Marisha (host): Welcome to the podcast. SLPs have submitted questions about the complexity approach, and I am so incredibly excited to tackle some of those questions with our guest today. I knew she would be able to break down the complexity approach for us, probably better than anyone else, because I've learned all of the practical implementation tips from her, I've been able to implement this with my caseload, and it's been an absolute game changer. And I learned all of this through her amazing site called SLPath, and it includes comprehensive courses as well as so many free resources that make the complexity approach totally doable and approachable, no matter what your caseload looks like. So I can't wait to hear all of her practical tips and strategies. So without further ado, I'm going to go through her bio real quick because it's absolutely amazing.

Marisha (host): But Jennifer Taps Richard is a speech language pathologist in the San Diego Unified School District, and an Indiana University graduate. She provides small group and classroom intervention to caseload and at risk students. And so she's in the trenches with us, but she's also a coordinator for the Phonology and Articulation Resource Center. She applies research and supports SLPs in articulation and phonological intervention through staff development, consultation and coaching. She also owns SLPath, like I said, a private company committed to promoting best practices and speech sound disorder intervention through online courses and intensive workshops. And I am so incredibly excited that she's here with us today. So let's dive in. And before we get into all of the juicy content, I'm curious because you're clearly a very busy person, very, very smart with free resources and ideas for all of this. But I'd love to hear a little about you and what led you to do so much work in this area.

Jennifer: Well, thank you. It's a pleasure to join you on the podcast today. I've just always found linguistics to be fascinating. I'm kind of a nerd where [inaudible 00:02:15]that's concerned, and it's just from the time I was in undergrad, when I first kind of learned the beginning of these principles, I was just kind of blown away by how impactful it can be to teach complex sounds to kids with speech sound disorders. And I remember the first time I heard about it, I was a junior in college in Indiana and by chance my father was in town visiting and he was picking me up after the class and, when I walked out he said that he thought I met like Bono or some other rock stars something because I was so excited about the potential for pursuing this kind of work and what it can do for kids to have them become more intelligible and therefore more successful in communicating. That's what kind of led me to be interested in this from the time I was 19.

Marisha (host): Yeah, that's such a great story. I love that. And I got a little bit of chills hearing it. Because you can totally see your enthusiasm for that topic. Help us understand a little bit. What is the complexity approach?

Jennifer: Well, I think most SLPs are familiar with the normative approach. The idea that we teach sounds and developmental sequence. That's been a historical practice in our field since really the beginning. Complexity takes this idea and completely turns it on its head. Instead, it advocates teaching phonetically complex non-stimulable and later mastered sounds. So this is a very different way of approaching helping kids to learn more probably about the sound system as opposed to one sound at a time. And, it's really based on about more than 30 years of research studies, and it applies linguistic principles to help SLPs re-target for action. So Judith Garrett, who's at Indiana University, my Alma Mater, was really the first proponents of this approach and other investigators have looked at this at other universities since then. And if people are interested in reading an excellent article that it provides an overview of the complexity approach, Garret's 2007 article is fantastic. And it'll give you a much bigger sense than I could give. It's due, during this session.

Jennifer: But essentially the idea is that we teach complex targets, especially complex clusters like /spl/ as in splat or /thr/ as in throw. And this helps kids to win more rapidly about the sound system, thereby helping them become more intelligible, more quickly. And essentially it's not just about the treat to targets, it's about the entire sound system. So it's really remarkable still after doing this for 20 plus years, I still get a thrill to see what kind of changes occur in a child's sound system after a few months of working on particular targets. But one other thing that I thought might be helpful when talking about what is the complexity approach is just to kind of share, the general findings of the body of work of complexity and then also how we've applied it in San Diego Unified in the trenches basically. So, the original research, it's all received, just like most biological research, individual services and the university clinic, the kids were around age four.

Jennifer: These are kids who only presented with phonological disorders. And that's to keep research more controlled and have fewer variables. Just like we often see with phonological research. So these are not kids who also have language disorders or other kinds of learning issues or behavioral issues. And that's pretty much the case with phonological research. But in looking at the original complexity studies, most kids received about 20 to 25 hours of intervention over 10 weeks and they increased an average of 20 to 25% in their consonance correct, which is a huge gain in a short amount of time. And that's because of these complex targets. But even with those original studies, Garrett and others, positive that these principles may apply to other populations as well, that we could potentially apply them to kids with concomitant language disorder, kids with autism, kids with down syndrome, kids with mild articulation disorders.

Jennifer: And in particular I wanted to highlight kids with concomitant language disorders, because researchers have estimated that as many as 60% in children with phonological disorder also present with expressive language disorder. That's a huge percentage of the population. And we of course want to consider that. But in our district we've had SLPs applying these principles for more than 15 years. And what we've really strived to do is go across populations and not only children with phonological disorders, but the kids we see in the real world. So the kids who also have language impairments, kids with autism, down syndrome, kids who stutter, I mean you name it, to anybody who needs to learn more about the sound system. And so what we've done is we've collected several case studies that have been submitted by various SLPs and not just students that I work with.

Jennifer: And I partnered with Jessica Barlow and Phillip Combats from San Diego State University. We took 32 of those case studies from this heterogeneous groups. So these again are kids, it's a very mixed group. So some kids had only phonological disorders, some kids had co-occurring language impairments. We had some bilingual kids, we had a child who stutters. And these kids all received group intervention in the schools just like we typically do. We typically provide and all 32 of these kids received treatment on complex two element clusters such as /fl/ or complex three element clusters such as /skr/. And so what we did with this data is, we had probe data before or at the beginning of the school year versus the end of the school year. And so we analyzed this data to look at a number of measures. And what we found was the children received about 20 to 25 hours on average of intervention, well in groups.

Jennifer: And each group had an average of 25% increase in consonance correct. So just like the original complexities studies. And this is different conditions. Because with the original studies, there was a very strict protocol. It was individual and so very controlled, which is what research needs to do, but the schools are messy. We have a million interruptions, but we can still achieve a very high level of increase despite that. And we even broke it down to different groups. So we found a comparable increased for kids who only had phonological disorder versus kids who had phonological disorder and language impairment. We had the same kind of increase for kids under five versus over five. Yeah. And he also had the scene changes for bilingual versus monolingual kids. And so essentially all this tells us that it's possible in the trenches to help kids rapidly increase their intelligibility when we teach them complex down.

Marisha (host): Oh, that's amazing. And I love the comparison of like the clean research and then the research in the trenches. That's amazing and so helpful.

Jennifer: Yes. Yes. And we were heartened just to see that we've had so many SLPs in our district applying these principles and people kind of apply it in their own way. And the way I also think of it is the target selection, which we'll talk about a little bit later, is kind of the science of it, but how an SLP actually goes about teaching is more the art. So that's the how. And so many people have creative ways of doing that, that I would have never thought of and vice versa. But what matters ultimately is as we'll talk about, is the target itself. That the target we select is everything.

Marisha (host): Yeah, now that's amazing. So helpful too. So why would you ... Because you've kind of talked about it a little bit, but why would you choose this approach over another approach?

Jennifer: Well, first I recommend applying these principles if we want kids to finish intervention as soon as possible and reduce our case load or maybe not reduce it but manage it a bit more effectively. But the way I always think about different methodologies for phonological intervention is that all methodologies work. But the difference is the treatment efficiency. So it's more efficient to teach kids, broadly about the sound system and deeply given complex targets rather than starting with early sounds and then building and teaching every early sound. If we were to teach every sound and then cluster and sequence that would take years and years. And kids really don't have that kind of time because we need them to be able to access the curriculum, we need them in the classroom and, if they have curriculum language impairment, we need just to support them with that rather than always working on sounds. And so, those are some really important considerations.

Jennifer: But another thing that I think is persuasive is, there is a very strong evidence base to support complexity. And so, inaudible did a paper in 2006, and he identified five major theoretical perspectives with regard to my biological intervention, including complexity as one of the five. And he found that there's almost more research on complexity than the other four major methodologies combined. So this is a significant body of evidence that's been amassed over more than 30 years. And so I feel like that's very persuasive as well. And then one other bit of information from research that I found compelling is a Baker and McLeod also did a review and they were trying to identify the number of studies that have been done for different methodologies. So they found that there was an eight to one ratio for complexity studies as compared to normative studies.

Jennifer: Even though the normative approach of teaching sounds in developmental sequence has been a historical practice, it really is very limited evidence to support its efficacy. I think those are important reasons we'd want to consider that. But, the other thing we want to think about is any target we select, we want that target to have maximum impact on the system. And so of course we want kids to learn dealers, like Angie, we want them to have those sounds. But if they learned those sounds in the midst of being highly an intelligible, that's not going to make a huge difference in terms of their overall intelligibility. So instead we want to consider different linguistic principles. So I'll just highlight briefly two language universals that are often highlighted with regard to complexity target correction.

Jennifer: So, basically, complexity is based on language universals, our laws. And what's so powerful about This is, not only do they apply across all languages of the world, they also apply to every individual speaker. And so, here's one example. So there's a universal that's been identified across languages that stipulates that affricates imply fricatives. And so that has two different meanings. So first, if we think of it from the perspective of an entire language, that means if a language has affricates, it also has fricatives. So one implies the presence of the other, but it doesn't go on both directions because a language could have fricatives but it could also not have affricates. So, the affricates are more complex and that they are implying the presence of the fricatives. We can also think of that regarding an individual speaker. So that means if a speaker produces an affricate in his or her system that speaker also has at least one fricative.

Jennifer: And so if a speaker potentially have difficulty with both of these sound classes by targeting affricates, we can predict change in fricatives as well. That's what makes it more efficient than teaching all the fricatives and then getting to the affricates. So, that's one language universal. And then one other I just wanted to share briefly, there's also universal that has shown there are three element clusters, and what I mean by that is a cluster like /spl/, /skr/ and those kinds of things. They implied both two element clusters that do not have /s/ such as /f/, /l/, /t/, /h/, /r/, and /w/, et Cetera, as well as two element clusters that include /s/, such as /st/, /sp/, /sm/ and so on and so forth. So essentially this tells us that by teaching a three element cluster, we can also help kids to improve two element clusters with /s/ and without /s/. So you can see that/ that would have a pretty big impact on the sound system.

Marisha (host): That has been magical. I feel like it's magic in my own practice, like I started working on /skw/ with a student, like that was one of our first targets. And then just watching the progress, it's magic! It's like raising arms, yeah.

Jennifer: It absolutely feels that way. And I mean, what you've done is great, and you could probably come up with a very creative way of doing stuff.

Marisha (host): We'll see about that. But yeah, it's been so amazing, and I'm just so excited to hear it broken down in a way, because it's not an approach that we hear about as much despite all of that evidence, which is really interesting.

Jennifer: Yeah. And I think, just from traveling around and meeting SLPs around the country, a lot of people have shared that in their graduate programs. A lot of programs are doing kind of a surface overview of many methodologies rather than going deeply into one or two. And, I kind of understand that impulse on one hand because you want people to be aware of these different methodologies, but then it really doesn't help us to be truly prepared or master them in order to apply them in the way that they need to be applied. And so I think, many people are exposed to it just like in a textbook as one out of 20 methodologies, when in fact I think it would be better to prepare people for our profession to go deeply into at least a few, be aware of them.

Marisha (host): Yeah. Like the top three with the most evidence. Then how to do it would be helpful too, just some idea.

Jennifer: Absolutely. Absolutely.

Marisha (host): Yeah, I agree. It's so amazing. And then so what types of students would benefit from this approach or who might not be a good fit?

Jennifer: Sure. Well, I think really the principles apply to all children with phonological disorders. And so that could be kids who are monolingual, multilingual, kids with concomitant language impairment, but as I shared earlier, SLPs in our district, not just me, but multiple people have applied this with kids with autism, with down syndrome, kids who stutter. Because, essentially every speaker has a real governance sound system. So we all have these linguistic roles that are applying in our systems. And so we can leverage that to help us to identify ideal targets for each speaker. And the other thing I wanted to mention about that is, I think normative studies give us good information about which sounds are relatively early mastered versus later mastered. But I also think that some of the study data could be unintentionally misleading in that, I think it's led us to kind of underestimate kids in terms of what they're prepared and what they're ready to learn, because their young children produce clusters.

Jennifer: So, there have been studies that show that two year olds produced clusters 50% of the time and three and a half year olds produced 75% of clusters accurately. So their young children are ready and it's within the zone of proximal development for even young kids. And there was also a really cool cross linguistics study from the McLeod and Crowe that just came out in the last six months. And they looked across 27 languages and they found that by age five children produce 93% of consonants correctly. So that's a lot. And so that means young kids are able to do that. But essentially the sooner we can introduce complex clusters into the system, the better. And kids can really learn these targets with the right support. Now one thing I didn't mention earlier when I was talking about three element clusters, there is one caveat that research studies have identified, is that to teach a three element cluster, the child already needs to have both the second and third sounds in the phonemic inventory.

Jennifer: So, for instance, if we're going to teach /spl/, both /p/ and /l/ need to be in the child's phonemic inventory, which means that the child not only produces this sounds but can use them contrastibly to help the listener understand. So like if a child said pink and think, that /p/ and the /th/ sound are being used contrastibly to help the listener understand the difference between a color and what we do with our brains. And so we would need to make sure that /p/ is a singleton and /l/ is singleton ran in the phonemic inventory. If not, that's a little too complex for that individual child. And we go to the next best thing, which is a complex two element cluster. But even if kids are not good candidates for three element clusters, they are still good candidates for complex two element clusters.

Jennifer: And there's no prior knowledge necessary for complex two element clusters. That's good news. But the one population, I am more conservative with applying these principles forward, are kids with childhood apraxia of speech. So at this point, there are preliminary studies that show that it can be effective, but we need more evidence. And because this population needs such individualized and intensive support, I think we should go with kind of what the the standard is at this point. But essentially we know even for kids with apraxia that the intervention is all about moving through sounds. And so when you're teaching a cluster, it's about moving through consonant sounds or into the vowel. So it's still kind of in the same vein. I would feel better if there were a little more evidence at this point. But I have applied it with a couple of my students with apraxia and they've had good outcomes as well.

Marisha (host): Wow, that's amazing. That's good stuff. So what are ... Because, in your course you go into a lot of detail, and I know this is another kind of big question, but what are your best practices for like getting started with an evaluation?

Jennifer: Sure. Well, I could probably talk for five hours just about this alone, but I'll try to focus on what really matters, kind of the big picture. So first of all, we need an in depth sample of those singletons and clusters in the target system. And, I want to highlight standardized test have a place. They do provide us a snapshot of the child with a snapshot of the child system, but they're not comprehensive enough to be the only evaluation tool. Because usually they only sample sounds once in each word position, and they usually don't cover all of the cluster, mainly half of the clusters. And so they really don't support strategic target selection. So we can think of them as a piece of the puzzle, but we need other tests or probes to supplement it to really be able to characterize a child's sound system.

Jennifer: So what I would suggest, there are several in depth independent samples that people could access. One is a probe that I created, it's called the In Depth Phonological Assessment, the IPA for short. And probably the IPA gives me yet another [inaudible 00:23:35], because we use IPA transcription when we're transcribing and then we can celebrate with an IPA inaudible factor. It's all over everyone's [inaudible 00:23:46]. But at this point the IPA, the target words, so the transcription firms are on SLPath for people to freely download the pictures but the target words are not there. And it's just kind of a long story. But, why not? But, I do want to tell you about two other probes that are available, where the pictures are available.

Jennifer: So, Jessica Barlow from San Diego State has allowed me to offer an SLPath, it's called the little PEEP and PEEP stands for Protocol for the Evaluation of English Phonotactics. And it is a pretty in depth probe and it might be ... And I know that it's longer than the IPA, but it provides really in depth information about child sound system. One other freely available probe is from how historical she created a cluster's probe and a probe for singletons that focus on complex singletons. So what she wanted to do was focus on the sounds that kids are most likely to have an air. So like her sample is not heavy on /p/, /b/, /t/, /v/, but it is heavy on /k/, /g/, /l/, /r/ and some of those later master type sound. So that would be a great supplement to a Goldman Fristoe or another kind of standardized test. But of course I do want to highlight if a child is bilingual, we do want a sample across both languages whenever possible.

Jennifer: Unfortunately, these days there are many free probes available online, including Doctor Barlow's Spanish probe, which is on his SLPath. But there are also several clubs available through Sharynne McLeod websites. She likes to stumble three probes and I just discovered a few months ago, there's a new website, it's called speakaboo.io, and they have ... I have lost the exact number in my brain, but it's something like 20 free probes of other languages. And they also have it where they have a video of the native speaker producing the words. So, I would feel comfortable as a monolingual SLP recording a child and then comparing it to see if it's the same or different from the adult model. And so it really could empower us to look at that as the child's other language. So some really cool things are available for those purposes. And, I can tell you about a few other things if you'd like?

Marisha (host): Yeah, go for it. Cheer up.

Jennifer: Okay, thanks. Okay. So, we also want to consider stimulability. And so a common phrase that is used in phonological analysis is in or out sounds. So in being sounds that either singletons or clusters a child produced at least twice independently. So for stimulability tasks we won't be sampling the end sounds, because obviously they're already stimulable, because they're in. So what we would focus on for stimulability task would be out singletons and out clusters, first specifically out singletons for this example. But Glaspey & Stoel-Gammon really came up with a cool way to look at stimulability, and it's really changed how I view things because when I was in graduate school, if a child didn't produce a sound, the way that I was training to elicit the sound is let's say that they don't produce ash in a sample.

Jennifer: So I'd say to the child, can you say, /sh/, and if the child said, /sh/, I would consider it to be stimulable. Glaspey & Stoel-Gammon take it a little bit further because they want to see not only can the child produced the sound in isolation, but also in syllables with the /e/, /a/ and /u/ vowels. And the idea behind that is /e/, /a/ and /u/ are different places. So front, back, high, low, and you can also see that if a child pairs one back vowel with a back sound, does that facilitate? So you can get a better sample of what the child can control and do. Furthermore, you are allowed to provide help. So you can say to them like, if you're trying to elicit /l/, you want to put your tongue up on the bumps and let it drop.

Jennifer: You know, so you can see what they can do with just a tiny bit of support. And if they can respond to that in the midst of this dynamic task, that says that they have a lot more potential, a lot more stimulability than a child who even given that wouldn't be able to produce it. So it tells us a lot more about what stimulable versus what is not stimulable. And so it's basically giving us a richer picture of modifyability and just seeing what a child could do with just a little bit of help. And so if you wanted to see the examples of these stimulability tasks, they are SLPath, there's one in English and one in Spanish. And so of course when in Spanish only has Spanish sounds. But I also want to highlight a few other tasks that I think are important.

Jennifer: We will definitely want a conversation sample as well as a single word test, because in a conversation sample, it gives us a better sense of overall intelligibility vowels, [inaudible 00:29:04], rate, those kinds of things that we wouldn't be able to pick up on it in single words. Because you were just saying one word at a time, you're not going to really be able to hear kind of the cadence of the child's productions. So that would give us a sense of that. And usually what I do in the conversation sample is I usually just describe it or just use it for, just a baseline to kind of hear a sample of where the child is in conversation. And only educationally do I actually go through and do a 50 inaudible sample where I score the percentage of intelligibility.

Jennifer: I do think that's valuable sometimes, but, I don't know that it's going to yield us information that's going to help us with target selection, which is the ultimate goal of a good assessment. But some examples of helpful tasks for conversation samples, as we know, we've all probably encountered kids who are telling us something and we want for the world to understand what they're saying and we just can't catch it for whatever reason because we don't have any shared context. So it's helpful to set ourselves up. So we have that shared contexts. So wordless books like the Frog, Mercer Mayer books, we're also ... This beautiful black dog called Carl. So there's a bunch of ... Where those books so the child could tell the story about Carl and his adventures or we could have toys with different target sounds, so that way we know kind of what the child is talking about. So I think those are some of the things that we want to consider, but then just a few more tasks if that's all right. Yeah?

Marisha (host): Okay.

Jennifer: So, there are three more tasks that I like to highly recommend because they can really help us with differential diagnosis. So a study from 2015 from Marie and her colleagues, they looked at ... I don't remember at multiple tasks, that would help to differentiate between kids with apraxia versus kids with phonological disorders only. So they wanted to see which tasks could help really to differentiate the two groups. And there were two tasks in particular that stood out. The first was a polysyllable task. So, they elicited, I think it was 25 different polysyllable words and they were not only looking at the accuracy of the sounds and the polysyllable, but they were also looking at stress patterns.

Jennifer: They were looking to see if kids deleted syllables or not and different kinds of things like that, that would be more enlightening for kids with apraxia and to have those prosodic kind of bears versus kids with a final manageable pattern. So that was a very, valuable task. The other one they found to be most informative was an oral mechanism exam. So looking at non speech movements including Dido kinetic tasks even though that's not really non speech. But there was a big difference in the performance between on the non speech and Dido kinetic tasks for kids with apraxia versus kids with monological disorders only. But one other tasks that I'd like to throw in to help with differential diagnosis is we know that kids with apraxia do better with well-rehearsed sequences. So things they've said a lot are usually a lot clearer than things are saying spontaneously.

Jennifer: So what I'd like to ask is to produce automatic sequences like the alphabet or count or a song and hear what it sounds like. To have them do that versus a spontaneous conversation. And it's been remarkable because I've only had maybe five kids in my whole career with apraxia. But one in particular, when she was in fourth grade, when I was doing her triennial, she had made a ton of progress. She was maybe 80 to 90% intelligible in conversation, but still definitely having some motor planning issues. But the clarity of her counting from one to a hundred versus having a spontaneous conversation was remarkable. I mean it just really jumped out and it's not that I needed that for deferential diagnosis at that point, but it was just really interesting to kind of see that.

Jennifer: But the last thing I wanted to highlight is we also want to consider the impact of this speech sound disorder on the child and his or her life. And so there's some really valuable tools that we can look out for this as well. This is another one is from Sharon McLeod and by the way, Sharynne McLeod's one of my phonology heroes. That's why you hear me saying her name so frequently, but one, it's called the Speech Participation Assessment of Activity of Children. She calls it the SPAAC for short and it's a series of questions for the child, the parent teacher, sibling friends. And so that way you can have some sense of what the child thinks about his or her own speech, how he or she feels about different talking situations with the parent observes in terms of confidence or frustration or strengths in general.

Jennifer: And so it can be really beneficial to gather that information. And then Sharynne McLeod also has a tool, it's called the Intelligibility and Context Scale. It's called the ICS. Yeah. It's a very elegant, simple tool. It asks the parent seven questions on a Likert scale. So one being never, five being always. So, the first question is how much do you understand your child? And then the next one is how much do immediate family members understand? How much to extended family? So it goes through seven different listening groups and, Sharynne McLeod and her colleagues have looked at this across different languages and cultures and they found that the parent reports are pretty accurate. Now, it doesn't allow us to calculate a percentage of intelligibility.

Jennifer: So if they're, getting three, which is an average of three which corresponds to sometimes being understood across different groups. It doesn't mean that it's 50% intelligibility, but it does give us some sense of how much the child is understood and how this might be impacting the child's life. But, I do want to mention that all of the tasks that I have described here, there is a report template on SLPath, a phonological report template, and all of the tasks are included in there. So if you wanted to see an example of a writeup, it's available for people to download.

Marisha (host): That is amazing. Like you just planned out my whole evaluation.

Jennifer: It's all good.

Marisha (host): And then I'll link to all of these different resources and show notes. So it's easy to find them.

Jennifer: Great.

Marisha (host): So helpful. And so, and I know this question is pretty much an entire course as well, but can you give us some initial tips on identifying targets for intervention?

Jennifer: Sure. Well, it all starts with this in depth assessment. So we need all that rich data, but looking at singletons and clusters. And so then what we want to do is an independent analysis. So looking at what sounds the child produced to communicate. Now, this is different than the kind of relational analysis approach, which is kind of comparing the child's production to the adult targets. And I understand, I mean there is a reason to do that. We do want to identify different patterns and things like that more for descriptive reasons, and just because if we can tell parents or let's say a teacher of this child is deleting, /s/ from words that might help them figure out what the child's trying to say in the moment versus, not sharing this kind of information. But the tricky thing with the patterns or the process analyzes is that they don't really tell us what to target, and they can sometimes send us in the wrong direction.

Jennifer: So for instance, a very common example is, let's say a child demonstrates fronting even a lot of the time they'd be not all the time, but the child produces /k/ and /g/ in a few words. So if we were to then target /krg/ to address the fronting, we do want the child to learn /krg/, but that's not going to have a big impact on the system. And so the patterns I think are more for descriptive purposes than for identifying targets. But the bigger picture is we want to look at the information we have from the probe data and identify the child's phonetic inventory. So that's, which sounds a child independently produced, just which sounds the child produced. And one thing that's important to consider, his kids do not restrict themselves to the target language sounds. And a good example of that and I think many of us have encountered kids with this, is like for instance, if a child is trying to say beach and the child says /b/, that's an affricate, it's just an alveolar affricate.

Jennifer: So we'd want to give credit to that child for producing that, because that child knows something about affricates. He or she doesn't yet know about the English affricate. But still we want to get credit for that. And so that's one thing we want to keep in mind. We also want to see if the child, or we want to identify the child's phonemic inventory. So which sounds a child is using contrastably? The earlier, I use the example of pink versus think that means the child is using the /p/ sound versus the /th/ sound to show the listener the difference in meaning. So you want to look at that. We also want to identify a cluster in mentoring, a stimulability like we talked about earlier. But once we have that information, we can utilize this to select targets. And so, really what we want to do is introduce these complex structures into the system to support intelligibility.

Jennifer: And so I'd really like to encourage SLPs to not be cluster-phobic because clusters are great targets and there are many ways to go about addressing this. So, essentially, and there's a document on SLPath that we can link to. It's called a cluster target selection document. The first choice to help kids learn broadly about the sound system is three element clusters such as /spl/, /skw/, /str/, /spr/, and /skr/. I think I've ... I don't know if I got everyone of those right there. And so that's our first choice. But again, the caveat is the child must already have both the second and third sounds in the phonemic inventory. And that's true for a lot of kids. And the two, three element clusters that are most likely to be viable targets are /skw/, because kids will often have /k/ and /w/ in their phonemic inventories or /spl/ because /p/ and /l/ might be, or often in a child's phonemic inventory.

Jennifer: Of course our beloveth friend inaudible prohibits many kids from working on /skr/, /str/, et cetera. But, more often than not, and I've been able to choose /skr/ several times in the last few years, and I love /skr/. So it does happen even in kids with highly unintelligible speech. So it is out there. There are some great /skr/ activities and books that I don't want to overlook. So that's really kind of our first choice, if that's viable for kids. But if they do not have the second and third sounds in their phonemic inventory, then we go onto the next best thing, which is a complex two element cluster. And essentially linguistic principles, I won't go into all the background, have identified five complex new element clusters that we want to highlight. So that would be /fl/, /fr/, /thr/, /sl/ and /sr/ and it's ideal, according to research that we choose among those two element complex clusters. It's ideal if the child does not yet produce either of the two sounds in the cluster.

Jennifer: So for instance, if /fl/ is a good target or for /fl/ to be an ideal target, both /f/ and /l/ should be out of the system in non-stimulable. Because not only are you going to teach kids new sounds, you're also going to teach them the sequence and how to put them together. But of course we always want to consider the child in front of us in different characteristics. So, sometimes when I've had kids who get easily frustrated or kind of shut down, if something's a little more challenging, what I might choose is, let's say that the child is stimulable for /f/, but non-stimulable for /l/. Then I'll teach an /fl/ cluster. And so the /f/ will give them a sense of success. And the /l/ will be a little more challenging and then I'm going to support them a lot. One other thing that I want to highlight about this target selection is teaching complex clusters, we're giving kids a lot of support.

Jennifer: This isn't about frustrating kids by any means. And there are a lot of ways we can teach kids how to blend sounds together. We could use principles from reading. So like when kids are learning how to blend sounds in words, we can use similar kind of strategies to teach them how to produce them as well. But by using those principles ... So again, first tries three element clusters, second choice, complex two element clusters. What I usually recommend is choosing multiple targets so that way we have different options, because we usually see kids in groups, and it's nice to have some different options. Maybe you start all the kids, maybe they all have a common /skw/ cluster, but then their second cluster is different. So you start them all together with the first one and then maybe you do different ones, once they kind of know how speech services work, that kind of thing. So, that's the gist of how target selection works.

Marisha (host): Yeah. So helpful. And I know that starting with those clusters can be a little scary, but it totally works. Like people around the world are doing it. There's lots of good evidence for taking that step into the unknown. So thanks for walking us through that.

Jennifer: And I do want to mention one other thing. The way we can use teaching three element clusters as a strength is because the child already has both the second and third sounds in their phonemic inventory. They already have at least some knowledge of the sounds. So you can take that as a strength and then build on that. And the other cool thing is to teach /skw/ they just need to have /k/ and /w/ in their phonemic inventory, they did not already need to have the /kw/ cluster. But again, you can use ... They already have some knowledge of /k/ and /w/, use that build on and say, "Look, you can already say the south and I'm going to help you to say them together." And so I think that's a way to approach it using that as a strength.

Marisha (host): Yeah, for sure. And then this is again another huge question, but do you have suggestions in getting started with goal writing if you're using this approach?

Jennifer: Sure, absolutely. And there's another document on SLPath. It's called sample phonological goals. But the tricky thing about this kind of methodology in terms of goal writing is the nature of language universals. So early I talked about how working on affricates can help kids to also improve fricatives. The tricky thing is we can make that prediction. So we can say that teaching affricates will help kids learn fricatives, but we can't predict which specific fricatives they're going to improve. And so I wouldn't want us to write a goal predicting that these fricatives will prove when in fact other fricatives improved. And so then it's looking like they didn't achieve the goal. So we need another way to kind of approach this. So, what I would recommend would be a few different ways of thinking. So we wanted to recommend goals that kind of capture learning across the system.

Jennifer: So not just about the treated but more importantly about the system. Because the end goal of treatment is to help kids become more intelligible, not to learn individual sounds or clusters. We can think of the cluster target as the vehicle for driving the change. But it's not the ultimate goal that, that's all we want the child to learn. But some examples of ways to approach this is we could write first a goal to add new singletons or clusters. So, you could say something like the child will produce, and I often will write eight or 10 new singletons or clusters in single words on probes and administer three times per year. And the exact number of singletons or clusters that identified depends on different factors. And this is in the document as well. If you have kids who have more non for more stimulable out sounds you're going to go higher.

Jennifer: If you have kids who have around more robust inventory, you're going to go higher. If we had kids who have other disabilities. So let's say kids are working on speech and language, it might go a little bit lower because we're going to need be targeting both at the same time. We also want to consider social emotional factors or motivations. So, all of those are going to be important considerations. So that's the first one is adding new singletons or clusters. The next one could be to just increase accuracy. So what we can do is calculate present consonants correct across the entire probe sample and let's say a baseline, it's 40% and then the next time it's 55% then the next time it's 70%. So you could say that the child will increase his or her present consonants correct to. ... I don't remember 75 to 80%.

Jennifer: It just depends on where the child starts, of course, in single words on this probe. And if the percent consonants correct is increasing, that means the child is learning more broadly about the sound system. So that will kind of capture that. But one last option could be to use a visual analog scale. So this is a new way of taking data for months and from colleagues. And this gets away from the historical practice in our field of just doing plus or minus for a target sound. So what he and his colleague suggested was to do more of a scale from one to five. And the example I like to start with is looking at /r/. So if we are trying to work with the child on /r/, rather than doing plus, minus because with, especially with /r/, everything is a minus until it's a plus.

Jennifer: And so it's not really capturing the progress of the child is making versus a one, two, three, four, five, of one would be a child who says wab for wrap. So it's a /w/ for /r/, it's a pure substitution and then a five would be wrup for wrap. But then two, three, and four are that in between. And I think we've all heard those kids who are like so close. It's like right there and that's a four probably. Or if you can tell it's got a little bit of our coloring, but it's not a true art that's number three. And so we could use this visual analog scale data also from old writing. So we can say, you know, at baseline, because a lot of kids at baseline for /spl/ for instance, will produce a /w/ and so for that scale, and we might do one through seven.

Jennifer: So you know, they're starting with an average of one out of seven. And so we'd say, by this time next year they're going to do six and a half out of seven ... So, which is essentially, I don't know, like in the 90 percentile or 98-90% area. So that way we can really see some change over time. But a couple of other little things, that I wanted to highlight about, writing goals and just what we might see as a result of that is the kids in the original complexity studies as well as the kids in our district, just with our anecdotal data, many of them did not generalize the treated target to untreated words on the probes. That's just a very common phenomenon. But as I shared earlier, the target is kind of the vehicle that drives the change, it's not the ultimate goal to.

Jennifer: I mean, we want them to learn /spl/ of course, but not every child will generalize, but if they learn more broadly about the sound system, that's a much bigger goal in learning /spl/. But in the one other phenomenon that commonly occurred in both the studies, and our data is kids will often improve singletons and clusters, but not necessarily at the same time. So if we re-administer probes every three to four months, it might be the first time through they are like, "I'm going to pay attention to the singletons and I had a bunch of them in my system." But then the next round they're going to be like, "No. And now I'm more interested in clusters. I'm going to add some questions this time." So if you see that kind of thing occur, it's nothing to be discouraged from.

Marisha (host): Yeah. So helpful. So many good tips. Do you have any suggestions because I feel like I'm ready, I've got my evaluation, I know what my targets are. I'm going to write some good goals. Do you have any tips for getting started with treatment and getting organized there?

Jennifer: Sure. I think as I shared earlier, we can choose multiple clusters for kids in a group, that's going to give us some flexibility in terms of maybe you start everybody with /spl/ or everybody with /skw/ before you would move on to another cluster. And that's going to make it easier for choosing books that feature complex clusters or activities that feature different complex clusters. I also like to make sure that kids have their own cards with the target. So either ones that, I have from other sources or where ones that they draw themselves because then they kind of own those cards and there's some ownership there as opposed to hear, just the words over here and they're just on these cards. I want them to be their cards. And so that way it's more meaningful to them, especially if kids like to draw, I love to have them draw it because then their version of splash is going to be different than another kid's version of splash and they're really entertaining little pictures too.

Jennifer: But the other thing is I tend to stay away from games just because the focus goes on the game rather than what they're actually trying to learn. But instead I try to focus on fun books and activities and that way we're getting some meaningful practice. So I've become kind of obsessed with books that feature complex clusters. So, I got my own group of clusters collection. And so two of my favorite books, one is from Mo Willems who does the Pinion Elephant Books. And there's one called Watching You Throw the Ball and has a lot of /thr/ words and it's the delight. And the nice thing about that book is you could also change ... So let's say you have a group where none of the kids are working on /thr/, some are working on /fl/, for the /thr/ kids, they can say throw, through, et Cetera. But the /fl/ kids could say fling or flung ... So you can always kind of modify it for an individual within the group.

Jennifer: And another favorite book of mine is Mr. Strong from the Mr. Man series, those little square books for the different characters. In Mr. Strong, I know they use the word strong 37 times. So there's a lot of meaningful models of it. And then the book is hilarious and there's a video on youtube. It's a married video by his great British broadcaster. And it's just hilarious. I have to confess, I've watched the video on my own, without kids sometimes. It is really funny. But there are also many activities that feature complex clusters. So, and if you've seen slap of before, but you can throw them against the whiteboard or like more metallic type of door, and they just really splat against it. And kids were kind of /spl/ forever with that, or those little mind-up animals where you wind it up and then they flip, like little flipping frogs. They've got pretty much every animal. You could do that for /fl/ for flip, flop, fly ...

Jennifer: There's so many different ways you could apply that. But the other thing I would really recommend is we want to make sure there's a solid foundation at the word level and make sure that kids are producing strong production of whatever the target is as soon as possible. Before we would jump into kind of mixed practice where we're using it in a sentence or in a story or during a game or ... I am not playing gamed. But during like a more fun kind of movement type of activity. But once kids have that strong foundation at the word level, I kind of have this mantra that I like to follow, that I want kids practicing in different places with different people for different purposes. So not just the SLP room, even though there it is a magical place, we want them practicing outside.

Jennifer: We want if we can go in the classroom and have them show off for a friend or for their teacher or make a little video of them that then we send the link to their parents. So they can see. Because we don't want it to just be, but what we're saying, we want them to use it with different people. But then, we can also play around with other conditions that kids will encounter in daily communication. So we know that kids aren't just sitting straight up in a chair with a piece of paper when they're talking. So we need to kind of create conditions that are more like what Golan counters. So that means for instance, we might have been doing different actions while they practice, like doing jumping jacks while they're saying a sentence or dancing while they're saying a phrase or something like that where they still have to self monitor the accuracy of their production, or you can have them play around with different emotions.

Jennifer: So, I guess this is my long way of saying I think we can have a core kind of group of activities that help kids to really focus on their target sounds, and then get into the more meaningful production like with these different movement activities and things like that. That can go a long way. And I don't think we need a ton of different activities and kids are just so thrilled to do a lot of things over and over again, especially if we find the right activity.

Marisha (host): Especially if we're having fun with it, I think they can do the same thing over and over again. And there'll be totally happy. I'm curious too, how do you move through the target? So like if you're starting with /spl/, do you stick with that for a long time, or do you move between different ones? How does that look?

Jennifer: Sure. Well, there's more of an art to this than like an exact amount of time. But, I typically stay with a target cluster, until the child can do it pretty easily in sentences or conversation, where you can just tell that it's mostly automatic. I want it to become essentially pretty easy or just fluent for them. And, again, that means not only in my room but also walking to and from their classroom, at the lunch area, other places like that. So I would say at a minimum I'd want them to be strong in words and sentences to 80% or higher just so we've got that foundation. Because that's what's really going to kind of induce that big change. But I do want to mention any original complexity studies. They are only taught at the word level. And again, that was to control for variables because it gets a lot crazier when you get up the sentence and conversation, but just studying that kind of thing.

Jennifer: But, word level was sufficient to create these significant changes in their systems. So some of our SLPs feel more comfortable just staying with the word level and, that's fine. But I think we can still expose kids to books and other things that model these words at a higher level. And, so I would kind of do that before I would move on to the next cluster.

Marisha (host): Okay. Awesome. And then do you have it ... I know it varies depending on each student, but I think that making progress on a cluster can sometimes take a little bit longer than a singleton. So do you have like a range of how much time you typically spend on ... Like if you're starting with the three element, and I know there's lots and lots of variables here-

Jennifer: Sure, of course. Of course. Well, I would say, some kids will get it within a session or two and that's not as common. But I would say it's not uncommon to take four to five weeks to really become solid and some kids really need kind of baby steps. So what I often do if I'm teaching at three element cluster, is because they already have the second and third sounds in there, at least some knowledge of them. I start by having them blend those two sounds and it makes sure that is really strong and automatic before I ever start to add /s/ into the mix. So if I am teaching /skw/, we make sure /kw/ is completely solid before we start adding /s/ into it. And so I think there are different ways like that, that we can scaffold that will make them successful and have that strong foundation.

Jennifer: And if we jump straight into all three sounds at the same time, it may not help them to pay attention to all the different sounds or be as precise with the different sounds. So, I think it's just a matter of being creative with different things. I've also just through trial and error, figured out that a lot of times we want to get help the kids get ready for the second sound before they even say the first sound. So like for instance, if you're teaching /fl/. I had a kid who, like a lot of kids would say and sort of fly was [inaudible 00:59:37], like insert that very well. And they are not very, very common. And so what we figured out together was I could have him put his tongue up in his mouth, getting ready for /l/, freeze it and then his teeth on his lip and then say fluff.

Jennifer: So that way he was already set up for the second sound and my core articulation was already kind of half in process before you even began the /f/. I also learned through trial and error that I'm way too verbose as you probably already heard today. But, I realized with some kids I was giving them this full blown explanation and what I realized I needed to do is just break it down. So now it's more like hung up, freeze it, teeth on lip, inaudible ... Even just a picture like I'll just even point and stop talking, just so the kid can focus on what he or she is trying to accomplish.

Marisha (host): Yeah. I love those treks. So good. Then the last piece, any tips for monitoring progress?

Jennifer: In terms of monitoring progress, I look for kind of three types of data. The first is how is the child doing with the treated words and targets? Like is the child getting to the point or from where you know there's a lot of limitation, there's a lot of placement, there's a lot of queuing to independence, either at the word level, sentence level or higher. And so this is where we could use the visual analog scale again. And so what I like to try to do, and I don't always do this with complete fidelity just because of the chaos that is working in the school. But if, let's say I've worked with a group of four kids, and so I see them twice a week. What I try to do is at the beginning of a session, each time is I just have one kid come over and then I use a visual analog scale. So I see how the child does with producing his words independently, and I score it and then we go into intervention.

Jennifer: And so the next session, the second child comes up, I was in his eight words, so each basically every two weeks I have this really rich data for each child, and it ends up taking a minute or two at the beginning of the session. And then over time I've got a lot of data, and we see this progression from ones to threes or fives or whatever the target number is. So that could be very rich information to either include in a goal that we've written or to share with parents and an IEP or, just to show some change over time. But then the other big piece of data that you referenced was, I like to be a minister of the probes every three times a year. Because that's really looking at the untreated sounds that the child sound system, so to see what kind of changes occurred and that's going to give us a sense of what singletons and clusters had been added.

Jennifer: That's going to help us to understand why the intelligibility is increasing and so on. And then the third type of data, it's just more observations, is the child showing more confidence, more independence, does the child understand how the sessions kind of work to go get? Like I always have kids get their folders, and they get warmed up right away when they get there. So they know that they're working or doing something from the moment we eat, we get there. I also, of course will check in with parents just to see, how's your child doing with talking, is your child showing more confidence or any frustration or anything else like that. I also talked with teachers, just in passing, nothing formal but just in passing. But, one of my favorite indicators of progress is when kids start getting in trouble in class for talking too much. And so that to me is the most robust kind of outcome. But those are some of the things that I'm looking at in terms of progress monitoring.

Marisha (host): Yeah. Thank you for that. And thank you so much for all of your time. I feel like we just got so much helpful information. If SLPs are wanting to learn more, where can they go?

Jennifer: So they can go to my website, SLPath.com and we do have over 250 free resources, many of which I've referenced during the podcast. And then we do offer free online courses for anxious CEOs, and then one additional course on a service delivery model. So there's a lot of learning that we can do in our pajamas and spread it out rather than everything at the same time, 10 hours or six hours at a time. So it's a lot nicer to kind of spread out the learning of it.