Winning and losing - kids' circadian rhythms

Why children’s brains respond differently to rewards and losses at different times of the day…
20 March 2020

Interview with 

Helen Keyes; ARU, Duncan Astle, Cambridge University


Brain schematic


Perceptual psychologist Helen Keyes from Anglia Ruskin University and cognitive neuroscientist Duncan Astle from Cambridge University told Katie Haylor about some neuroscience news stories that caught their eyes this month. For Helen, timing is everything! She’s been looking at a paper about how children’s brains might respond differently to rewards (getting 50p) and losses (losing 25p) at different times of the day…

Helen - We know quite a lot about circadian rhythms. For example, humans tend to sleep at night and be awake during the daytime. And there are other changes like our body temperature and blood pressures vary through the daily cycle, with our lowest body temperature being at about 4:30 AM. But there are changes also in our hormones. So our cortisol and testosterone levels follow a daily cycle and they peak at particular times of the day. And this can impact on our neural patterns and even on our behaviour. So these authors we're looking specifically at children aged 7 to 11 and how their brains responded to rewards and losses taking into account these circadian rhythms. And they all took part in what we call a doors task. And this is a very simple computer game where you have two doors in front of you and you choose. And if you choose the right one, you win 50 pence and if you choose the wrong one, you lose 25 pence.

So the scientists used EEG, which records voltage differences across the brain, to measure the brain's responses to rewards and losses here in this game. And they did this across different time slots across the day for these children. And what they found was that for older children, so pre-teens, responses to gains were much stronger than the responses to losses in the early evening, so after 5 PM. And this was driven by the response to losses being really dampened. And so when the pre-teens experienced a loss, they lost 25 pence, their brains didn't really respond that much to that loss after 5 PM. However, they still responded quite highly to getting a reward. The opposite pattern was found for younger children. So children closer to age seven, researchers found that after 7 PM their neural responses to losses were greater than their responses to gains.

Katie - Does this have anything to do with bedtimes, do you think? I think it does perhaps fit in with what we know about bedtimes. So there's a phenomenon that we call "eveningness" in teenagers who have a preference for later sleep times. And we can certainly think about the earlier sleep times for younger children. I think it would be no surprise to parents that seven year olds respond very strongly to losses after 7 PM!

Katie - I've come across this concept of the "witching hour". I think it's - what is it - six till seven?

Helen - There you go. It fits right in with what we know about the witching hour in younger children, but it's really interesting that when these children become closer to adolescence, so when they enter the preteen phase, these younger teenagers might be experiencing greater urges to engage in rewarding behaviour later in the day. So considering that their brains aren't responding much to negative experience or to losses, we could conclude that these pre-teens are liable to just not be that sensible later in the day. And this might be something maybe that we should bear in mind when dealing with preteens or adolescents, but it might be something that we maybe shouldn't point out to them.

Katie - Do you think this study has any applicability in the learning environment?

Helen - It may well have applicability in the learning environment. We do know that there are peak times based on hormonal changes for alertness for example, and there are peak times for when we are best at committing things to our long term memory. So absolutely we should perhaps be taking these things into account in our education system. What times of the day match up quite closely with when adolescents' brains in particular are responding at their best.

Katie - Any advice for parents off the back of this study, do you reckon?

Helen - I think you should probably put your seven year old to bed as close to 7 PM as possible! And perhaps if your seven year old is responding quite negatively to losses, maybe use more rewards in the evening time, but I think in general it's what we might already have known for quite a long time that preteens, moving into adolescence, you perhaps respond not so sensibly to situations. And interestingly, this might be even more so the case in the early evening.

Katie - Did they look at individual variation (in circadean rhythms)?

Helen - No, they didn't. And they'd, what would have been really nice would be if they had measured actual cortisol levels on the variation and how they mapped on to these changes, and they didn't measure that. So this is a very broad, vague study, but looking at 188 participants is pretty excellent for an EEG study. So I think we can be fairly confident and that they're fairly robust findings.

Duncan presented us with his own paper this month, in which he’s been MRI scanning the brains of children, in order to see if it’s possible to predict any cognitive difficulties...

Duncan - For a long time, lots of people have been interested in what is it about the developing brain that can give rise to different types of cognitive difficulty, like memory problems or listening problems. And the results have been super inconsistent. So, you know, if you choose children that have a particular diagnosis, like attention deficit, hyperactivity disorder, if you look through the literature, you can find studies that find brain differences all over the brain. And it's not really clear why that is. And so we wanted to see whether, if you scan the brains of lots of kids, and we had 480 kids in our study, could you predict the kind of cognitive profile that they have?

Katie - So what did you find out then? Can you, via MRI, categorise kids in this way?

Duncan - If you did a standard MRI scan of children's brains, and you use that to find information about grey matter in different parts of their brain, then that information is significantly related to the kind of cognitive problems that the child has. But not very much. So if you had the MRI information, you'd be about 4% better than chance, at predicting their cognitive difficulties.

Katie - Pretty low, right?

Duncan - Yeah. So it's significant because we've got 480 children, but it's a tiny, tiny amount. And that maybe goes some way to explaining why there's such inconsistency. And that's because rather than thinking: Oh what's the brain area for ADHD, or what's the brain area for a memory problem? Maybe the answer is that there is no brain area. And there's this interesting concept called equifinality, which is the idea that you might end up at the same end point, so the same kind of characteristics, but through multiple different brain roots. And that's what our data seems to suggest. Because when you looked at it, you could see that kids could have very similar cognitive problems but really quite different brains.

Katie - Do you think that language might be a bit of a factor here? I'm wondering how easy it is to put kids in particular boxes, say of dyslexia or dyspraxia.

Duncan - Do you mean the language of the labels?

Katie - Yeah.

Duncan - Totally agree. And that's why in this study we intentionally took those things out of the equation. So we had lots of different assessments of different skills in the children, and that's what we were trying to predict. So rather than trying to predict the label, for the reason that you say, that it could be that actually these things don't map on very neatly to underlying difficulties. We tried to predict specific types of cognitive difficulty they might have instead. And so having shown that there was this really poor relationship between the MRI scans and children's cognitive difficulties, we started to wonder why that was. And then we thought, well maybe, whether or not you've got more or less grey matter in this part of the brain, or that part of the brain, actually isn't really very important. What maybe is much more important, is how well those brain areas are connected to each other, and how they're organised. And so then we used a different type of brain scan called a diffusion tensor image, which sounds quite fancy.

But really what that's designed to do, is to measure the white matter tracts or fibers, that connect different parts of the brain. And with that information we were able to produce a, kind of, wiring diagram for each child's brain, showing how well these different areas are coordinated or connected with each other. And when we did that, we found that there was a really clear relationship between a particular type of brain organisation and children's cognitive strengths and weaknesses. And that was how well the brain areas were organised around hubs. And so a hub is just a very highly connected brain area. So if you imagine something like the tube network in London, King's Cross would be a hub, because it is really important for so many journeys and it's so well connected, versus something like Russell Square just down the track, would be a peripheral node in that network. And so what we've found is that the kids who had poorer cognitive abilities, the amount of gray matter in the different brain areas really didn't count for much. But how well those brain areas were connected to hubs counted for an awful lot.

Katie - So what relevance does understanding more about these hubs have in terms of helping kids who are struggling then?

Duncan - I think in the long run, what it means is that we have to radically rethink how these different types of difficulty might emerge over time. So rather than thinking that it's, sort of, the underdevelopment of, you know, brain area A, or brain area B. If it actually turns out that there's a more general global property of children's brains, which seems to be related to doing really well in terms of cognitive development, then a whole different set of mechanisms in terms of the genetics of those brain areas, in terms of the kinds of environmental influence are really much more important. And another second thing is that this key principle about, or being organised around, hubs cut across different diagnoses. So regardless of the child's diagnostic label, whether they had an ASD, or an autism diagnosis, for example, or an ADHD diagnosis, this principle held true throughout. And I think what it drives home is that relying on a child's diagnostic label to think about the ways in which you might help them, doesn't make great sense. Because as far as we can tell so far, those labels don't really map on neatly to underlying mechanisms. And so it's much more likely that interventions that are generally good for everybody or that are tailored to a child's specific difficulties, irrespective of their label, are much more likely to be effective.

Katie - This concept of brain connectivity, or crosstalk between different bits of the brain. It seems to be a bit of a recurring theme in some neuroscience, and certainly when I have a chat to people like you, do you think this way of looking at the brain has applicability beyond learning difficulties?

Duncan - Absolutely. So for instance, there's lots of work in things like schizophrenia, and other mental health difficulties. They seem to be sharing similar properties. Hub organisation seems to be a really important characteristic of an efficient and healthy brain. And so it's kind of an emerging area of science, that seems to cut across lots of different areas. And in aging, for example, it may well be that hub organisation is really important for continuing cognitive health into old age. Which is actually kind of encouraging to us that it seems that there are some more general principles, because rather than thinking of all of these different types of difficulty, like learning difficulties, or mental health difficulties, or aging as being totally distinct and separate, they might have some similar common underlying principles. And the organisation of the brain around hubs might be one of those principles.


Add a comment