On transdiagnostic development and learning, with Dr Joni Holmes

Joni is a cognitive developmental scientist at the MRC Cognition & Brain Scientist, University of Cambridge. She specialises in learning difficulties and developmental disorders. During this podcast she chats to Sue about a piece of work exploring how communication, behavioural and cognitive difficulties interact and co-occur in children with and without traditional diagnoses of learning-related disorders.

You can follow Joni on Twitter here.

The paper discussed in this episode is:
Mareva, S., & Holmes, J. (2019). Transdiagnostic associations across communication, cognitive, and behavioural problems in a developmentally at-risk population: a network approach. BMC pediatrics, 19(1), 452.

Transcript

Sue: 0:09

[Podcast jingle][ringtone] Hello? Oh, it is recording. I see the little figure. Okay, great. I will do my little spiel and then I'll introduce you. Nice. Okay. Here I go. Hi, I'm Sue from the Salvesen Mindroom Research Centre at the University of Edinburgh, and I'm recording another PsychologiCALL, um, which is our sort of little podcasty type series that we're doing during the COVID-19 lockdown to try and, um, contribute to the debate that's going on at the moment about kind of children and young people's wellbeing and learning, and maybe give people something slightly more intellectual to think about while they're stuck at home. Um, today I'm talking to Joni Holmes from the University of Cambridge, and she's going to talk to me about a paper that she did, um, using a network approach to look at trans-diagnostic associations across communication, cognitive, and behavioural problems. So it's a very wide ranging paper, it sounds like. Um, so hello, Joni. How are you?

Joni: 1:12

Hi, Sue! I'm good, thank you! Very nice to talk to you!

Sue: 1:13

Good, it's lovely to talk to you too. Thank you for giving me some of yout time!

Joni: 1:18

You are very welcome!

Sue: 1:18

Um, so what did you, as I say, this paper seems like it covers a lot. Is it possible to summarize what you found?[Laughs].

Joni: 1:26

Well, I'll do my best![Laughs] So we found that symptoms of communication, behavioural, and everyday cognitive problems don't fit- fit into the neat boxes that we might think they once did, when we think about specific developmental disorders. What we actually found was that rather than these symptoms fitting into these neat boxes assigned to disorders, the symptoms interact dynamically, and that means that they cause one another and they kind of reinforce each other. And really, we found that the sort of symptom specific idea of diagnoses doesn't actually hold, when you look at data from children who are struggling at school.

Sue: 2:08

Oh wow... And could you give me a sort of um... specific example of the kind of interactions you're talking about, you know, even just as a sort of metaphor to help us ground that a little bit more.

Joni: 2:22

Yes. So if we think about, um, say for example, symptoms of inattention, so you struggled to pay attention, for example, in a classroom situation, um, and you find yourself easily distracted and often people would assume that that would be related for example, to a disorder like ADHD. And what we found is that it's not just related to other symptoms of ADHD, but the inattention is related to cognitive problems and also to some kind of language and communication problems as well. So we find that these symptoms don't fit neatly in categories that we would associate with a specific disorder. And in fact, they can co-occur or even cause and reinforce symptoms that we would have thought were associated with different disorders.

Sue: 3:12

Right... Because this is quite a superficial reading, but you know, let's say you're struggling to kind of, um, pass some of the language that's been used in the classroom that might make you more distractible. Cause you're not sort of super following what's going on. And that in turn could feed back into your problems in the kind of communication domain, right? Is that, is that...

Joni: 3:34

Absolutely, yeah. That's absolutely it, yes.

Sue: 3:39

Great, I'm always very excited when I get that sort of thing right!

Joni: 3:41

[Laughs] You did a much better job than I did for that![Laughs]

Sue: 3:47

[Laughs] It's very easy to summarise someone else's work, I find.[Laughs] Um... So this, what brought you to this approach, right? Because this is quite an innovative question to be asking, given that so much of the research literature and kind of practice is based on the kind of diagnostic categories that we, that we like to think of as being quite sort of robust categories, right?

Joni: 4:11

Yeah. I think, well, one of the big drivers is that when we look at, basically, the research literature or sort of a clinical manuals we might use to understand why a child is struggling, they do group symptoms according to these very sort of discrete disorders. Um, but when you actually go out and work with children, or if you are a child in the classroom yourself, or you're a teacher, you might actually observe that children don't present with neat sets of symptoms. So it's very common to encounter a child who is showing symptoms we might have traditionally associated, for example, with both ADHD and dyslexia. And so that doesn't really fit with what the research literature tells us, or the clinical manuals. So I was really motivated to run this study to think about... And if we go out there and we get lots of data on lots of different symptoms associated with these different disorders, then can we run some kinds of analyses to show what actually happens to the symptoms that children do present within the classrooms? So it's really about sort of modeling the reality of how symptoms might be expressed, um, that might reflect better what children tell us themselves about their problems, and also about what I've observed in my own research studies, and about what parents and teachers have observed in their children.

Sue: 5:35

Um, that's a great motivation for doing a study. Um, so how did you go about doing this then? It all sounds quite complicated, I'm guessing you needed a quite large number of children to participate in this study.

Joni: 5:50

Absolutely! We were very lucky. So we were able to use some data for a big cohort study at the Center for Attention, Learning and Memory. And what's really unique about the data there is that it's been collected from the large sample of 800 children who have a real mix of different kinds of problems. So some children have diagnoses of dyslexia, ADHD. Other children have no diagnosis at all, but what they all have in common is that a health or an education professional has recognized that they're struggling at school. So we know that we ended up with a sample of about 800 children who were struggling at school. And then we have data from their parents, and they have rated the children on a whole range of questionnaires that might be use at school or in clinical practice, that assess their communication skills, that measure behaviors related for example to ADHD, that measure how well they might use their cognitive skills in their everyday lives. So we were able to pull all of that data together to look at how the different symptoms were related.

Sue: 6:58

Um, so what about the analysis techniques then that you employed? So it mentioned a network approach in the title of your paper, and I wondered if you could try and, um, help us understand what that looked like.

Joni: 7:12

Yes. So a network approach is a way of modeling an interaction between some variables. In this case, we were using the symptoms of these different kinds of problems that children might have. The method that we used is based on graph theory. And it sounds really fancy and really complicated, but if you... To put it really simply, it allows you to show how every symptom that you've measured is related to all of the other symptoms that you've measured. Um, I was personally really lucky cause I had a very talented PhD student working with me who run all of the analyses. So my understanding is- is not as detailed as hers, but I was very lucky to have her working on this with me. Um, we actually managed to then, when you've got sort of... You almost have like, um, almost like a network map that shows you how these symptoms are related. So it's a bit like, um, maybe like a tube map where you have all the stations and then you can see how they're all connected to each other. And just like, when you look at a tube map, you can identify maybe a tube station that's most closely related to lots of other stations. So for example, King's Cross, you can do that with the symptoms as well. So you can look at, well, is one symptom really central in this map or in this network, and what is it related to? And what that might tell you is that it's a core symptom that might be driving kind of the expression or activation of other symptoms in the network. They can tell you some quite important information about what symptoms might be really crucial. Um, and some people have suggested that you might have to use these as targets for intervention and support. Another thing that's really interesting, which was particularly interesting to this study is that you can use methods to then carve up the network or the map into clusters of symptoms. So you can see how different symptoms cluster together. And then you can say, well, the way that these symptoms are clustering do they map onto the traditional diagnoses that we have. So for example, do all of the symptoms related to ADHD g roup together? So it was really interesting in that way. And actually we found that that wasn't the case that when you c arve your m ap up and look at how symptoms naturally group together in this big group of children w ho are struggling at school, you see that the clusters don't map onto traditional diagnostic groups or groupings of symptoms.

Sue: 9:40

Um, so I mean, there's so many interesting questions I shou- could ask now, but I'm going to try and, um, rein myself in![laughs] I suppose the question that I'm most interested in from that is, is what that means for our kind of diagnostic practice, right?

Joni: 9:56

Yeah!

Sue: 9:56

So, you know, I guess you could argue that... the, the... the challenge of mapping is just that, you know, clinical diagnoses have some sort of clinical validity, but it doesn't map on to, you know, the way that parents, um, perceive their... their children's kind of strengths and challenges, or the way that they are manifesting in education settings. Right? So it's just a lack of sort of clinical into, um, daily life mapping.

Joni: 10:26

Yeah...

Sue: 10:26

Or maybe it means something a bit more profound, you know, that this, this, um, group of features that we have put into a bucket and labeled"ADHD" is not actually really a bucket that exists. You know, it's not a natural category, it's just something, but the, that seems to hang together with the observational methods that we once had available. And now we need to be, you know, radically rethinking those categories. So what, what do you think about that sort of debate?

Joni: 10:57

I agree quite strongly with your latter argument there.

Sue: 11:01

Right.

Joni: 11:01

That... That the sort of features don't fit into these neat buckets or categories. Um, and you could argue that perhaps it's sort of clinical observations playing out in a different way in everyday life, and parents' own perspectives being different, for example, to a clinical kind of observation, but there are many other studies about there, um, that have also shown that these sort of diagnostic-based criteria don't really map on to what you see in the real world. Um, so based on that and sort of my own understanding of the samples where I've been working with for a number of years, um, I think it's time for kind of a, a radical shakeup of, of how we define children's strengths and weaknesses. Um, and I think that there's a real challenge to that though, because... of the language we use, the vocabulary we use, is very diagnosis-based. Sort of our cultural systems for assessing and supporting children are very diagnosis-based. And so we can show empirically, we perhaps need to move away from these sort of hard boundaries between different disorders. Maybe we need to reframe things. Maybe we need to think about it in a different way, but there's a huge challenge in trying to do that because I think it's gonna require a lot of, uh, uh, a big cultural shift in how we do assess and support children. Um, and so, I mean, for one example at the moment you have, some children are referred through health services that we support, um, and some assessment we support, and other children are referred through education. So we've even got two completely different, different systems that currently deal with children's strengths and weaknesses. And I think one of the starting points might be sort of harmonizing that into kind of a single unified system, and then moving away from these hard kind of diagnostic cutoff and boundaries, and thinking more about the sort of, the space and dimensions upon which children have particular strengths or weaknesses. That's a huge challenge, I think.

Sue: 13:10

Yeah, absolutely. And I, I think the point that you raised there about the way that these diagnostic labels have such, kind of socio-cultural meaning is, is a really big deal, because as you say, you know, that there are key that we use to unlock a certain amount of service provision and understanding right...

Joni: 13:29

Absolutely! Yeah!

Sue: 13:29

... in an... In an ideal world. I mean, it doesn't always, um, unlock as much as we might like, but, um, but it's also about people's identities and sort of self-understanding...

Joni: 13:40

Yeah!

Sue: 13:40

... And, you know, um, and recognition of, of why they might be finding something, a struggle or, you know, that sort of thing. So it's really, it's very important that we don't whip that rug from under people's feet, isn't it?

Joni: 13:55

It's very important. And I think that what we will have to do in sort of, I know it's going to take a long time and it will require a huge shift in people's thinking is that we still have to ensure that we think about the individual and people around them, and how they can identify with why they might be struggling or what the problems might be that make life difficult for them. Um, and I guess that one way towards that is that you have to develop a new system that still has those very important elements to it.

Sue: 14:31

Mmmh, absolutely. Well, that sounds like, u m, you've got some more work to be g etting on with, in that case Joni![Laughs]

Joni: 14:46

[Laughs] Yes! Certainly![Laughs] It's a lot of work Sue![Laughs] I can't do it single-handedly!

Sue: 14:46

Part of a lifetime approach to revolutionising the way that we understand and support young people. But, you know, I've got your back Joni, I'm right behind you![Laughs]

Joni: 14:57

[Laughs] Well, some elation lies ahead!

Sue: 14:59

[Laughs] Absolutely. So, so obviously we're going to want to... This is going to be such a smooth segway I'm going to do right now. We're going to want to bring along some early career researchers, PhD students with us on this quest to the, to the new future. So I wondered if you wanted to finish by giving any kind of message to them in terms of, you know, what advice you would give to people who are, you know, maybe kind of starting out in the field.

Joni: 15:25

So yeah, I think it's probably solid advice that people hear all the time, but I think there's no such thing as a silly question, um, or a stupid question. It's like when you... If something comes to mind that puzzles you about the talk you've heard or, um, something you've read, that question that you might think is not worth pursuing, it could unfold into something really valuable. So always question things and do pursue your own interests as well, um... I think sometimes we get drawn towards certain lines of research because we think they might be trendy or exciting at the moment, but I would say trust your own instincts about what interests you and what might be an interesting question to ask.

Sue: 16:17

That is great advice. And I think, you know, things like the work that you've been doing, that you've been talking about this now is obviously the result of, you know, someone being willing to ask a question that maybe felt like it wasn't a sort of question you're supposed to be asking, right? So...

Joni: 16:34

Yeah, absolutely. It's a question that might challenge so many things on so many fronts, but it's, you know, it could yield some very important, um, results or data or information that could enable us to help and support young people in better ways.

Sue: 16:51

Ah, Joni, such an inspiration! Um, thank you so much for your time. It's been a real pleasure to talk to you!

Joni: 17:00

Thank you!

Sue: 17:00

And for anyone who's... Thanks!... anyone who's listening, you'll be able to find out more about the work that Jony talked about today by following the links on the podcast page, which is at ed.ac.uk/salvesen-research. So thank you very much, Joni! Bye!

Joni: 17:21

Thank you Sue! Bye![Ringtone]

Sue: 17:21

Okay, we did it! I thought that went quite smoothly![Podcast jingle]