former grad student, looked at this in his dissertation and showed that eyeballing conditioning can modulate KV 1.2. So you can modulate KV 1.2 and affect eyeballing conditioning or eyeballing conditioning itself can modulate KV 1.2. I'll show you the one other figure here, data from one of those experiments where we had rats that undergo eyeballing conditioning and we measured KV 1.2 expression in their cerebellar cortex. And we found that it was increased in cerebellar cortex. Unless those rats were actually expressing the learned eye blink response. It's as if this increase in KV 1.2 occurs right before they're about to show that they've learned. We also had rats another group that received unpaired tones and eye stimulation. They never show the learned eye blink to the tone because they're learning an anticipatory eye blink to the eye stimulation. And interestingly, they also show an increase in KV 1.2 in cerebellar cortex. And here's the figure of the actual data. These are all compared to no stimulus control groups. These are guys that are just in the conditioning environment. So rats that undergo eyeballing conditioning but are not expressing the learned response and rats that undergo unpaired stimuli, stimulus presentations both show an increase in KV 1.2. Which is what the y-axis is here. But not rats that at the point we measured it or expressing the learned response. Their KV 1.2 is at seemingly baseline levels. So to summarize what I think my lab's contributions are to understanding cerebellum, understanding the physical basis of memories, and our use of eyeballing conditioning, which continues. So I think we've contributed some data on learning related activation of different neuron types in the cerebellum in different key regions that are important for eyeballing conditioning. We've used eyeballing conditioning to examine cerebellar function in a couple different disorders. And most recently, the last five or six years, we've really been focusing on the cellular mechanisms of cerebellar dependent learning. And that's an area we continue to collaborate with Tony Morialli's lab. And that's really become the major focus of my lab's use of eyeballing conditioning. So I'll give you a moment to think about that. So I'm just going to very briefly touch on other labs recent work using eyeballing conditioning. We're not the only ones. It's not as large as the fear conditioning community, but there are a number of other labs that are using eyeballing conditioning to answer interesting questions about the neurobiology of learning and memory and cerebellar function. There are labs using, in this case, this lab uses mice to look at how cerebellar activity is connected to eyeballing conditioning. What eyeballing conditioning does to cerebellar activity to try and understand cerebellar function better. John Freeman, who's been using eyeballing conditioning for many years, is now expanding into how the amygdala talks to the cerebellum. We know, of course, many of you know the amygdala is a key region for learned fear, and he's looking at the interaction of the amygdala and the cerebellum using eyeballing conditioning. Those are non-human animal studies, but folks are also using eyeballing conditioning in humans. This is a group looking at how participants diagnosed with autism undergo eyeballing conditioning and what that might reveal about cerebellar abnormalities associated with autism. And then a colleague, Bill Hattrick, who's at Indiana University, uses eyeballing conditioning to understand cerebellar contributions to schizophrenia, and he's done a lot of work with patients with schizophrenia. So final thoughts. So eyeballing conditioning's been very, very good to me. It's been great. We've learned a lot using eyeballing conditioning. But is it the be all and end all? No, of course not. So I want to leave you with a couple final thoughts on what's missing out of this account. What else do we need to understand about learning and memory? So William James said way back in 1890, now the slightest reflection will convince anyone that there's no conceivable ground for supposing that with the mere re-excitation of a group of neurons, there should arise the conscious addition that it's a re-excitation. This could not end out with any reference to the past. The gutter is worn deeper by each successive shower, but not for that reason brought into contact with previous showers. So we're not capturing anything about the sense of the past necessarily about memory. Going even further, one of my favorite books about memory from way back in 1932, Frederick Bartlett wrote, remembering is not the re-excitation of innumerable fixed lifeless and fragmentary traces. It's an imaginative reconstruction or construction built out of the relation of our attitude towards a whole active mass of organized past reactions or experience. Memory and all the life images, all the life of images and words which goes with it is one with the age old acquisition of distant senses and with that development of constructive imagination and constructive thought. We're in at length, we find the most complete release from the narrowness of presented time and place. So Iblen Conditioning is a useful model to get at some of the basic core aspects of learning and memory but it also does not get us very far in figuring out the physical basis of the reconstructive nature of memory, for example. So there's a lot to learn still. So I'm gonna just finish with a number of acknowledgments because I have a lot of people to thank. I've worked with many, many great folks at UVM since I arrived in 2003. I'd especially like to acknowledge my second family over in the Department of Psychological Science, been great colleagues and friends over the years. And I wanna especially acknowledge my four other faculty colleagues in the Biobehavioral Cluster in Psychological Science who I've learned so much from over the years and been so inspired by Professor Mark Bouton, my good friend here, Professor Bill Falls, Professor John Hammack who's back here and Professor Donna Tefexes. And I wanna also acknowledge my other UVM faculty collaborators who have just been really great to work with and other folks I've learned quite a bit from Professor Tony Morielli, Professor Betsy Hoza and Professor Jean Delay who I haven't collaborated in a research way with but as Bill mentioned we collaborated on the creating the neuroscience major which was really one of the kind of the thrills of my time here seeing that just grow incredibly. And then finally the biggest thank you probably is to all the students who have worked in my lab over the years. I've had just amazing luck with really hardworking, creative, smart, fun to work with students. There's a list there, it's incomplete. If you're out in the audience and I didn't mention you, it's just because my own memory is so poor but I just a big thank you because I never would get anything done without my terrific students. So thanks for your attention. Thank you.