(beeps) (light upbeat music) – I wanted to sort of pitch or present the work today in the context of the effects of stress because of the high burden that mental illness has on young people. That is as many as 40% of young people today experience a mental illness. The majority of these are related to anxiety and stress. And if we don't treat these early, they can go on to lead to chronic illness, both psychiatric as well as physical illness. So the questions that we've been asking in our laboratory over the years is how are these early life experiences impacting emotional well-being and the underlying brain circuitry involved in that well-being? And ultimately, what we need to do is to use this information so that we can facilitate and enhance healthy brain development and also, of course, well-being of young people, which is then gonna be well-being of our society. But if we first take just a simplistic look at all the changes that are happening in post-natal development, what you'll see is that there is dramatic changes just in the number of synapses throughout the brain that develop regionally. So we see a proliferation and a subsequent pruning of these connections.
Simultaneously, their changes in nerve chemicals and neurotrophins that are absolutely essential to development and learning. And with all these regional changes, it is co-occurring with increased myelination throughout the brain, which is making the connections stronger. So we've been focused on all these dynamic changes and trying to understand with just one picture using human imaging and focusing on one circuitry that's really important for our ability to process emotional information. So emotion regulatory and emotional reactive processes. And during this time, during childhood and adolescence there are major changes in this circuitry, and we think that they will, are one pathway for us to seeing how early life experiences are reflected in them and how this continued development may also be a window of plasticity of sorts where we may have the biggest change.
Now, in human studies, one of the ways that we try to look at this, at these circuitries in the behaving human brain non-invasively, is to use simple stimulator cues that we present. So here are, here's an example of a cue with emotional information. We examine how these cues impact both brain and behavior. These cues can be positive, smiling faces, or negative, or they can be neutral. But what we see is even if you are in the laboratory performing this task, if we ask you just to press a button whenever you saw one of these cues, you would be significantly longer in detecting a threat cue, a fearful cue, than you would be to a neutral or happy as you see here. So we have a longer latency there. It's adaptive when we see cues that have some uncertainty or ambiguity. I don't think for a minute when I present this cue to you that you are threatened by that.
However, we learn over a lifetime when we see cues like this that there might be potential danger, and what's very important for us to understand is in this context, is it a threat? So if a bear were to come into the room, I might have an emotional response. But the circuitry is important if I saw a bear in a zoo to know that that bear, in that situation, I would be safe because they were behind a cage or wire. So now, if we look at these responses in this change in our responses depending on the potential threat of information in our environment, if we open up the brain and look inside, the areas that tend to be related to that delay in our approach behavior in such situations involve the amygdala and the prefrontal cortex.
And hopefully, what you note from this slide is that the amygdala, the more active it is, the slower you are to respond. The amygdala is very important in picking up the emotional significance of information in the environment and has been associated with emotional reactivity. In contrast, the prefrontal cortex, which has direct projections to the amygdala based on elegant animal work and more recent human work, that area is less active when you're really slow to respond. And the more active it is the quicker you are. That is, you see a potential threat. You are able to understand in this particular situation with repeated presentations of it. Get over yourself. Alright, already. Nothing's gonna happen to you. Now, I've just made a sort of claim that over time, nothing's gonna happen to you. Well, let's look at that in the brain and what systems are absolutely essential for that. What's important when we look at with those repeated presentations how well you habituate to these cues is this inverse coupling between the prefrontal cortex and the amygdala in this circuitry.
And this inverse coupling or negative connectivity seems to be changing radically from childhood to adolescence. And it is when you are not able to habituate that response that is associated with heightened anxieties, so I just want you to focus on this quadrant for a moment. And basically, if you look over at the Y axis, that's amygdala habituation, and it's a negative score. That means not only was there not a change in how active the amygdala was to these cues but that it was actually sensitized and it was increasing over the course of the experiment. And that's related to the highest level of self-reported trait anxiety. Now, some of these pictures had a lot of data, and I think data slides and pictures are worth 100 or 1,000 words, but movies might tell a simpler story. So this is an example of an individual who reports low anxiety, and they're being presented with these cues of potential threat.
So you'll see, there's bilateral activity in the amygdala with repeated presentations, but then, it's like the system begins to return to baseline, and it turns blue. That represents from the increase going back down to baseline. In contrast, when we look at an individual with high anxiety, we basically see a similar pattern at first with this bilateral activation of the amygdala, but then, it stays up and it stays up and it stays up. And it doesn't return to baseline. So it's that vigilant state of that anticipation of threat and an uncertainty of what these cues potentially mean. So, two different paths that we have taken in trying to understand these two very different neurosignatures here is to look at differences in genetic factors that may explain this, but in the interest of the symposium today, also, in the environment or experiences that we have. So how does early life stress impact the development of the circuitry? And we have used really an extreme example of early life stress. It is an unfortunate but a natural occurring one.
And that is children who grow up in the orphanage. Now, in the orphanage experience, all the orphanages vary. But there's always gonna be some fragmented caregiving because of the high ratio of so many children to a single caregiver. So what we've been interested in is as these children are adopted to families in the United States, we've been trying to understand how this disregulation of their needs not being met, how there's not attunement in the child's needs with what a caregiver can actually provide to them, how that impacts their ability to regulate self. But particularly today, I'll talk about how to regulate emotion. So again, we use these simple cues, and we ask children who have been adopted from the orphanages as well as a comparison group who have not been adopted, who live in the United States. They all have moved. The adopted children have been here for at least two years to make sure they get acclimated to their new home and their new culture and environment. And what we do is we present these cues of potential threat.
But in the task we tell 'em, you know, ignore these. Try not to pay attention to them. And you find, in the situations in which you have these cues relative to situations where you present a smiling face, that the adopted children are much slower in anticipation of one of them occurring. So even if it's a neutral face on the screen, if they know they're about to see a threatening face, they're really hesitant to respond. But if it's to a happy face or a smiling face, you don't see a difference between the two groups. Now, this is paralleled in the brain by enhanced activity in the amygdala. And that's shown here on the left and the right side. But what I think is important about these findings if first, this is greater activity in the children who were adopted from orphanages abroad relative to the comparison group. The area in blue is actually more active in the comparison group. This is a part of the brain that's important for attention regulation and emotion regulation.
So one is being able to regulate their emotions and ignore. And the other is quite reactive, that group. But more importantly, when you see images like this, and you have these neurosignatures, it is how does that relate to their actual behavior when you get them outside of the scanner environment or the laboratory. And so, we measured how the behavior between caregivers and their adopted children when they had been separated briefly and then, they were reunited. And we looked at the amount of eye contact they had with their caregiver when they were reunited. And basically, we see that the more active the amygdala was the less eye contact they had with the caregiver. Also, the less eye gaze they actually held on the faces that they were presented in the experiment as well, too. Now, when we have natural occurring experiments such as this, we don't really have control on preexisting conditions that may be there, genetic or environmental.
So we've actually turned to use mouse models to try to see if we can control for the genetic and environmental confounds that might otherwise explain what we're seeing. And so we borrowed paradigms that have been developed by individuals like Regina Sullivan at NYU and also Tallie Baram at UC Irvine to, and do sort of a fragmented caregiving of a dam to her pups. And so I wanted to show you those movies. On the top is a dam where we've taken away the nesting material. She has a little bit but not sufficient, so she is running around, oh, this is not in real time. (laughs) (audience laughs) She's a little bit slower than that. So basically, I don't even know if you can see the control dam. So she has all the nesting material she needs. And she's spending the majority of her time grooming and feeding, nursing her pups. But you can see the mother above is trying to pull together the nesting material and sometimes her pups over to her corner for the nest.
So basically, you're seeing this fragmented care because she can't attend to her pups when she's doing other things, which we thought might, in the slightest way, mimic some of the fragmented care we see in the orphanages. And if you look within a two hour period, you see a significant difference in the amount of time that the dam is spending with her litter relative to controls. But quite frankly, if you look across a 24 hour period, they're spending almost as much time with physical contact, but it's very fragmented in that contact. So there's not that attunement between the caregiver and the pup. So then, if we look at their ability to regulate their emotions, not the dam, but now her offspring who've been given this fragmented care, basically, what we see is a paradigm in which we've trained them that a nozzle will lead to them having access to condensed milk. And mice love condensed milk.
And so we put that nozzle in a novel cage with a bright light where there's potential threat for mice. And we see that when we put it there, there's a difference between those mice that grew up with the stress dam relative to the mice whose dam, their mother, was not stressed. And you don't see any difference in terms of how quickly they move to the nozzle in their home cage. And then if we look at the brain using C-Fos activity as an index, we see heightened amygdala activity in this group relative to the controls. And so this gives us a bit more confidence that what we're seeing in the human data that I presented is not as much associated with maybe preexisting conditions as it is with the early life stress because these parallel quite nicely. Now, with the mice, unlike our human children who are usually adopted by super parents, if you do nothing after that, you see that there's persistent affects of that early life stress so that they're still showing heightened amygdala reactivity in adulthood even though there's continued development of the circuitry. Now this persistent effect in these mice is somewhat reminiscent of work that Nim Tottenham has done at Columbia in collaboration with Dylan Gee whose now a colleague of mine at Yale University.
And she's actually shown that if you look at this frontolimbic circuitry, that there may, there appears to be something similar to a premature closing of a sensitive period of neurodevelopment of this circuit such that if you just look at healthy children, what I described before, there are drastic, significant changes from coupling between the prefrontal cortex, and it becomes inverse or negatively related in adolescence. What we see in children who have grown up in the orphanage is you're already seeing those changes early. And so now, Nim is trying to follow and see just how rigid does that make the individual when they go through adolescence, which is an even more stressful time of life in meeting so many challenges. And I just want to end with one more study, an area of work that Dr. Tottenham is following up on, and that is showing how important the caregiver is.
And so, in these experiments, where we're showing these very simple stimuli in the scanner while we're taking pictures of the brain and watching how they perform games, if you simply put a picture of the face of the caregiver along the screen where they're performing the task, and you can counter-balance that with a face of a stranger, she sees that that's associated with decreased activity in the amygdala, so it's a decrease in that emotional reactivity just by having that parental cue present there. And also, there is an increase in the inverse coupling with the prefrontal cortex that is typical more of adulthood, but we're seeing the parent has that ability to regulate. So I hope what I've shown you or illustrated is just one small set of experiments that are being performed where we can show that early life stress can lead to persistent changes in brain and behavior, particularly in terms of emotional capacities.
And it highlights, too, this last bit of work, the importance of having very early interventions and also, the importance of the caregiver in helping to develop a healthy brain. And also, in terms of enhancing emotional well-being now and hopefully, for that individual. So I just want to end by thanking so many individuals who have come through my laboratory over the years, the majority of them fellows who are stellar stars now. And to also thank you for your attention. – I'm really delighted to be able to participate in this wonderful symposium. What I would like to tell you today is one particular aspect that is absolutely essential for a normal brain development. And that is to have proper parenting. We've heard from the wonderful talk from BJ Casey right before me how important proper nurturing from caregivers is for normal mental health of the infant that then become adult. And so the question we've been very interested in is what makes a parent a parent? If you think about parenting involve one, two, sometime multiple adults that take care of an infant.
The relationship is completely asymmetrical. There is really a very helpless individual that requires a lot, a lot of care for a very, very long time. And so what makes parents be parents? It's a, I think, a very interesting behavior, long-term behavior, that has a lot of emotional components. And so this is a behavior also that is displayed by many animal species. It is absolutely essential for the development of and the survival of the species. So the idea is that there maybe some genetically pre-programmed neurocircuit in the brain, and what we are interested in is really trying to understand the neural basis. What are the specific neurons that are involved in the control of parenting behavior, in the display of parenting behavior? And when is it that this behavior actually goes wrong? And there are a number of circumstances in which the behavior goes wrong. One, this is one of, I think, one of the most outstanding, really impressive, or very surprising slide related to mental illness.
That's the number of psychiatric hospitalization after childbirth. You can see right at the time of childbirth this enormous peak in psychiatric hospitalization of women, and this are patients that suffer from post-partum psychosis. This is very quite rare. It requires hospitalization because these women have obsessive thought about harming their child, their children, and therefore, they require very intensive care. These very way less severe from, which is post-partum depression that affect 10 to 20% of mother and 5 to 10% of fathers in the US. And this also comes right after birth and result in a inability to emotionally connect with children and in effect impaired parenting. Now, there are quite a number of risk factors: stress, life circumstances, prior depression, and sensitivity to hormone changes. As it turns out, right at birth, there's an enormous change of hormonal levels in the young mother.
You can see here progesterone, estrogen, oxytocin, prolactin is really an enormous and very sudden drop in hormone level. Some people call these hormonal cows, and women, in particular, that are sensitive to these hormone fluctuation have really a difficulty in coping with those changes. So how do we understand the positive and the negative regulation of parental behavior? Well, let's turn to animals. So in animals and human as well, the primary caregiver is usually the mother, and in animals, and in mammals in particular, this makes a lot of sense because in mammals, the mother nurture the fetus, the embryo, in utero for a long period of time an animal is involved in lactation, so really involve enormous maternal resources in nurturing the infant and therefore it makes sense that mom continued to nurture the infant through parental behavior. What about males? Well, in most mammalian species, a very large subset of mammalian species, the males actually attack the infant. Attack them sometime, or very often actually kill them. And this infanticidal behavior has been widely observed in many animal species. And it is thought that this is actually an evolutionary-driven aggressive behavior of the males in order to gain access to the females that are not accessible when they are nurturing their own infant.
And interestingly, this is a behavior that is absent in monogamous species in which both the male and the female are nurturing their infant and therefore, there is not this conflict of access of the male to the female. Now, infant neglect and aggression is also occasionally displayed by stressed female. This is seen in animals as well as I mentioned in post-partum depression and psychosis in human. Now, we're not working with human. We're not working with primates. We're working with mice, and in mice, there is as in this animal species I mentioned, a very clear sexual dimorphism in the behavior of animal, male and female towards infants. Females even virgin females or sexual inexperienced female, spontaneously take care of infants. They will build a nest. They will lick and groom the pups, retrieve them to the nest and huddle around the infants. Virgin males in contrast will spontaneously attack infants, wound them, and kill them through infanticidal behavior.
So very different set of behavior. However, males are not always infanticidal, and in fact, strikingly, males after mating with the female, are no longer infanticidal. And the video I would like to show you here is from, if you wish, a certified infanticidal male. This male attacked infants three weeks earlier, and then, we had that male mating with a female. And then, we're testing the behavior of this male now with these pups. And, as you can see, the male has built a nest, and now, is retrieving these pups one by one, and I should mention by the way that these are not the pups of that particular male. This behavior will be displayed towards any infant that is, to which they are exposed. So all the pups are now collected to the nest, and as you see, this is a really good dad. (audience laughs) He is checking that he has not forgotten any pups. Okay.
Nobody left. Good. (audience laughs) Let's go back to the nest and then take care of the pups. So this is a really fascinating switch in behavior that really indicate that even in infanticidal males there's the ability for these males to be parental. So how is this happening? Well, evolutionary speaking, this is actually not that surprising because if you look at virile animal species, insects, fish, amphibians, reptiles, birds, and mammals, they are species in which the female is always the one handling the progeny. You can see the little eggs here and little larvae over here. He's a tadpole on the back of that frog. But very similar species, genetically very similar, show animal in which there's bi-parental care, so both the male and the female handle the nurturing of the progeny, and similarly, also very similar species genetically have the male that exclusively takes care of the progeny.
For example, this beetle here with the little eggs here or this male amphibian with a tadpole or here this fascinating monkey, the Titi monkey, in which the male is actually providing exclusive care of the infant. So what's happening here? Well, in mice, what's happening is that there's a dominant male that mates with the female and the female have a communal nest. These male is always parental. However, subordinate males that do not have access to the female, are infanticidal will attempt to attack the male, kill the pups, and that, in turn, enable them to mate with the female and at that time, they become parental. Now from a neuroscience standpoint, this is fascinating because what it shows is that the brain has actually two types of circuit. One driving infanticidal behavior that is displayed by virgin males and one that drives parenting behavior that is displayed by females and fathers.
So the question we are very interested in is what are these neurons that control parenting and infanticidal behavior? And I would like to tell you today is how are we able to identify neurons that are necessary and sufficient to drive parenting behavior both in males and in females? And so, what we use is a way to identify molecularly neurons that are active during the certain behavior. And the idea is that when a neuron is firing action potential, they're also changing their gene expression. And in particular, they turn on a particular gene, a transcription factor called C-Fos. And so if we have a female or a male interacting with pups, then neurons that are involved in parenting will fire and therefore, they will express this gene C-Fos. So if we look at the brain of parenting animal, both males and female, compared to infanticidal male, we find that there is a sub-population of neurons in one particular area of the brain in the hypothalamus, an area called the medial preoptic area in which you see this very dense collection of neurons that express this gene C-Fos after the animal has interacted with pups, both males and females.
So this is very interesting. It's actually not that surprising because already in the 50s, 60s, 70s, a number of groups through lesion experiment have shown that maternal behavior requires the function of the medial preoptic area. But what we really would like to know is which precisely, what precisely are the nature, the identity, of these neurons that are C-fos positive? And in fact, this area, the medial preoptic area, fulfill a lot of functions. It's involved in mating behavior, thermal sensation. All sort of other functions than just parenting. So knowing precisely and specifically which other neurons that express C-Fos is something that we really need to know if we want to understand the control of parenting behavior. So we tested the number of candidate genes, and we found to our delight one particular gene, the gene that in code for the neuropeptide galanin has been very nicely co-expressed with C-Fos when the animal are parenting. So interestingly, this, the number of galanin cells in the medial preoptic area is identical in males and females in virgin males and in virgin male and mothers and father.
In other word, these neurons are there in the male and the female brain irrespective of whether these neurons are parenting or not. Now, galanin is a peptide expressed in many different areas. It has been involved in many different functions such as nociception, sleep, thermoregulation, et cetera. So we don't know whether galanin has any role to do, to play in parenting behavior, but the fact that we know one genetic marker gives us genetic tools. In other words, we're able to use a genetic line that express this enzyme called cre. Think about the scissor if you wish that enable to activate specific molecular tool to manipulate the activity of neurons. And so the experiments that we've done is to inject a conditional toxin into the preoptic area of the galanin cre mouse, and what this does is that it enable us to kill specifically this galanin neuron present in the medial preoptic area. No other neurons outside of the brain or within the MPOA.
Nothing else than the specific galanin expressing cells are affected. What's happening when we test the behavior of females and fathers when we kill galanin neurons in the medial preoptic area. The affect is very striking. The ablation of these MPOA galanin neurons entirely abolishes both maternal and paternal behavior. And actually, elicits infanticide. So there seem to be a role of these neurons, not only in driving parental behavior but actually inhibiting infant mediated aggression. Now, we try the other experiment, a sufficiency experiment, which is if we now take an aggressive male and activate artificially galanin neurons in these infanticidal males, what's happening? What's happening, and we performed this experiment using a technique called optogenetics that enable to express a light-activated channel in the neurons and then using an optic fiber to activate these cells, and what we found when we shine light and activate these galanin neurons is that infanticidal male now are longer infanticidal and instead display paternal behavior.
So what it shows is that this specific population of neurons are both necessary and sufficient for the control of parental behavior in both males and females. Now, the parental behavior is a complex behavior, and what we'd like to do now is really to understand this behavior in mechanistic terms. So these neurons as we've seen are able to control parenting. Parenting means a lot of things: grooming, licking, crouching, retrieving, nest building, and then inhibiting infanticidal behavior. But these neurons do all of this according, obviously, to the presence of an infant but also, according to the physiological state. Male, for example, will either trigger this behavior or not trigger this behavior according to whether they are fathers or not. And so, what we would like to understand is how are all these different displays being performed and what is the role of this environment in the activity of these neurons? And the first set of experiment that we've done is really tried to understand whether these galanin neurons really are involved in all these different part of parenting.
And so, the experiment that I'm gonna show you here is what is called bulk imaging or volume imaging of these galanin neurons. The idea is that these neurons now are expressing genetically encoded indicator of neural activity. So these neurons are now gonna emit fluorescence when they are active and only in this particular area. And when we have an optic prompt that enable to visualize the intensity of the signal. And so it's called bulk imaging because we don't have cellular resolution. In other words, we're looking at the activity of all these neurons together. And so what you see here is a female and here a bunch of pups over there. And the animal has an optic fiber here and what you can see here is the level of calcium. And as the female approaches the pups, you can see the level of calcium going up. And she's licking the pups, and indeed the level of calcium stays high.
And then as she brings the pup to the nest, and start licking the pup, then the level of calcium is really going up the roof. Interestingly, we found that all of the pups composing parental behavior involved this galanin neuron. So that's quite interesting. Now, I'm gonna show you another video in which this female here has a very similar type of behavior as you can see. She's grooming something, licking, and then, she's gonna bring that thing to the nest exactly as she has done with the pups, but as you can see here, the level of C level is completely flat. What's happening? Well, although the behavior looks very much like parenting behavior, actually, she's handling a fish cracker, so that's not.
(audience laughs) And in these circumstances, galanin neurons are totally silent. Again, although the behavior looks exactly the same, what's happening in the brain, the significance of this behavior is completely different and involved in very different parts of the brain. So, in summary, we started by the idea that parenting is an essential behavior and that in different species it's displayed differently by males, by females, sometimes both, and that we were really interested in going to the cellular and molecular basis of this behavior. And we saw in this really interesting system in a mouse in which animal are or not parenting, and we found this fascinating immunocell type, this MPOA galanin neurons that now really give us entry into how parenting behaviors control. What are the different regulations of these neurons? What are their sets of projection? What type of input do they receive, and how all of these changes in both during development in males and females and during mental illness. And I would to thank the people, the wonderful collaborator in my lab who've performed this work.
Thank you. (audience applauds) – I tend to get unsolicited feedback, such so that should be a very short talk. (audience laughs) Isn't that a kind of contradiction of terms? You mean they found one? (audience laughs) And there's like a book a month that sort of piles on in these unflattering portrayals. The Primal Teen, that's actually not too bad, but Mom, I Hate You: Get Out of My Life, but First, Drop Cheryl and I Off At the Mall, Now, I Know Why Tigers Eat Their Young, or right to the point, Yes, Your Teen is Crazy. But with the technologies such as magnetic resonance imaging, we can for the first time look under the hood at the living growing brain. And what we've found is that not only do teens have brains, (audience laughs) but they're good brains. They're as they should be. They're not broken. And I'd go so far as to say if they weren't the way that they are, we wouldn't even be here.
Evidence from that comes from a kind of unlikely source that I'll get to in a minute. So the teen brain's different than the brain of a child. It's different than the brain of an adult. It's not just halfway between. It's, you know, kind of its own distinct entity, and it's been exquisitely forged by evolution to have certain features. Behaviorally, the big three are increased risk taking, increased sensation seeking, and a move away from parents to peers. And I think these are really deeply rooted in our biology cause it's not just humans. All social mammals have these three features. And so, we're probably fighting mother nature by trying to eliminate these. And this is always very speculative to argue these ways, but one idea is that it helped us get out of the home, which is a really irrational thing to do, right? (audience laughs) People love us. They feed us. They protect us. It's a good gig, right? But it turns out it works better if we do. Less inbreeding, it just sort of, not morally right or wrong, it just works better if this happens.
And so these features, they evolved at a time without firearms, without, you know, high speed motor vehicles, without designer drugs and stuff. Some of these issues are kind of the stone-age brain in a computer age world aspect. But I think that these behaviors have virtues as well. When I was I was a the NIH, the Smithsonian Museum was sort of close. They had this exhibit, The Hall of Human Origins, which I really like, but kind of not particularly featured. A little placard on the floor looked at the relationship between brain size and climate change. And the last big increase in brain size, 500, 800,000 years ago, but what I thought was intriguing was that what would correlate is the change in climate, not the degree. So before seeing this, I thought, yeah, it got really cold. You had to be super smart just to stay alive long enough to get food and reproduce. But this is subtly different.
Everybody in this room had ancestors whose brains were good at adaptation. And were really good at it in terms of, even compared to our quite close, genetically close, rather in the Neandertrumps or Neanderthals. (audience laughs) I'll pause. We can edit that later. (audience applauds) And there's a, we can tell an enormous amount from teeth and fossilized teeth, which is actually redundant. Everybody's teeth right now are fossils, calcified cells, but they work like trees. So they have rings. So tree rings you can, this was a wet year, a good year of growth, the rings are wider. And across many different species, the rings get closer and closer as you mature. The rings stop, and you're done growing, done maturing. And so when you find these fossilized teeth, if you find a fossilized Neanderthal tooth of a 12-year-old, and then check the rest of the cave, he's gonna be with his children, not his parents.
And this is often portrayed as surprisingly rapid growth in the Neanderthals, but I think that's the wrong way to look at it. What's surprising is our protracted growth. We're the outliers by far. It's one of the most distinctive things about us. And even across, like, crows and many other species, the longer you're under protection of your parent, the more complicated your food gathering, your communication, you know, problem solving. Crows are actually really smart as an example, but similar crows in size and stuff that don't have this protective maturation don't have those abilities as well. And it doesn't work to just keep your kids at home 'til they're 40. I don't think on an individual basis it doesn't work. It's an intriguing trade-off I think that we keep options open. We keep our brains changeable, see what the environment's gonna to be like.
We can live on the North Pole. We can live on the Equator. We can even live in outer space for a little while with technologies that are being made and developed. And so this is a good thing, I think, in terms of this ability to keep options open for a long time, but it's really being put to the test with the digital revolution. And this is just in my short career, it's a game-changer. In the way that we interact and like we're doing at this moment with ones and zeros and the lights, the projectors, you know, it's changed everything. It's changed the way that we learn. You know, content that's on internet. I mean, the greatest minds on the planet are a click away, for free. It's amazing. It's magical. The way that we play, and the way that we interact with each other. And so, I've been fascinated by this interaction in terms of the, you know, biology of this changeability and the technologies that have taken over, in a sense, so of almost 11 hours a day of screen time.
And 30% of that time more than one device. And so the usual question is is it good or bad? That's the wrong question, right? Almost any interesting question is it depends in terms of in what ways and how it depends and what it depends upon. But I think that this is an opportunity in terms of to influence adolescence. One of the tragedies of my profession is that it's almost a 10 year gap between onset of illness and treatment. It's, we need to do better, and I think perhaps new technologies can help us get there in terms of by monitoring things like social media activity. Maybe even just movement to best aid harnessing these technologies in an ethically appropriate way to help us recognize mental illness so that we can intervene while the brain is still more changeable. And so a lot of the debates around this that it's just not natural, right.
We evolve to talk to each other, to be with each other, share smells and touches and everything. And now, we're looking at screens for a big part of the day. But a kind of argument to that is reading's not natural either. Reading's only about 5,200 years old. So most humanity, nobody read. So I don't think that by itself is a good argument. It kind of makes the point that the whole aspect of this is the changeability. 10,000 years ago, hunting, gathering berries, that's the same brain in terms of that's a blink of an eye in evolutionary terms. But our brains are amazing. We can adapt. A lot of us spend a lot of our day with symbols, you know, words, and that, and that's so different than, you know, what our ancestors did. And so my career has basically been this in terms of trying to understand this plasticity in terms of how to optimize the good and minimize the bad. And this, kind of, how do you help people with mental illness is the fundamental question. And so, kind of that notion of what do we know? How do we know what we know? What don't we know? Why don't we know that? But, you know, my first assumption is the brain's involved.
I hope so. If it's like the spleen or something, I'm gonna feel like a complete fool down the road. But I think, you know, that's a good reasonable assumption. And Professor Jernigan began this journey. BJ and I started together each following down that path of looking at the brain and how the brain changes in both typical development and in illnesses. It's kind of a non-creative design actually. But scan kids, you know, when they're young, and follow them as they grow through life. See how they're doing at school, at home. See what sort of influences are on the brain for good or ill. And at the NH, we did about 10,000 scans, half the kids healthy, half the kids with different illnesses. And what we've found were, it's nuanced, but like, the brain doesn't mature by getting bigger and bigger. By first grade, it's already 93% of an adult size.
It matures by becoming more connected within itself and more specialized. And this idea of being more connected, there's many ways you can approach this, but white matter is one of them. So this insulating material that you'll get one to two percent more of in the fourth or fifth decade. The brain was able to communicate amongst itself faster. It's not very subtle. It's like a 3,000 fold increase in bandwidth. It's, I think, underlies a lot of the remarkable behaviors that we can do. But it's not just a matter of maximizing speed. It's all about the timing. And so fire together, wire together, the meaning, all the information in these patterns. But more and more we're understanding that's the progression, and if we look at different parts of the brain like letters of the alphabet, as we go from an infant to child latency, teenagers, emerging adulthood that these letters become words, the words become sentences, the sentences, you know, paragraphs metaphorically.
And this all goes up in adolescence. The brain almost, no matter how you measure it, whether molecular, EEG, blood flow, it's just, you know, it becomes more connected. And this is a kind of a fresh look in terms of this idea of graph theory networks, and it gives us a whole new look. So for something like schizophrenia, before we'd be like, is this chunk bigger or smaller or a different shape or size? But looking at the same MRI scan and the same data and looking at how it's interconnected, then we can discern old from young, healthy from ill, cause they bring it back. Not perfectly, but it's really exciting for someone like me. I can't do the math, but to be a consumer of it in terms of that, by looking at this connectivity, it gives us a whole new perspective on these illnesses. The other process is the gray matter process.
And the one, two punch is over produce and then war or fight it out, so it's how almost all complexity in nature arises. Engine of evolution, over produce something non-random selection, and it has great, you know, potential. So it's constantly ongoing. It's not like you only over produce during childhood and only prune during adolescence, but we see this upside down U type of curve where, as we specialize, the brain actually becomes smaller. So after around 10, 11, 12, your brain doesn't get bigger. It gets smaller but leaner, meaner, more specialized based on what is demanding of it. But it's not all parts equal. The prefrontal cortex involved in controlling impulses, long-range planning, it's particularly late to settle down.
Some 25 to 30, and that combined with the hormonally activated puberty activated limbic system, the path of rewards, this imbalance creates a lot of the specialness of teen behavior aspects. But again, this is how it should be. If the prefrontal cortex was already done, like 11 or 12 and stuff, then, we wouldn't be as adaptable. And so I think this is the tension or the trade off. The other place to start in terms of that, these illnesses happen at different times and not perfectly. There's always variation, but Alzheimer's doesn't happen when you're three, and autism doesn't start when you're 60. That characteristically, certain illnesses tend to emerge at certain ages. And that's puzzling, you know. Why is that? In terms of, and when you start looking at this, so much happens in adolescence.
Not a lot, most. So up to 75%, and I still don't know the answer. For 25 years, I've been like, why? Because the early answer is oh, teenagers are stressful. It's stressful time of life. Kids have their parents killed in front of them or a lot are starving to death in war-torn countries. Enormous stresses, but they don't get schizophrenia. And so that never rang true. And I still don't know, I don't know the answer to this. Why do things happen when they do? And so just one example is for schizophrenia. All of the findings you see in adult schizophrenia, you could predict, what if typical teen changes went too far? Is not causal. They already had schizophrenia. But it sort of, so far without exception in terms of both the MRI changes but also the molecular changes. And so, it's just, this point, it's intriguing.
It doesn't help me help families with schizophrenia. But I think these are the kinds of clues that we're starting to understand. So in the specialization process, in typical development, it's about 7% from ages 12 to 17. In schizophrenia, 28%. So it's not subtle, you know, a four fold difference. And so understanding the typical development, I think is key, but about half of what I deal with as an illness, isn't an illness, right. Pregnancy is not an illness. But it's a big deal, right. Relationships, car accidents, incarceration, you know, life decisions, this happens during adolescence. And it's frustrating, you know, as a physician, it's like there's no insurance forms to check in terms of for these very real issues that aren't an illness. And this kind of notion, is the glass half empty or half full? Because this changeability could be a great opportunity making it even more tragic that we aren't recognizing the illnesses when they occur.
And my final sort of analogy is to use Michelangelo in terms of this is a very famous painting of his that by design should look like a brain, a cross-section of the brain. Yeah, he wrote about it himself and stuff, and it's sometimes called the original synapse. You know? (audience laughs) A little neuroscience humor. But it's not like that. It's much more like his other expression of art, sculpting. And we start with this block of marble and life experiences. So then, we eliminate parts. So me might be born with different chunks of marble, from sizes, from genetics, but within each if we knew what we were doing, if we could guide this process, you know, there's masterpieces. And, you know, in I think almost everybody. And we don't know very, we don't know what we're doing yet very well, and it's like, most of the illnesses emerging, less than one and a half percent of the funding has been adolescence. Until now, finally, now, we have this project that for the first time is going to really do this right.
11, 12,000 or, you know, kids, 19 sites across the country to understand what matters. How does the brain grow in health and illness? Looking at, you know, everything we can think of frankly, in terms of, influences on this. I'm gonna brag for San Diego a bit in terms of there's these 19 sites across the country, but the coordinating center for the quantitative core and the neuropsych core coordinating all the centers are both here in San Diego as well. What a good deal for us in terms of the opportunities to try to understand, you know, what matters in teen's lives. And so the technology is a big part of it. How can we get a better sense of internal and external environment with the sensors, with devices they're already using, they're already wearing cause this is the crossroads in right, this is where people, you know, make big decisions about their direction in life. And there's this kind of notion that teens are messed up and they're misguided and stuff, and it's dangerous. I feel bad that I've, cause I've.
.. (audience laughs) Oh, I see, people, I don't know. So this is in the crossroads. And what happens is even teens themselves buy into this, right. Like stereotype threat and stuff. If you think that you're not capable and stuff like that. It matters. Most teens do well. You know, they'll get through this. They'll do well. But I think we do a disservice by, you know, selling them short, and I think that, you know, we really need to recognize the huge upsides of this. Much more than the downsides. That if we can figure out what we're doing, what matters, we can really make a big difference. Thanks. (audience applauds) (light upbeat music).