[00:00:05]
>> So I'd like to talk about some of the things my students and postdocs u.c.s.f. and I have been working on for the last few years and I'm going to actually try to give 2 talks I succeeded in doing this in Telluride last summer at this workshop on our morphic engineering that we've been running for 25 years actually.

[00:00:32]
And so I'll I'll give it a try but if it gets too late I'll just let one of let the knock at the 2nd part of the talk so we know that neural plasticity really depends on configurations of the circuits that excite neurons and that result in responses of excited Tory neurons and this is the anatomy of that sustains this is something we we looked at a long time ago in early life where we could see this reconfiguration of the circuits.

[00:01:13]
But. More recently we've been studying adult plasticity which is a different thing and the interesting thing about. Both plus to city and development activity dependent plus to sit in development and both of these forms of adult plus to city that we've seen is that you know despite the fact that the plasticity is actually effected by changes in the excited Torrie circuits all these forms of plus to city are actually triggered and cause to happen by the actions of specific types of inhibitory neurons and in fact in the 2 stories I'll tell today they are different types of inhibitory neurons and have a turnarounds are different from the excited Tori neurons because instead of being born and going straight up into the cortex they're born in a completely different place from where they excite you Tory neurons are called the ganglion a cabinet says and they have to migrate by a very long indirect route to get to the cortex where they differentiate and the different kinds of inhibitory neurons are born in different places and so the 1st story I'll tell is a involves of inhibitory nor on called Keane or ons which is not very important person nor on its value active intestinal peptide these neurons in addition to creating the inhibitory neurotransmitter.

[00:02:48]
Secrete they secrete a peptide called v.i.p.. And they come from one place in early development and then the 2nd story involves a much more common kind of intern or on which comes from a different place embryonic Lee but the plasticity essay we use in a lot of these x..

[00:03:14]
Parents' is derived from. Really work slightly more than 50 years ago by David you will and Thorsten these all who show that there is a a rapid form of activity dependent plasticity that takes place only during a critical period and early life if one eye is allowed to see normally and the vision of the other eye is blurred or a clue did then what happens within within a week is a rearrangement of the connections to the cortex that serve the deprived eye and a loss of response to the deprived in the cortex that is lifelong So just this brief period of unequal vision during early life can have lifelong consequences and until the last 15 years or so one or 2 percent one to 2 percent of American children would suffer this loss of.

[00:04:19]
A loss of decent vision in an eye that was with if they were born with the eye pointing at the nose or with a droopy I later with congenital cataract they suffered from this problem clinically although unless they lost their good I didn't. Compromise there like very much and this this rapid plasticity during this critical period in early development is associated with the pruning of the axons that go from the lateral genetical that nucleus to convey the jewel in but to the primary visual cortex and what we saw is that the axons normally make lots of connections in a big bush like this and after a week of visual deprivation during the critical period where as deprivation earlier or later the critical period has no such a fact these bushes get pruned and lose about half of their branches and half of their synaptic connections and the effects on visual responses past the input layer of the cortex these this direct input goes to a particular layer of the cortex Layer 4 but if you look above there in the upper layers of the cortex which is the 1st stage of purely cortical processing the plasticity is much more rapid So within $1.00 to $2.00 days most neurons lose their ability to respond to the eye that was closed and only respond to the eye that was open and it turns out that there is a sort of millimeter scale anatomical rearrangement of connections.

[00:06:05]
That correspond that also takes place not in the input layer but in the layer of cortex above there that core that happens about as rapidly as this physiological change so we studied all these things in cats and in ferrets and a little bit monkey surlier but to understand the mechanisms of what's going on with this activity dependent plus that city with the only tools that we had available which was essentially pharmacology and lesions.

[00:06:38]
Behavioral tools we made very little progress we just couldn't couldn't figure out what was going on we had these very interesting phenomena that were even clinically important but. But it was pretty hopeless and so. Largely in part stimulated by a wonderful graduate student Josh Gordon who is now the director of the National Institute of Mental Health but when he was in the lab he said Come on let's really work on genetically altered mice we have molecular genetics we can probe mechanism.

[00:07:17]
In a in a straightforward way if we use my niece and and so we started using mice and mice do have 2 eyes they can see you with both of their eyes so they have a bi Knockhill are part of the part of their visual system and in fact when you look at the anatomy you can see that there's a part of the visual cortex that gets input from both eyes 5 a lot of places as well as the part that's over on the side which can only be seen through one eye and when Chris Neil was in the lab.

[00:07:55]
Post-doctoral fellow what we found this when we looked carefully at the receptive fields of neurons in the mouse visual cortex that in the upper layers there were many neurons that looked like they could have come from the classic $962.00 paper that people in weasel wrote about the cat really the earliest discovery of how the cat visual system worked which turned is also true of the monkey and of people which there are these clear elongated cells with the long gate of receptive fields which respond to oriented contours in the world things are simple cells there are also other kinds of cells called complex cells and essentially all of the exciter Torrie neurons in the upper layers were more or less like this and the only neurons that were not selective for the edges for the orientation of edges of contour.

[00:08:53]
Are shown in blue here those were the hit Tori neurons so. You know knowing that the mounts visual system had a lot of similarities to the mammalian to the higher mammals that we're interested in. We looked and this was Josh's work at activity dependent plasticity and what we found was even though the mouse has overwhelmingly input from the contra lottery law in much less than it from the epsilon or lie in the body not killer zone if you closed one eye for as little as 4 days you would cause a change in the balance of input from favoring the contra ladder ally to favoring more the epsilon or ally and this phenomenon had a clear critical period you did the same 4 day deprivation earlier or later and nothing happened.

[00:09:49]
During this sensitive week there was a big change and we spent a lot of time to sect ing the. Chemical signaling mechanisms that cells that change to make this deprivation happen and to allow it to recover so actually if you open the deprived during the critical period things go back to their initial state and that depends on one's signaling system and so on and the nice thing in the last few years as we've been Act actually been able to watch this happen while it's happening in 2 photon my cross copy we can see the inputs disappear and some new inputs reappear over the course of a few days.

[00:10:36]
Watching if we put in a molecular marker that shows one of the the proteins that's important in the post an optic density or in the priest and terminals we can actually see this plasticity but like all things. Nora science devising a measurement technique ended up being crucial for understanding what was going on in biology I mean the big advance while I was a graduate student was my Ph d. advisor and I created the 1st computer driven optical display where you could randomly interleave visual stimulation record the responses and get quantitative measures of plasticity which everything else had been done by hand just estimating what was going on by listening to the sounds of the neurons is used in related by hand and our advance was to automate this an instrument that so we could get quantitative measures so the next technique that we invented.

[00:11:44]
Allowed us to get high resolution maps and measure responses rep reducibility in a noninvasive way in the mouse visual cortex using repeated presentation of a stimulus and analyzing the spectrum of responses so it turns out that when. Blood in the capillary bed in the brain loses oxygen because of the neurons in that area are more to more active the optical absorption of the him a glow in molecule changes and if you eliminate with the right length of light and one of the wavelengths that works is 610 nanometers what you see is the cortex gets darker so this is the response to a horizontal bar moving up in the brain moving up in the visual field and you can see this wave of darkness spreads through in the several of the early visual areas and so this technique allows us to make quantitative measure months of over.

[00:12:52]
All responses in the visual cortex that are you know the. Post-doctoral fellow who was working on it's that we measured $47.00 mice in a row and the signals vary in intensity by less than 5 percent and so they're highly stable in an animal if you don't cause plasticity it's the only thing that goes into the brain through an intact skull is light just a little bit of red light that much too much too little intensity to give rise to Photodynamic damage and having this technique.

[00:13:32]
If we restrict the visual stimulus to the by an ocular zone we could measure the response there one eye and measure the response separately through the other eye and calculate the balance of the inputs to the 2 so technical advances and I think that's important in a biomedical engineering program being able to make good quantitative measurements of something instead of.

[00:13:57]
Non quantitative measure months it's really important to be able to do good biology. So we were very interested in this plasticity that takes place during early development but all the work until that time had been done in anesthetized animals and we know there's no plasticity under anesthesia and it's these are completely prevents the plasticity So we were really interested in studying alert animals and fortunately an apparatus was being built the 1st instance of it was in David tanks lab at Princeton that allowed us to study the visual cortex of alert mice and it was a scaled up version of Something had been that had been done at the mock Spock Institute for Biological cybernetics in Tubingen with flies so they had a fly's walking on a ping pong ball they use that to study up to motor responses in the fly and the scaled up version is is this.

[00:15:06]
This is a mouse with its head fixed walking on a styrofoam ball that's floating on air so if the mouse pushes on the ball instead of moving his head which is fixed to stainless steel play. The ball just goes down and the bounce is actually really loves this he's very happy walk around sometimes still stand there sometimes if you've given some angel air pasta or a sunflower seed it will gobble it up and and yet his head is fixed so this allows us to use silicon microprobes to record your aunt's in his brain or to use to photon my cross copy to record signals optical signals from the brain and this let us study the alert mouse and much to our.

[00:15:57]
Satisfaction the properties of the neurons that we had previously measured in anesthetize mice were pretty much the same. Same in alert mice but one thing was really different and at 1st we didn't realize what was going on and we just said boy responses are a lot more variable in the alert mice but that turned out not to be the case what was the case was that when the animal was walking or running the responses were just a lot bigger and so we referred to the state in which the brain was put during locomotion as the high gain state because if you plot response as a function of stimulus orientation.

[00:16:42]
The red line shows what you would see in an anesthetized Now animal which is pretty much what you see it when an animal is quiet and alert like he stops running for a minute for a few seconds and you measure you can measure this response and then as soon as the animal starts to walk or run the response becomes much larger but the tuning is exactly the same so it's exactly as if you turned up the gain on an amplifier so there's no change in the spontaneous activity evoked activity is much bigger but the with the tuning the orientation selective any of the with the tuning doesn't change so this was a high gain state of the mouse visual cortex and it was a state of cortex it wasn't that anything coming from the from the eyes was changed although we worry about this because when we made simultaneous recordings from the lateral genetic ulit nucleus which is the pathway which is in between the eyes and the visual cortex there was no such change the responses in the nucleus or quantitatively unchanged.

[00:17:53]
During locomotion whereas the responses from the cortex were were much greater so this this was really interesting to us and we're interested in. Well what are the pathways that convey this signal to the cortex that the animal is walking or running and makes the responses have much higher gain and so to study this we we read a bunch of old papers about the midbrain locomotor centers so a Russian Orlowski along with some Swedes found that there was a place in the mid brain near the the put on killer Ponting tech mental nucleus where if you put an electrode in stimulated electrically even in they were studying cats even in a cat where the cortex have been removed the legs would start to move in the animal would start to walk or or if you stimulated harder would start to run now wasn't clear at that time that there were actually cells here that were doing that because it could have been that you were just stimulating some axons that were passing by but of course nowadays we can infect the cells with absence and so we can opt to genetically activate the cells with light and there are 3 different classes of cells there and we separately to genetically transfer acted them activated them with the light and made the cells fire look at the blue light as soon as the blue light comes on the animal starts to run blue light goes off he stops blue light comes on again he starts to run and stuff and so this shows that the midbrain locomotor Center actually is a locomotive center there are cells there that when you turn them on it makes the animal start to walk or run.

[00:19:49]
And not surprisingly the cortex went into the high gain state but actually this experiment was completely out informative for our purposes because you don't have to opt to genetically stimulate the midbrain locomotor center to get the cortex to go into the you know into the high gain state it goes into the high gain state whenever the animal walks or runs by itself what wasn't formative however was that we could turn down the intensity or the frequency of this up to genetic stimulus to a point where the legs no longer moved so the animal just stayed there quite a lurch but the cortex still would go into the high gain state so it wasn't these descending connections from this that make the legs move that were turning on the cortex it was one or another of the a sending connections and since there were a bunch of we next turned to the cortex itself to try to figure out what was going on.

[00:20:55]
And in the mouse at that time and now still there were a bunch of. Mouse lines that in which different classes of cells would express a recombination which allows you if you have a class of cells that expresses a recombination you can express a protein in those cells which only in the presence of the recombination gets expressed so I mean it's called You express a phlox construct which has locks piece cites.

[00:21:37]
Surrounding a stop code on and when that stop code on is cut out whatever is after that gets expressed and so what we wanted to do and what we did was one by one took all the create lines that were available all the with creepy combinations and expressed a red protein.

[00:22:00]
T.v. tomato in the cells wanted to time and then we made images of the cortex with 2 photo microscopy to record the calcium signals in these neurons so when neurons are active in fire action potentials calcium enters those neurons and there are Flora fores to the common one is.

[00:22:24]
It's called it's you don't need to know the chemistry. But essentially you the cell twinkles bright. With this calcium Flora for whenever it's active and so we look to see are there any cells in the visual cortex that become active in the dark when the animal starts to walk or run because you know in the light and the animal seeing things all kinds of activity is there in the visual cortex but in the dark most of the so.

[00:22:55]
Aren't responding to anything so our screen was to look for cells that were activated by locomotion in the dark and it turns out that there was a minor population of inhibitory cells that express the peptide value active intestinal peptide. That tracked locomotion in the dark so here's a measure of the running speed of a mouse as a function of time over.

[00:23:22]
You know what 8 minutes 450 seconds and here's the calcium signal inside a v.i.p. neuron and all of the do this that so you can see that these 2 signals are almost the same so that they actually track v.i.p. even in the dark tracks locomotion and the v.i.p. cells are sitting at the top layer of the visual cortex than 97 percent of the other cells in the visual cortex are like this that you care about locomotion at all so initially this was very puzzling because.

[00:23:58]
The phenomenon we had to explain is why did the neurons in the visual cortex respond so much more strongly to visual stimulation during locomotion and here what we found was an inhibitory cell and the mystery was solved I thing when. My colleague my present colleague although he was sending a go at that time Masa most gone Zeani and roughly used its lab in New York had a similar finding that the output of the v.i.p. cells is not to be excited Tori pyramidal cells of the visual cortex it's instead v.i.p. cells their output goes to some out of Staten cells which is another kind of inhibitory neuron and this amount of statens cells exert very powerful input on the visually responsive pyramidal cells so when the mouse starts to walk around these cells are active they inhibit the cells the cells no longer inhibit the excited Tory cells and so the specific visual inputs that drive the cells and make them selective for a particular visual stimulate those specific inputs just produce more spikes so if that idea were true then at least.

[00:25:18]
Some of the some out of step in cells in the this amount of setting create line labels actually 3 different kinds of cells that have in common that they express amount of Staten but are otherwise different from one another and about a 3rd of the some out of Staten cells show this behavior which is when the animals are running they become silent so that's consistent with this thing so if we were right that the v.i.p. cells were conveying this signal of locomotion to the cortex and putting the cortex in the high gain state we ought to be able to take a mouse which is just sitting there and it's k..

[00:25:58]
And turn on the v.i.p. cells and mimic the effect of locomotion and we minutely mimic it in these experiments not quite completely but pretty well so here's the response as a function of orientation of a cell when the animals just standing there. Without locomotion and this is the response when he when we opt to genetically activate activate in Ops and in the v.i.p. cells that makes the p. cells fire a lot and on average the responses are greater when the animal when we activate the v.i.p. cells even without locomotion so we can partially mimic the activity of locomotion by activating the eyepiece though and more importantly if the animal is actually walking or running Can we turn off the v.i.p. cells and block the effect of locomotion and this turned out to be a very frustrating frustrating experiment because there were centrally 5 techniques people had for transplant leap turning off cells and we tried them you know one after another and none of them were successful in and recorded the activity of the v.i.p. cells and the v.i.p. cells get very powerfully activated by the commotion through a pathway that I won't talk about now and none of the things that turn off cells were powerful enough to overcome this excitation So finally you know The Post who was working on this came and to my office you know when the 5th one failed we're feeling really bad we wanted to finish this work and publish the paper I said let's just blow the blanker away so that's what we did we took.

[00:27:57]
Our microscopic field are we labeled our v.i.p. cells with a red dye t. to May fluxed t.d. tomato that we expressed under the control of the the IP promoter so we could tell which cells were v.i.p. cells we put a chemical in that let us record the calcium signals and we measured.

[00:28:22]
We measured how much does locomotion increase the response of the most of the cells not counting the v.i.p. cells which as you know the 97 percent of the other cells that are activated by visual stimulate and you know it was almost doubled the responses. To locomotion we knew that would happen anyway then we went in and one by one focused our laser on the v.i.p. cells and blew them up I mean Photodynamic Lee damage them we said and and after they were damaged we just repeated what we've done before and measure the visual responses of all the other cells.

[00:29:08]
During locomotion while they were stationary and during locomotion locomotion had no effect anymore. Changing the activity of the cells of the you know instead of almost doubling it it really didn't change it at all so with the v.i.p. cells going on this. You know we really block the effect of locomotion So this was this was very exciting to us I think it was was quite strong evidence even though it wasn't reversible.

[00:29:40]
You know that if the v.i.p. cells aren't there anymore. Locomotion doesn't work so it really was strong evidence that the v.i.p. cells were the pathway by which locomotion puts the cortex in this high gain state so given that the cortex was. More active we wondered this was in normal adult mice.

[00:30:06]
Is there is there more plasticity if there's more activity you would think with more spikes there be more plasticity maybe and so we used a phenomenon that's relevant to the human condition of amblyopia who have compromised vision through one critical period in early life and so we deprive mice through their critical period for a couple of more months and then we open the deprived eye and measure the magnitude of the responses to their deprived die over a period of recovery and what happened is exactly what happens in human amblyopia with one of my colleagues in the author Mala g. and neurology departments we've looked at adult human amblyopia who lose their good eye and measure their vision and their bad guy and it recovers to some extent but it never gets more than about halfway to the normal range and that's what happened in the mice so we asked Suppose we give these mice visual stimulation while they are during this period of recovery if the if vision responses are enhanced during locomotion doing enhance the recovery and indeed that was true so if we allow the animal to see a specific visual stimuli during recover recovery the response to that stimulus recovered almost to the normal range within a week whereas if we just let the animal have locomotion with a great screen or if we let it have the visual stimulation actually for twice as long but in their home cage so they didn't have an opportunity to run around that was just like not giving them any special.

[00:31:57]
Treatment at all it was just like leaving them in their own cage they didn't recover very much. So we had kind of 3 stories we we showed running enhanced as responses we show that there's this circus involving the eyepiece ells in the cortex and I haven't told you about the subcortical parts of that circuit that's activated by running to enhance responses and we show that running enhanced as plasticity and presumably through the enhanced responses but but you know it wasn't clear whether this enhanced plasticity directly or maybe it's just you know aerobic exercise which you know you read Runner's World and actually the New York Times Science Science Tuesday column and you know exercise is good for the brain exercise is good for you so maybe it's a completely indirect effect that and handsome as this plasticity and not you know and not affect where we're thinking about and fortunately for us around that time.

[00:33:10]
Tom suit off Tom suit of Stanford had come up with a wonderful molecular agent that we could use to silence v.i.p. cells so tetanus talks and you know if you've got tetanus it's fatal because you don't breathe anymore because you can release an optic transmitter and make your muscles contract but he made a fluxed tetanus toxin like chain.

[00:33:40]
Which you can. Use to target to v.i.p. neurons among other things he hadn't used it with the IP neurons and silence their output so they can't release neurotransmitter anymore the cells look perfectly happy anatomically they look normal but they don't release neurotransmitter anymore and the. The effect of running the responses on average is just abolish the measuring responses while they are stationary or running.

[00:34:13]
Shows no change but the responses are normally selective for visual stimuli these are just a few examples so we silence the activity of the v.i.p. neurons and did our essay of recovery while the animals saw a visual stimulus stimulate stimulate during locomotion and normally they recover pretty well and if we block the output with tetanus toxin the v.i.p. block the output of the v.i.p. cells with tetanus toxin they they didn't recover so well they recover just like control Donna Mills which I should've put in this graph.

[00:34:56]
So this was consistent with the idea that we had before that the v.i.p. cells that their activity putting the cortex in the high gain state really did enhanced plus tips that easy. And if we up to genetically activated the v.i.p. cells it turned out in adults if you normally I said you know if you close one eye and it's outside the critical period there's no effect and that's indeed what we saw in the controls but if we opt to genetically activated the v.i.p. cells during this period of time we could get an effective monocular visual deprivation so blocking the v.i.p. cells prevented plasticity activating the let the cortex have a lot more plasticity and I see I've been talking much slower than I should of and I know I have to finish in the next 7 minutes so I'm going to go pretty fast.

[00:35:57]
Fortunately when we were doing these recovery experiments we used 2 different visual stimulation. Some of the animals saw a spatial temporal bandlimited noise stimulus like this matched to the spatial temporal frequency response of the mouse visual system. And some of the animals instead saw a traditional visual neurophysiology stimulus which is a sine wave gratings moving at different orientations with different spatial frequencies and so on and it turned out that the recovery we saw in this essay was specific for the visual stimulus that we used if we tested the animals with the noise pattern the animals that had experience with the noise pattern during locomotion recovered beautifully the animals that had experience with the gratings didn't their response to the noise pattern didn't recover at all it was just a control levels and similarly you know Similarly if we the animals that had experience with the gratings their response the gratings improved.

[00:37:09]
Mislay their response to the noise pattern didn't so we wondered just how specific is this plasticity that's produced in the adult visual cortex by locomotion. And is it specific only when recovering from pathology like the effect of monocular deprivation so we looked at just can't can you does this form of plasticity really just enhanced responses in normal intact down a moles measured with this completely noninvasive technique using the intrinsic signal imaging and indeed.

[00:37:49]
We had 3 different stimuli in this experiment some of them saw vertical bars some of them saw Harz on the bars some of them saw the noise pattern and in each case and these were just normal animals they saw it for an hour a day for 5 days and then we measured again an hour a day for 5 more days and then we let them sit in their cage for a week in measuring them again and what we saw is there is a stimulus specific and Hance manner of response that's only for the stimulus that the animals had see and it depended on locomotion because if we did exactly the same thing but turned off turned way down the air under the styrofoam ball so they couldn't run anymore and they they gave up trying to walk then.

[00:38:37]
There's really no enhancements no significant enhancement whereas this is the enhancement if they were allowed locomotion. So we were wondering is this plasticity specific to individual cells and so we did some more to photon calcium imaging and this is just a picture of what the calcium imaging looks like you can see that the different cells twinkle at different times when they're activated by a visual stimulus.

[00:39:09]
Yes. And. You know this is just these are the primary data for these experiments so what we did in the next experiment was allowed some animals to see a gray screen during locomotion some animals saw 45 degree bars some animals saw 135 degree bars and we looked at individual cells and you know if we measure the calcium responses of individual cells these are the responses to bars moving in different directions before and then after 5 days of stimulation basically the animals were seeing a great screen there was basically no change on the other hand if you look at a cell that was selected for 45 degree orientation here's this response as before and this is $45.00 degrees moving in one direction moving in the other direction this is the response and then after 4 days 5 days of an hour a day of stimulation with $45.00 degree lions during locomotion the responses were very much greater so there was a there was an enhancement that.

[00:40:25]
Was really quite dramatic and if you took cells that originally responded to orientations near 45 degrees or 225 degrees the the 2 directions their response was very dramatically enhanced by stimulation with 45 degree lines whereas these cells were not enhanced by stimulation with 135 degree lines and the opposite was true with cells that were originally selected for 135 degrees but I actually like to look at a plot like this and it's the same thing if the original preferred orientation were near the orientation we stimulated with during locomotion.

[00:41:09]
You see there is a dramatic announcement of these cells whereas the cells. Were the cells whose initial preferred orientations were different were much different from 45 degrees weren't enhanced at all by by the 45 degree stimulus and the same thing with 135 degrees and and this is animals that saw a great screen you know this the fluctuation here shows like the noise in the measurement so the conclusion is that there is this rapid and persistent increase in visual cortical responses and.

[00:41:48]
And even with the noninvasive method or with a slightly invasive method the calcium imaging what we see is a cell a single cell specific change in response. And what I'm not going to have time to do because I know you guys have a class at 1215 actually take 3 minutes and ask what's special about the high gain state here you have an animal who he sees something repetitively and his response is for the cells that the enhancement is proportional to the original response that stimulus if it responds well to that stimulus it will be enhanced by 50 percent and that an answer will persist for weeks what's best about that state and one question is is there actually more information in the visual cortex about the world when the animals in the state and so we can measure information by asking if we record from a 100 nor on simultaneously at the mouth his brain can we tell which stimulus the animal is looking at so your ability to predict with a support vector machine or linear discriminant analysis or one of the data analytic techniques your ability to predict the visual stimulus on the basis of the response over a narrow period of time is a measure of how much information there is and the representation in the visual cortex and so we measure the information.

[00:43:25]
And there's an increase in information in cells in all of the layers of the visual cortex during locomotion but of course there are more spikes so the interesting thing is is there some change in the pattern of response in the cortex and if you look at the distribution of responses while the animal stationary and the distribution of responses while the animals walking or running they're not completely separate there's a region of overlap where you have the same number of spikes in the 2 cases but in one case it's locomotion the other case is stationary and so.

[00:44:01]
When we analyzed the information and information increases down what we see is even in cases with the same spike same number of spikes there is more information when the animal's walking or running than when a stationary so there is some change in the organized pattern of activity so how do you make this kind of real world thing well you can measure you know the longer an animal looks at something the more information there is in in the spike trains that we record so we can ask suppose you get this amount of information when the animal's walking or running How long does the animal have to look at it while he's stationary to get the same amount of information and in $100.00 milliseconds of response during running it takes between $3500.00 milliseconds to get the same information about what's out there in the world when the animal stationary so.

[00:45:04]
I'll skip this you know we were trying to figure out what Which of the things that piece cells release is responsible for this plus does that and it turns out it's not the peptide and we actually think. Can any you guys read this this is in front 2 or in German gothic letters.

[00:45:30]
This is what it is so you know when I learn German I could read German novels and stuff but if I read an old German paper which is printed in this fine it would take me forever I have to go letter letter letter letter this perceptual darning lets you learn these alphabets Now when I look at this it just looks normal as comprehensible is this so we're studying whether my use whether this system is the one that is responsible for perceptual learning in mice and we don't know yet so I'll leave it that so I think this enhancement phenomena is a really interesting thing it's it may be the basis of perceptual learning and it's been a very satisfying thing to try to discern this circuit in the mouse's brain which is turned on Fortunately for us it's turned on by locomotion but this isn't our brain too and these v.i.p. cells are exactly the ones that are receiving input from the frontal cortical areas that are involved in planning what we're going to do now exist and so I think this very same circuit is it's likely to you know it was just lucky for us that locomotion activates it in mice but other things activated too and put the cortex in a state where it's set up to have more plasticity bigger responses and more plasticity than it happens in couch potato mode so the one thing I want to do is show you a picture obviously I could only tell one of the stories.

[00:47:13]
Show you a picture of the people in the lab who did this work. And. Mostly they have their own laboratories out Chris Neil is in Oregon you food is Singapore who you did the work on the piece elves Chris did the work on alert animals. Anyway we can eco did the work most of the plasticity experiments Maria daughter lot who has her own lab at Purdue now is did the work on the representation of information and so on Anyway I've been blessed with wonderful people coming to my lab who done great things I don't exactly tell them what to do but I try to make them understand that what I think is the most exciting and interesting thing to do.

[00:48:06]
Try to get them to think that too and then. Then they're there industry and creativity is released and they do things much better than what I would have done by myself so thank you. Thank you. Thank you I'm happy to answer questions if anyone has some I know a lot of you have to get a class so.

[00:48:35]
Yes So I think. It. Was. A yes. Yeah. Yeah. I mean I am too I mean. Basically So I mean one of the things that we thought might be true but turns out not to be true is the faster you run the faster the world passes you by and maybe the accumulation of information would be more and more but it's not true it's binary it's either locomotion or not and so what I think is really going on is that that.

[00:49:32]
Locomotion is only one of the things that turns a v.i.p. cells on and I think you know focal attention or importance things coming from the front of the brain back to the visual cortex also turn the v.i.p. cells on it's just that we don't know how to control those.

[00:49:49]
And so we were just fortunate we have a switch to turn on and off that let us probe the circuit but I think what this circuit is really doing is is activating the high gain state and the plasticity system. In these other circumstances so but you know that's kind of and I mean that's kind of bullshit at this point but but I do think it's very likely to be true and I'm trying to persuade you know the technology that we use in my east is only available in the way we did it in mice in mice because of all these genetic altered laws ends however with viruses you can now do this sort of thing in other animals in David Fitzpatrick slab in Florida has been doing it ferrets it's clearly going to happen in monkeys and so on so we will be able to probe these systems in similar ways in brains that are much more like the human brain pretty soon so it's not in it was inaccessible when we started doing these things in fact the whole world of genetically altered my you systems neuroscience for the 1st 5 years of this we could all meet in one room a quarter the size of this and we did every year we would meet a Cold Spring Arbor in meetings that I'll see no Silva organized.

[00:51:25]
You know and it was air Condell and me and Stan McKnight you know there were like only like 5 or 6 labs and we all exchange mice and exchange 3 agents and helped one another. And it seemed like it was it might never be the case that we're going to be able to use it in.

[00:51:46]
In ultimately in people and certainly in large animals but that's that's clearly not going to be true the advances in the virus technology in the last couple of years of made it clear this is going to happen so. You know. When they. Start It's a really interesting question and I wish we.

[00:52:17]
Mean they're just 4 of us in the lab now so I I'm not sure that I hope somebody looks at that because I think it's a very interesting question and I have no idea we we're doing the only experiment doing we're doing the plasticity the enhancement as a function of age one of the slides I skipped over because this is a big concern for me is what about aging you know old mice have this plasticity capacity for plasticity that you know young vigorous mice have when they're $56.00 days old and it turns out all my niece if you do the same amount of experience only have about half as much plasticity but if you do it for twice as long eventual a they get to the same level of an answer as as young mice do but unfortunately for old mice it goes away much more rapidly whereas in young mice it lasts for weeks as long as we've measured.

[00:53:19]
The story is so anyhow we haven't done the babies yet critical period ones and so we don't know it's entirely possible that the babies are always in the high gain state and you don't need locomotion but. Ask me in 6 months. Yes I. Speak very loud I don't hear very well.

[00:53:48]
Yeah. That's an interesting story so there's a there is a group in Italy which has case reports from a small number of stew from a small number of patients that it works very well to enhance recovery from amblyopia. Nobody's done a proper blind controlled study so I don't know if it's true or not.

[00:54:24]
I have been I have waited. Before doing that to try to get some indication from human electrophysiology whether the same thing is going on in human visual cortex because one of the things that's associated with this high gain state is a big increase in gamma oscillations that you know sort of 50 to 80 Hertz.

[00:54:51]
In the mouse brain and so I've tried to measure gamma oscillations in people on a treadmill and it's the gamma signal it's small and there's a lot of artifact and we haven't succeeded in doing this because if one were going to do a clinical trial I think you would 1st want to have evidence that this phenomenon actually exists in the human visual cortex but anyhow in Italy.

[00:55:18]
There are only positive reports no negative reports so I think it might work but. Anyone else yes. I don't know I don't know I mean it's puzzling because. You know if you if you look at as what are the indications of familiarity with the visual stimulus So Nicole rust.

[00:55:57]
And Jim De Carlo have done some at mit have done some experiments with monkeys recording from the end for temporal visual cortex where cells are selected for complex objects and asking you know if you show them 50 things and then you show them some more some of which were the same as the earlier ones what happens always in their studies and other studies I know of is that the response to the object that they've seen before is less than than it was Whereas what we're seeing is an enhancement and it may be a difference between my use and monkeys it may be a difference between and for temporal cortex and the want so I don't know but I we haven't we haven't seen anything that makes it worse I mean it's it's sort of interesting when people talk about cortical state what they usually meant were different stages of sleep and then awake and anesthetized where the other 2 states it's now clear there are at least 2 cortical States in my use the During alertness when they're capable of responding and I don't know anything bad about the high gain state but it.

[00:57:14]
But we haven't measured anything that would let us as it's an interesting thought yeah. There's. A. Whole. More. Yeah yeah yeah. Yeah I mean it's. A lot of the you know the mice there are these wonderful experiments that the 1st ones were done decades ago but essentially if you have a dark a looming object a dark spot that gets bigger or something that's the most terrifying thing in the world to mice up about like an aerial predator you know but it can just be a dark spot that expands and the mice will either freeze or they'll if there's a hidey hole for them no run into their little how you hole and not be eaten by the owl and that's it turns out it's not done by the visual cortex it's done in the mid brain so there are built in mechanisms for at least that kind of predator escape behavior that don't involve the cortex you know you can take out the visual cortex Meisel still do this so whereas the kind of things we've been investigating all require the activity of the visual cortex.

[00:59:02]
So it's functional arguments I mean I think a huge hole in this is an area where engineering is going to make all the difference in the world. Is we really we really don't look at natural behavior and what would be wonderful be to record lots of neurons in the brain in unrestrained mice in a natural environment where they're running around encountering things and we could tell what they're looking at and so on I think this is this is going to be the future of neuroscience but it's it takes a lot of new technology to figure out how to do this in in a meaningful way and.

[00:59:52]
You know we're just we're just not there yet but I think I think you know for me I was really worried that the mice we were using which for 200 generations have been raised in cages these mice don't need to see that people feed them right they all they need to do is be able to smell and to find the right place on the opposite sex to mate and that's that's really so I was really worried that we were looking at this totally vestigial to generate visual system and Chris Neil and Jan and his post doc did the most wonderful experiment after he went to its lab in Oregon what they did is they made an enclosure about a metre metre and a half square that mice couldn't climb out of they ordered mice from Jackson Labs so they had a cage of 5 my use and they got 5 crickets little crickets like this you know threw them in the mouse cage the night before after the mice came from Jackson the mice they shared I don't know whatever they did you know they ate the crickets they like eating crickets the next day you took one of these mice out that had been born in a cage from 200 generations of mice that are always been in cages you put the Malise down in the enclosure you put a cricket there and the mouse is like a guided missile he loves crickets he goes he navigates he chases the cricket and the cricket moves the mouse goes through.

[01:01:30]
And catches the cricket and eats it of course and so it's very clear even in these genetically altered mice that you know have been outside the natural world for a long time they still have visual pretty they're really good at this. And that depends on visual. Techs not just so.

[01:01:56]
It made me a lot happier about working with our laboratory mice who who haven't lost all capacity for doing mouse like things but I think it is a good level professor at Goettingen who who was a post-doctoral fellow in my lab she has done these experiments raising my in a in a and enriched environment which is like a 3 level mouse condominium 2 meters by 2 meters by 2 meters with lots of floors lots of hiding places bunches lots of social activity and where the technician hides the food in different places and changes the hiding place every couple of days so this is much more like a natural world for the mice and her mice have dramatically more plasticity than the mice normal cager it might use and if we could all afford it I think you know because it's very expensive it takes up a lot of space and a lot of time but the mice in these and rich environments.

[01:03:09]
Have much more plasticity than the. The mice we see so I I think are my use might be like the Romanian orphans you know you know this you know in Romania where they prohibited birth control and and how they were when the remaining fell apart there were huge numbers of orphans an orphanage was basically kept in cribs with no social stimulation they were just given food and their development was tremendously there was a u.n..

[01:03:49]
Delegation to sort of try to give therapy to to these children and they were dramatically impaired for the rest of their life by this very early social and sensory deprivation. Dekalb the psychologist was one of the people who went in to see them so so I think our cager and mice really are.

[01:04:15]
Not showing all the capacity that mice in the natural world probably have of course most mice in the natural world get eat but. So yeah. You know. Great Ok.