We are the Association for Child and Adolescent Mental Health, or ACAMH for short, And this is ACAMH Learn. Welcome to Mind the Kids. I'm Dr. Jane Gilmour, honorary consultant, clinical psychologist, and child development programme director at UCL. I'm Umar Toseeb, a professor at the University of York focusing on children and young people's mental health and special educational needs. In each episode, we select a topic from the research literature, and in conversation with invited authors, sift through the data, dilemmas, and debates to leave you with our takeaways for academics and practitioners. Today, we'll be discussing genetic and environmental influences on neurodevelopmental traits in young children. This episode is called Neurotypes, Genotypes, and Phenotypes. Well, Umar, I know you are really interested in this particular paper for a variety of reasons. It's one of your special interests, I would say. So I want you to begin by telling me a little bit about why you are keen to discuss this. So my interest in this paper specifically is because we're working on something very similar. So I was like, oh yay, somebody else is doing this. So I want to know what they've done, and then whether what we're finding is similar to what they're finding. And it came about really-- the reason why we started doing this kind of work was we applied for a grant and we were at the very final stage, and for some reason, the funder decided to send it out for review again. And the reviewer said, oh, I don't like this. These are the reasons why this is not a good project. Why don't you do this other project? Which is, I think, a bit of a rubbish thing to say as a reviewer. It's kind of like, can you review this project rather than the one that you think we should be doing. But actually the one that they thought we should be doing was this intergenerational transmission stuff. So that didn't get funded. But then I was like, you know what? I'm going to take you up on that. So I'm going to write a whole new proposal based on that reviewer's feedback. And yeah, then we reapplied and we got funded with that. So that's why this paper specifically. But I think why genetics in general? I think it started from an interest in methods. So I'm always interested in big data analysis and I'm always interested in learning new and analytic methods. And I work in a department where a few other people do behaviour genetics type stuff and I'd not really done it. So the opportunity came up to go on a training course, and I was like, yeah, let's do it. So that's my methods reason for liking this kind of stuff. But also my psychology head is, well, we can't understand development if we don't look at the whole picture. And the whole picture is genetics and the environment, and how they work together. And I think this is really interesting because I think for those who are in the area of research, of genetics, are very comfortable with the ideas of it being something that allows us to discuss the environment as well. But for those that may be adjacent to the area, let's say, the idea of genetic influences can sometimes imply that this is something that's not amenable to change. But of course, it's much more nuanced than that, isn't it? Yeah, absolutely. And I think that's also one of the things that I was quite worried about when I started work in behaviour genetics, was the potential negative interpretations of any findings. Like how the science that we do might then be used in the wrong ways, and that is a risk. But then I read this book. One of the things that they said was, if scientists who want to make a difference in a positive way to help children's development, give up this space, as in like, don't do this research, then the people with ill intentions will occupy the space and then do research that advances that negative narrative or whatever. So I was like, OK, I'm ready to carry on doing this work, but it's always very difficult to try and communicate behaviour genetics to people, because there's always a negative connotation based on recent history in this field. Yeah, I mean, last week, I was training some would-be mental health professionals about Tourette syndrome and tics, which is another neurodevelopmental disorder which is relevant for our paper today. And there was certainly a lot of caution about even telling families that there was a genetic component to Tourette syndrome and a sense of I would say, it's almost embarrassment or self-consciousness about, how do I discussed that with a family. And of course, that will be conveyed to the family as if somehow there's a shame within that. It's something that runs in families. That's OK. And knowing that, knowledge is power. And I think that's really what you're saying about inhabiting a space and using that information for good purposes to inform and expand ideas about the influences on various traits and expressions so that we can empower parents and they can make some use of it, too. So there's lots of different ways to think about it. Certainly something that even now, when I was quite surprised, because even now in terms of training, it was seen as something to be thought of as a cautiously. And I was quite surprised to hear that in the mental health professionals I was training. Yeah, I think there's also, I think, it's a genuine fear of being erased as a population. So if for example, there's a, I don't know, the autistic community, I'm just using an example, might feel that genetic research into autism might lead to pre-screening at birth or prior to birth, and then people might screen out certain populations. And I know it's not the same type of genetics that we're going to be talking about today, but the Down syndrome, in some parts of the world, that is something that might be happening. So I think if we look at those examples, then I think that is a real fear and a genuine concern, as in, could this research on the genetic influences on neurodevelopmental traits and conditions lead to the erasure of certain populations? But I think that all the more reason, as you say, for us to inhabit that space. Because if, A, you can't decode the content and understand the literature and have a conversation about what you can and cannot extrapolate from the data, then we will not be able to protect the population from any negative applications of the data. So I think really it has never been more important as the technology moves and as there is potentially polarisation in terms of how we may use the data for us to inhabit this space. So I think we should fly the flag proudly that we are interested in the application of behavioural genetics and explore it because it exists. And we need to know about it in order to make use of that data. I'm really interested because I know that the paper we're going to discuss today, as you say, has got lots of different ways of exploring these neurodevelopmental traits in this instance. And Laura is going to tell us a lot more about that in a moment. But this is a population study, so it's not a clinical group, which, again tells us something else, a different nuance, if you like, about the genetic influences in the population, not simply a clinical-referred population. I think, but it works on the assumption, and Laura can maybe elaborate, neurodevelopmental conditions are just the extreme end of neurodevelopmental traits that exist at a population level. So I suppose by investigating traits, or symptoms, or characteristics, you are then probably tapping into a similar population-- sorry, a similar type of characteristic or behaviour, even if it's not a clinical sample. There's so much to it. I'm wondering, because we've got so many questions, we're raising a lot of big themes and specific questions, if we should invite Laura into our conversation, and she can help us explore and discuss this really interesting paper. Yeah, let's do it. So today, we're joined by Dr. Laura Hagemann from the Norwegian Institute of Public Health. Laura is the lead author of the paper Direct and Indirect Genetic Effects on Early Neurodevelopmental Traits, published in the Journal of Child Psychology and Psychiatry. Welcome, Laura. Thank you for having me. It's great to have you, Laura, and I'm going to dive in with the first question, in part because of the discussion that we were having earlier. Now, there are some very interesting and complex ideas if you're new to the area of behavioural genetics in your paper. So how could you summarise the findings to the parents who took part? So I think maybe just for some background first, I would start by saying, if we're interested in epidemiological questions around neurological conditions, whether that's what sort of factors lead to development, what sort of factors may lead to better outcomes, or some of the heterogeneity that we see in these conditions. If we're doing those studies in observational data, something that is important to consider is that families share both their genes and environments. And so in an observational study, if you're looking at one or the other environmental effects, maybe they're getting influenced by genetics and genetic effects. They might be getting influenced by some of the environments, and so it can be very hard to piece apart these differences. So the goal with our study was that-- we have this cohort, MoBa, and it's a wonderful cohort because we have lots of early developmental measures. And then we also have genetic information on the children, the mothers and the fathers. So our goal was to use all this and use two different methods to try to start to piece apart, how much are these environmental factors potentially biassing genetics and how much do we see evidence for potential environmental factors when we control for genetics? And so what we found, and I think the main takeaway from this paper is that the genetic influences on these early traits is very complex. And if we're doing these sort of studies, we really need to be considering the fact that parents and children share genetics and that these might be impacting our findings. I think we find evidence for what we call direct genetic effects, so that's the child's genotype on their own traits or their own behaviour. So that might be mum has red hair and child has red hair? So the genetic-- Yeah, mum has red hair, so she's passed the genes for red hair to her child. And so then her child has red hair. And so in this study for example, we see that, if a child has genetic variants that are associated with ADHD, at age 3, they have reported higher in attended levels. What we also though, find, which is maybe a little bit surprising, is that we are seeing effects from the parents' genetics, and so we call these indirect genetic effects. And what that means is that basically the parent's genotypes are associated with the child's traits over and above what they transmit to their child. So, for example, I guess, red hair doesn't quite make sense, but maybe the mum didn't pass down genes for the child's red hair, but for some reason, the child still has red hair. So what it means is that these indirect effects are of theoretically indexing environmental pathways because the mum's genotype or the father's genotype is influencing their trait, and then that trait is influencing the child's behaviour. And the key point there, though, I think, is theoretically, there's lots of other things that may complicate that association and may not be what we think of as a trait. And so that is where further research, I think, is needed to piece apart what exactly we're picking up in these estimates. So let me get this that right. Sorry, apologies. I promise you, I will let you get into your nerd fest with Laura about the methodology. But just so I understand this. So let's say, for example, that the mum's genetic or her genotype influences the sort of environment that she will create for her child. So it's influencing the parenting environment. Is that a fair way to summarise the theoretical pattern? Yes. It could be one pathway. I think typically the example people use, and it's maybe easy to think about, is if parents have maybe a higher genetic propensity to educational attainment, let's say, then they might have higher educational attainment in general. And that may lay-- they can give their child more resources, and maybe they'll be doing more things like reading early, be more invested. And that would also lead to higher educational attainment for the child, even if that child was adopted and didn't share any sort of genetics with their parents. But maybe we'll get into more, is that it could also be like with the mother, maybe if you're talking about prenatal effects. So it's not something that's actually like a behaviour in the mother. But if there's maybe genetic effects that influence certain medical conditions that might make it more likely for a premature birth. For ADHD and autism, we know that's associated with having those conditions. So that could be another sort of pathway in the indirect effect space. I think it's a good example of how investigating genetics can help us to understand the environment, because you're using genetic data to understand the transmission of difficulties from parent to child via the environment, potentially. And I also think it's worth unpacking here what you mean by propensity for educational attainment. So my understanding is that at the population level, using data from millions of people, researchers investigate hundreds of thousands, if not millions of genetic variants to see if they are associated with how many years of education has been completed by a person. Then from those millions or hundreds of thousands of variants, they look to see which ones are statistically associated with educational attainment, and then use that analysis to generate a score for each individual. And actually, that score for an individual is meaningless on its own, but at a population level, when you map it across hundreds or thousands of people, then it gives you some sort of rank order. And so when we're talking about propensity for educational attainment, that's what we mean. But then we have this polygenic score. Now, lots of people who are listening might have heard of various consumer organisations where you might have got this as a Christmas present or something where someone's like, oh, here, do like a mouth swab, send it in and they'll look at your DNA, and they'll give you an indication of your risk for various different conditions. And that's what we call a polygenic score, which is what you've used in your analysis. I wonder whether it's worth having a chat about, what is the utility and the usefulness of a polygenic score, either for an individual who has got a Christmas present and is sending it off, or also in this context where parents might be like, oh, can we use a polygenic score to test my child's risk of autism or ADHD, or whatever it might be? Yeah. I mean, I think it's very important to separate what we know with polygenic scores, like what they mean at the individual level and what they mean at the population level, because those are two very different things. I think, like you said, at the individual level, it's pretty much meaningless. I think it's really important to get to maybe the general public is that these are explaining very, very little percents of variance. So when you look at it in yourself, it's theoretically, you're measure on how much genetic variants that have been associated with this trait. But since it explains so little variance, it's meaningless. And if you were to make decisions based on that, let's say you were looking at polygenic scores for ADHD, and maybe those in the higher end would get more screening or something. Somebody at the lower end could very much still be just as likely to get ADHD, and then they're missing out. So using polygenic scores in that way could not just be meaningless but also have negative consequences, I think. So yeah. And they are probabilistic. So I think it's not an exact science. And then there's also the issue, in my view, we work in this framework of medicine where we have, you have this gene, therefore, you will get this condition. So when it comes to neurodevelopmental conditions, that's not the case. It might increase your risk when compared to other people in the population, but then the environment is still important. And, that risk might not manifest into a condition or a set of traits if the environment is a certain way. Yeah, no, 100%. Because all these conditions are extremely polygenic, which just means that it's the effects of many, many genes and that they're not heritable-- they're very heritable, but they're not 100%. So there's always going to be effects from the environment. And yeah, I think sometimes when I'm trying to explain this to friends and families, I always say, you know, your genes are going to influence your behaviours for these sort of conditions, but they're never going to determine anything. And I think that's something to very much keep in mind when you think about the genetics of these conditions. That's a very nice phrase. And I love the way that you are highlighting the idea that if polygenic scores, for example, are applied and misunderstood, actually, we are doing those who are missing out on the screening, let's say, an enormous disservice. So it's about understanding the power and the lack of what the data can show. So it's a really powerful point that you're making. I was really impressed by the way that you describe your data. And it's complex data, but you describe it in a very cautious way and very thoughtful way, I thought in the paper. And you describe the idea that the findings may not be generalizable to other groups, and you raise the idea potentially about different heritability patterns in your developmental traits for groups without European ancestry. Can you say a little bit about that, about the data that is around? Yeah. Well, that question was because unfortunately, we just based on QC of the genetic data, we had to limit our samples to those of European ancestry. And so this gets back to the point that we were talking about, the roles of the environment, because heritability is just in its definition, the proportion of genetic effects that are explaining variance in a population. So if you change that population and there's different environmental factors in that population, the heritability may change. And so if we think about-- I mean, this is maternally-reported traits. But if you're thinking maybe about clinical diagnoses, there may be different environmental factors, or diagnoses, biases, or things that may affect different groups of individuals differently, that could then impact heritability estimates. So, for example, a cultural expectation of what typical behaviour is and what's acceptable or not? Yeah, exactly. For maternally reported traits, that could definitely have different cultures or have different expectations of how they interpret these questions, that could 100% in those indirect effects then show differences. I read a book, I think it was by Adam Rutherford, who does a lot of stuff on the genetics of race, et cetera. And I hope this is correct. I think what he said in the book was, there's more genetic diversity within Africa than there is outside of the continent of Africa. And the reason he was making that point was like, we investigate a lot of genetic influences within a Northern European sample, but actually that quite genetic pool is quite homogeneous, and there's lots of diversity in other parts of the world. And actually, this is part of one of the reasons why I've resisted the temptation to do a swab and send it off for a consumer company. Because I'm like, well, any polygenic score that they generate, even though I know what the drawbacks are of polygenic scores, will be based on analysis of mostly a White Northern European population. And the extent to which that applies to me as a person of South Asian origin, it adds another layer of error. And it will be even less meaningful for me than it would be for a White Northern European person. Yeah, that's completely true. And it's a big issue in the field that I think everybody knows about. And there are still the efforts to try to make it better, because it's not just our sample when we're doing these polygenic scores, but also the GWAS that you make the scores. And yeah, I think, I believe the recent Schizophrenia GWAS where they had more from different ancestral groups, and they show that combining, you're getting more power to actually find the variants that are more causal for schizophrenia. And then you're potentially reducing noise about these variants, because in GWAS we don't really know why things are associated with that. So yeah, it can only help. And it's important if we want to understand the development of these cushions and be able to get results that we can use to help all people with ADHD and autism, not just European samples, so Europeans, yeah. And you talk about indirect effects, and you've already mentioned this a bit, so let's just unpack it, genetic nurture. So the idea is that parents pass down half of their genetics to their child. And then the other half that isn't passed down is then somehow correlated or associated with the child's outcomes. And that's the indirect effect in genetic [INAUDIBLE]. Can we talk through an example of how that process works say for example, for ADHD traits? Yeah. I think, in terms of an actual example. I think, we're just guessing what the mechanism could be. But let's say mothers-- we found one weird association. It was the mother's educational attainment. Polygenic score was positively associated with higher hyperactivity levels in the children. And so what that would mean is that potentially that mothers who have higher liability or propensity to higher education, they have higher education in general. And something that's captured in having higher education in the environments that may create means it's influencing children or is associated with children having these higher activity levels. This is where I was saying, that's the theory and we're very cautious to interpret like that because these are maternally-reported traits. So one other thing we could be picking up is potentially mothers with higher educational levels, maybe they have different expectations for their children's behaviours. So when they're answering these traits, maybe that will influence, and then maybe they'll be more likely to report these traits. And so that's something where it could look like this environmental effect, but maybe it's actually something with what others are perceiving or how they're genetically influenced traits are, how they interpret things. I think that this is you've described it well there. A lot of this is, we know that these things are associated and that we're trying to figure out why. And in the trying to figure out why, we then add more complexity to a statistical model. And you keep adding stuff, and you keep adding stuff, and you eliminate stuff. And you're like, well, it's not this, and it's not this, and it's not this. And I think it talks the complexity involved in doing this kind of work where we're not quite-- like you said, we're not quite sure why SNPs, as in genetic variants, are associated with certain outcomes. And part of the point of this kind of research is, well, we know these things are related, but we need to figure out why. And that figuring out why is a process of elimination. And then we hopefully get to an answer, where it's like, we think it might be this because it's definitely not this, this, and this. And Laura, you highlighted one of the more puzzling findings, but there were other findings that made sense. Do you want to explain, giving another one of those examples, because I think there were others that for me as a clinician, I thought that would make sense? Yeah. So I think most of the direct effects, all were in directions we expected. So for example, we saw that the ADHD or variance associated with ADHD were associated with higher inattentive levels at age three. We also saw that variants that are associated with autism are associated with language and motor development and social communication, so those make sense. I believe we saw some effects where-- and I'm actually forgetting exactly which domain-- but there the indirect effects, so lower educational attainment was associated with the higher of behaviours or difficulties. And so that makes sense if you think about potentially that means people of lower socioeconomic status, or maybe have less resources, or less access to early interventions or things like that. So I think that made sense. But yeah, we are picking up some more counterintuitive effects as well. Can we talk. developmental stage? So investigated age three so early childhood. And I think what you found is mostly in line with what I would expect. So I think if we think about the proportional variance are explained by genetic effects versus the indirect effects. So let's call the direct genetic effects the genetic effects, and the indirect effects the environmental effects. So what you found is what you'd expect, I think, where the environmental effects are stronger if we compare them to the genetic effects. If we put them both together, the environment effect seems to be stronger. Am I right for most of it? Sort of. I think for some traits, yes. For hyperactivity that's what we saw, as well as the repetitive behaviours and interests. For a couple domains when we did a variance decomposition, and for attention, we saw about equal effects from the child's and the parent's genomes. And then for the other three we saw primarily direct effects. But what's also important is that in all of these model, there's a large bit of unexplained variance. And so that could still be environmental effects, just not ones that are influenced by heritable traits in the parents. So that's one. But what we do see in what you get, is that those are relatively higher indirect effects than what we've seen in other studies in MOVAS that have done this at older ages, where they've seen primarily direct effects using some of the same methods as well. And you put this in your paper, I think the other studies looked at later time points, as in middle childhood. I think what I was going to ask, and I think what I was trying to get at was, in early childhood, you might find that the environmental effects are stronger than the genetic effects. But when you get to middle childhood, it seems to the balance seems to shift. And I would hypothesise that when you get to adolescence that then the genetic effects are much stronger than the environmental effects. And that seems to chime with this developmental approach, where genetic effects become stronger through development. Yeah, it is in line with a decreasing effect of also passive gene environment correlation. So another way to say that, yeah. If we were talking about some of the reporting factors, there's some evidence that maternal reporting has less of an impact as children get older as well. So either two of those things are in line with my findings there, yeah. And this is more because also we're doing a developmental approach. Does this support early intervention? As in, if we're saying the environmental effects are stronger earlier in development. And by the time kids get to adolescence, then genetic effects are much stronger, then are we then saying we should be intervening much earlier? Not necessarily with neurodevelopmental traits, but with any sort of indicator. Yeah, I was actually just thinking about this this morning and some of the implications. And I think you probably would want to be modelling these things longitudinally and not just looking [INAUDIBLE], because it's somewhat of a different question of, are these effects affecting this behaviour at this time point, and does that affect continue on, or is it specific to that time point? So does it affect the developmental trajectory, or does it just mean that at this early age, children's development on these traits or neurodevelopment conditions at least, there's going to be more variance maybe at the earlier age. But when you get to the older age, kids who are on that extreme, it's going to be more obvious than it's going to probably be in a real mental condition then. So I don't know, and I think that's something that you'd actually specifically need to model. But it's a very interesting question. Is it that it actually might change the expression if that were the case? In individual differences, it could change the expression and the trajectory if the intervention were placed at the right developmental stage. It's really interesting. Yeah, it has very different clinical implications. I think this is the point at which I'm like, OK, so I can do behaviour genetics cross-sectionally, and then I can do behavioural stuff longitudinally, but trying to combine both of them. So doing gene environment stuff longitudinally. For me, the math doesn't matter anymore and my brain stops computing. So this might be the point which-- unless you have more questions, Jane. Do you know what? I think it so interesting, I could go on and on. Partly because I've got data envy, because that data set is so extraordinary. And there may well be opportunities to do exactly what you talked about in terms of modelling longitudinally. So we must be able to come back and think about this in more detail. When you do that, Laura, and you have the answer to those questions theoretically or actually, that'd be wonderful. Thank you so much. It's been great, Laura. So it's been a fantastic discussion. And I've definitely really enjoyed it, and I hope the listeners have too. Great. Thank you, and thanks for having me. Yeah, thank you so much. That was great. So that was a very interesting discussion. Laura's paper has so much in terms of principle to understand about behavioural genetics, but also neurodevelopmental conditions, developmental issues, genes and environment. There is so much to understand. So I wanted to talk about this in real gritty detail. Think about an example, because that helps me consider how I've interpreted this data. So maybe we could talk about this example, and you can tell me what you think of my interpretation. Let's do that. Yeah? Yeah. So I wondered because I thought the example of using attention hyperactivity might be a good example. So if we were considering a young child, say, age three, who had traits of ADHD. What we're saying here is that their presentation is influenced both by the genes they've inherited from, let's say, mum. Because that was the findings and also the environment that they are experiencing, which is the influence, at least in part by mum's genes as well as other factors. So mum's parenting, if you like, is creating an environment that is driven by genes. Are you happy with that so far? Yes, I am. Theoretically. One thing that I'm unsure about is, so the environment is influenced by mum's genes and child's genes? So the child helps to create their own environment, and the mum and the dad will help to create their environment. But I think in the context of this, yeah, I think we're interested in the environment that the mum's genes help to create independent of the other stuff, I think. Because the findings were that the indirect effects from mum's genes were particularly influential in attention and hyperactivity. So if we were thinking about that in terms of one theoretical way of understanding the data, and it's really important to be clear that it's not the only interpretation and it's a theoretical understanding. But if we thought about a 3-year-old child with traits of ADHD assessed using questionnaire data, not clinical assessment, we wouldn't establish a clinical diagnosis of ADHD. But an assessing team might have a strong hypothesis that ADHD is around and early support would be put in place on that basis. So we know the NICE guidelines indicate that parenting groups would be the first line of treatment for under-fives who with query ADHD. Let's think about what the parent training groups would be like. For example, it might be Webster-Stratton's Incredible Years Programme. So this helps parents offer clear routines and a calm emotional context. So why these targeted parenting behaviours are not diagnostic features of ADHD. They're certainly adjacent to it, and they have executive function skills at their heart. And for the older children, depending on the presentation, environmental modifications would be indicated. And that includes regular routines, scheduled breaks, consistent strategies. So my point here is, in both of these examples, so parents who may share traits of ADHD are invited to change their practise and offer predictable, calm systems at home. So you could argue that on the basis of the findings of this paper, we're offering intervention to precisely the right population because of the genetic influences. Or you could see in different terms and say, we're asking parents of young kids with traits of ADHD to do the very thing that they find most challenging because they have ADHD traits themselves. That's the way I see it partly in my clinical experience. So the other thing to say, just before you come into this, and I think you're going to explore it in different ways, potentially. But it's also worth saying that if we use that theoretical model of influence on ADHD, it's also worth saying that there are instances when the influence on genetic traits can be a gift to a family in terms of the parenting environment offered. So, for example, parents with neurodevelopmental traits could model their tried and tested strategies to their children in order to help them navigate the neurotypical world. So it may be an additional challenge, but it may also be a gift, I think. Would you agree with that positioning or potential positioning? Yeah, I think so. I think that makes sense to me. I think the way I see it-- we use ADHD as an example. When we think about ADHD traits, we know that it'll be a combination of genetic and environmental effects. And what this kind of work does is just to try and give an estimate of what those proportions look like at a given time point. So I think what this would suggest is at the age of three, the child's genetics influences whether and to what extent they manifest in ADHD traits and characteristics. But the parents' genetics also do that. And why the parent's genetics do that, is because the parents help to create the environment. Now, I think the controversial bit, or the thing that might be seen as a bit controversial, is if we're saying the environment is important in ADHD traits in children, so the environment the parents create, does that then mean that ADHD is a learned behaviour and a learned trait, and that it should be changed? No, I don't think it does. I think it means-- I think. So my observation would be you might have a vulnerability to be distractible, let's say. And that in a particular environment that distractibility will potentially be more impactful on your functioning. But in an environment that's very calm and very routinized, that distractibility would not impact in the way. And in fact, we know there's lots of wonderful things about being neurodiverse and having an ADHD profile, which is extraordinary for the person and for the world. And that distractibility probably has great adaptive reasons. If you're looking at lots of different things, you're going to find eventually something that's interesting and fruitful. So this is not about dialling down the idea of ADHD in and of itself. Certainly very much of the neuro-affirmative position, but it's also recognising that if you are extremely distractible, you are not going to be able to engage in social and educational experiences in the way you might otherwise. So that's the way I would look at it. And it's certainly the way I would recognise it in lots of families I work with who have a young child who's got traits of ADHD. And indeed, then may subsequently be diagnosed with that, fulfilling criteria for what that's worth. But also, they are living in an environment, which can be somewhat unpredictable. There can be a lot of big emotions, a lot of high expressed emotion. And that can in some instances mean that the vulnerability that they have in terms of ADHD traits will manifest in a way that is more impactful. The reason I asked the question about early intervention was the case that we've made to a funder is that we are going to do this analysis, and we're doing this analysis to understand who and what should be the target of early intervention. Because this does give us an indication if it should be the parent, the child, or a joint thing, which it probably it. The what is a bit more difficult. So I don't think we can figure out what based on the data that we've talked about today. But there is data out there where you could include measures of shared book reading, or spending time with your child, or going to the park, or whatever it might be, parent-child activities, into statistical models. And then see if various indices change, and from that infer that, this should be the target of interventions. I think from this, what I would argue is, it gives us an idea of who might be the target of interventions and when if you do it developmentally. But what that intervention should look like might be a different research study and a different method. I think my other takeaway, more for the general listener who isn't necessarily going to be interested in the specifics of the method that we've talked about, is polygenic scores. Genetic risk for any sort of neurodevelopmental condition or any sort of psychological trait really, is not deterministic and it's meaningless on an individual level. We are not near and I don't think we will be near in any way in the near future of saying, these are the genes for autism. These are the genes for ADHD. Now, we're going to be able to identify at birth, or pre-birth, or early on, who is going to get autism and who is going to get ADHD. That's just not how it works, and that's not how it's going to work. I can't see a future where that is how polygenic scores are applied. Not because I think that that's not what researchers are doing, that's just not what the science is. That's just not what the method is about. You can't use it-- I'm trying to think of an example of having a tool and then using it in a completely different way than to what it's intended. And it's not just you're using it in a different way to what it's intended, but it still works. No, to say that these polygenic scores would give you an indication of who's going to have autism, it's like using a tool in a different way to what it's intended, and it doesn't work in the different way that you're trying to use it. And that's what I would say is the take-home message when it comes to polygenic scores. Brilliant. OK. Well, from a clinical point of view, and I really enjoyed the way that Laura was appropriately cautious about the findings, and talked very carefully about theoretical implications and so on. But certainly on the basis of the findings, we could say that environmental interventions are certainly indicated because the environment influences the presentation of these neurodevelopmental traits in some cases. And also, it's worth holding in mind that in some instances, at least, parents of children with neurodevelopmental traits may need additional support to make those appropriate environmental shifts in the family home precisely because they share the same traits with their children. [MELLOW MUSIC]