#007: How to Implement the Complexity Approach

Hey there. Welcome back to the podcast. It's Marisha here today, and it'll just be me because I wanted to take a minute to just take in the awesomeness that was in the podcast last week. I am still here with my head spinning because they're so incredibly helpful. I just wanted to take a second to talk about how I'm implementing this and how I broke down all of the information that Jennifer Taps Richard shared about the complexity approach.

I just wanted to do a quick recap of why I'm even bothering to figure things out, and then I'm going to go through a case study with a student that I work with. I obviously changed the name, and I changed things up just a little bit, but I will talk through the process, how I figured things out, like the tools that I used, how I got everything organized, all of the nitty-gritty practical tips, just continuing on the amazing information that Jennifer Taps Richard provided for us last week. So, if you didn't join us last week, I highly recommend that you go back to that episode because it is filled with so much practical information.

I wanted to take a quote from Jennifer Taps Richard, and it's based off of a ton of different research. If you go back to the show notes from last week, you'll see all of the different citations, all the good nerdy stuff, but one quotation I pulled out or one quote I pulled out was that, "Children who are taught complex sound often learn treated and untreated sounds due to the relationships among sounds. So, for example, if a child is missing many sounds and is taught a three-element cluster like STR as in strong, it's predicted that he or she will also learn some missing two-element clusters: affricates, fricatives, and stops." So, how Jennifer says this is that three-element clusters imply two-element clusters. Two-element clusters imply affricates. Affricates imply fricatives. Fricatives imply stops. So, if we target a three-element cluster, it's going to change the student's sound system so that it makes those other targets that I just listed easier.

Conversely, if a child is missing many sounds and is taught a stop like K or G, which is something that we frequently do, it's predicted that K or G will change but not any of the other sounds. Teaching these complex sounds leads to rapid change and gains in intelligibility, so it is the epitome of working smarter as an SLP when working on articulation goals because by targeting one more challenging piece, that all trickles down and helps students acquire those different sounds. There's a strong evidence base. Like I already said, check out last week's show notes to review all of the amazingness and to learn more from Jennifer Taps Richard.

Now, let's dive into this case study. I started working with a preschooler. She was four years old at the time. Now, she's five. We were starting with her assessment. She scored on the first percentile on the Goldman-Fristoe Test of Articulation, so the GFTA-3. Based on that testing, she was able to produce bilabials like pa, ba, m, wa. I'm not used to articulating these, but I usually just write them out. Then, she's also got her alveolars like ta, da, m, and she's got the velar k and fricative F, so she's missing her voiced velar, so the g, and she's also missing the m, and as well as some fricatives because she only has F, so we're missing V, esh, and then voiced and voiceless T-H. She's missing liquids, L and R, as well as affricates, but she is stimulable for L, esh, j, and then voiced and voiceless T-H.

I found this out because Jennifer Taps Richard has this amazing tool on her site, which I will link to, but I got a really great stimulability probe, and I just love how it was set up. It was really easy to identify which targets I wanted to test for stimulability, and it gave me a really nice framework, clear data to go off of that, so I will link that stimulability probe in the show notes. Then, I also used her assessments just to get more data, get a better baseline. There were two other assessments that I used, one that looked more closely at clusters, and then another one that gave just a more overall assessment, but it just gave me some really nice data, and I was able to go off of that.

This was a re-evaluation, so I had been doing some work with the complexity approach already, and we're seeing some really nice gains, but she does produce all sounds for 78% of blends and clusters. However, she's still using some substitutions like she's substituting W for L and R, so some gliding going on there, but she is marking ... A year ago, she wasn't using any blends or clusters, but she is marking almost all of them now. There's just 22% that are being reduced, which is a huge jump from last year where she was marking zero at them. She was just producing all of the blends.

So, the things that I noticed most, which you might be able to guess already in terms of phonological processes or the patterns, she's gliding in 75% of opportunities. She's stopping in 33%, which has decreased significantly. She's using stridency deletion in 24% of her productions. She's vocalizing in 93%, struggling with that R. She's using palatal fronting in 58% because she's struggling with the sh sounds, esh. She's fronting 17%, which decreased significantly and then reducing clusters in 22%, which also decreased significantly, so she's made a lot of progress, but her speech is still really difficult to understand.

Her parents understand about 70%, but unfamiliar listeners still really struggle, and her intelligibility is as low as 50% with people who don't know her. I've observed it varying a little bit more between those, but I really liked getting this feedback from parents and just from observations. Then, they have another really nice assessment to give just to assess the impact, which I think is something that's really important to measure, so I will link to that impact assessment in the show notes as well.

I collected all of that information, again, like I said, using those two assessments, the stimulability probe and that impact assessment, and I was able to put together all of that data. I used Jennifer Taps Richard's phonological assessment and treatment target analysis form. I went through that process, so if you want to head to the show notes, that might be a good idea, so you can follow along with me. Then, you can check out the different assessments that I gave as well, but if you head to slpnow.com slash ... I'm trying to remember the number here. Slash seven, so head to slpnow.com/7 if you want to follow along with me. These resources are on Jennifer's site, SLPath. She just has so many amazing resources. I just wanted to link the ones that were most helpful in this particular process. All of the resources are absolutely amazing, but I just wanted to link the ones so that it's really easy to find those, but I definitely encourage you to check out SLPath.com and go through her entire resources section. There's a dropdown with so many different free resources, and it's absolutely amazing.

Hopefully, by now, you've had time to head to slpnow.com/7 and then scroll down and click on the phonological assessment and treatment targets analysis form, and let's just go through the different steps. I started with part one. I entered her phonetic inventory. I circled the phones that were in this child's phonetic inventory. Then, I wrote the ones that were out, and that's what I listed previously. Then, I did a phonemic inventory, and I'll also link to this assessment. That's one that I forgot to mention, and so I was able to circle the phonemes that she's discriminating against or that she's using in a meaningful way. If she says walk for ... like walk, the action of walking and walk for rock, she's not using R to differentiate the meaning. She's not contrasting that phoneme, but if she were to say lock and walk, like lock for the object and walk for the action, then L and W would be in. She has to use that at least twice, so she would get one point towards L and W being in, so if she says ... What's another good L and W word? So, yeah, I can't think of one now, but if she uses those contrastively in two different situations, then those phonemes are in.

The assessment walks through it pretty easily. It's pretty simple to figure out, and if that's something that you guys have questions about, maybe we can do an example together if you're joining us for the live Q and A, and of course, that'll be over, but if you go to the speech therapy course, you'll be able to access that Q and A, and we can walk through it a little bit more. It's just a little bit hard to do it as we're just with the audio here, but that's ... I know Jennifer explained this really well last week, so you can also head back to that episode.

We're through the first two steps. We've got her phonetic inventory. We have her phonemic inventory, and now, we're going through the word-initial cluster inventory. I'm circling all of the word-initial clusters that happened or that occurred at least twice in the sample. I went and circled the clusters that were in and that I heard at least twice. Then, I went through and gave the stimulability probe that I mentioned, and I listed the out phones that were stimulable and the ones that weren't stimulable. If you click open the stimulability probe, it's super easy to see how that works, but again, if you have questions, we might be able to do a video demo during the live Q and A. If that's possible, I'll totally share that with you too.

Then, we have to go through some different steps. We've mapped all of that information out, and then we need to go through some steps to actually select the targets. The first step is to determine if any three-element clusters are appropriate targets. S doesn't have to be in, but the second and third phonemes do have to be in. For example, if we're trying to target scratch, the student has to have K and R as an in phoneme. Oftentimes, that won't be the case, but luckily, we have some other ones. We have SPR, STR and SKR, and oftentimes, those aren't in because a lot of our students don't have R, but we also have SKW. In many of our students, it's possible that they would have K and W in their inventory, so that's when that could be in or SPL. If they have a P and an L, that can be in as well. She wasn't using the L consistently yet in splat, but her P and L were in and K and W were in as well, so we could potentially target SKW and SPL.

Okay, so now, we're going to jump to step two. We get to determine if two-element clusters are appropriate targets, so with these, they don't have to be in. We can choose something that's a little bit harder, and we obviously don't want to target something that's already in, so if they're already producing SM, we would cross that out and not work on that sound. Then, on the sheet, it lists the clusters according to their sonority difference, so we'd look at their minimum sonority difference. The sheet says to identify the minimum sonority difference produced by the child. Then, we would cross out all out clusters that have a sonority difference that's equal to or large than the minimal sonority difference of the child's in cluster. For example, if the child's smallest sonority difference cluster was KL, which has, and I'll refer to it as an SD now, sonority difference of five, we would cross out all of the clusters with a sonority difference of five or larger, so we wouldn't target any of the clusters under SD equals five and SD equals six.

Okay, so that was a mouthful. Note that the child does not need all clusters with a particular sonority difference, so one representative cluster is sufficient. If you end up crossing out all of the clusters, then you would just go to step three. With my student, she had SP and ST and SK in her inventory, and those have a sonority difference of minus two, and so if I'm following that rule for the clusters that have a sonority difference that is equal to or larger to the minimal SD, then I would end up crossing out the entire target pool because she has all of those.

Yeah, so then, I am jumping ahead to step three. If you didn't do that, the document is really clear on what to do through the next steps, and it just helps you break that down. Then, with step three, we're working on selecting singleton targets. They have a nice little table where we'd list out all of the out phones based on the phonetic inventory analysis that we did in the first step. We would cross out all of the stimulable sounds based on the stimulability testing. We cross out all of the early acquired sounds. This is the complete opposite of what we would normally do. We aren't targeting any of the early acquired sounds listed on the document. Just for the sake of your ears, I won't list them out.

Then, for all of the sounds that are remaining, we would circle the ones that lead to the greater system wide change based on the language laws. Within that packet, she lists all of the different implicational laws. Three-element clusters yield two-element clusters. If we have a choice, then we'll want to pick a three-element cluster over a two-element cluster. Clusters with a small sonority difference yield clusters with large sonority distances. Then, clusters yield singletons. Clusters yield affricates. Stridency contrast, so like TH versus S yield liquids, which oh my goodness, isn't that so exciting if we can get out of targeting R? Liquids yield nasals. Affricates yield fricatives. Fricatives yield stops, and so on and so forth. I won't go through the entire list, but it's such a helpful resource because it can really help us explain why we're choosing any target. We would just go through that list and identify the targets that would yield or lead the greatest system wide change based on the laws that she have listed. It's so simple, it makes so much sense, and it's just amazing.

Then, based on those sounds, we would pick the sounds that occur most frequently. She'd list the order of English consonant frequency. It's all right there. It's so incredibly easy, so we'd just go through that list and prioritize the target that's the most frequent. With my example, let's look at the list. So, she was stimulable for a number of different sounds, but she was not stimulable for G, then NG, ng, V, which is so strange to me. I'm like it's just so fascinating how the results come out, or I bet she was, but the stimulability probe didn't include those earlier sounds, but she was stimulable for L and sh and j and then the voiced and voiceless TH, but she was not stimulable for affricates, so that gives us some pretty good information. We might want to start with some affricates with her.

So, if we go back to that target and to the implicational laws, if we start with an affricate, that leads to fricatives. So, she is missing some fricatives, and if we start with affricate, that'll lead to help her ... There's a number of articles listed there. That'll help her develop her fricatives without us having to do anything. Fricatives also yield stops, so it'll help us fill in all of the sounds that she's missing just by targeting that affricate first.

I also decided to leave in the SPL because she is stimulable for L, but it's not completely in her system, and I knew that three-element clusters in play, two-element clusters and that by targeting that more complex cluster, we would be able to influence changes across the student's system, and that would help us work smarter. Then, I could target SHR and THR because those two-element clusters imply affricates, which kind of trickles down that whole hierarchy that I just listed, so that could be an option because SHR and THR are pretty hard, so this is something that I tried for a while. The student was totally open to doing SPL. I just had to slow things down a little bit and give her a little bit more support, but we were able to make it really fun.

We did lots of splat things, and she picked that up really quickly. SHR and THR were really tricky because it had lots of tricky sounds in there. Sometimes, we'll try it, and they'll just be a little bit harder to do, so it's still on our list, but we are having more success with the affricates. So, that's what we're currently focusing on, but we will ... because we just had to take a break from those two-element clusters with the R in them, but we will likely come back to those because I think those will give us a lot of bang for our buck and really influence a lot of change in the system.

So, that's how we went through the process of target selection and how that all worked. Then, in terms of therapy, which is my favorite part, if you are working on singletons, then that's really easy. It's easy to find decks of cards that include those singletons, and you can really kind of go for that and make some good progress there. I ended up creating my own cards because I've needed better sets of cards to work on the three-element clusters because it was really hard to find just activities with SPL, and I really wanted to be able to use the assessments that I used that I mentioned at the beginning of this talk for progress monitoring purposes, and I really wanted to be able to look at generalization, so I wanted to make sure that I wasn't using words that were in that assessment, again, so I would have a way to measure generalization. That made it a little bit tricky because there aren't a ton of child-friendly three-element clusters, like words that make sense, so that narrowed down the list a little bit, but the good news is that we don't need a ton of targets. We can just go through them, and make it work.

A lot of times when I'm working on the complexity approach, I'm working with younger students, so I wanted to have cards that were really big, so I made those cards. I printed them out. I have a laminated version that I keep in one of those, like it's an iris box. I'll add a picture to the show notes, but it just has different boxes inside of a larger box. I like it because they're all labeled by the targets that I'm currently working on in therapy, and I just have them organizes, so it's really easy for me to pull them out and grab them for therapy, and so that's been working really well. I like having the laminated ones, so we can play different games because I made two copies of each card, so we can play like Go Fish and Memory and all sorts of games just to switch things up.

If you don't have time to prep, that's totally cool because I also created black line version so that you can just print and go. When I do this, I print on card stock. If I can get my hands on some colored card stock, that's what I do. It's kind of nice because then, if we have different sets of cards in play over time, parents can then ... they'll know when a new sound is coming home. I really like being able to have students create their own cards, so I'll print out the black line version for them. They might color them a little bit or decorate them, and then we'll cut them out. It's a great activity again for mixed groups.

This is really similar to what Shannon was talking about during the Cycles podcast, which was two episodes ago. A lot of the ideas that she shared would totally apply to this, and we just happened to have a similar system, but we print out the black line versions. The kids get to prep them. It's a really great activity to kind of keep hands busy, keep students engaged, get them thinking about their targets. It's awesome if we can target the same target within a group, but if not, we can totally make it work using some of those strategies that we talked about. Yeah, we have them make their cards. They get ownership of the cards. We can put them in the classroom for practice. We can send them home for practice, and yeah, I just put them in a little envelope with a note on how things went and what we're doing and what they can do at home and just giving different ideas. We just run from there.

So, that's one thing that's really fun. I'll give some more specific ideas based on what we were doing for some of the more specific sounds. So, we started working on SPL, and like I said, I worked on this with a four-year-old, and she totally got it. She was generalizing within a couple weeks, which is absolutely amazing. You might be cringing, thinking about targeting SPL with such a little kiddo, but it's totally possible. I was skeptical at first, but it's just been so amazing to see her be able to do this and to see the growth that has come by using that approach and seeing the change in the system overall. It's magic essentially. That's what it feels like.

When we're doing SPL, we did a couple of different activities with that as well because splat is a really good word, so we did some different activities with Play-Doh, and I made balls, and she had to splat the ball. She had to say it correctly before she could splat, and we just went and got a bunch of repetitions doing that. Then, we also read a lot of Splat the Cat books, which also is a great SPL target. I really like making books. One example, we made a book called Splash because she loves water, so we made a book about all the different places that you can splash, so you splash ... We made it really simple. Every page just had a Google image. We made this in the session. We did it together. We worked on putting the book together, and then she got to take it home, but we did splash in the tub, splash in the pool.

We found like a play think, and we just found a bunch of different places you can splash. It was tied to something that she was really interested in, and we got a ton of repetitions. I was able to share the book with the parents when we did the Splat the Cat books. They're really easy to find in the library if you don't want to part with one of your own books, but that's another really great suggestion to share with parents. Even if the students aren't producing it on their own, that could be a good thing just to have the parents emphasize that and just giving them exposure to lots of SPL clusters. Yeah, so that's what we did with SPL.

I also found some other books from Jennifer Taps Richard's book list. She has a really great list of books that you can use, and then she also has a list of activities. I'll link to that, but you can ... There's a bunch of games that have to do with splish and splash and playing with frogs. There is a splat frog or different splat toys would be super fun. I have one of those splat balls, and we just throw it, and it splats on the floor. That one is really fun. Yeah, she also lists Splat the Cat doll, which is super cute, and you can talk about his different parts like Splat's tail, Splat's ears, and then who's that Splat?

One thing we did to Ashley, one of the last things we did with SPL was we used the Toca Tea Party app. It's one of the Toca Boca ones. She loves doing the birthday parties and the tea parties, and there's always three spots. If you have a Splat the Cat doll or any kind of stuffed animal, you can name him Splat or Splish or Splash or whatever they like, but we would go around, and this is more of a reinforcer. You don't get as many repetitions when you're doing something like this, but it's a good way to kind of work on things overall, but you can talk about I ... This could be good for multiple language goals, but like I would like juice. Splat likes juice. Splat likes cookies. Splat likes cake. Splat wants more. You just have a ton of conversation around Splat. Think it's Splat's birthday, so he gets all the attention. You just keep asking Splat questions and talking about Splat. That was one thing that we did with SPL, so actually, lots of different ideas.

There's another cute idea that Jennifer shared about ... because splendid is also an SPL word, so you could make a book about splendid things. She recommended butterflies, so you could do one splendid butterfly, two splendid butterflies, or you could do their favorite food like splendid cookies, splendid juice, splendid pizza. Then, another cute thing is like playing doctor and putting on splints. There's a splint on the arm, a splint on the leg, and all of those different things. You can even model them getting hurt and say, "Uh-oh, the doll went splat," or, "The toy dog went splat. He needs a splint." Those are just some different ideas on how I got started with SPL.

As you can see, I just kind of went with what the student was doing and kind of followed her lead, her interest, and all of those different pieces. When we did THR, we did a lot with throwing because that's really fun. Again, one of my go-to's is always using the cards, and then we get to earn one of these different types of activities. Some other really fun words are three. You could do this with different games, like games that have a lot of different pieces or if you're doing a paint dauber activity, you can be like, "How many do you want? One, two, or three?" Then, hopefully they would always pick three or if they're purposefully going to pick last because three is hard to say, then you can say, "Okay. One, two," and then have them say three.

Again, lots and lots of throwing, going through things is really fun. There's a bunch of Melissa and Doug toys that have thread where you can, for the girls especially, making necklaces and threading the beads, and they have to say thread to be able to add more pieces. That's really fun. You can make a book with three of their favorite things on each page, so you can have three books, three cats, three dogs and whatever their favorite things are. That's a really fun way to set things up.

Then, Jennifer has a lot more ideas for THR on her site. She has a lot of ideas for all of the other sounds as well, so it's a super amazing resource here. She even has some more ideas on just some other ideas for clusters in general or any articulation really. You can do scavenger hunts looking for sounds. You can do flashlight games. I used to do this when I was in the school too. We would dim the lights a little bit, and they got to use a flashlight to find their sounds. Sometimes, they might get a little bit off-task, but you can get them to be motivated to find all of their sounds first. Maybe they have to find it and bring it to you and say the word before they can start looking for the next one and have it be a little bit of a race potentially. It really depends on the group, but it can be really tricky or really fun with the right group.

You can also use the word during the activity, so you're getting close to whatever card they're getting close to, and if they're above the word level, they can say, "I found," and then whatever the word was. They can also use some other grammar targets too or create different sentences and talking about where they found it. The options or the opportunities really are endless, like when we were working on SHR, we use the same combination of activities. We start out by doing a lot of targeted practice with the sound, and then if we do good work, we get to jump into something different like creating a book, reading a book, playing a game.

When we were working on SHR, we had like a shrimp game that we did. Some other targets were shrink. We made stories about shrinking things. We made pictures shrink on the iPad because you can take a picture and then make it smaller. We talked about things that were shriveled. We made a book about shrieking. So, things that were scary, we could shriek. We would say like, "Oh, no. There's a spider. The girls shrieked," and something like that or, "That makes me shriek." We can also do just different shredded foods like shredded lettuce, shredded beef. It's a really good taco activity, so that one's really good. With all of the shredded foods and the shrimp, we can make a menu with different things, lots of good activities with that.

Then, the other target that I mentioned, this won't be comprehensive, it's just the ones that we went through, but we were also working on the different affricates. We talked about chips and chickens and cheese, and we did a lot of the same types of activities that we've talked about. If you're having a hard time coming up with good words ... Another one is chase. That's a really good one. Chair, chest. You can do a treasure game. Anyway, I just got a bunch more ideas for the CH thing that we're going to use in therapy, but yeah, so if you are having trouble coming up with good words, we can make up names for characters too. If you're not finding a lot of SHR words, then we can come up with a name too. Ideally, we would want to come up with words that they could use, but if you're just really struggling to keep students engaged, you can make up some names. Bonus points if it's an actual word, but if it's just a little bit easier for the student to understand or to come up with stories and different activities around that, then that's totally fine kind of like we did with Splat the Cat. We definitely talked about what splat actually means, but it's totally fine to switch things up and get creative.

We don't have to follow all of the rules. There some really great things that we can do to make this fun, so yeah, that's the process that I went through with the evaluation, the different measures I collected, the different therapy ideas, how I organize my cards, and then like I said, with the cards that go home, I just have envelopes. I bought some colored ones that were a little prettier and that would stick out. I just add in a quick note so they know what's happening and then just to know what to expect. Yeah, so that's what we do with that. Then, we can share those with the teacher as well if they're not being returned when we send them home. That's also why I like to have a version of my cards laminated and ready to go because if the student return their cards, then I at least have something to work with, and it's not end of the world.

Yeah, so I just keep all those cards in a box and then have students create their own cards, so that they get ownership of it. We make a lot of books using our words because we get a lot of great repetitions there. A lot of times, we're working on multiple goals. I really like the idea of having focused sessions. Depending on the student, like if it's a 30-minute session, we might only be able to do 15 minutes of the sound work and then 15 minutes of language, but I agree with ... Shannon was talking about how she does this with the cycles. If she has students working on both types of goals, she'll separate the sessions, so sometimes, it'll be articulation, and sometimes, it'll be language. She just plans that way. She just does one or the other. That's generally what I do, but if we're creating stories, we can throw in a little bit of language there too, but the focus will always be ... During our articulation sessions, the focus is on the articulation, but I really like that setup. That's been working well for me as well.

Yeah, so those are my ideas. Then, Jennifer talked about last time, we want to do progress monitoring. I just get so excited, and this is probably a weird thing that I do, but if I'm feeling unsure of what I'm doing or if I'm making enough of a difference, it's just really helpful to be able to give that assessment again. They always make progress. Even if it's just a small amount of progress, there's always some progress if we're giving the assessments every few months. It's just really cool to see that growth and to see the students making progress on sounds that we didn't even target. It's really amazing.

That's another really important piece, is just to continue collecting data and seeing whether you're just kind of keeping track of the progress and getting that proof that things are changing and that change is happening to that system, and really being able to look back and compare the different assessments is really powerful and exciting. So, definitely plan on doing that if you're implementing this approach or any other approach really. These are all just really great resources that you can use to implement this on your own. You can totally start it tomorrow. The target selection that we went through, it's not too crazy. There's a document that walks through exactly what you need to do, and it tells you why you're selecting what you're selecting. It might be something that is different than your district is used to or it's different from what your colleagues are used to, but there is so much evidence.

If in the IEP, you can just pull snippets of the rationale and some of the evidence for why you're doing what you're doing, there is a mega boatload of articles that you can cite. Jennifer makes it really easy to find them in her resource, so it's not something that'll take you a ton of time. I know it can be scary to do something different and to venture into the unknown, so to speak, but there is a lot of evidence around this approach. Like I said, Jennifer's resources make it really easy to find that evidence, and you can easily pull that to support what you're doing in your IEPs, and you'll sound ridiculously smart when it comes to your parents and, yeah, you'll have everything that you need to back up what you're doing. You'll be able to collect evidence along the way. That shows that it's happening, and that change is occurring and that you're influencing that system and helping them graduate from speech sooner, which is everyone's ultimate goal, right?

Yeah. That's what we've got for today. I would absolutely love to hear your ideas and whether you've implemented this, what your experience was with it, all of that good stuff. I'd love to hear if you have other activity ideas. Pass it all along. You can do that at slpnow.com/7. If you scroll all the way down, there's a place to leave comments, but then you can also refer to all of the links there to get an easy overview of the different materials that I mentioned as well as the link to the speechtherapypd.com course, and you can earn ASHA-CEUs for listening to this episode, which is really exciting. Like I said, you can find all of that information at slpnow.com/7. Thanks for joining us.