Developmental Language Disorder with Dorothy Bishop: History, Advocacy, Terminology, Co-morbidities and so much more!
Note: If you listen to the podcast on this website (seehearspeak), its best to use the Chrome internet browser. Best listening comes through the apps listed below because you can 'follow' the podcast to receive notifications of new episodes and see time stamps that correspond with the podcast transcripts.
Special Guest Contact Info:
Contact Info for Professor Bishop’s Frequent Collaborators:
For the Episode 15 Transcript, Click "Read More" below
00:13 Tiffany Hogan: Welcome to See Hear Speak Podcast Episode 15. In this Episode I talk with Dorothy Bishop about Developmental Language Disorder: History, Advocacy, Terminology, Co-morbidities and so much more!
This conversation is 1 in a 5 part series on Developmental Language Disorder (known as DLD) released this week in honor of DLD Awareness Day, which this year is on Friday Oct 18th.
Thank you for listening! And don’t forget to check out www.seehearspeakpodcast.com to sign up for email alerts for new episodes and content, read a transcript of this podcast, access articles and resources that we discussed, and find more information about our guests. Also don’t forget to subscribe to this podcast and leave a positive rating in apple podcast or wherever you are listening.
01:15 Tiffany Hogan: Well, Dorothy, Professor Bishop, I appreciate you joining me for the SeeHearSpeak podcast. I’ll have you start by introducing yourself!
01:30 Dorothy Bishop: Well, it's great to be here and I'm really excited by the podcast! But just to say who I am, I'm Dorothy Bishop. I’m a professor of developmental neuropsychology at the University of Oxford in the UK, with a particular interest in developmental language disorder.
01:43 TH: Great, I'm very excited to speak with you as someone who's studied multiple communication disorders such as developmental language disorder, dyslexia, ADHD and speech sound disorder. I've very much appreciated your study of the intersection of many of those. And for the listeners, I would like you to tell us about those disorders that you've studied and how they relate to each other. And in particular for this podcast series, I'm very interested in your description of developmental language disorder or DLD and how DLD relates to other disorders such as autism, speech sound disorder and dyslexia.
02:20 DB: Okay, that's a big chunk! But I'll do my best. So I started out way back in the 1970s, I believe, studying children's language disorders. And we had some special schools for children with serious speech and language disorders, which I visited to do my doctoral studies. And I think at the time I thought that this was a very clear cut condition. But I think the longer I've studied children with language disorders, it's become really clear that it's not a very tightly defined disorder with very clear boundaries. Many children have problems that start veering into the autistic spectrum. And there's sometimes real debate as to whether a child should be diagnosed with autism or with DLD.
03:10 DB: Then you've got the whole issue of children who have problems with speech sound production. And again, when I started out I think that was thought to be quite separate, but what I kept finding was that there were children who were classified as having speech disorders but when you assessed them, they actually had problems with oral language and comprehension and things like that as well. And so, again, there was some overlap. You can have, obviously, a speech sound disorder that doesn't involve broader problems with oral language. But this is the thing, you can also have of course a language disorder that doesn't involve a speech disorder. But there's quite a lot of children who are in the middle there with a mixture of the two and similarly with reading problems.
03:55 DB: I've done quite a lot of work looking at literacy as well as oral language in children with DLD. And typically we find pretty poor rates of literacy, and in some children really serious problems. But again, it's not everyone. That's one of the things I've been interested in is to see how far you can distinguish children who have oral language problems with written language problems. And then what about these children that actually do learn to read despite oral language problems? And then on the other side of the children you have problems with learning to read but don't have oral language problems with them. So I think the main thing is that... I think a lot of our ideas of these very distinct disorders have come because each type of disorder is studied by one type of professional that tends to just assess one type of thing. And it's only when you start really assessing them more broadly, you find that quite often there's multiple things going on. It does indeed even extend further. So I certainly think attentional problems, motor problems are not uncommon in children with DLD.
05:02 TH: I've also found that interesting too because, like you said, it has to do with the professionals that are studying each of these impairments. And even in our curriculum we have a class on speech sound disorder and then we have a class you take on language disorder, and a class you take on autism. So it seems that when I was a student, I really had this feeling that they were quite separate. But immediately, when I started clinical practice, I realized that there was... That there's a whole child there. And then it can vary along those continuums. I used to also joke that… because I study also the intersection between language and reading… I would say when I was at a reading conference if someone said, "What do you study?" I would say language disorder. And then if I was at a language conference, I would say reading disorder. And that was very acceptable. So I do think it's like there's these... It's rare to have someone who studies all of the intersections and I really appreciate that about your work. You've started a public awareness campaign around DLD, and I'm wondering if you can tell us a bit about the impetus, the story behind that advocacy.
06:09 DB: Yeah, I mean I think it was really that after literally decades of studying children with language disorders, I was getting fed up with the fact that people were saying this is quite common. So there was Bruce Tomlin's epidemiological study coming out saying that 7% of six-year-olds have this. And yet nobody knew about it. This was just ridiculous. I found if I was trying to get funding, the research councils and people who give you research funding, you could be pretty sure that they would have heard of autism or dyslexia but they've never heard of children's language disorders. And just the lay people you interact with... I have my sort of taxi driver story. I was in a taxi, and I had my aha moment when the taxi driver said, "What do you work on?" I tried to explain about children language disorders and he was utterly baffled. But then I had to say, "Well, it's a bit like autism." And he knew autism. "And it's a bit like dyslexia." And he knew what that was. And I found myself thinking, why is it that we work on a condition that is completely invisible in the general world?
07:10 DB: And I also I felt it was really a matter of being important to advocate for these children and their families, because if you're not recognized it just makes life much harder to get resources, as well as for researchers to get research funding. So that was really why I started thinking we should do something. And fortunately I was really lucky that I had some colleagues who felt very much the same and we were all good mates anyway. And so we got together and formed this group to create what was originally called RALLI, the Raising Awareness of Language Learning Impairments which had me, Maggie Snowling, who worked more on reading problems, but has also sort of gone over the boundaries. And Gina Conti-Ramsden, in Manchester and Courtenay Norbury in Royal Holloway and Becky Clark, who is a practicing speech and language therapist.
08:04 DB: And we were able to get together with one of Becky's friends who was actually working in public relations, who gave us really good advice about how to run this campaign, which was completely differently from what I would have had in mind to run the campaign.
It’s a very loud noise, but I'm afraid it's rain on the roof!
08:23 TH: Oh, that's okay. [laughter] No problem. You know what, we do this podcast rain or shine.
08:29 DB: Yeah. We are actually in a building that is not the building I started my career in. Because that was condemned because of asbestos.
TH: Oh no! [chuckle]
08:40 DB: And we all had to move out. So we're in a makeshift building, which is quite nice, but does have this roof that every time it rains, you get very loud noise. So that what's happening at the moment.
08:49 TH: I think it's the environment clapping on your DLD efforts. So now, you had RALLI start, but now it's RADLD. So there's been a change?
08:58 DB: It's now RADLD! Yeah. So what happened was, we realized... This again, we were really helped by Natalie Orringe, who was this PR person who advised us. She would ask us a lot of questions, and she'd say things like, "Well, if you wanna have a PR campaign, you need to be able to sort of say, one in so many children has this condition, and these are the defining characteristics." And I realized that people wouldn't agree about this. And even worse, couldn't agree what to call it. So at the time I started, it was really “specific developmental language disorder” that people talked about. Some people in those days talked about “developmental aphasia.” And then about, in the 1980s, people started to use “specific language impairment,” particularly in the US, and I think Larry Leonard had an effect on that.
09:56 DB: And I moved to that, because everybody seemed to be using SLI. But then there was a backlash against SLI from some people who started to think it was really a bit of a misnomer. Because the problems were not nearly as specific as the textbooks would have you believe. And so, even when we started RALLI, we called it language learning impairments, because we didn't really... We weren't really committed to SLI, but at the same time, we didn't know what else to put. But we had as one of our goals to sort out the terminological mess. And we hadn't done that. We started trying to raise awareness, but we still had this massive problem that the terminology was so confused and confusing. So that was like the next step that we felt we had to deal with, and so we did a study to try and get consensus.
10:45 TH: So that probably was one of the biggest barriers to public awareness then was really just getting consensus. So, can you tell the listeners about the process by which you got that consensus?
10:55 DB: Yeah! So that was... This was a whole succession of quite chance events, in fact, that led to this. One was that I met a person who's a professor of... I can never remember the precise title, but something like, she's a family doctor, professor. Oh, she's gonna kill me for not remembering the details, but it's Trish Greenhalgh. She's a very wonderful professor working in the medical field, who is here in Oxford. In fact, when I first met her, she was elsewhere, but she has since moved to Oxford, it was great. And she does a lot of research on things that are more qualitative, using qualitative as well as quantitative methods. And we were discussing this and that, and she was asking me about what I do. And I said, we have this problem. I'm working on a condition that's common, but nobody can agree what to call it. And she said, "Well, what you need is to use the Delphi method," which I had never heard of.
11:57 DB: But she had used it herself, and could give us some advice on it. And so, with Maggie Snowling who's also in Oxford and my statistician, Paul Thompson, and me. We decided to run a study where we would use this method. It is a process for achieving consensus where you start with a whole set of statements, about which there might be some debate, and get a lot of experts together to rate them, and give justifications for their ratings.
12:30 DB: And you then can collate all that information and feed it back to people. So they can really see on their own, are they a bit of an outlier or where there is consensus. If they are an outlier, it's up to them to possibly update what they've written, and try to persuade people, or to perhaps go with the flow. And the beauty of this method is, there's two things that are very good about it. One is that you can do it online. So people can take their time to think about what they want to say. And the other is, it's anonymous. If I had previously thought, we might have some meeting and try and get some consensus building in a day with a whole load of important people who knew about this stuff in a room. And I think I could already tell that would be a disaster because you'd have very important people with very opinionated views sort of dominating and trying to argue their corner, and you'd probably just end up with a lot of disagreement.
13:27 DB: Whereas with this method, it's more reflective, and people really can't pull rank. Everybody gets heard. And we deliberately went for a fairly wide-ranging panel of experts that included predominantly speech and language therapists, as we call them, or speech language pathologists, as you would call them. And predominately from the UK, but also quite a few people from other English-speaking countries, including experts from the US. And then some people who were coming from education, psychology, medicine. And then some people representing advocate groups for families affected with disorders. So there was quite a broad spectrum of opinion. And we managed to go through two rounds of this Delphi process twice, so we started just thinking we're going to agree on who is it we're talking about. So what are the defining conditions? And then we moved on to try and decide what label should we use.
14:26 DB: So it was reasonably successful. It was quite remarkable. We had about 56 people, I think, on the panel. So to get these people coming together was quite a feat. But I think the main thing people did agree it was important even though... The one thing that was really controversial was the terminology. I mean, that was the hardest thing. We could get people to agree about a lot of other things, but getting people to accept that we should have a common label. Everybody agreed that that was really important for communication, but there was sort of quite a lot of battling about what that should be and it was much, much harder. But arguments were made in all directions.
So, we did end up, though, with developmental language disorder or DLD as the agreed term, which has been accepted by most of the people who were on the panel. I'm very pleased to see has been sort of picked up quite widely around the world. I think it has had the, to my mind, rather an unanticipated benefit. Not only are the researchers and the clinicians agreeing more, but I think it has seemed to give a huge boost to just general recognition of the condition as well, which is great.
14:49 TH: Absolutely. And how, to give the listeners an idea, how long did that process take?
15:53 DB: It wasn't too bad. I mean, the hard bit was the thinking bit. The actual doing it, I mean, once you've assembled your panel, we did this on a shoestring. We didn't actually have proper funding for it.
16:03 TH: I wondered about that.
16:05 DB: We used some existing, I mean, little corners of funds. But, I mean, fortunately I had funding for me and for my statistician. So we were able to do it very economically. I mean, setting up the questions and setting up the simple survey software that you can use to set up the thing. And then you really don't want to give people a very long time to give their responses. You want to give them enough time that they're thoughtful about it but I think we said to people, "If you can respond in three weeks or so.” Then we had to look at the results and sort of tried to integrate them before feeding back. That was the thing that I did with Maggie Snowling. So, we didn't put in our own views but we tried to absorb the views of other people, and identify what were the sticking points. Quite often when people disagreed it was not that they really disagreed. It was more that the way something had been worded was sort of making them interpret it one way or another.
17:05 DB: And so it was quite often possible to get better agreement just by understanding the reasons why people were going one way or the other. That took a bit of time. But I think we got both phases. I mean, we did phase one, and then we had a pause, and then we did phase two. I think we did it in about 18 months.
17:28 TH: Wow.
17:29 DB: The actual time for getting people to respond wasn't more than about three weeks once it was set up, but it was more, and having to go through the second round and think about it and then write the final papers which had 56 co-authors. So you can imagine, it was not trivial getting everybody to agree.
17:42 TH: Wow. I think 18 months is lightning speed considering what you were tackling, getting terminology after decades, and getting people to have a consensus. Also thinking about the multiple countries that came forward with this consensus and the multiple authors. I think that's phenomenal.
18:00 DB: And people were saying to us that we should, should we extend it to other countries? Non-English speaking countries? And, of course, in theory, the answer is yes. But we felt it was enough of a problem to just get the English-speaking people to agree. But there's now some people in other countries where other languages are spoken who are using it as a model.
18:15 TH: Oh, that's good.
18:17 DB: They will come to their own consensuses or not, as the case may be.
18:20 TH: Oh, that's great. I didn't know that. That's really nice extension. I think these papers also serve as such a nice literature review on the key characteristics of DLD. You had to synthesize so much work that was out there, and I'm wanting to talk about a few of those key pieces from my point of view. And, of course, please raise any other key pieces you think. But, one thing that struck me was how you tackled the early diagnosis aspect, and what you found about the variability of language, and the reliability maybe of diagnosing DLD early. Can you tell us a little bit about that?
18:57 DB: Yeah. I think one of the things that we found was that there were quite a lot of people, particularly those coming from the education side who were… they thought one of the barriers to talking about DLD is they really don't like the word disorder. And I think there's some justification for that. You stick the label “disorder” on a child and it really seems medical. It seems serious, and so on.
19:20 DB: So, I think we came quite rapidly to the conclusion that we were only going to include under DLD, children for whom there was a sort of persistent problem of some severity, and had a real functional impairment. We would have to rely on what we knew about prognostic factors to identify those early. But there is quite a lot of longitudinal literature that allows you to pick up which of the children that are likely to have something severe and long term.
19:47 DB: So, we really didn't want to have the sort of classic late-talkers who might just be a bit slow to get going. And there's also quite a lot of pre-school children. I mean, I've done longitudinal studies where we've started with them at four, and there are children who are presenting with particularly expressive problems who by the time they're five-and-a-half they're pretty much in the normal range. And so we thought those sorts of children, we're not saying they shouldn't get help or they don't exist, but we felt they should not be categorized as the same as DLD. We felt the word disorder was just one that should be reserved for the longer term problems.
20:23 DB: So, that was a sort of interplay really between the label and the diagnostic criteria, was that we felt that we should focus on children who had longer term problems without... So then it may be possible to identify these children very young. But really predominantly, if you've got a child who's got, for example, notable comprehension problems at a very early age or who's got very limited speech production. Like coming out with one-word utterances at the age of three or four, would suggest that you probably are picking up DLD very early.
20:59 DB: But if a child is just perhaps not producing fully intelligible long sentences at the age of three, I think we would say we wouldn't call that DLD. But not say that it doesn't merit some attention but that it's not the same thing. So we would use some... And people have said what terminology would you then use? And I think I would be inclined to just say language difficulties or something more descriptive, rather than something that sounds more like a diagnostic category.
21:30 TH: In the US, I know that it's commonly used developmental delay as more broad in the educational setting. So that's, I know, there's definitely differences in educational setting diagnostics versus the clinical setting and research setting.
21:46 DB: Delay in the UK tends to mean them intellectual disability.
21:50 TH: Yeah. That brings me to my next question, actually! I think from my point of view, one of the most controversial aspects about the papers and the literature that was gathered, was this idea of nonverbal IQ. So how does nonverbal IQ play into the diagnosis of DLD or SLI? Can you speak to that for a moment?
22:12 DB: Yeah, I mean, so I was sort of brought up with the idea of specific language impairment or specific developmental language disorder as being very much defined in terms of a mismatch between nonverbal and verbal abilities. So that the child had to have a normal range, nonverbal ability, and then a significant language problem. I think what worried me for a long time, was that this did seem to be an artificial category, in that it didn't really reflect the children I was seeing very often. In fact, there were very few children who met that sort of criterion.
22:50 DB: More often, you would have a child who had clearly got significant problems with language, and the nonverbal ability would be in the normal range, but not massively discrepant with it. And these children in the UK were quite often finding themselves in a bit of a no man's land. Because they didn't have the big discrepancy, the funding agencies who provide intervention would be saying, "Well, no, they haven't got the right sort of problem.” Then they'd be left with no intervention at all. And there were two bits of evidence that concern me here. One is that there really is no evidence that these children respond any worse to intervention, than children where there is a big discrepancy. And indeed, their prognosis from longitudinal studies I've done is that bit worse, so you could argue these children are more in need of some sort of help.
23:41 DB: And then I think the killer for me was when I did twin studies. I was interested in the genetics of language disorders. And I would find two identical twins, one of whom had the big discrepancy and one of them didn't, but they both had a language problems. And they seemed very similar, in many ways. Clearly, this would really go against the idea that it's a completely different sort of condition, depending on what your nonverbal ability is. So I think, now what we've come up against the people who are saying, "Well, are you therefore just saying that any child with any severe intellectual disability or something would be included?" And we're not. Because we do make a distinction if there's a child has some sort of syndrome, something like Down syndrome, or if they meet criteria for intellectual disability, which is more severe. And that tends to mean that the child is not likely to be able to carry out activities of daily living independently and things like that. You would not say that's DLD.
24:41 DB: But it's more these children who may not do well on a nonverbal IQ test, but are still within normal limits by any regular psychology assessment. They just don't necessarily have the big mismatch with language skills that we would say, "They've got a language problem." As far as we know, there's no reason to suppose they won't respond to intervention. Why should we ignore their language problem because they don't meet this mismatch criterion? So that was really the logic behind it. And it was seen very much... I mean, this was a matter of some debate in the CATALISE panel, but it was seen very much as an equity issue that there were children who are being denied intervention, because they didn't meet a rather artificial criterion. The criterion that had been imposed by researchers who had nice tidy minds about how children should be. But we're actually therefore excluding quite large numbers of children who we felt did merit some sort of help.
25:38 TH: That makes sense. And I think, I've been studying these data myself. I have been struck by the continuous nature of the data. I mean, we expect these conditions or these abilities would be on a continuous measure. So, if you have a study where you say, "Okay, this is a child with SLI," and to be a child with SLI, you have to have, let's say, if you're being more liberal, and these are the studies I've traditionally done, is that you say, "They don't qualify for intellectual disability." So they have a nonverbal IQ at 70 or above. But then when you go out and find these children, it's hard to really see a difference between a child that has a 70 IQ, 71 versus 69, that's the arbitrary cut point that you make, or even more, so many studies make the arbitrary cut point of 85, for instance. So then we have kids with 86 versus 84 on a nonverbal IQ.
26:31 DB: But to be honest with you, it's not even then. I mean, when I came in the field, it wasn't so much that you had. I mean, we've all moved to that sort of cut point of 85 or 80, in most of the cases. But when I came into the field, it was you actually had to have a measurable mismatch.
26:44 TH: Oh, I see, a discrepancy.
26:46 DB: If you measured the two things on the same scale, you would have to have, I mean, ICD, one of the classification systems, says you should have a 15 point discrepancy. So if your language skills came in at about sort of 70, you'd have to have a nonverbal ability of at least 85. And if your language skills came in at 85, you'd have to have a nonverbal ability of about 100. So there was this notion that you actually had to have a measurable discrepancy, rather than just meeting some arbitrary cut-off. Now, I think most of the children that we are now including, who might not had been included under some previous definitions and more got nonverbal abilities in the '80s. But certainly in the UK, you would have people arguing they should not have help, because they didn't have this big mismatch.
27:27 TH: It's almost like that, it goes back to the idea that nonverbal IQ is a sense of potential, right?
DB: Yes. Yes.
27:35 TH: So if you're at your potential, and I know we see this parallel in dyslexia and luckily, we've moved away from that as well.
27:41 DB: Yes. Exactly. It's exactly the same arguments for children with reading problems where again, it can be much easier to get help for kids who have poor reading in the context of an IQ of 100 than children who are poor readers in the context of an IQ of 85.
26:43 TH: And I know as a clinician, it was so frustrating with when I worked with children who had a reading disability. I would just have to wait a few years till their scores unfortunately dropped in reading to then see that discrepancy, and that just wasted that time. So, yeah, I definitely recall that period of discrepancy, which is slightly different than the arbitrary cut point aspect too.
27:56 DB: I think the main thing is that we felt that language is not determined by nonverbal ability. In fact, sometimes it's the other way round, to some extent, because what you certainly do see is as children get older. Unfortunately, children with language problems often the nonverbal ability does decline. And I think it's that the content of nonverbal IQ tests tends to involve more implicit verbalization, even if it's not explicitly a verbal test. And so it's not sort of some constant measure of your potential in any way, shape, or form. I think the other thing that's affected this debate is that I, like many other people, have been quite interested in trying to find what are the core characteristics of children's language problems. And if you want to do that, it's quite useful to have a discrepancy definition, because then you can say what's the correlate of the language problem. And be fairly confident that it's not just something due to more general nonverbal difficulties.
29:12 DB: But that's the researchers' definition, which isn't... People have taken definitions that are useful for research, and then sort of applied them or they have certainly again, been picked up and used in a way to deny children services, which has been really quite pernicious.
29:31 TH: In the US, I don't necessarily think that children are denied services, at least for the speech and language. They have this category, speech language impairment, and they can as speech pathologists qualify children. But I think what has happened here, is that you don't have a connection between the research findings, and what's happening in practice. So if you ask a clinician, have you read the latest research article on specific language impairment? Does that affect the children you see? They say, "I don't have those children on my caseload," or "I'm not sure if I have them on my caseload," because even to make matters worse, here we called it SLI, speech and language impairment. So then it was even more of an issue of making that connection. And I think here, we also have the issue of how to even diagnose language impairment based on multiple different assessments and different arbitrary cut off points. And I interviewed on Elena Plante on the podcast to talk a bit about that. But I think that's also been a huge issue of what tests to use and what criterion to use.
30:35 DB: And one of the things that I've had a lot of people communicating with me is they think because we've done the CATALISE thing that we can come up with some sort of absolutely worked out protocol for diagnosis. And of course we can't, it's far too early. I mean, it just... One of the things that our exercise really emphasized was that although we do have some good standardized tests, there's still... We don't really have very good measures of functional impairment. And that is a key thing, according to our definition. I mean, I have seen children when I'm just routinely testing kids that do badly on a standardized test, but actually, nobody's worried about them. And they do seem to be fine. And possibly just a bit inattentive or something
31:17 DB: Then you have other kids who have quite severe problems, but might pass those tests. But you feel the tests are just not quite, particularly kids with more sort of pragmatic difficulties, may pass standardized tests and still have problems. So I think, we need better measures to to look at the whole range of how children are using their communication skills. I think the core idea we had was it is not just gonna be particularly... I would say I'm a great fan of standardized tests often revealing things that you might not have been aware were going on, particularly when children have got comprehension problems. But at the same time, I would be very averse to sort of just saying, "Well, if you score below this, on this test, you've got a DLD." I think you'd need to look at it in a broader way than that.
32:00 TH: I do think that the next frontier is that a functional outcomes and trying to understand and quantify more thoroughly the functional outcomes. But the listener may think, "Okay, well, there was this consensus that occurred, and now we have this public awareness campaign, which is gaining a ton of momentum across the world, the RADLD campaign." I've been a part of the North American version, DLDandMe.org. But that would I think, give a misrepresentation because to me, the debate is still continuing a bit whether to call these children DLD and SLI. So I wondered if you could talk to us a little bit about why you think that debate is still continuing and what are your thoughts as we move forward?
32:44 DB: I think it's still continuing, in part because there is no ideal solution. I mean, I've written about this. We’ve come up with DLD. DLD has tons of things wrong with it as a term. And I think one of the problems is that children are very heterogeneous. It implies that you've got a clear cut single disorder. It's really describing a sort of broad area. And I think the other thing is that there's people who've been researching SLI for years who've got a lot invested in that. And one of the things, arguments I've heard is, well, if we drop the term, we lose all the insights we've had from studying SLI. I think that's a valid argument too. Because I think, in a sense, the way we've defined DLD will encompass SLI as traditionally informed. So anything that was true of SLI is likely to apply to DLD. It's just that DLD is a broader category.
33:38 DB: But I think that yeah, it's really problematic trying to get the terminology right when you're dealing with a condition that is very fuzzy on boundaries. We don't have a very tight definition of it anyway, and the children are varied, and some of them have more additional problems, some don’t. I think it's understandable that you can... That's what we had going through our CATALISE exercise and, in fact, we have got everything from the exercise online open, including all the comments on all the questions. I don't think anybody's ever looked them up, but we have got them on open sites framework so that you could actually read, although we did have to anonymize things to make sure people who were responding were not identified, but you can read the sorts of arguments that people were putting forward for and against different terminology. These are intelligent people with very good views, so it's certainly not simple to just say we're gonna get a clear answer. I was very uncertain before we started this exercise how it would end up, and in fact I was quite keen to invent a completely new term, like “language learning impairment,” and everybody hated it, so it didn't go anywhere. But I thought that might be one way forward, to just drop everything from the past. But I think that probably would have been even worse, and that would have also cut us off from the medical profession.
34:00 DB: Because in the DSM, it is language disorder. And in the ICD, which is used more in Europe, which is the World Health Organization, they have developmental language disorder. So we are more in line, at least, with the medical profession with the DLD. But SLI, I can see why some people are wedded to it. I think it's mainly for more historical reasons than anything else. Although, I do think that some people also feel that there is this difference between the children with a very specific problem and those in the broader group who might be included in DLD. I would give no evidence for that, because I just don't think it's there from any of the research domains that I've looked at.
35:40 TH: I think talking to clinicians is helpful to think about from their perspective. It's not that we are arguing that these kids don't exist, there's not problems, that we don't need to create awareness, that we don't need to make sure that they're treated. It is more of an issue of semantics and how to bridge that gap between research and practice. So these kids exist, that's not what we're debating. I did hear a clinician tell me that once, and so I think that's one thing to always kinda keep in mind when I do when I do these presentations. I also try to link to the research when I work on my papers now that involve children that would more qualify from a traditional SLI diagnosis. I always say, developmental language disorder without intellectual disability. And then I have a footnote to try to tie to the research. I say some of the studies these children have been called SLI, some are DLD, refer to the CATALISE papers. I do think it's on us as researchers to bridge that gap between the work and make sure it's pushed forward in the appropriate way.
36:45 DB: But I would be clear that DLD is always without intellectual disability.
TH: Yes, yes.
DB: If you have intellectual disability it would not be a case of DLD and I think there's a misunderstanding among some people...
TH: Yes, it is, it is!
36:57 DB: Yeah, I think we've got where everybody in... But we say specifically you would exclude intellectual disability, you would include autism, you would include children with a known biomedical syndrome.
37:08 TH: That's really helpful, Dorothy, because I think as a researcher myself, I get confused about, do I say without to make it really clear? Or do I say with if I include with the children?
37:20 DB: Then it wouldn't be developmentally, then it would be language disorder with intellectual disability. So we make that distinction between language disorder, which is the overarching category, in which DLD sits as a sub-category. But you can then have language disorder with intellectual disability, with autism and so on. So it's emphasizing that a child may have a language disorder with some other condition. Again, this shouldn't be seen as a reason to deny them intervention, provided there's some effective intervention there. There's a sense sometimes that with this condition as opposed to many other conditions there's a tremendous desire to keep out as many children as possible and it's a bit weird, really.
38:03 TH: Like pure blood you need or something. I guess, that then that means to me that all the research done on specific language impairment, that included children with a cut point that was maybe more liberal in nonverbal IQ, like 70. Those are just the kids that have DLD now, 'cause thats the definition. And if it 85, though, that's where it gets a little tricky. Because that's not really intellectual, as some states have SLI with 85 as the cut point.
38:30 DB: But they would still... They would be to my mind, they would be very much classic DLD, they haven't got the... To me this is the problem, to me SLI was meaning a real discrepancy, not just meeting some lower threshold. It morphed into something different in the many years that I've worked in the field. I think that's because if it didn't you wouldn't find any children. Very, very few cases who have that mismatch. That’s how it's started, it started that you actually had to have a mismatch.
39:02 TH: And I never worked in the mismatch in terms of a researcher so I never... Luckily never had to deal with that. As a clinician I did, but when I came back for the PhD, that had dissipated a bit. So what do you think is on the horizon for research on DLD? I mentioned the functional outcomes, what do you see as the big research in clinical questions on the horizon?
39:20 DB: I think there's a huge neglect of adults with DLD. We know very, very little about them. There's a really strong difference between autism and DLD in this regard. We have done some follow-up studies of kids that I saw when they were younger, and it's fascinating. But when we've tried to follow people up with DLD, they're very hard to find, they disappear into the woodwork, you really can't locate them, they don't tend to respond to correspondence. Whereas people with autism are much more engaged, and families of kids with autism are much more engaged. I think that these are people who find communication difficult. They often have poor literacy skills, and it's really hard to engage with them. But I think we do... One of the things that I'm increasingly being asked about after the RADLD campaign is how do we identify people with DLD in adulthood? And what are their needs and what are their outcomes? So I think there's a handful of longitudinal studies, but that we have tended to think of this is a child condition.
40:33 DB: And quite a lot of speech and language therapists I think assume that they just sort of get better as they get older, but in fact there's quite a lot of adults out there who have these persisting problems. We need to know more about what are their needs, and how can we help them function in society.
40:46 TH: I think that makes a lot of sense. I'm also very interested in their self-esteem, and how it affects their choices over time. I think the shame that can be associated with having a language disorder, I would like to see more research in that area, as well.
41:05 DB: I mean, the other area I'm particularly interested in is really just getting better methods for intervention. So I've spent many, many years trying to determine are these really auditory problems at heart? So influenced by the work of Paula Tallal, and come to the conclusion that there often are some auditory problems, but they don't seem to be causal as far as I can see. They seem to be just another associated problem. I've been interested in looking at sort of memory problems, and then I've sort of come to the point of thinking, well, actually, we're continually trying to find something that might explain the poor language skills. But actually, this is a problem with language learning. It’s much more interesting to do studies which actually look at the process of learning itself. If we did such studies, and we could understand what are the factors that really determine what makes something easy or difficult for a child to learn, this could also be of huge benefit for targeting our interventions better.
42:00 DB: I think that's, that's where there is a huge need. I don't think I'm gonna solve that one before I retire, but I think that would be a good line to take, investigating manipulation. I mean, we're starting trying to look at things like if you are even just doing a boring vocabulary learning task where you're trying to teach a child new words. We don't know really basic things about how the timing or the spacing of presentations affects children with language problems and what is the optimal way, should we bang on and teach them one thing and only go on to something else when they've learned that one thing, which would be very boring, but might be more effective. Or should we give them mixed items and so on? And there's a bit of literature on this in things like second language learning, which I think will be quite interesting to hook up with. But it may be that for children with language disorders, different conditions are needed and that maybe exactly why they've got a language disorder that they don't, perhaps, follow the normal course of events when they're learning to learn new words or indeed learning grammatical constructions.
43:11 TH: It does seem like our intervention now is based on more. It’s more dosage. But I agree, I've been very interested in the research and have done my own research on word learning and the characteristics influence word learning. But also, all that work on statistical learning and what that looks like, working memory aspects. And I have a colleague here, Yael Arbel, who does work on feedback processing. So, even do we tell them they're wrong, do we tell them they're right, how are they processing. I think there’s a lot of really new work in that area.
43:45 DB: And I think it's weird that I think what happened was, again, I've got this very long perspective over many, many years. When I first got interested in the field, I got the impression that such intervention as was going on was very much of the rote learning. That was completely abandoned because people found you could teach children by rote learning, but it didn't generalize to the real world at all. So people went to a much more naturalistic sort of approach and embedding intervention in naturalistic interactions and things like that.
44:12 DB: But in a sense there are some things where it might be useful to go back now, especially that we've got computers that can make rote learning a bit more fun. There are questions about whether some things might be, such as, learning a new vocabulary, there might be a role for treating it in a more rote learning fashion. But if so then you would really want to know under what conditions would that optimize. And certainly, I'm not advocating that we should bore children to death, learning things that they then can't use in real life. I think you always need a mixture of things.I just feel I’ve misconceived SLI or DLD, as I now call it, for many years, as possibly a perception, auditory perceptual problem. Now, I think, no, it was a learning problem all along. The problem is learning stuff, and we have very little literature on the process of learning.
45:03 TH: Yeah, I like the work. Also, it's looking as you mentioned almost like the contrast of explicit versus implicit learning. Liza Finestack's doing some interesting work in that area. I love this approach of what works for whom, I think it could get even that complex. We probably won’t the silver bullet, but have a sense of the multiple silver bullets that could work for individual children and what that combination looks like. It's very complex and I think we're often limited by as humans, by our own understanding about these complexities, right? Like thinking, you know, once we get to three and four dimensional space, it can be very tricky. I like to think that makes a lot of sense to think about the intervention more, and the adults, and the functional outcomes. So we're getting close to the end of our discussion. I always ask two questions of every guest and I wanna start with the first one. That is, what are you working on now that you're most excited about?
TH: Or what direction are you taking that's really getting you excited?
46:05 DB: Well, I am interested in doing some stuff on what we've just been talking about. We’re trying to develop some online games for looking at... Particularly at comprehension, which is where I started, looking at language comprehension in kids. So, we're playing with that idea. The idea would be, if we can get things that work online it will be much easier to see large numbers of kids. But we have to make things fun, we have to make them fun games.
46:31 DB: So we're working with a platform in the UK called Gorilla, which has been developed and is used massively, now by psychologists for running psychology experiments. That makes it easier to sort of set up experiments for kids, so that's fun. And then, the other thing I've moved much into, as you may know, is sort of the reproducibility crisis and methods. I spend a lot of my time now giving talks on how we can overcome problems with the limitations in methods that increasingly, certainly in psychology and in biomedical science, are recognized to have really held us back. Or misled us over the years. As a psychologist, I'm getting really interested in how human psychology affects how we do science. So, we have things like confirmation bias, where it's much easier to interpret and remember information that agrees with our preconceptions.
47:31 DB: Whereas to be a good scientist, you should be requesting your preconceptions. So, I think there's this wonderful conflict between how science should be done, and how it actually is done. It all has a lot to do with human psychology. So, that's another direction I'm sort of veering off into when I get free time, which I don't very often. [chuckle]
47:50 TH: Oh, I love that. And that's actually right in line with what I was thinking about, how we're limited by our human capacity to do science. We have to do it through our filter of being human.
48:00 DB: Yes, and I think it's particularly difficult for people working in a field like speech language pathology, where people have the best of intentions. You come up with a new intervention, and you want to help children, and you want it to work, and it can be really quite difficult, then, if you do a trial and you find that it's not effective. I think that it's really hard to do trials that show things are effective. I think it's difficult to combine being a caring, compassionate practitioner, sometimes, with being a sort of really rigorous scientist. But I think we need people to do that, because otherwise we'll just carry on doing things that we hope work, which don't actually work.
48:42 TH: Absolutely. Wow, yep. I've been interested a lot in implementation science, and I think it's similar. In some ways, trying to tackle this in a slightly different way of thinking about what are the factors that help research that is replicable into practice, and how to create science that maybe takes into account the clinical practice barriers.
49:04 DB: Yeah. And I think the other thing is that you have to be careful that you sometimes get the feeling that researchers feel very superior. They come along and sort of want to hand down to clinicians some great truth, without really realizing how it is in real practice. So I think it's a two-way street. But I think, when I first started out in the field as a full-time researcher. I trained as a clinical psychologist, but I've done research for most of my career. I had this belief that if you did some research and published it, that the practitioners would somehow pick it up by osmosis. And I was quite shocked to find that things that I thought were well-established were not sort of generally understood by practitioners. They of course don't have time, energy, whatever to be reading the literature. So, there is a huge need for this sort of interchange.
48:42 TH: Absolutely. Now, my next question is, what is your favorite book from childhood? I have to ask this, because of course, as a person who loves language and literacy, I'd love to know know if you have a favorite book from childhood? Or now, even.
49:55 DB: Well, I was thinking about this, I was very much an avid reader. I was a real bookworm as a child, but probably not terribly discriminating. In fact, I came from one of those homes where we didn't have many books but we had a really good library. So I remember trailing down to the library to get books, but I was thinking one of my favorites probably was Little Women.
TH: Oh, yes!
50:32 DB: Which I think is quite nice because... I mean, the great thing about books, of course, is they take you into this completely different world. I had no understanding of the background of Little Women or anything else. [chuckle] But I liked the women and I liked Jo, of course, who was the sort of bookworm. So that sort of stayed with me, and I think that was a great book for girls, because it was a book about friendships and sisterhood, but also just being sort of resilient and tough, and doing what you needed to do. So, it was good role models, but not too syrupy. Probably if I went back now, I'd think it is syrupy. I don't know.
51:09 TH: Oh, it's a big one here in Boston, as you can imagine. So... Oh, that's a great book.
51:13 DB: And in adulthood, I discovered, as many adults did, I discovered JK Rowling and Harry Potter. I think the woman's brilliant. I read that once, one of those on a plane, when I had a long plane ride and I was a bit tired. I just went straight through. And so much of what she writes about is capturing the battle of good and evil, and many things that we now see in politics, but also journalism going wrong. Although it's a children's book, the themes are really very universal and important. But I do read grown-up stuff too. I like Margaret Atwood. I like quite a diverse range of authors.
51:55 TH: I also try to make sure I read, too. I think it's so important to keep up on reading. And I also loved Harry Potter, I was one of those people that actually stood in line at midnight when the books came out. I was actually in my 20s at that time, and I was standing there with 13-year-olds. [chuckle] I didn't have a kid with me, and I would go home and read it all night. So, I agree, Harry Potter is an amazing series, and still is so, so relevant to issues nowadays.
52:25 TH: Thank you, Dorothy, thank you so much for your time, I appreciate it. I can't wait to get this out to the listeners, thank you.
52:29 DB: Well, thank you for inviting me and thank you for having this whole series. I'm so impressed with what you're doing. You seem to be on fire, in terms of these podcasts. [chuckle]
52:41 TH: It's an interesting hobby, isn't it? [chuckle]
DB: It is, it is! Well, long may it flourish.
TH: Thank you!
DB: And best of luck with it.
52:30 Tiffany Hogan: Thank you so much.
52:55 Tiffany Hogan: Check out www.seehearspeakpodcast.com for helpful resources associated with this podcast including, for example, the podcast transcript, research articles, & speakers bios. You can also sign up for email alerts on the website or subscribe to the podcast on apple podcasts or any other listening platform, so you will be the first to hear about new episodes.
Thank you for listening and good luck to you, making the world a better place by helping one child at a time.