Scatterplots and Speech Therapy

I have been looking for an opportunity to try out this neat spread sheet for creating scatterplots as an alternative to the standard bar graph as a way of presenting the results of a treatment trial. This week the American Journal of Speech-Language Pathology posted our manuscript “A randomized trial of twelve-week interventions for the treatment of developmental phonological disorder in francophone children”. In the paper we compare outcomes (speech production accuracy and phonological awareness) for the four experimental groups in comparison to a no-treatment group using the standard bar graphs. Weissberger et al disparage this presentation as “visual tables” that mask distributional information. They provide a spreadsheet that allows the researcher to represent data so that the underlying individual scores can be seen. I am going to show some of the speech accuracy data from the new paper that Françoise and I have just published in this form.

In our trial we treated 65 four-year-old francophone children. Each child received the same treatment components: 6 one hour individual therapy sessions targeting speech accuracy, delivered once per week in the first six weeks; followed by 6 one hour group therapy sessions targeting phonological awareness, delivered once per week in the second six weeks; simultaneously in the second six weeks, parents received a parent education program. The nature of the individual therapy and parent education programs was varied however with children randomly assigned to four possible combinations of intervention as follows: Group 1 (Output-oriented Individual Intervention and Articulation Practice Home Program); Group 2 (Output-oriented Individual Intervention and Dialogic Reading Home Program); Group 3 (Input-oriented Individual Intervention and Articulation Practice Home Program); Group 4 (Input-oriented Individual Intervention and Dialogic Reading Home Program). The Output Oriented Individual Intervention and the Articulation Practice Home Program components focused on speech production practice so this was a theoretically consistent combination. The Input Oriented Individual Intervention and the Dialogic Reading Home Program included procedures for providing high quality inputs that required the child to listen carefully to those inputs with no explicit focus on speech accuracy; the child might be required to make nonverbal responses or might choose to make verbal responses but adult feedback would be focused on the child’s meaning rather than on speech accuracy directly. This combination was also theoretically consistent. The remaining two combinations mix and match these components in a way that was not theoretically consistent. All four interventions were effective relative to the no-treatment control but the theoretically consistent combinations were the most effective. The results are shown in bar graphs in Figures 2 and 3 of the paper.

Here I will represent the results for the two theoretically consistent conditions in comparison to the no-treatment control condition, using the Weissberger Paired Data Scatterplot Template to represent the pre- to post-treatment changes in Percent Consonants Correct (PCC) scores on our Test Francophone de Phonologie. The first chart shows the data for the Input Output Oriented/Articulation Practice intervention (Group 1) compared to the no-treatment group (Group 0). You might be surprised by how high the scores are for some children pre-treatment; this is normal for French because expectations for consonant accuracy are higher in French than in English because consonants are mastered at an earlier age even although syllable structure errors persist and may not be mastered until first or second grade. The important observations are that the difference scores for the no-treatment group are tightly clustered around 0 whereas the difference scores in the treated group are spread out with the average (median) amount of change being 7 points higher than 0.

Group 0 v 1

Next I show the same comparison for the Output Input Oriented/Dialogic Reading intervention (Group 4) in comparison to Group 0. In this case the median of the difference scores is 9, slightly higher than for Group 1, possibly because the pretreatment scores are lower for this group. In any case, it is clear that a treatment effect is observed for both combinations of interventions which is striking because in one group the children practiced speech with direct feedback from the SLP and parent about their speech accuracy whereas in the other group direct speech practice and feedback about speech accuracy was minimal!

Group 0 v 4.

Do these scatterplots provide any additional information relative to the traditional bar charts that are shown in the AJSLP paper? One thing that is clearer in this representation is that there are children in Group 1 and in Group 4 who did not respond to the treatment. Randomized control trials tell us about the effectiveness of interventions on average. They can help me as a researcher suggest general principles (such as, given a short treatment interval, a theoretically consistent intervention is probably better than an “eclectic” one). As a speech-language pathologist however you must make the best choice of treatment approach for each individual child that walks into your treatment room. Providing an evidence base to support those decisions requires access to large research grants for very very large multi-site trials. There is only so much we can learn from small trials like this.

I hope that you will check out the actual paper however which includes as supplemental information our complete procedure manual with a description of all target selection and treatment procedures, equally applicable to English and French.

Using Orthographic Representations in Speech and Language Therapy

Word learning, and in particular, productive word learning is associated with three important processes in the phonological domain: first, the child must encode the acoustic-phonetic form of the word in the language input; second the child must transform this representation into a lexical representation, generally considered to take on a more abstract phonological form; finally the child must retrieve the representation to reproduce it. The first process is reliant on speech processing abilities that have been shown to be impaired in many children with speech, language and reading deficits, as shown by for example by Ben Munson and colleages (@benjyraymunson) and Nina Kraus and colleages. Phonological encoding is enhanced by access to repeated high-quality but variable inputs as shown by Richtmeier et al for normally developing children and by Rice et al for children with SLI. The majority of children with SSD have difficulties with encoding: we have a paper in press with the American Journal of Speech-Language Pathology showing that speech accuracy in these children can be improved with an approach that focuses largely on the provision of intense high quality input – I will have more to say on this subject when it (finally) emerges in print.

The second process, forming a phonological representation and storing it in the lexicon, involves articulatory recoding which can be a serious problem for children with severe SSD, accounting for deficits in speech accuracy (especially in association with inconsistency), nonword repetition, word learning, productive vocabulary, word finding, rapid automatic naming, and other phonological processing skills. These children are often diagnosed with motor planning disorders but I have pointed out previously that the problem is actually at the level of phonological planning. I have further pointed out the very close relationship between speech planning and memory. Children who are having difficulty with phonological planning may not show the same benefit from a therapy approach that is focused on the provision of high quality inputs. Therefore a new paper on the use of orthographic inputs to teach new words caught my eye.  Ricketts et al taught children with SLI and ASD as well as younger and age-matched children with typical language to label nonsense objects with new names, using a computer program. For some words, the children were exposed only to the object–auditory word pairing; for others they saw the object, heard the word and saw a printed version (orthographic representation) as well. All children found it easier to learn the new words when they were exposed to the orthographic representation along with the auditory word.

This study reminded me of the research we are doing with children who are referred to our clinic with an apraxia diagnosis due to inconsistent speech errors. So far, 40% of those children have difficulty with phonological planning rather than motor planning as revealed by the syllable repetition test, as I have explained in a previous blog. We have been using a single subject randomization design to compare the relative efficacy of two treatment approaches with these children. The Phonological Memory & Planning (PMP) intervention pairs the phonemes in the target words with visual referents that include letters as shown here. Imitative models are avoided and the child is encouraged to create their own phonological plan and produce the word using the visual symbols when necessary. An alternative treatment, the Auditory-Motor Integration (AMI) Treatment is quite different with a heavy emphasis on prior auditory stimulation and self-judgments of the match between auditory inputs and outputs. A third condition is a usual care CONtrol condition focusing on high intensity practice. In all cases we teach nonsense words paired with real objects, with the words structured to target the children’s phonological needs in the segmental and prosodic domains.

The results are assessed by applying a resampling test to probe scores and then combining p-values across the children. These are the statistical results (F and t tests by resampling test) for the Same Day Probe Scores, with p-values combined across the 5 children who have proven to have phonological planning problems in concert with a severe inconsistent speech disorder:

TASC PMP results Aug 2015

The results in the third column show that all of the children obtained a significant treatment effect. The findings in the remaining columns pertain to planned comparisons with positive t values being in the expected direction. The combined p values indicate that all treatments are significantly different from each other and inspection of the mean scores across children show that the pattern of results is PMP > CON > AMI. The result is made more interesting by the fact that the pattern of results is the exact opposite for children with a motor planning disorder. Tanya Matthews and I will compare these two subgroups with data and video during our presentation at ASHA 2016 in Denver this coming fall.

Session Number: 1429
Session Title: Differential Diagnosis of Severe Phonological Disorder & Childhood Apraxia of Speech
Day: Friday, November 13, 2015
Time: 1:00 PM – 3:00 PM
Session Format: Seminar 2-hours

For now, the take away message is that learning new words involves (at least) three important processes: encoding the sound of the new word, memory processes for storing and retrieving the phonological representation and motor planning processes for planning and programming articulatory movements prior to production of the new word. There are published studies showing that intervention procedures targeting each of these processes help children with speech, language and reading difficulties. Increasing frequency of high quality input improves quality of the acoustic-phonetic representation. Pairing phonological segments with visual symbols helps with storage and retrieval of the phonological representation. High intensity speech practice with appropriate stimulation and feedback improves motor planning and motor programming. The trick is to figure out which children require which procedures at which time.

Thinking about ‘Dose’ and SLP Practice: Part III

Continuing my discussion about the concept of ‘dose’ as applied to speech therapy, I finally get to the heart of the matter which is the issue of the optimal ‘dose’ of speech therapy to achieve the desired outcome which in our context is generalization of a phonology goal to untreated words. In previous blogs I discussed the definition of ‘dose’ in terms of the number of effective teaching episodes and the need to identify the effective ingredients of your intervention beyond the therapeutic alliance. Here I will discuss ‘dose’ specifically, as in how many effective teaching episodes are enough to achieve a good outcome in phonology intervention?

Let’s begin by returning to the pharmacology context from which the concept of dosage is borrowed. How is the concept helpful to physicians? First, it is important to know the optimum dose (or dose range) for average patients so as to avoid harming the patient. If the prescribed dose is too low the patient may not improve and the continuance or worsening of symptoms and disease will be harmful for the patient. If the dose is too high the medication itself may be toxic and harm the patient directly. Second, the patient’s response to the medication is diagnostic. If the maximum safe dosage has been prescribed and the patient is not responding favorably the physician must seek the reason: Is the patient complying with the prescribed treatment regimen? Is the patient doing something else that interferes with the effectiveness of the medication? Is the health care system administering the dose as prescribed? Does this patient respond to medications in an individualized fashion, such that a switch to another medication is required? Is the diagnosis wrong such that an entirely different treatment is called for? I will describe the research on appropriate dose in the case of meaningful minimal pairs therapy (applied to preschool aged children with moderate or severe phonological disorders) and we can consider whether these questions are relevant in the speech therapy context.

The method of meaningful minimal pairs is a uniquely linguistic approach to therapy that has the goal of changing the child’s production of an entire sound class. The procedure has two key components: (1) teaching the child pairs of words that differ by a single phoneme; and (2) arranging the environment so that the child experiences a communication breakdown if both words in a  pair are produced as a homophone. (SLPs and researchers usually get the first part right but often forget the second!) The method is directed at the child’s phonological knowledge and therefore should not be applied until after phonetic knowledge of the contrasting phonemes in the perceptual and articulatory realms has been established.

There is a lot of research involving this method and at least two papers have carefully documented the dose that leads to generalization from trained to untrained words/targets. More than 50% generalization is the outcome of interest because we know from other studies that you can discontinue direct treatment on the target pattern at this point and the child will continue to make spontaneous gains. The two papers that I will discuss have the further benefit of allowing the reader to count the “dose” precisely as the number of practice trials. The papers also provide information about the number of sessions and the number of minimal pairs over which the practice trials were distributed.

Weiner (1981) demonstrated that the method was effective with two children, using a multiple baseline design and treating deletion of final consonants (DFC), stopping of fricatives (ST) and fronting (F). Four minimal pairs were taught per target pattern and use of the pattern was probed continuously for treatment words and on a session-by-session basis for generalization words. The results do not show that much difference across target patterns but the response across children was markedly different with one child showing much faster progress than the other for all targets. For example, Child A reduced DFC to below 50% in treated words after 120 practice trials and in generalization words after 300 trials. On the other hand, Child B required 200 and 480 trials respectively to reach the same milestones for DFC. Furthermore Child A was able to accomplish many more trials in a session (e.g., 400 practice trials over 5 sessions for child A or 80 trials/session vs. 570 practice trials over 13 sessions or 43 trials/ session for child B). Despite this large variance in rate of progress across children, the study suggests that an SLP should expect a good treatment response with this method after no more than 500 trials.

This finding was replicated in a larger sample (n = 19) by Elbert, Powell and Swartzlander (1991). In this study a behaviorist approach was taken to the treatment of the minimal pair words in contrast to Weiner’s procedure that emphasized the communication breakdown as an important part of the procedure. The children were taught one pair at a time in series and the study was structured to determine how many children would achieve generalization to untreated words ,at a level of at least 50%, after learning 3, 5 or 10 pairs of words. They found that 59% of the children generalized after learning 3 pairs which took an average of 487 practice trials (range 180 to 1041) administered over approximately 5 20-minute treatment sessions; 21% of children needed to learn 5 word pairs (1221 practice trials on average) and 14% needed to learn 10 words pairs (2029 practice trials on average) before generalization occurred. This left 7% of children who did not generalize at all.

How can we use these data about dosage in our treatment planning? There is a lot of useful information here. First, we know that it is possible to achieve 80-100 practice trials in 20 minutes. Therefore, if your treatment sessions are 20 minutes long you can target one phonological pattern and if they are 60 minutes long you can target 3. Second, they show us that children do not usually generalize in under 180 practice trials (and I would argue that the data indicate that it is number of practice trials rather than sessions that is important). What harm might arise if you provide a child with the government mandated 6 annual treatment sessions, targeting three patterns, but failing to achieve more than 100 practice trials for each target pattern across the 6 sessions? We can predict that the child will not start to generalize before the end of the block and therefore will not continue to make spontaneous gains after treatment stops. When the next block begins the child may be discouraged and less cooperative with the next SLP. The parent may become discouraged and seek out complementary or alternative interventions that are even more useless or harmful than speech therapy provided with insufficient intensity!

What if the child has achieved more than 500 practice trials and has not generalized? At this point you have more than enough reason to reassess your diagnosis and/or your approach. Child B in Weiner’s study for example did finally achieve many practice trials but did so slowly because he was unable to achieve the recommended intensity, producing much fewer than 80 practice trials per session. This child also failed to generalization after 500 trials for one of his targets. Perhaps this child was lacking in the necessary prerequisites such as stable perceptual and articulatory representations for the target phonemes. Or, perhaps the child viewed the communication breakdowns to be the SLP’s listening problem rather than his own speech problem and thus a disconnect at the level of the therapeutic alliance was hampering the child’s learning.  What about the children in Elbert et al who did not generalize at all? It was eventually revealed in the paper that these children presented with many “soft signs” indicative of both speech and oral motor apraxia. Therefore, continuing to almost 3000 practice trials for these children was most assuredly harmful, given that they were not benefiting from the approach and they were deprived of the opportunity to experience a treatment approach better suited to their needs.

I am hoping that this example in the specific context of minimal pairs intervention demonstrates that the concept of dosage can be very useful in speech therapy. We need much more research that establishes typical ranges of ‘dose’ for optimum outcomes for any given intervention procedure that we use. Then we need to track these dosages as we apply procedures in our interventions. It is important to remember that the dose is not the number of sessions or visits by the child or family to the SLP. Rather, the dose is number of learning opportunities experienced by the child. When the child is not learning and we know the child has experienced the optimum dose of practice trials, we can adjust our intervention procedures with greater confidence. We can also set evidence based goals for our clients and document objectively their progress with respect to these expectations. In addition to these benefits for individual clients, this kind of information will allow us to evaluate the efficacy of our service at the program level with an objectivity that is currently lacking. Imagine if a government or an insurance company suggested that they save money by reducing the dose of our medications below effective levels! We should not allow this solution to be proposed to reduce the cost of speech therapy services. The only way to protect ourselves and our clients is with more research and greater specificity about how our treatments work. We must know the right dosage.

Thinking About ‘Dose’ and SLP Practice: Part II

I have been talking about whether it is helpful to think about dose-response relationships as an important aspect of treatment efficacy. During a recent @wespeechies exchange, we discussed whether this “medical” concept should be applied to speech therapy. One objection raised was the idea that treatment efficacy is “all about relationships” and therefore the dosage of specific inputs was not all that relevant to outcomes. In psychotherapy, objections to manualized care protocols that prescribe specific procedures for defined cases are also based on the notion that treatment efficacy is determined not by the specific ingredients of the treatment program but rather by common factors, as I discussed in a previous blog. One of the important common factors is the therapeutic alliance. How important is the therapeutic alliance to treatment outcomes? And does attention to the therapeutic alliance preclude thinking carefully about which procedures to use in which amounts with a given case?

In psychotherapy the therapeutic alliance is defined “as agreement on the goals and tasks of therapy in the context of a positive affective bond between patient and therapist.” Even when working with children, this can be an important aspect of the treatment program. For example, McCormack, McLeod, McAllister and Harrison describe children’s experience of speech impairment in a paper entitled “My Speech Problem, Your Listening Problem, My Frustration…”. This qualitative study illuminates multiple facets of an SSD and further shows that the child’s perspective and the adult’s perspective on the problem and the solution are often not aligned. Shifting the child’s attention to the role of his or her speech problem in communication breakdowns will require a genuine, caring, sensitive and trusting relationship between SLP and child. Establishing common goals and motivating the child to try new tasks to achieve those goals will also be highly dependent upon the therapeutic alliance between child and therapist.

To understand how the therapeutic alliance impacts on therapy outcomes we must return to the psychotherapy literature because I am aware of no scientific studies in the speech therapy arena that have addressed this issue directly. In mental health services, the strength of the therapeutic alliance is measured by asking clients questions about their relationship with their therapist in three domains, specifically goals (e.g., We agree on what is important for me to work on.), tasks (e.g., I agree the way we are working on my problem is correct), and bond (e.g., I believe my therapist likes me).  Very large sample studies have shown that the relationship between therapist and client accounts for about 20% of variance in outcomes. However, the relationship between outcomes and the therapeutic relationship is reciprocal: if the client gets better, they have more trust in the therapist’s guidance regarding goals and tasks. Therefore, the therapeutic relationship is theoretically independent of the techniques and procedures that the therapist uses, but in practice these variables may be related.

To put this in the speech therapy context again, Francoise Brosseau-Lapré and I are in the process of publishing the results of our RCT, Essai Clinique sur les Interventions Phonologique. We found that an input oriented approach (procedures focused on perceptual and phonological knowledge with very little articulatory practice) was as effective as an output oriented approach (all procedures focused on articulation practice) for improving children’s articulation accuracy.  Therefore, when working with a very shy child who does not like to imitate or indeed, talk at all, during speech therapy, you and the parent and the child might all agree that the input oriented approach is the ideal way to work on the child’s speech problem. Initially the therapeutic alliance might be high but what if the implementation of the approach is not competent? We find for example that it is actually quite difficult to teach students to implement the procedures (focused stimulation, error detection tasks and meaningful minimal pairs procedures) correctly. Furthermore we found that when procedures are mixed and matched in a way that is not theoretically coherent (for example, input oriented procedures in the clinic but an output oriented home practice program), we observed very poor outcomes. It is probable that in cases of poor implementation, outcomes and the therapeutic alliance will both suffer. At the very least, as I have found previously, parents are able to identify poor speech outcomes in their children even as they report good relationships with their child’s SLP.

This discussion reminds me of a very interesting article about teacher effectiveness that was circulated on twitter by @KevinWheldell. Gregory Yates makes the distinction between good teachers and effective teachers. Similarly SLPs may be readily judged to be good on the basis of personal and moral qualities such as warmth, caring, friendliness and conscientiousness, all of which contribute to positive relationships with clients, coworkers and their institution. Effectiveness requires the skillful application of specific techniques and procedures in relation to client needs however and can only be measured in reference to client outcomes. More about this in the next blogpost in this series.

Thinking About “Dose” and SLP Practice: Part I

A debate arose on @wespeechies about whether cumulative intervention intensity concepts, especially dose, are “too medical model” for speech-language pathology practice. Several objections were raised and I am going to address them singly in independent posts. One point of discomfort was the sense that talking about dose implies that the SLP does something to a passive patient in a context that is incompatible with both the biopsychosocial model of health care provision and collaborative models of service provision in the schools. These objections have been raised before in print, for example by Alan Kahmi although in his commentary he does not actually discuss dose but rather scheduling of treatment sessions which is a different concept altogether.

I want to defend the importance of the dose concept recognizing that one can of course rename everything if “medacalese” offends you, my dear readers. Our patients have become clients and the dose may be referred to as a “teaching episode” or a “learning opportunity”. Nonetheless, as Lise Baker said, thinking about dose forces us to identify the essential active ingredients in our interventions and structure our efforts to ensure that the child is receiving those active ingredients in the right amounts at the right time. The analogy to medical (pharmaceutical) treatments does not preclude the application of the biopsychosocial model at all and may in fact enhance our effectiveness in that regard.

Consider as an example, a kindergarten aged child with unintelligible speech who finds himself in daily conflicts, sometimes physical, during craft time that is deliberately structured by the teacher to encourage sharing and cooperation by the children. The value of the biopsychosocial model is that it forces us to think beyond the impairment level. What factors contribute to the difficulties that the child is having during craft time? Besides his inability to express himself clearly he may be lacking in social strategies for solving the conflicts that arise, having so little experience with successful conflict resolution. He is also likely to be very anxious in this situation and lacking in self-esteem generally. The SLP may elect to bring in another professional to help the child with the anxiety and self-esteem issues (personal factors), council the teacher to change the environment during craft time to reduce the opportunities for conflict (contextual factors) and to engage a teaching assistant (TA) support the child during craft time, and to personally attend craft time weekly to encourage the child to use words such as “black, blue, glue, please” more accurately in the classroom. Now we have a collaborative model that appears to be compatible with the biopsychosocial framework but will these interventions achieve the goal of improving the child’s competence to verbally resolve social conflicts? Unless the TA, the teacher and the SLP all have a clear idea about the what the active ingredients of the treatment are, it is quite possible that the treatment will not be effective, regardless of the number of times that the TA is available to mediate the child’s experience during craft time.

An excellent document entitled Making Best Use of Teaching Assistants points out that TAs tend to prioritize task completion over other goals. Therefore it is likely that the TA will sort out the materials for the child during craft time and help him to assemble them thus avoiding any possibility for conflict. The teacher will be happier and the parents will receive fewer distressing reports but the child’s self-esteem and sense of isolation will suffer further and bullying by other children may actually increase. What would the active ingredient of the intervention program be and how would we count the dose? The active ingredient is not support by the TA and the dose cannot be counted as successfully completed crafts. Rather, the goal is successfully negotiated “sharing” of craft materials and therefore the dose should be counted as the number of opportunities to use new strategies to obtain materials from another child. The child will need a means of communicating his needs clearly, strategies for resolving conflict that do not involve grabbing or hitting, reinforcement for engaging appropriately with the other children, and supports from the entire class that do not isolate or stigmatize him further. My readers are more qualified than I to work out the details of the intervention but it should be clear that intervention intensity is not the number or duration of supports provided; rather it is necessary to document the number of times the child practices specific behaviors that will eventually result in satisfactory levels of independent participation. It is also necessary to teach the TA to specifically ensure that these learning opportunities occur and to support the child’s achievement of the necessary skills in an appropriate fashion. Thinking clearly about the active ingredients of the intervention facilitates the success of the consultation process.

My point is that the intervention is not the SLP’s conversations with other staff or even the child. Currently, IEPs are often written in terms of applying units such as the TA or SLP to children (or teachers) for specified periods of time rather than a specific description of what the child needs to do in order to achieve successful functioning in the school environment. We spend a lot of time determining what the child (or school) is entitled to and not what the child needs. The intervention for our imaginary child is the number of times the he uses the relevant sentences in appropriate communicative contexts, experiences error and self-corrects, initiates interactions, and verbally resolves conflicts. From this perspective, thinking about doses of intervention units is the antithesis of passive actions on the child – it is all about the child’s opportunities to practice and learn the skills necessary to participate in his every day environment. I will come back to the issue of how best to promote learning during these teaching episodes in another post.

Support for Speech Perception Interventions in Speech Therapy

I am writing a third blog on this strange experimental protocol in which the talker produces a syllable repeatedly and the talker’s speech output is altered in a systematic fashion so that the talker hears him or herself say something that does not correspond to their own articulatory gestures. I am fascinated by these experiments because they are a window onto feedback control which is essential for a successful speech therapy outcome. Initially in traditional speech therapy the SLP is providing a lot of external feedback about the child’s articulatory gestures (knowledge of performance feedback) and the correctness of the child’s speech output (knowledge of results feedback). But given that the SLP cannot follow the child around outside the clinic room, eventually the child must learn to use self-generated feedback for speech motor learning to occur. Can children use auditory feedback to change their own speech?

In a previous blog, On Birds and Speech Therapy, I discussed interesting work from Queen’s University  suggesting that toddlers do not use feedback control like adults do during speech motor learning.  These researchers found that adults will compensate for perturbations of their own speech by adjusting their articulation to get the desired auditory feedback. In contrast, very young children do not compensate in this way. I suggested that this may be because toddlers do not perceive speech with the same degree of precision as adults. This hypothesis was supported by another study in which speakers of French and English did not show the same compensation effect to a perturbation that made their vowels sound like a French vowel. The English talkers did not respond to a perturbation to which they were not perceptually sensitive (see Feedback Control and Speech Therapy Revisited).

Recently, I was delighted to find another study involving children provides even stronger confirmation that perceptual representations play a key role in the child’s ability to use feedback for speech motor learning. Shiller and Rochon (2014)  randomly assigned 5- to 7-year-old children with typical speech to two training conditions: the control group received speech perception training for the /b/-/d/ contrast; the experimental group received speech perception training for the /ɛ/-/æ/ contrast. Prior to and subsequent to this training both groups experienced the perturbation experiment: both groups repeated said “Beb” while their own speech was altered to sound more like “Bab”. Prior to perceptual training, both groups showed a small compensation for this perturbation in the feedback of their own speech. After speech perception training the experimental group showed twice as much compensation as before whereas the control group showed no change in the amount of compensation.  The results show that children can indeed use feedback for speech motor adaptation; furthermore, this ability improves as perceptual boundaries between phoneme categories become better defined —with age or with training.

The conclusions of the study are very gratifying. Citing my own work on the importance of speech perception training as a strategy to facilitate speech production learning by children with speech sound disorders, the authors conclude:

“The results of the present study complement this work nicely, demonstrating that improvements in children’s auditory perceptual abilities do not simply improve motor performance, but also alter the capacity for auditory-feedback based speech motor learning—a process that is central to the clinical treatment of speech production disorders.” (p. 1314)

No surprise that I like this study a lot!

Top down or bottom up target selection with toddlers?

A new paper on the consonant repertoires of toddlers confirms the close relationship between early speech and language development: Sotto, C. D., Redle, E., Bandaranayake, D., Neils-Strunjas, J., & Creaghead, N. A. (2014). Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders, 49, 1-12. Specifically these authors examined the relationship between consonant repertoires at 18 months and performance on the MacArthur-Bates Communicative Development Inventories: Words & Sentences (MBCDI: WS) at 18, 24 and 30 months. Although inventory size was not significantly correlated with vocabulary size and use of grammatical markers at 24 and 30 months, the presence of fricatives in the inventory at 18 months was associated with higher mean scores on the 24 and 30 month language tests in comparison to language test performance for toddlers who did not produce fricatives at the earlier age. The discussion nicely covers the many intervening variables that might account for this relationship. (Clarification added in response to reader questions: the children in the study were normally developing).

I was pleasantly surprised to find that all the raw data is presented in the paper so that the consonant repertoires for each of the 37 toddlers at 18 months could be examined directly. This allowed me to check whether these repertoires conformed to the expectations of the implicational hierarchy as described by Dinnsen et al in an older paper (the hierarchy is derived from earlier work by Jacobson I believe): Dinnsen, D. A., Chin, S. B., Elbert, M., & Powell, T. W. (1990). Some constraints on functionally disordered phonologies: Phonetic inventories and phonotactics. Journal of Speech and Hearing Research, 33(1), 28-37. The hierarchy takes a structural approach assigning the child to different levels on the basis of the phonetic feature “contrasts” that are present in the phonetic repertoire as shown in the table below, keeping in mind that the child does not have to use the phones contrastively; there need be only representatives of the opposing feature classes present in the repertoire. The hierarchy is said to be implicational because if a child produces the feature attributed to one level of the hierarchy it is theoretically impossible for the feature contrasts at any lower level to not be present. Therefore the presence of voiced and voiceless phones (Level B) implies the presence of labial and coronal place as well as obstruent, glide and nasal manner features (Level A). I examined each of the repertoires in the Sotto et al paper and determined the highest level represented for each child and noted any violations of the expectation of the implicational hierarchy. The results are shown in the table below.

Implicational Hierarchy

I find it to be very interesting that one third of the toddlers in this study do not meet the assumptions of the implicational hierarchy. For example if we consider the children who produced a Level C phonetic repertoire, in other words, the 24 22/37 toddlers who produced fricatives and or affricates, we find that 6 of these did so despite lacking glides (4 children) or nasals (2 children) in the inventory. In Dinnsen et al, 40 preschoolers with speech sound disorders were described and the 1 child that failed to meet the implicational assumptions of the hierarchy was assumed to have a “deviant” phonology. However, (as I have suggested for Dodd’s so-called “deviant” categories as well), deviant behaviors have a funny way of showing up in very early language development.

What are the implications of these data and my re-analysis for speech and language therapy? To be honest I am not sure. One thing that I am quite sure of is that the findings do not support the “complexity” approach to target selection – the idea proposed in Dinnsen et al is that treating Level E phone contrasts will cause the lower level contrasts to appear is if by magic because of the internally specified implicational relationship between the contrasts as if they were all linked together like Christmas tree lights. I have shown that this approach does not in fact work (see Rvachew & Bernhard, 2010). This is not to say that one should take the opposite approach by structuring your treatment approach to introduce features in the opposite direction, working your way up the levels from A to B (and contrary to some myths in circulation I have never recommended this). I think that the important thing to keep in mind is that children at this stage of phonological development are developing their phonology and their lexicons so as to enhance lexical contrast  rather than phonological contrasts and therefore one needs to take a language based core vocabulary approach in any case.

The problem that I have is that I have never been comfortable with the idea of picking any words that are “functional” and teaching them even if the child hasn’t a hope of approximating the word (those poor children with speech problems named Clarence are fortunately few!). I prefer to take a systematic approach introducing new phones to maximize the probability of success and the implicational hierarchy shown above is maybe not a bad start if you have a child with no consonant repertoire at all (this does happen sometimes). Another possibility is Daniel Ling’s approach (teach all manner classes at once at one place of articulation and then move to the next place, leaving voice contrasts to the end). On the other hand, given the broad variety of strategies evidenced by the children in the Sotto et al study maybe the notion of taking any words that are important in the child’s environment is both ecologically and theoretically defensible. I will come back with some summaries of Dodd’s work on the core vocabulary approach another time but I think that this is a problem worthy of more rigorous and empirical study with larger samples!

Introduction to the Wait Times Benchmarks Project

1. Introduction to the Wait Times Benchmarks Project

Access to speech, language, swallowing and hearing services is a critical concern across Canada. One indicator of the urgency of the problem is lengthy waits for service after a need has been  identified in one of these areas. Of course this issue is not restricted to communication health as attested by the Wait Times Alliance.  The Alliance was formed by doctors in 2004 to provide solutions to the problem of long waits for medical care in Canada’s publicly funded health service. Sadly, long waits for speech, language and hearing services are not specific to Canada, as reports in Australia and the United Kingdom have highlighted similar concerns to those raised by families by of children and adults who need services in Canada.

Although access to service is a multifaceted problem there are many reasons that wait times in particular invite a common focus by clients, service providers, funders, and politicians as the essential issue to target for improvement. The recent report by the Wait Times Alliance (Time to Close the Gap, Wait Times Alliance, 2014) lists several:

  1. it is established that many other countries with universal health care have succeeded in providing timely access to service and therefore we should not tolerate long waits when they are clearly not necessary;
  2. it can be shown that long waits for necessary services impose a significant burden on patients who are waiting as well as on society in general; and
  3. long waits for service impair health system performance such that improvements to wait times should result in gains for the system as a whole.

These considerations are as crucial for speech, language, swallowing and hearing health as for any other sector of the health care system. One step toward improvements in wait times is the development of benchmarks that indicate the maximum time that an individual should wait for service after taking into account the likelihood of significant clinical consequences should the wait  be longer. The Pan Canadian Alliance of Speech-Language Pathology and Audiology Organizations has committed to establishing reasonable wait times benchmarks as the first step toward reducing wait times for services.  A series of ad hoc committees recommended benchmark wait times for different diagnostic categories (see the Speech-Language and Audiology Canada (SAC) website). These wait times are being reviewed and reformatted according to a standard template and released publicly to the clinical community one at a time along with a published paper that provides the scientific foundation for each benchmark. The Benchmark Wait Times for Pediatric Speech Sound Disorders was released at the SAC Conference in May 2014 _and the associated Report was published in CJSLPA in Spring 2014 .The revised Benchmarks for Pediatric Language Disorders will be released soon and the Benchmarks for Fluency disorders are in progress.

In addition to releasing the benchmarks and the associated scientific reports, SAC will be providing additional information about benchmarks and their use in this blog which will be cross-posted to the SAC website and We will be inviting feedback and participation from the SAC membership or other interested commenters with each release. The schedule of upcoming blogs is shown below. We hope that you will follow the blog and consider commenting or contributing to this conversation.

Upcoming Posts

2. What is a Benchmark?

3. Approaches to Developing Wait Times Benchmarks

4. Evidence Based but not Evidence Bound

5. Use of Benchmarks by Clinicians and Policy Makers

6. Potential Advantages of Having Wait Times Benchmarks

7. Potential Disadvantages of Having Wait Times Benchmarks

8. Strategies for Achieving Wait Times Benchmarks

9. Factors that Impact on the Achievement of Wait Times Benchmarks

10. Role of the Pan Canadian Alliance and SAC in the Achievement of Wait Times Benchmarks

Wait Times Benchmarks for Speech-Language and Hearing Services

An important statement in the Universal Declaration of Communication Rights (International Communication Project 2014) is “We believe that people with communication disabilities should have access to the support they need to realize their full potential”. Even in those countries where speech-language pathology and audiology services are well established, long waits for service can be a significant barrier to communication for many children and adults. Twitter is a powerful tool for sharing knowledge and strategies for problem solving. This week on @WeSpeechies (see WESPEECHIES) we can share international perspectives on perceived appropriate wait times, actual wait times and strategies for reducing wait times for services around the world. When sharing information about this topic please identify yourself and provide general information about the nature of your clients and service sector while respecting privacy and confidentiality of specific individuals and organizations.

Q1. Approximately how long do your clients with speech-language needs wait for services? #WeSpeechies
Q2. Do you work with established expectations for wait times? How were the wait time benchmarks determined? #WeSpeechies
Q3. Do you think that clients with speech-language needs should have a guaranteed wait time for service? #WeSpeechies
Q4. What kind of criteria for deciding who gets served first are most fair? #WeSpeechies

Dose Frequency for Effective Speech Therapy

I am writing to address a specific question that has come up: in order to be effective when treating an “articulation disorder” how many trials should the SLP elicit from the client per treatment session? This is an important question and it is surprising that so little research attention has been directed at uncovering the answer. This is a question about what Warren, Fey and Yoder (2007) refer to as “dose: number of properly implemented teaching episodes per session”. We could be talking about the number of presentations of a model or perceptual responses by the child when conducting an “input oriented intervention” but in this blog I will restrict my comments to those interventions that are focused on obtaining speech responses from the child and therefore the teaching episode involves practicing a speech behavior such as a sound, syllable, word or phrase and each elicitation is counted as a single dose. In speech therapy the question of optimum dose frequency (how many trials per session of a given length) comes up most often in the context of Childhood Apraxia of Speech (CAS) where it is generally believed that practice intensity is particularly important. Recently, Murray, McCabe & Ballard (2014) reported that studies on approaches for CAS typically involved 60 to 120 trials per session whereas studies on approaches for phonological disorders typically involved 10 to 30 trials per session. The closest I have seen to an experimental investigation of dose frequency is the single subject experiments conducted by Edeal and Gildersleeve-Neuman (2011) in which low intensity (30 to 40 trials/session) versus high intensity treatment (100+ trials/session) was compared within two children with CAS. They concluded that “Both children showed improvement on all targets; however, the targets with the higher production frequency treatment were acquired faster, evidenced by better in-session performance and greater generalization to untrained probes.”

I don’t see any reason why a higher intensity intervention would not also be a “good thing” when treating children with a phonological disorder and indeed this is what Williams (2012) concluded when she reviewed data from her lab. After a quantitative summary of treatment outcomes for 22 children who received her multiple oppositions intervention she recommended a minimum dose of 50 trials over 30 sessions with anything less being ineffective and higher doses (70 trials or more) being necessary for the most severely impaired children. In this case the children received 30 minute sessions twice per week.

Recently we have been conducting single subject experiments with children who have CAS and although treatment intensity is not the primary focus of attention in these studies my doctoral student, Tanya Matthews, and I have been looking at the relationship between dose frequency and outcomes. In the figures shown below the children’s “next day probe scores” (an indicator of maintenance of learning over a short-term period, expressed as proportion correct) are shown as a function of the number of trials completed (top chart) as well as the number of correct trials in each session (bottom chart). There is not much variability in the number of trials per session because we put a lot of pressure on the student SLPs to keep this number high. However the number of correct trials varies quite a bit depending upon the severity of the child’s speech delay and whether it is early or late in the child’s treatment program. The lower chart shows that next day probe scores are better if the number of correct trials in each 20 minute practice session is above 60. The number of correct trials never goes above 80 because we are working to keep the child “at challenge point” so if the child begins to produce more than 80% correct trials we make the task more difficult. However, if the child is producing many errors it does not really help to keep the response rate high either because the child is just practicing the wrong response anyway.

So to sum up, notwithstanding the rather poor quality and quantity of the data, my impression is that dose counts: regardless of whether the child has a motor speech disorder or a phonological disorder it is important to achieve as many practice trials as you can in a treatment session but it is also a good idea to ensure that the child is achieving accuracy at the highest possible level of complexity and variability during practice as well.

Number of trials by probe score


Get every new post delivered to your Inbox.

Join 148 other followers