I'd like to begin my talk the way I usually do, and that is acknowledging all the wonderful women, brilliant women I've had the opportunity to work with and Min to also highlight Mike Posner who really took a chance on me and helped position me to lead an institute, the lead to much of the work that I'll talk about today. But throughout my career I have had stellar mentors who have challenged to me as well. It's given me opportunities to help build confidence in myself, which we all need when we pursue science, which there's so much uncertainty around science and there's the highs and lows. So this is John Richards and Randy when I was at USC and then I did my post-doc and gt rep reports branch at the NIMH. The title today is twofold. That is, I want to talk about cognitive neuroscience in an age of discovery and how the field itself has emerged. And it seems like every single day there's a new tool, technique or form of analyses that can provide new information or new animal model that can help constrain some of our interpretations that we're seeing in the human. But I also have been fascinated by age of discovery, and that's this period of adolescence. So while the brain changes over the life course, even beyond 75 Randy, to help us meet the challenges of each developmental phase, which really helps us to discover and learn about the environments. But that also helps our survival. Even though that's occurring across the developmental life-course. It's particularly obvious to me during adolescence when an individual has to learn to negotiate their environment without the buffer or protect turf protection of the caregiver so that they can independently to be functioning adults in society. I'll be talking a fair amount about developmental science and particularly this period, how it informs our understanding of adolescence. And there are three key questions that I'll present a brief examples of how we're beginning to address each. But I have a reputation for being opportunistic and highlighting former trainees and current trainees work. And so these questions are giving me some leeway than being able to discuss work from separate those individuals. The first to start with, how is cognitive neuroscience changed? Well, I could go down memory lane and we can actually even talk about memory in some of the first studies that we were able to carry out that in fog functional magnetic resonance imaging with children. And I was positioned in just the right place at the NIH with one of the authors of the first paper that was coming up and coming out that gave us hints that we began to non-invasively look at function with MRI techniques. And this paper came out in 95. It basically used an n-back task to assess working memory. So they either look for a target through the sequential presentation of stimuli, or they did a two back version of that task and they'd look for repeats of stimuli. And what we saw in this huge sample of six children, nine to 11 is activity and dorsal lateral prefrontal cortex. When we looked at the contrast, the two back versus the 0 FAQ. We then went on to look at reproducibility of that finding, again in very small samples. And to see if we could reproduce that finding early on when this technique was emerging. And again, showing dorsolateral prefrontal cortex as well as the frontal parietal network basically was activated. But I look back at this paper and we didn't even do a conjunction analysis to see if there was overlap in these regions at the time. And the field has really changed dramatically since then. And now, if, if you luck for open access data sets and you look around the world, we're seeing more and more emerge where we have the availability of data that we've just not had before. And we're beginning to get much more representative samples. Though. All of these blue areas are areas where studies are ongoing where at least 500 US are being followed and brain imaging data is being collected, as well as other phenotypic data. And this was a paper that I had to update since 2018 when Monica Rosenberg first published it. Because there have been additional studies, they really are popping up. It seems like almost weekly. But the advantage of these big data and our accessibility to them is they may speed up scientific discovery. And I'll have to say during the pandemic, having access to data when we couldn't all collect data during that time was very important. It also hopefully is helping to democratize data access. And importantly for our field, enhance reproducibility and replication. And again, as we get larger and more representative samples, we can begin to better inform policy as well. So one of the studies that I've been involved in and I'll share just a couple studies today, is the adolescent brain cognitive development, or ABCD study. This is the largest study in the United States. Looking at brain development. Currently we have almost wealth also participants who were recruited when they were nine and 10. They're being followed for 10 years. Right now. They're in their mid teen years and they're being followed across 21 sites, which I've depicted those sites on the map in the upper right corner. And we're getting a lot of very phenotypic data. But that also includes brain imaging data that is both structural and functional in nature. And so using data like this, I'll just illustrate from some, a lot of his work. We can now sort of revisit what we did with that small little night team that'll be published at 995. Looking at working memory can look at behavioral and neural signatures of working memory and these children. And and so what Monica did is she defined working memory by using a list sorting cast that was part of the NIH toolbox. And it's a task that we're collecting outside of the scanner. And you can look at the how it relates to a number of other toolbox measures. And also cognitive abilities, which include not surprisingly, fluid intelligence, short-term memory, reading vocabulary. Many of these associations that Randy has examines early in his career and I know his GAAP paper. I can't remember the year it was published has like four to 5 thousand citations at this point. With regard to those associations of these cognitive abilities with Tom and with working memory. It's also the case that performance on this list sorting task correlates with performance on the end back, which is another working memory task. So what we were able to do is to use that task. And the stimuli were different emotional faces and places. And to look at a contrast that compared memory load. So comparing the two back versus the 0, that contrast there you're seeing a frontal parietal network and then seeing how that neural signature collected on the n-back task relate to performance on the list sorting task. And here you can see those regions again, frontal parietal network is popping up and it's related to performance on the list sorting task based on activation collected during this n-back performance. And you don't see this association if you look at other contrasts with this task, such as comparing emotional and neutral faces, you also don't see the same associations when we look at other tasks that we collected. Gans, which were the stop signal task, and the monetary incentive, the late task. Now, this is, it's important to note that the other opportunity we have is to look at replication or reproducibility. And so when Monica looked at the first release of data, that was on, the first group of nine 10-year-olds recruited. And then the second release of the data that was on a slightly larger group, we see replication of these results. And again, we consistently see the lack of an association and the neural signatures related to other tasks or contrast with regard to our working memory measure. Now, I don't have time today, but perhaps when I have individual meetings or Q and a, something else that we're beginning to look at is using representations, similarity analyses to look at neural representations and the similarity among them and how they can help explain one of the findings that we're seeing on this working memory task. With regard to higher performance for faces during working memory, but subsequently lower performance for faces. When we look at long-term memory of items that were in that working memory task. And how we're focusing on regions that encode these to be able to predict the long-term memory that was done. That measure was done outside of the scamper. And another aspect of ABCD I won't be able to talk about, which is going to tease you. Is work by Erica Bush who did her undergraduate work with Jim Haxby at Dartmouth. And we're taking advantage of the large twin dataset that we have as part of ABCD. And so she's looking at the heritability of unique Functional Connectomes based on fine-scale connectivity patterns and looking at that during development. Instead, what I'll highlight today is I'll focus on some published work and sort of bring us forward again, taking a bit of a historical approach to presenting these data and emphasizing what we've learned from a different number of imaging modalities, structural and functioning modalities. So if we just look at structural changes that are occurring from childhood into, through adolescence into adulthood. This work, coming out a 2D graph reports branch those. This nice pattern. The sensory motor regions reaching adult cortical thickness before Association regions like the prefrontal cortex, you may be able to see that better. Looking down on the brain and the static image where those sensory motor regions that allow us to see and to act or developing before prefrontal regions that are associated with us. Using those abilities and seeing and acting and, and, and planning our actions based on goal directed behaviors. And this maps nicely onto elegant post-mortem human work, but also pasco Lachish at Yale, His work looking at synaptogenesis and subsequent pruning. And so we're seeing differences in the, at least the plateau. And these areas that in part map onto that work. Though, a number of people have suggested that it's an index of synaptic pruning and fine tuning of connections during this time. Now all the action is not happening at the cortical level. Let's be clear. And this is particularly the case when we think about adolescence. We might think more in terms of emotional centers of the brain that might be changing. And this is old work out a cherry Jerne against loud when it leaves the solid was a graduate student with her, but basically showing, yes, there are changes in the prefrontal cortex, but there's subtle changes. There's still taking place from childhood into adolescence. And Sarah Jane Blakemore has shown this just looking at three different regions, the amygdala, ventral striatum, prefrontal cortex, and showing you're reaching close to developmental asymptote earlier in these regions. Well before what we're seeing with regard to volumes and the prefrontal cortex. Now, my work is focused more specifically on functional imaging. And we are like warriors as adults looking inside the behaving brain. And those individuals who have chains, of course, news alert, they do have a brain, although sometimes there are actions. We might question that. But in fact, their brains are probably we assume the braids or have been really evolutionarily hardwired across different phases of development, but particularly adolescents to help us meet the challenges that we're anticipating. Hearing each of those developmental windows. That again, there are many challenges that the teenager faces intellectually, physically, sexually and socially. So much that they have to learn before they can go out and be an independent social functioning member of the society. And so I want to highlight some work that avionic our van completed when she was a graduate student. Because very early on, We are hearing our colleagues and the lay public talk about how adolescents had no prefrontal cortex. And that's why they took all these risks and did all of these things that sometimes could lead to fatal consequences. And since adolescents are very attuned and value money, we use money as a reward. And since we are interested and reward, we used paradigms at the time that bill directly off of some of them Schultz's work looking at probabilities of reward, expectations and also the magnitude of reward. And today I'm just going to focus on manipulations in the magnitude of reward. And what those studies showed is an adolescent specific effect, where teens had an exaggerated response and receipt of large relative to small reward. A pattern that we didn't see in children or adults. And a pattern that we don't see in areas of the frontal cortex, which is a bit more protracted and it's a pattern with age. Now, importantly and trying to understand how this might relate to adolescent behavior, she also looked at endorsement of engaging in risky behavior. That's all an association between the magnitude of response to reward and that risk-taking. This adolescent specific sort of heightened sensitivity to reward has been shown in a number of labs, band Luna's and Pittsburgh. Also Evelyn Crohn's in Holland and Hamline. And I had the opportunity to work briefly when she was a graduate student and I was at Pittsburgh and she was there. And she's gone on the pelvis. Just a beautiful longitudinal study looking at age continuously. And this is work with Barbara Brahms. And they had over 250 participants, each tested at least three times. And so you see this peak and this response to gains versus losses and the mid teens to early 20s. But the question is, if you have this sensitivity reward, how can you use that? And benefiting the adolescent? And there's beautiful work by Juliet data DO that focuses on that. There's some earlier work that we examined when we were trying to understand when to self-control break down because it doesn't make sense to describe their behaviors being all about the prefrontal cortex because children clearly have less developed prefrontal cortex, but they don't engage in the same behaviors. To how can we use this heightened sensitivity to cues or that might be related to positive outcomes and particularly for adolescents, social cues are important as there's a strong peer influence during this time. And we use Nim stem that were developed by Mary former graduate student Nim Tottenham. And so what Leah show is that just like with monetary reward, when you presented these positive social cues, you saw an adolescent specific response through the smiling faces that was not observed for either the children or the adults. And importantly, this pattern parallels the degree of impulsivity that we see in adolescence, that, that parallels that activity in the ventral striatum. We see that they make many more false alarms to rare smiling cues than they do neutral cues. That this is a specificity to the, the type of cue that they're being presented bit. And we don't see that pattern in children and adults. And if we look more broadly, what we see is like a stepwise, if we plotted this continuously almost of a linear change and recruitment of lateral prefrontal regions had been associated with response inhibition, few lesion that area that have difficult time on impulse control tasks. And what we see is they are activating that area more, I'm correct. Trials when they're younger than when they're older. And their performance and products as a function of age across all stimuli. We just see the specificity when we compare the positive social cues to neutral once. We can also look at how well we can regulate ourselves. And this is work in collaboration with the late Walter Michel and Kevin Ochsner. And it's really driven by John Silver's who's now a faculty member at UCLA. Just looking at reappraisal of repetitive cues. And I haven't had breakfast or lunch. So this chocolate chip cookie is looking very appealing to me right now. But what you can see on this graph is that with age, we get better at regulating our cravings. And if we use these strategies to imagine them as far as opposed to a close, we can do even better. And if we look at the neural correlates of that, while we see is that with age you see less activity. Have this area, the ventral striatum that I've been talking about with age, and greater activation of frontal parietal systems implicated in control. Now, everything's not fun and games. And Todd Hare actually wanted to look at the response to negative lead valence emotions. And what he observed is that adolescents show this heightened and work variable response to cues, a potential threat and he's fearful faces. Chris Mok had shown that previously in adolescents compared to adults, he showed the specificity of that change during this period. And importantly, he also showed that this magnitude that we're seeing with each repetition and presentation. The stimulate, what you would expect is that those repeated presentations you go all right already, like nothing's happening to me. Even though fearful faces usually suggest there's some danger around, there is none. But individuals to show continued activation, but these repeated presentations of those fearful faces and some who even showed sensitization or increases over time endorsed higher rate anxiety. And we'll come back to that in a minute. If we look at the neural circuitry, this ability to, for lack of a better term, habituate that amygdala response. We see that's associated with increased our, this inverse coupling between the medial prefrontal cortex and the amygdala. And that fits nicely with subsequent work out of Nim Tottenham lab. And with Dylan, she has her graduate student, bill, and it's also a post-doc of mine where they show that you get this switch or slip and positive coupling between the amygdala and prefrontal cortex and supports the quench a negative or inverse coupling between these regions. And if we take this one step further, we can look at how can you regulate this with cognitive appraisal. Again, another study with John Silver's. And again, what she shows this with h, we get better at reappraising and filling less negative when we view images, particularly when we're reappraising those images. And with age, we see less amygdala activity with reappraisal. And what's been found is that lateral prefrontal cortex was mediating this change. Association between age and the amygdala. But the most exciting data that, that I felt and this research was really how different underlying circuits might be changing across development. But possibly in a hierarchical way. That when she looked at more carefully, what she saw is that the association between the ventral lateral prefrontal cortex and the amygdala. That, that was moderated by connectivity between the ventral medial prefrontal cortex and amygdala. Again, I think it's maybe consistent with developmental changes that we're seeing in the circuitry with age. And may suggest that the development of cortical subcortical circuits may need to mature to an extent to help instantiate subsequent development of more cortical cortical circuitry. Here involving lateral and ventral medial prefrontal cortex. Though, we had a relatively simple model, the imbalance model that was based on changes in circuitry and a lot of avionic our finance early work. But it was very similar to a dual system model, almost as if we were talking about the prefrontal cortex and subcortical systems as orthogonal. But in fact, it was more like for you tricky expands communication between spot, the more Balkanized part of the brain prefrontal to use John Cohen's terminology. And then emotional centers that were more like Captain Kirk who often made decisions no, fly by night. And that can sometimes seem illogical to Spock. But it wasn't that they were working separately. It was how they were connecting and speaking to one another. And Spock would rarely speak to the captain unless he was doing something illogical or the captain called on him to thinking about how these systems may need to change radically in order to initiate the subsequent projections and communications back to them. Though, we've developed the hierarchical model that extends isn't balanced view. And there's evidence both that we've been looking at animals and humans that suggests that we might be predominantly subcortical connections changing before we ultimately see these critical, critical development where circuits are better able to communicate with one another, to drive behavior and to help regulate behavior. Though. This suggest now that we have different stages within adolescence where a team may look very different from a 17 and 18 year-old or a 19 and 20 year olds, which I'm going to call late adolescence. And I might refer to it as young adulthood at some stage. But leave Summerville had this beautiful article and neuron. The title was in search of the neural signature or brain maturity, which for neuroscientists, that gives you kind of the Willie's because maturity sort of sounds like stability. And we think of the brain is plastic throughout the life course. But She illustrated in this paper is depending on the imaging modality, the point in time where you saw developmental asymptote really varied. So if you look at functional connectivity or cortical thickness or how white matter tracks are changing. You see that they're just can extend well beyond 18, which is when legally we consider an individual adult in this country in terms of when they can sign documents without needing an adult or caregiver and a cosine with them. But the one thing that you also see across these studies is that the prefrontal cortex does tend to show the most and extended or contracted development over this time regardless of modality that you're looking at. But because it's connected to other nodes and regions of the brain, if it's changing, surely they are, but where they're just under threshold in terms of being able to detect those changes. Though. And sort of sticking with this theme of Windows, the brain mature. As part of a collaborative project, Tim Brown wanted to take all the neuroanatomical data that we're collecting on a sample, it was close to a 1000, but I think he had useable data, or almost 900 participants from three, that should be three to 20 years. And basically showed if you throw in all the anatomical data, this would include diffusion-weighted images, t1, t2. And you look at how those vary with age. Can you come up with a model that can predict true age? And well, I think this is interesting that we can predict true age. The graph I love is the one that shows just how dynamic development is. Because at each age there are different measures that are contributing. Do that model and defining what aging individual is. And so I don't know if you can see this carefully what they see bleed just a punchline. Gray matter signal intensity within subcortical regions appears to be strongest. And it's stronger predictor in this model during early and middle childhood, we are as diffusivity within these regions, the strongest predictor during late adolescence. So it's all variable. Important to keep in mind. We can also look at functional connectivity and see where does the brain look like. It's a relatively mature. And this is early work by Nico Joseph Bach, where he came up with a maturational index based on functional connectivity. And basically looking across networks and how connections among these networks changed, strengthened and weekend with age. He showed this functional connectivity maturation index that you see to the right. So basically you have a good idea looking at these data in terms of how a child's functional pattern may be different adults. But another thing I just want to note is you're continuing to see this change and not really reaching a somewhat of a plateau until the early twenties. Now both of these studies that I've just mentioned, one looking at anatomical and another looking at the functional connectome at rest aren't really looking at the behaving individual. And so maybe what we need to think about when we look at the brain is to treat it like a cardiologists with the heart and use like a stress test. But what we want to do is probe those networks and emotionally arousing situations that teens and adolescents tend to find themselves. So I wanted to describe a study in collaboration with Damien fair and Mark Rudolph, where we probed functional networks under conditions of emotional arousal and see how they change as a function of age and whether or not our model worked across development when we looked at these different emotional states. And so basically, what Mark did is he developed a model where he trained to settle in the majority of the subjects, about a 148 between the ages of 826. And p was train them during no real state of arousal. They were just asian it in a task, but they didn't think anything was going to happen to them during that task. And he came up with a model that could predict true age of the individual. Bill it relatively well, it perform relatively well. Do we then look in the independent sample and to see how well this model would work in a neutral state, R and a negative and positive arousing state. And the negative state was hearing the subversive sounds that we had shown was associated with hike up antic skin response and also was rated as a person. And the positive state, what is winning? Thinking? They were going to win up to a $100, but they didn't know when they get know how much. I had nothing to do with our performance. And the same thing's true for the negative state that had nothing to do with their performance. They didn't know when or how loud the sound would be. We told them it was computer-generated. And if we'd look at how the model did it, it did okay, and it clearly did better for the neutral state. But it's due unless, well, in predicting the true age of an individual when they're under emotional arousal. And so, while we can learn a lot from these models, what I always like to ask is a0. So like rid of the modern fail. And what age are we not, this is a model not really capture what we're seeing and the changes in the brain with age. And if you look, what was happening is during this window, the adolescence, where the motto was failing. And this is plotted relative to the neutral data where it was failing was in predicting adolescence to be younger than they were based on the functional connectome during these emotionally arousing States. But something out, you'll notice that there's a lot of variability here. So there are a number of individuals who were showing this lesson chert functional connectome. And so the question is, how is that related to behavior? And so Mark and Damien wanted to look at risk preferences. And so what I'm going to show you is self endorsed reported risk preferences for individuals. If it's an open bar, it means that they were predicted to be younger. They truly were based on their functional connectome and intellectual rousing state. And if they're closed or build that someone who was predicted to be older. And what you can see is generally individuals who are predicted to be younger or showing higher risk preferences. But this is particularly true during this period of 18 to 21 year. So what we're beginning to refer to as late adolescence. So continuing to show changes in the brain. Also preferences and behavior that are changing and extending into the early 20s. Though from the cognitive neuroscience has been helping us should again, to understand how the brain is really dynamic and non-linear in the way that it's changing. The little child's brain is not just a smaller version of an adolescent and then add a less than a smaller version of adult. They're really these nonlinear changes that we're seeing better dynamic. We think that those changes are helping to underlie nuance changes in behavior that help with the demands on the individual at each of those developmental stages and they're clearly going to continue into adulthood. Though. I also just want to briefly address how cognitive neuroscience is beginning to inform interventions for our treatment of the developing brain and mind. And this springs back an important point of how cognitive neuroscience has really been moving into trying to predict behavior and outcomes. So let me just start with early identification and prevention. That's one example. And one real risk factor in the United States is that of obesity. And if you look from the 1980s to 2010, we see that there's been a dramatic increase. And even if we look more recently 2018 stats that I had, it's at 42%. Adults are obese in the United States. So this begs the question for understanding risk for obesity and how we might be able to prevent it. Because typically a risk factor for adult obesity is associated with childhood obesity. And Christine wrap, you want to join my lab as a post-doc and she arrived with just this wonderful program research where she was looking out reward and control circuitry that had been implicated in not only obesity risk for obesity, but also for our alcohol abuse and other things. But here she was focusing on the influence of food ads on our behavior, on providing calorie information in terms of what we would eat and also genetic predispositions. But really the question I had for her is sort of what are the possible mechanisms driving this? And so when she had arrived, there had been a paper that had come out by Stephanie fault and that just really captured my imagination, our imagination in terms of this area of the nucleus accumbens in the rodent. This is the same ventral striatum area that I've been talking about in humans. Don't want equivalent. And, uh, basically showing that depending on what you eat, that can really impact the brain with regard to neuro inflammation in that region of the nucleus accumbens, which could then lead to subsequent behaviors related to additional poor eating. So let me just unpack this in a model and a simple way. So if you eat this saturated high-fat diet, that leads to increase in neuro inflammation of the nucleus accumbens, which is then associated with the mice compulsively seeking sucrose. And they also showed depressive like behaviors. And so it looked like. It's a vicious cycle in terms of what we eat. Now you are what you eat. And there's been a whole lot of work about relationships between the gut and the brain. And so I mentioning this because the imaging techniques are evolving. So we began to look at not only macro-structure, which I've been talking about so far, well, we can begin to look at microstructure. And so based on advances there are happening with diffusion-weighted imaging where we usually think about how water molecules are diffusing more easily a longer track as opposed to across tracks. We can also look at that diffusion within a cell relative to outside of a cell. Gonna get more restricted diffusion with NSL than an extracellular space. Let me see if I can unpack this a little bit more. So here, the more neurons and glial cells that you have, the more restricted diffusion relative to the hindered that's associated with extracellular space. And this has been histologically validated and a wonderful paper. I wipe and under stills team. Several years back. Though. The reason why we're interested in this is neuro inflammation leads to reactive gliosis, though, that is this increase in glial cells and that area of neural inflammation. So our question was, is the cell density as measured using RSI, is that cell density in the nucleus accumbens related to weight gain and children. And so looking at also cortical several subcortical regions, what Christina found is that the nucleus accumbens was associated with individual difference in waist circumference. This is also true for body mass index. There reasons I could discuss and why waist circumference might be better at this age. But more importantly, the nucleus accumbens was associated with change in waist circumference after one year. And this is just showing you the spatial specificity of these findings and cell density within the nucleus accumbens. And then the next question we had is dietary fat. How is that related to waist circumference? We see an association there. But Christina went on to show that the cellularity in the nucleus accumbens was mediating this association between dietary fat and waist circumference. These data together are consistent with the animal work, suggesting this vicious cycle that is associated with what we eat, its effects and impacts on the brain That's subsequently can impact behavior. That again leads to eating poorly. So the findings can inform identification and prevention strategies to some extent. But I wanted to highlight this because it's one of those advances in neuroimaging tools that allow us to really look at the microstructure and see how that's predicting outcomes. We can also begin to look at how we can use neuroimaging data for diagnostics or treatment. Now I want to highlight work from two of my former trainees, MD, PhD students. Connor was a faculty member at the time that Andrew was a graduate student with me. And with over a thousand scans, individual with depression and controls, they wanted to see if they could identify. Biomarkers are subtypes of depression just based on neural signatures. And I just want to say from the get-go because I think it's really important that these types of depression cannot be differentiated based on the Hamilton D, which is the typical used often as a clinical outcome measure. However, the subtypes were associated with different clinical symptomatology. So if I just highlight a couple, anhedonia was associated with one biotype, whereas anxiety was predominantly associated with another. But the reason why we want to be able to begin to explore different neural signatures that might suggest subtypes of disorders that have very broad phenotypes, you can have too much or too little sleep, eat too much, less. Is it that's going to inform treatment. So a subset of these individuals had been treated with TMS. And what Andrew was able to show is that certain biotypes were more responsive to this transcranial magnetic stimulation. Both in terms of the response rate and also their improvement in depression severity. So that's one example where that build is beginning to go. There's much more work to be done. This is done in adults and I just want to highlight some developmental work that's been in collaboration with Charles Black, Francis Lee and really driven largely by ship on cattle, who's now on the faculty at the University of Washington. And this is in the area of anxiety and stress related disorders. Because they're so prevalent in young people with estimates of as many as one in four being impacted. And it also builds on early work that Katie Thomas showed with our group where there was an exaggerated amygdala response and anxious children to those fearful cuz I showed you before. And the magnitude of this response was associated with severity those symptoms. But the data that I presented to here that was a more recent study for the begs the question of, is this is an exaggerated response or is it just reflecting that the pattern of activity may not be habituated or decreasing with repeated exposures to these bases. Because again, what's hot it's shown is that ability for that observation that we see a decrease in response to these cues with repeated repeated presentations is associated with less trait anxiety. And those individuals who don't show a decrease or actually show an increase endorse higher trait anxiety. And also, just to remind you that this in Bob's ventral prefrontal and amygdala functional connectivity. So sort of beg the question for us and thinking about how we could drill down a little bit deeper with animal work. And again, a number of collaborative studies because it's circuitry has been implicated in fear extinction. And for your regulation in mice. And also, we know that facial expressions can be interpreted in many different ways and that can be influenced by our experiences and also our cultures and also across development. Looking at faces is not really that control because adults are going to have more exposure to faces than a young child. But just by the nature of age and experiences as social beings that we are. But also individuals with anxiety. Given the heritability of many of the anxiety disorders, a child may also be experiencing more fearful expressions from their mothers. They go out and explore two. So we wanted to try to control for that using fear conditioning. And this is work by several wonderful graduate students and postdocs, just to name a few. Shivani again, Stephanie, you and also Hartley, now a faculty member NYU. First they showed that it's easy to learn for her memory. And to form that it just takes a few associations of an aversive sound with a queue where you get these nice differential skin conductance response to the cue that was associated with aversive stimulus relative to the one that was not. The same thing is true with mice. It only takes a few trials for them to very quickly learn that a tone is associated with a foot shock. Though, we then wanted to look at this pickiness. How well can they habituate during childhood, adolescence and adults? And how can they extinguish that fear memory when you only present the conditioned stimulus alone. And this was compelling work both in humans and mice. Are team showed that adolescents have to manage acute fear extinction. Though in the mice, the amount of freezing did not change very much over time. They were still freezing as much. So there's a different score. Also an adolescence, they were still showing elevated skin conductance responses to cues that had previously been associated with an aversive stimulus even though it wasn't being presented. So this begs the question of given that cognitive behavioral therapy that uses exposure components to that therapy really builds on the basic principles of fear conditioning. Though, we thought if adolescence as a time in which we're seeing diminished cubed fear extinction. Does that mean we're going to see less efficacy of CBT that has exposure component and energized hill looked at this, I first just want to mention that Rick Richardson's lab had shown this the year before and rodents. So we're just extending that unannounced but also showing humans. So there have been several replications in humans as well. But what Andrew dry style showed is using data from a study that was published in the New England Journal of Medicine. And I only had a few weeks of exposure therapy as part of that CBT trial. But he saw diminish nonsignificant, but a diminished effect size and the adolescence relative to the pre adolescence. And that study was a developmental one. So you'd look in the adult literature to find an equivalent effect size and seeing that it was a bit higher for adults. This work has actually led to a number of groups trying to think about how can we bypass or use compensatory alternative circuitry to help diminish cute. They are memories that might be important for individuals with anxiety and stress related disorders that have a specific trigger that we might want to target and desensitize. And so they've Johnson and I, building on Marie MMPI and Daniela Schiller's great work when they were with Jolla to enlist belts and looking at bypassing the prefrontal cortex. And rather than with extinction, you basically our teaching them. You're learning a new memory. Now, that tone that needs to be associated with the shock is now safety. So you have the fear memory and the safety memory and they're competing. But what Murray at and Daniela we're trying to show is you can actually change the memory itself just based on the timing of when you do the extent extinction. When the memory seems to be more malleable and changeable. We know memories are dynamic, they're not static, and so changing the memory itself. There's also been work by Siobhan where she paired extinction, a queue with placing them back in the chamber in which they were originally condition and got a bigger bang for our buck. That is putting them back in the original situation in which those memories formed. And I think this is important with regard to virtual reality approaches that are beginning to be used in the clinic to try to help with PTSD symptomatology. And then Dylan G has been using safety signals. So during extinction, CIO pair of safety signal with the Q that was associated previously with an aversive events. As you're seeing this nice decreased galvanic skin response, but also an increase activity, eventual hippocampal circuitry. So the last I want to just briefly mention is there's treatment in terms of Madison and there's treatment and how we treat individuals in our legal system. And unfortunately, many youth come into contact with our legal system during this extended period of adolescence. And I'll just highlight there, you see it's in the early to mid teens through the early twenties. And as I mentioned before, age of majority model assumes that you have full adult capacity by a team. So that suggests that by 18 we have cognitive capacity at that point. This doesn't mean that individuals younger than 18, just based on what we know from psychology and nurse psychology that they have, don't have the comparable ability adult to make some simple decisions. We know that from very basic cognitive task. But Ellie hypothesis was that you'd see more protracted development of cognitive capacity and these emotionally arousing situations. And basically to do that, she used conditions both rewarding and negatively affecting. And what she showed is that one was the sustained threat, that loud sound, and another was the sphere cues. And she saw the 820 one year olds, just like teens, were significantly different and their cognitive performance relative to individuals over 21. And when you looked at the neural correlates of that, you see less activity and lateral prefrontal regions and increased activity and limbic cortical regions in these two age groups. This begs the question of when this adolescence and, and also how should we be treating young people in the justice system? And there are a number of examples of how the system is changing. We have younger adult courts developing across the country. We have experimental young adult units, which are less about being punitive and more about preparing the individual to be released. And it's showing promise in terms of positive interactions with officers and inmates. But I think the point here is that in the eyes of the law, a young adult is, is not a juvenile, but in the eyes of developmental scientist, they are similar in juveniles in very important ways. And there are also efforts to extend the death penalty from 17 to 18 or 19 in terms of expanding. So I just want to end by saying, I hope I've given you a flavor of how the field is changing with a few examples of how it's changed, particularly since 995. And the importance of how these discoveries are beginning to inform policy that can improve the lives of young people. And by improving their lives. We're also improving our society. And to just end on using these big datasets. With them come a lot of opportunities, big, big, big responsibilities. And we have to be aware that the social structure in which these individuals live and had been assessed is typically not captured in these data. Into always remember, the brain is plastic and we have the potential for change. Though, just to acknowledge, support. My fabulous lab and my son, who was getting me a unique insight into the extended period of adolescence. I think here, pretty much out of time, but maybe we can do a couple of quick questions. So Sherman Jones had a question earlier, he putting the chat trim and if you're there, do you want to unmute and ask a question? Yeah. Can you all hear me? Yeah. Awesome. We have mentioned how dietary changes affect the brain. There are essentially maybe withdrawals if you type pattern where the great, It's all just wondering if you wear. Whereas our research on how your diet changes your microbiome, how you kind of like vial, that's the kind of small how your diet effects, I'm sorry, what? Your microbiome right now there's a whole area of research that I'm fascinated by, but it's not an area that have expertise in. But thank you for mentioning that because we really are learning that just like we don't separate brain in mind, we cannot separate the brain and the gut rate and the line. So very important. Thank you, Chairman. Good to meet you virtually. I had a really quick question. So you showed a lot of data on what is essentially U-shaped curves, though children look one way and then in other lessons something happens and they look very different than in adults actually getting more and more steam to each other. Now I was wondering, do you think somebody's things that are culturally specific? I mean, at least the one on the Saudis was in mice. So it's not culturally specific, but it seems to me that maybe some of them are in other cultures. You may not see that, that distinction, I think that's really important. Larry Steinberg isn't a lot of work cross-cultural. We're looking at risk-taking and the influence of peers. And he's shown across a number of, of countries that you see this inverted U in terms of sensation seeking and reward. And also shows, I can't believe this. And mice, mice who are with a peer or a cage made spend more time by a nozzle that they can get ethanol or basically drink than an adult when they're with a cage mate. And he's been a whole lot of time but allowing these mice to grow together. But I think that really speaks to changes that we're seeing in behavior. And the brain data I think are speaking for themselves is getting us away from thinking about growth and thinking about dynamic processes, Amman, networks, and notes that are changing during this time. And apologies that I was trying to highlight as many of my former trainees as possible, because I really do love the Q and a. That they do is absolutely great. Maybe it maybe last question. Thank you. That was the thing for that answer. So is Shen pack the question is, do you want to unmute or if not, I can also read a question. Sure. Yeah. First of all, thanks so much for that wonderful talk. My question was, in terms of the structural functionalist signatures of changes in the brain during adolescence, are there any studies which focus on, like subspaces are subsets of regions or functional connections to come together rather than individually changes. And the regions are individuals set up connections between regions. So looking at clusters as opposed to, so, it really depends on how you postulate the brain, right? So you can look at network, so you can look at regions. Or we began to work with Jim Haxby using hyper alignment. And we're beginning to move from sort of regions or parcellation too, vertex to vertex to get this really fine grained functional connectivity. And we haven't used that to look at the changes during the adolescent period, but we have begun to look at that in the developmental population, looking at heritability versus reliability and how that is related to predictions of cognitive performance. Not sure that really addresses your question, but I, I appreciate it. And maybe we can e-mail and have a discussion subsequently. That's way less like what I show you. Thank you.