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.


What’s in a Name? Does SSD Smell Sweeter than DPD?

Francoise and I are engaged in the writing of two books currently. I am taking the lead on the revision (mostly cosmetic) of Developmental Phonological Disorders: Foundations of Clinical Practice (the “big book” that we call DPD for short) while Francoise is taking the lead on the writing of a new undergraduate text that will prepare readers to tackle the “big book” at the graduate level or to use the DPD text as a handbook in clinical practice. We still haven’t figured out what to call the second book! Introduction to Speech Sound Disorders? Introduction to Developmental Phonological Disorders? Introduction to Articulation and Phonological Disorders? Some combination of the above? We notice that many of the undergraduate text books now have very long titles because the authors keep adding terms as they become “fashionable”. I have just arrived (in my revising) at the introduction to Part II of the DPD text in which we explain our preference for the term Developmental Phonological Disorders. Even though this text is copyrighted to Plural Publishing Ltd. I am going to reproduce it here in case any of my readers would like to weigh in. We have been told that instructors cannot use the big DPD text because it is not titled with the preferred American term ‘speech sound disorders’ and I have met readers who were very surprised to learn that we covered articulation disorders and motor speech disorders in the book, as if the children with these problems did not have developing phonological systems! So much complexity here – I look forward to your thoughts in the comments or on twitter or by email. Here is the text from our book, Part II Introduction:

Developmental Phonological Disorders as the Diagnostic Term

As we discuss the application of the ICF framework in the context of DPD, we must be begin by unpacking the term “developmental phonological disorder” and justifying this choice of terminology to describe this health condition. Since the dawn of our profession, many terms have been used to describe children who have unintelligible or inaccurate speech, with all of the terms reflecting the tongue-in-cheek perspective of Compton (1970) who compared the diagnostic role of the SLP to that of a “TV repairman”! The diagnostic term that is applied specifies the “part” that is presumed to need fixing, either “articulation,” “phonology,” or “speech,” with these terms all in current use although, historically, earlier usages focused on articulation problems and current preference in North America is to refer to “speech” as a cover term that is presumed to include both the articulatory and phonological aspects of the child’s difficulty. We feel, however, that “speech” is too broad a term because it is often used as a cover term for difficulties with articulation, stuttering and voice in epidemiological studies, as seen in Chapter 7. Furthermore, in the developmental context there is no possibility of separating articulation from other aspects of phonological knowledge. Children who appear to have a motor speech problem called childhood apraxia of speech have significant difficulties with various aspects of phonological processing (see Chapter 7 for further discussion of this point). Returning to the topic of cleft lip and/or palate, this structural disorder that might appear at first glance to cause a purely articulatory problem, actually results in speech patterns that are best described and treated with phonological approaches (Howard, 1993; Pamplona, Ysunza, & Espinoza, 1999). Therefore, it is our preference to identify the central issue as being in the child’s developing phonological system, stressing as we do throughout this book, that phonology comprises interlocking components at multiple levels of representation.

The diagnostic term also requires one or more modifiers that indicate a specific type of phonological problem. We use the term “developmental” to simply denote that we are referring to children whose phonological systems are still developing. Furthermore, as shown in Chapter 7, the most likely causal factors in the majority of cases are interacting genetic and environmental variables that impact primary neurodevelopmental processes. The modifier “functional” was used for many decades, sometimes replaced with the phrase “of unknown origin,” to differentiate problems that had a known biological cause from those that did not and were therefore presumed to reflect an unexplained failure to learn the required articulatory gestures or an unexplained delay in the suppression of phonological processes. We reject these terms on the grounds that distinguishing between biological causes that are currently known and those yet to be discovered is nonsensical and that, furthermore, we cannot force a pure demarcation between biological and environmental causes. For example, so-called functional speech problems are indeed associated with sociodemographic disadvantages (for discussion, see Shriberg, Tomblin, & McSweeny, 1999) but these sociodemographic conditions are themselves associated with biological causal-correlates such as increased risk of otitis media, fetal and child exposure to parental smoking, and low birth weight. Furthermore, environmental variables and biological maturation are reciprocally related as discussed in Part I: maturation of brain function in areas associated with language and reading development is driven in part by exposure to high quality language input. In another example, Noble, Wolmetz, Ochs, Farah, and McCandliss (2006) demonstrated that socioeconomic status significantly moderates the relationship between brain function and phonological processing even when phonological abilities are controlled across advantaged and disadvantaged groups. The nature of the relationship is such that high quality inputs for children in advantaged homes buffers them from the ill effects of poor phonological processing abilities, allowing them to achieve higher reading levels and higher activations in areas of the brain important to reading than would be predicted on the basis of their phonological processing abilities alone. Disadvantaged children show a correspondence between brain activation and reading ability that is linearly predicted by their phonological processing skills, however. These kinds of studies support a dynamic systems approach to phonological disorders and highlight the joint causal influences of intrinsic and extrinsic factors on children’s linguistic functioning (issues that are revisited in Chapter 7 when we discuss approaches to the subtyping of phonological disorders). For these reasons we prefer the modifier “developmental” rather than “functional” or any other term that strictly demarcates biological and nonbiological causes of phonological difficulties.

Finally, there continues to be some controversy about whether the problem should be referred to as a “disorder” or a “delay.” In fact, as we discuss further in Chapter 7, some classification systems explicitly differentiate between children whose speech appears to be delayed by virtue of having characteristics similar to younger normally developing children and those whose speech has characteristics deemed to be atypical. We argue as we move through Part II that the diagnostic and prognostic implications of this distinction are uncertain and that the delay-disorder classification exists more on a continuum of severity than a sharply delineated categorical distinction. With respect to those children who are deemed to have a “disorder” on the basis of “atypical” speech errors or learning processes, it is our impression that the child’s behaviors are only “atypical” in the context of the child’s age or overall profile. For example, inconsistent word productions are often considered to be atypical and yet we showed in Chapter 4 that variable word productions are fully expected in the earliest stages of word learning. Therefore atypical behaviors reflect heterochronicity in developmental trajectories across cognitive-linguistic domains within a child rather than fundamentally different learning processes across children. As to those children who appear to have typical but delayed patterns of speech error, we take the position that some children’s delay is severe enough that it places them at risk for current or future activity limitations and participation restrictions. Consistent with the position of the ICF-CY (McLeod & Threats, 2008), the problem in this case deserves the appellation “disorder”. Furthermore, to be consistent with the dictionary definition of the word “disorder”, this appellation justifies an intervention to change the child’s rate or course of development so as to synchronize function among different developmental domains or to align function with expectations for activities and participation.

Ultimately, this brings us to the diagnostic term developmental phonological disorder (DPD), which corresponds to one of the superordinate categories in the Speech Disorders Classification System as originally formulated (Shriberg, Austin, Lewis, McSweeny, & Wilson, 1997). DPD can be contrasted with normal (or normalized) speech acquisition, differentiating those children whose speech development is progressing as expected from those children who, at ages younger than 9 years, are producing more speech errors than would be expected for their age. Nondevelopmental phonological disorders denotes those cases where the speech difficulty has its onset after 9 years of age. Speech differences arise from cultural and linguistic diversity and are not considered to be a speech impairment (although a speech difference may overlap with a coexisting health problem and may have functional consequences for an individual’s participation in some environments). The outcome of the initial assessment of a child who is referred due to concerns regarding speech accuracy or intelligibility should be a diagnosis with respect to one of these 4 major categories. Subsequent to an initial diagnosis of DPD the SLP may also diagnose a specific subtype of DPD, as discussed in Chapter 7.

We point out here that throughout Parts II and III we remain focused on those cases where the child’s primary difficulty is with speech (and/or language and/or reading). We do not specifically cover secondary phonological disorders in which the child’s speech delay is directly associated with impairments of sensory systems, cognitive deficits, craniofacial anomalies or other developmental disorders. The assessment and treatment procedures to be described are applicable to children with secondary speech delay with modifications to take these specific developmental conditions into account however.


Compton, A. J. (1970). Generative studies of children’s phonological disorders. Journal of Speech and Hearing Disorders, 35(4), 315–339.

Howard, S. J. (1993). Articulatory constraints on a phonological system: A case study of cleft palate speech. Clinical Linguistics and Phonetics, 7, 299–317.

McLeod, S., & Threats, T. T. (2008). The ICF-CY and children with communication disabilities. International Journal of Speech-Language Pathology, 10, 92–109.

Noble, K. G., Wolmetz, M. E., Ochs, L. G., Farah, M. J., & McCandliss, B. (2006). Brain-behavior relationships in reading acquisition are modulated by socioeconomic factors. Developmental Science, 9, 642–654.

Pamplona, M. C., Ysunza, A., & Espinoza, J. (1999). A comparative trial of two modalities of speech intervention for compensatory articulation in cleft palate children: Phonological approach versus articulatory approach. International Journal of Pediatric Otorhinolaryngology, 49, 21–26.

Shriberg, L. D., Austin, D., Lewis, B. A., McSweeny, J. L., & Wilson, D. L. (1997). The Speech Disorders Classification System (SDCS): Extensions and lifespan reference data. Journal of Speech, Language, and Hearing Research, 40(4), 723–740.

Shriberg, L. D., Tomblin, J. B., & McSweeny, J. L. (1999). Prevalence of speech delay in 6-year-old children and comorbidity with language impairment. Journal of Speech, Language, and Hearing Research, 42(6), 1461–1481.

Conversations with SLPs (3)

I have had an interesting conversation with some orthophonistes from the Québec community who have their own very excellent blogs, Cuit dans le bec   (Mélissa di Sante and Marie-Pier Gingras) and Langage et cie  (Marie-Pier Gingras). They asked my opinion regarding a debate that arose on their sites and the issue is so interesting that I am replicating my thoughts here, first with English examples and then finishing up with a French example although the evidentiary basis for making diagnostic decisions in the French context is much weaker.

The question that arose was which type of normative data is most appropriate when making decisions about whether a child’s speech is within normal limits or not – age of acquisition norms for segments OR age of suppression norms for phonological processes? The specific example given was the case of a child who misarticulates /f/ – does it matter whether the child stops this fricative (i.e., /f/ → [p]) or makes some other kind of substitution error (e.g., /f/ → [s]). Would this difference in type of error impact on the severity rating for the child’s speech delay?

As it happens I am about to administer the final exam in my fall course on Phonological Development and most of the exam questions will require the students to make exactly this judgment – given a sample of speech from a child of a given age, is the child’s speech within normal limits or not? I teach my students to use and integrate across multiple sources of normative data. It is essential to consider these multiple sources with no one source being more important than any other, specifically: (1) a standardized measure of the total number of errors; (2) phonetic repertoire; (3) segment level age of acquisition norms; (4) phonological process norms; (5) information about atypical versus typical errors; (6) normative data about the development of prosodic structures from a multilinear perspective where available; (7) intelligibility and other “whole word” measures of speech production. In many cases all of these sources of data will line up with each other so that you could use any one and get the same answer but very frequently there are discontinuities between sources of data and therefore you must integrate across these different perspectives and use your clinical judgment. I will provide some examples of cases where the different types of normative perspective do not provide a congruent result.

English Example 1

Male child aged 36 months produces /f/ → [p], /v/ → [b], /s, ʃ, ʧ, θ/ → [t], /z, ʤ, ð/ → [d] and has no fricative sounds in the phonetic repertoire. The Iowa-Nebraska norms indicate that the expected age of mastery for these phonemes is 3;6 for /f/, 5;6 for /v/, and between 7 and 9 years for the remainder. I have known SLPs who would tell parents that there is no need to be concerned about a child like this because the child is younger than the expected age of mastery for all of the phonemes in error but in fact this would be the wrong conclusion – because indeed the type of errors produced by the child does make a difference. We know that a child at the age of 24 months is expected to have fricatives in the phonetic repertoire. Furthermore, stopping of fricatives should be suppressed (occurring with less than 20% frequency) before the age of 30 months.

English Example 2

On the other hand, let’s say we have another male child aged 36 months who produces /f/ → [p] and /v/ → [b] in the word initial position but /f/ → [s] and /v/ → [z] in the word final position;  /s, ʃ, θ,/ → [s] in all word positions and /z, ð/ → [z] in all word positions. Furthermore Percent Consonants Correct in Conversation is 80% and whole word accuracy in conversation is 80%, scores that are within normal limits according to published norms. In this case a conclusion of age appropriate speech could be warranted because the error phonemes are not expected to be mastered at the age of 36 months and the errors are all of the type that are typically observed in children of this age and the stopping that is observed is not pervasive.

English Example 3

A male child produces errors on the following consonant targets on the Goldman-Fristoe Test of Articulation: /k/ (i,m,f), /ɡ/ (i,m,f), /ʃ/ (i,m,f), /ŋ/ (m), /s/ (i,m,f), /z/ (i,m,f), /sl/, /sp/, /st/, /sw/. Fronting of velar stops is pervasive in the child’s speech. In this case, because the child is 5;0 and he has made 20 errors, a percentile rank of 29 is obtained which might suggest that the child’s speech is within normal limits warranting a discharge recommendation except that fronting should be suppressed by age 3 years and fronting in the coda position in particular is atypical. Therefore this child should be eligible for an intervention despite his GFTA score.

English Example 4

Male child age 5 produces errors on the following consonant targets on the Goldman-Fristoe Test of Articulation: /ʃ/ (i), /ʧ/ (i), /ʤ/ (i,m), /l/(f), /r/(i,m), /ɡr/, /kr/, /tr/. These 10 errors would yield a percentile ranking of 48 and the child’s speech could clearly be judged to be within normal limits given that the expected age of mastery is 6 years for the first 4 consonants and 8 years for the /r/. However, what if the child’s speech intelligibility in conversation was found to be only 80%? Then this child would have what is called an intelligibility-speech gap because his consonant accuracy is within normal limits but his connected speech intelligibility is below the expected 100% intelligibility. This gap can be observed in children who have experienced chronic otitis media and may reflect imprecision in the production of consonants and vowels and a reduced vowel space. It is possible that some kind of intervention might be warranted in a case like this.

English Example 5

Male child aged 6 years consistently produces initial and medial /l,r/ → [w] and final /l,r/→ [o]. There are no correct liquids in the child’s repertoire. Here again the difference in normative expectations for phonological processes versus segments must be kept in mind. Mastery of /l,r/ is not expected until ages 7 and 8. However, gliding and vocalization of these liquids is not expected after the age of 5 years. Even a 24 month old child is expected to have a liquid in the phonetic repertoire.  Therefore this child’s speech is concerning even if the GFTA score is within normal limits and the child is younger than the expected age of mastery for these phonemes.

French Example

In French it is very important to take word and phrase structure into account but unfortunately we don’t have as much normative data as we need in order to interpret child performance reliably. We know for example, that when naming simple words, all consonants should be at the level of at least customary production (in fact most are mastered) before age 5 years in French. Therefore if we take the case of the 5 year old male child again, if the child is producing the words “giraffe” → [ziʁas] and “éléfant” → [elesã] then these productions are not age-appropriate. On the other hand, if the child is producing the words “giraffe” → [ʒiʁaf] and “éléfant” → [elefã] then these productions are correct and age-appropriate. What if the child produces the simpler words correctly but says “fromage” → [kʁomaʒ] and “framboise” → [kʁãbwaz]. Surprisingly these strange looking cluster productions might not be concerning in French. Firstly, error frequency is strongly conditioned by syllable stress and position. Therefore, the /f/ targets in “giraffe” → [ʒiʁaf] and “éléfant” → [elefã] are easier because they are in the stressed word-final/phrase-final position. Secondly, the /f/-clusters in French are vulnerable to spreading of the Dorsal feature in /ʁ/ to the initial obstruent, especially in non-final contexts such as this. Although this error is less frequent than simple reduction (e.g., “framboise” → [ʁãbwaz]),  the spreading error occurs with greater than 5% frequency, and is therefore not atypical in French. One thing to keep in mind is that although the segment acquisition norms suggest that consonant acquisition is more or less complete by age 48-53 months, phonological development continues through age 7 years just as in English because mastery requires consistency in the long words that are characteristic of French and in connected speech. French does not have lexical stress per se but phrase level prosody is important. The child may be able to label the picture as ‘fourchette’ but a phrase such as ‘J’ai perdu ma fourchette hier.’  puts the /f/ target in a more difficult context. We need phrase level normative data for French and studies that manipulate syllable complexity, syllable stress and word length variables systematically before we can be really clear about the normal course of phonological development in French. Nonetheless, testing children’s consonant acquisition in the context of words that are consistent with French word structure is important and therefore I would examine Percent Consonants Correct on a screener such as our Test de Dépistage Francophone de Phonologie in addition to considering the segment acquisition norms in MacLeod et al which are based on words with relatively simple structure.


 In English all of the normative data used in this discussion is available in our book:

Rvachew, S., & Brosseau-Lapré, F. (2012). Developmental Phonological Disorders: Foundations of Clinical Practice. San Diego, CA: Plural Publishing.

 The specific data referred to includes:

Austin, D., & Shriberg, L. D. (1997). Lifespan reference data for ten measures of articulation competence using the speech disorders classification system (SDSC): Waisman Center on Mental Retardation and Human Development, University of Wisconsin-Madison.

Cahill Haelsig, P., & Madison, C. L. (1986). A study of phonological processes exhibited by 3- 4- and 5-year-old children. Language, Speech & Hearing Services in Schools, 17, 107-114.

Schmitt, L. S., Howard, B. H., & Schmitt, J. F. (1983). Conversational speech sampling in the assessment of articulation proficiency. Language, Speech & Hearing Services in Schools, 14, 210-214.

Shriberg, L. D., Flipsen, P., Kwiatkowski, J., & McSweeny, J. L. (2003). A diagnostic marker for speech delay associated with otitis media with effusion: the intelligibility-speech gap. Clinical Linguistics & Phonetics, 17, 507–528.

Smit, A. B. (1993). Phonological error distributions in the Iowa-Nebraska Articulation Norms Project: Consonant singletons. Journal of Speech and Hearing Research, 36, 533-547.

Smit, A. B., Hand, L., Freilinger, J. J., Bernthal, J. E., & Bird, A. (1990). The Iowa articulation norms project and its Nebraska replication. Journal of Speech and Hearing Disorders, 55, 779-798.

Stoel-Gammon, C. (1987). Phonological skills of 2-year-olds. Language, Speech & Hearing Services in Schools, 18, 323-329.

In French the following sources were considered:

Bérubé, D., Bernhardt, B., & Stemberger, J. P. (2013). Un test de phonologie du Français: Construction et utilisation. Canadian Journal of Speech-Language Pathology and Audiology, 37(1), 26-40.

Brosseau-Lapré, F., & Rvachew, S. (2014). Cross-linguistic comparison of speech errors produced by English- and French-speaking preschool age children witih developmental phonological disorders. International Journal of Speech-Language Pathology, Early Online, 1-11.

MacLeod, A. A. N., Sutton, A., Trudeau, N., & Thordardottir, E. (2011). The acquisition of consonants in Québec French: A cross-sectional study of preschool aged children. International Journal of Speech-Language Pathology, 13, 93-109.

Rvachew, S., Leroux, É., & Brosseau-Lapré, F. (2014). Production of word-initial consonant sequences by francophone preschoolers with a developmental phonological disorder. Canadian Journal of Speech-Language Pathology and Audiology, 37, 252-267.

Rvachew, S., Marquis, A., Brosseau-Lapré, F., Paul, M., Royle, P., & Gonnerman, L. M. (2013). Speech articulation performance of francophone children in the early school years: Norming of the Test de Dépistage Francophone de Phonologie. Clinical Linguistics & Phonetics, 27(12), 950-968. doi: doi:10.3109/02699206.2013.830149

Conversations with SLPs (1)

I have been enjoying conversations with SLPs about Developmental Phonological Disorders, discussing questions either about the book or about their patients in relation to the clinical advice offered in the book. I am going to share the gist of some of these exchanges on this blog from time to time, with the permission of the parties in involved.

Gabrielle has been reading the book, valiantly page by page, with a view to the implications for her primary clientele which is students with dyslexia as well as speech and language disorders. She has two interesting questions regarding Chapter 7. The questions relate to the discussion of the perceptual abilities of children with DPD generally and the results of Rvachew and Jamieson (1989) in particular.

In Rvachew and Jamieson (1989), the participants who were preschoolers with phonological disorders were required to identify the words seat and sheet via a picture pointing task when presented with the words live-voice, in order to be enrolled in the study. Even though the children were able to complete the live-voice task, about half of them were unable to identify synthetic versions of these words. I relate these results to Nittrouer’s finding that some children have immature cue weighting strategies such that they identify these fricatives based on the formant transitions between the fricative and the vowel rather than attending the steady-state spectrum of the fricative noise. Gabrielle asks: “What is the difference between a spectral cue and a formant transition, and if the kids can hear the difference during “live” speech does it matter that they can’t use spectral information during synthetic listening tasks?” The way in which these fricative sounds are perceived in words by adults and children is described in pages 43 to 45 (see also Nittrouer, 2002). Like all speech sounds, the perception of fricatives requires the integration of many acoustic cues that are spread across the word.  When the child attends primarily to one cue (the dynamic formant frequency changes between consonant and vowel) and ignores other cues (such as the peak frequency of the fricative noise) there are consequences for both perception and production of these phonemes. The child can achieve accurate perception with live-voice stimuli under some conditions with an immature cue-weighting strategy but this strategy will be unreliable, leading to perceptual errors under difficult listening conditions (unfamiliar listeners, fast speech, background noise). The consequences for speech production are greater however; if the child believes that the primary cues to the /s/-/ʃ/ contrast lie in the dynamic transition between the fricative and the vowel, the child is not attending to a critical piece of information that will help him or her learn to manipulate tongue placement and groove width in order to produce /s/ with higher frequency centroid frequencies and /ʃ/ with lower frequency centroid frequencies. Testing the child with synthetic speech that isolates these cues highlights the fact that even though the child can perceive the contrast in some circumstances, the child’s perceptual knowledge is incomplete. Developmentally, perceptual and productive knowledge of phonological contrasts is acquired gradually, with perceptual knowledge leading somewhat but iterative gains occurring over time in both domains.

The second question was “about auditory processing difficulties (on page 551 and 552), as different from speech perception difficulties it was my understanding that often “auditory processing” evaluations include speech perception tasks. Was the original concept of auditory processing based on non-speech tasks?” Indeed, in this section I am making a distinction between the hypothesis that auditory processing deficits cause speech or language disorders (such as Tallal’s proposal that temporal processing difficulties are a causal factor) versus the hypothesis that speech perception deficits play a role (as in Ramus’ proposal about phonological representations). Traditionally, auditory processing skills would be attributed to lower levels of the auditory system, midbrain and primary auditory cortex for example, whereas the formation of phonological representations necessary for speech perception takes place at higher cortical levels (see discussion of dual-stream model on pages 127-130). Of course the development of the system is influenced by all kinds of auditory input and perception of speech requires the entire system working together. However, certain nonspeech tasks such as gap detection or the binaural masking level difference are supposed to measure functioning at lower levels of the auditory system. It is true that batteries of tests that are supposed to measure central auditory processing involve speech input and, at least to me, often appear to be measures of language processing. It is not for me to explain the choice of tests in those batteries (although you may enjoy this RALLI video in which Dorothy Bishop discusses the controversy about central auditory processing disorder as a diagnosis). My point was that it is important to not mix up the literature on the auditory processing skills of children with speech and language deficits with the literature on the speech perception and phonological processing skills of children with speech and language deficits. The unreliable findings on auditory processing have no bearing on the more consistent finding that children with speech production difficulties have problems with speech perception.

I welcome interesting questions such as these and invite more from pre-practice and practicing speech-language pathologists.