Traditional Therapy – Single Word
Traditional therapy for articulation disorders has been around forever. It is the technique of choice for the great majority of practicing speech therapists. This is a technique based on the single word. Therapy stresses sounds in the initial, medial, and final positions of words. There are multiple tests currently on the market; Goldman-Fristoe, Photo Articulation Test that are based on the single word. Graduate students learning the profession are taught by PhDs that articulation therapy is conducted with single words. Homework is handed out with lists of words. The failure of traditional therapy as a useful viable therapy technique is glaringly evident in the state of the profession as experienced by students in the public school system.
The lack of success that traditional therapy has had with severe articulation disorders has led Speech Pathologists to incorporate a variety of extraneous nonarticulatory strategies. These include blowing into straws, snake sounds, and oral motor and strengthening exercises. It was stated by Banotai, 2007 in a review of Gregory Lof’s ASHA presentation that articulation is not about strength and tone but speed and agility. These nonarticulatory strategies are counterproductive and show a lack of understanding of how speech is produced.
Coarticulation Therapy – strings of syllables built up into a functional phrase
The literature on Coarticulation has been around for decades.. Kozhevnikov, V. and Chistovich, A. (1966) established the fact that the single word could not be broken up into discreet movements. They also stated that the acoustic signal could not be separated into distinct phonemes. It should have led to the single word level of therapy being discarded. Basically /k-ǽ-t/ does not say “cat”. The syllable was also established as the basic unit of production. Individual phonemes are produced as part of a syllable and these are built up with other syllables thus making a longer utterance. The syllable can not be split up into its component parts. There is no definable boundary between the consonant and vowel. The movement (Mannell, 2005) or transition between the two phonemes share characteristics of both. During the production of every syllable, involving every utterance, the movements for each phoneme are overlapping. When the movement is too complex, fast or due to other factors, assimilation or other phonological processes can occur (Barnes, 2007). Epenthesis can occur to simplify a difficult movement in the production of a syllable. When one is selecting a target for therapy, the syllable has been put forth as the appropriate unit to begin with. The rationale is that children recognize the syllable easier than they do the individual phoneme and that children under 5 years of age resist segmenting words into units smaller then the syllable (Barnes, 2007).
I have been a working therapist for 28 years. I have worked in the school systems (with different cultures: urban, rural, the Navajo nation, and currently in St. Croix, V.I.) and recently ended 17 full time years in Home Health (geriatrics) in Phoenix. My sole choice of therapy for severe articulation (phonological) disorders has been Coarticulation Therapy. I have used this therapy for Dysarthria, Apraxia, and literal paraphasia. It’s the only therapy that attempts to fix a child or an adult’s sound system using syllabic strings that mimics how the body/brain produces speech. Coarticulation occurs across word boundaries. An example of this would be the expression, “sun stroke”. After living in Phoenix, Az. the last 25 years, the expression is appropriate. If you say the two words separately then lip rounding only occurs on the “tr” in stroke in anticipation for the “o”. But if you say the two words together in a normal rhythm then the lips start to round during the transition between the “u” and “n” in “sun” which is the previous word. Using the principle that Coarticulation is always happening in running speech, that it crosses word boundaries, and that the basic unit of production is the Cv syllable, I will now discuss how to apply this version of Coarticulation Therapy to fix three specific phonological disorders. They are stopping, t/s, alveolar assimilation, t/k, and gliding of liquids, w/r. In addition weak syllable deletion will be briefly discussed. This technique stresses the motor movement as place and manner are ever changing.
Stopping – t/s
The first thing to do is to elicit the /s/ with a flow of air for approximately one second. It is important to listen for /ts/ or /st/. Point out the proper tongue tip position. Once this is established have the child (client) produce /ais/. Do not accept /aits/ or /aist/. Next is the production of /aisi/, a nice functional vCv phrase. Always stressing tongue movement, air flow, and no stopping of air. Multiple productions of each are necessary. Then have fun with “I see you”. Eventually a string of vCvCvCv is built up. It’s always possible and/or beneficial to produce strings of /sisisi/, /sasasa/, etc. Then quickly go into /aisisΛm/. If a /t/ is produced for a /s/ or adjacent to a /s/ then extract the syllable and rebuild it. When controlling and building up the syllabic environment the continuum should go from other fricatives to continuants to stops and to /t/. An example of a continuum of phrases is:
1. I see some ice, /aisisΛmais/.
2. My face is so nice, /maifesιzonais/.
3. My soup is in a cup, /maisubιzιnəkΛp/. It’s important that there is no pausing in the phrase so as to experience the effects of blending all syllables and to have Coarticulation happen across word boundaries. The direction a phrase takes expands exponentially with experience. This then makes articulation therapy interesting for therapist and child alike.
4. I sit on a seat, /aisιdΛnəsit/.
5. I see a star in a sky, /aisijəstaιnəskai/. Lastly, /s/ in a vCv syllable has to be in the environment of /s/ in a vCCv syllable containing /t/.
Alveolar assimilation – t/k
Eliciting /k/ can sometimes be a challenge. If this can be manage efficiently then quickly have the child produce /kəkəkə/, then /ekəkə/, then /mekəkə/, /mekəkûki/, /aimekəkûki/. It goes from “I make a” to “I make a cookie”. Errors in the syllable (t/k) are isolated and the string is rebuilt. Lastly, the child is to produce phrases such as: “I take a turn, I take a turkey, I take a taco, I cut a key, and I cut a cake”, so that /t/ and /k/ are placed into sequence. All sequences are produced with a natural prosody and rate with no stopping. In the above examples the target phoneme is practiced in nonmeaningful strings where the motor movement and place of articulation are pointed out. Once the continuous flow of movement is established then with the addition of one more syllables, language is introduced and the phrase is made functional. It is also important to note that the string initially builds up “I plus a verb” unlike the single word level which usually uses object names. In 28 years of practicing this therapy I’ve never used single words. A claim not many can make.
Gliding of liquids – w/r, distortion of /r/
/R/ consists of two parts. The first is the vowel and the second is the movement of the tongue into the iconic retroflexed position for /r/. When a child mispronounces /r/ they first distort the vowel then miss the movement. In order to fix this phoneme in speech it is necessary to teach the vowel /a/ (eventually you need to teach the rest of the /r/ controlled vowels) and its general tongue placement then teach the retroflexed movement for /r/. Now have the child produce /ar/ and have him pay attention to the change in sound and movement. Once /ar/ is established and the child can demonstrate this with 100% accuracy then I teach /ar ju/, “are you” and /ju ar/ “you are” with a slight pause between the two words, just so /ar/ is perfect by itself though /ar/ is being placed into a question and/or a statement. Then I teach /arə/ stressing the correct vowel and tongue movement. This is practiced until the movement is easy and Coarticulation is happening. The next step is /juwarəbכּi/ - “you are a boy”. From this sequence it is possible to extract /arə/, add a Cv syllable to get /arəri/ which has sequential /r/ movements to practice the rapid tongue movements. This gets plugged back into the phrase to become /juwarəridər/ - “you are a reader”. The last /r/ controlled vowel to be discussed here will be the schwa, /ə/. A very important unstressed syllable when fixing /r/. You teach the sound and tongue placement for /ə/ and then the movement into /ər/. This is practiced until it is automatic. To this is added a vowel and we now have /ərΛ/. Consonants are added to the back and front and you get /ərΛv/ and then /dərΛv/. This is practiced and as always stressing the movement and sound changes. This gets plugged back into the main phrase and you have /juwarəridərΛv/books – “you are a reader of books”. Another short example of the use of /ər/ is: /ərιz/ → /mΛthərιznais/ -“mother is nice”. Going into the /ə/ and coming out of the /r/ is stressed. You thus teach and achieve Coarticulation across word boundaries. This last point is very important and can not be emphasized enough.
A byproduct of building up the phrase one Cv syllable at a time and controlling the syllabic environment is that this decreases the incidence of weak syllable deletion and gliding of medial consonants. Both of which have a major negative effect on intelligibility.
Phonological processes happen in the string of syllables when meaning is being communicated. If this is so, then where is the logic to using and working on different positions in single words?
Finally, I hope that this establishes Coarticulation Therapy as the choice of therapy and perhaps the only choice of therapy for those who pride themselves as users of techniques based on current literature and speech science.
Siegelman’s paradigm for a shift in the conduct of articulation therapy
Banotai, A. (2007), Reviewing the Evidence. Advance, Vol. 17, Issue 36, Page 6.
Barnes,L. (2007), Syllable Constructs of Preschool Children and the Implications
Regarding Speech Therapy, Journal of Education and Human
Development, Volume 1, Issue 2.
Kozhevnikov, V.& Chistovich, A. (1966), Speech: Articulation and Perception,
U.S. Department of Commerce.
Mannell, R. (2005), Coarticulation, Department of Linguistics, Macquarie
University, Sydney, Australia, An Internet manuscript.
I have given the word “Run” a prominent place in the title. What is meant by this word and how does it figure in the therapy technique that I will be presenting? A RUN is the sequence of syllabic utterances that end in a predetermined sentence. The final utterance contains several productions of the target phoneme in an environment that supports its correct production. Additional RUNS are then set up. They contain a series of ever more complex and varied environments. Beginning RUNS contain Cv syllables in strings of increasing size. “Making a RUN” is about programming meaning on a level that is easily handled by the child’s language system. The child’s ability to be able to produce a specific phoneme or syllable and to suppress the targeted phonological process in incremental steps is at the center of this therapy.
When a child presents with the phonological process of stopping (t/s) it is necessary to establish the feature of frication in the syllabic sequence. From experience, it is far easier to elicit or teach a vC syllable where C is /s/ than it is to elicit a Cv syllable where C is /s/. The child who substitutes t/s will usually produce t/sh and will probably produce p/f and reduce a ch into /t/ and a /j/ as in jar into a /d/. The child will in the initial stage of therapy have a tendency to produce vts, vst or stv when asked to produce a vC or
Cv syllable. Once vC is fixed and therapy moves on to vCv /aisi/ or /aise/ the tendency to produce “vstv” needs to be addressed. The child has a history of stopping the air by moving the tip of the tongue to the area around the alveolar ridge. It is this movement that needs to be negated. One technique I have used with 100% success is to teach /ais/ - pause for one second then produce /hi/, the pause is then modified and shortened while pointing out tongue position and movement. Patience is employed during this procedure as it will take several sessions to actually achieve a proper vCv syllable. Eventually a vCv syllable is achieved and therapy moves on to a vCvCv sequence. The heart of this therapy for all its variations is the vCv syllable. Once this is achieved everything else is downhill, so to speak. The first strings that are presented to the child contain a predominance of /s/. One sequence would be /sisΛ / then /sisΛm/ then /aisisΛm/ then /esI/ then /esIz/ then /fesIz/ then /aisisΛmfesIz/ There are five fricatives in the last sequence. Four of the five fricatives are /s/ or /z/. The sixth consonant is a continuant. There were no competing consonants presented. Another sequence would be: /ais/ then /aise/ and this syllable could be developed with the previously cited technique of /ais + he/ if the child produces /aiste/. Then / saise/ then /jεsaise pause εs/ (Yes I say s). When the string is being produced there is no stopping in the sequence. One last sequence before a vCCv syllable is introduced could be: /εs/ then /εsI / then /εsIz/ then /εsIsowizi/ (S is so easy). One more therapy tip is to prolong the /s/ in one of its positions in the string.
When correcting the phonological process of final consonant deletion traditional therapy would deal with this problem on the single word level by having the child produce a set of minimal pairs /so/ and /sop/. A pair of pictures would accompany this. One picture is of a person sewing and the other a bar of soap. Therapy on the single word level using any technique whether it’s minimal pairs or cycling through a series of steps accompanied by auditory bombardment is ultimately ineffectual because it does not control for nor take into account the effects of coarticulation. The blending of movement and sound is left out of every previous type of therapy. Siegelman’s Paradigm (Siegelman, 2008) takes this into account since it is based on coarticulation. The word level is never introduced as it is in all previous therapies. The important idea here is that a phrase or utterance is being constructed. The phonological processes of deletion of unstressed syllables, gliding of medial consonants, and open syllable can be addressed with the following steps. I would like to point out that phonological processes happen in the string of syllables that is spontaneous speech. A phonological process is a descriptive method of what is happening and it establishes a rule system to make comprehensible the anarchy that is the speech of an unintelligible child. Through the application of coarticulation, the construction of the vCv syllable and the blending of words, the debilitating effects of phonological processes are negated. The final consonant in a single word for example the /t/ in /pHt/ where the /t/ is released is not the same sound as in the sequence /pHdın/ (put in). This is the reason why traditional therapy or therapies that call themselves phonological therapy but end up using single words fails. It is the reason why therapists have often said, “The child is intelligible at the word level but difficult to understand in sentences or spontaneous speech”. To say that a therapy uses phonological reasoning but ends up back on the obsolete idea of fixing the single word when trying to increase a child’s intelligibility in conversation is an oxymoron. A similar phenomenon exists when the regular education or special ed. teacher teaches reading and employs their version of phonics. The child is made to sound out the word “p-I-t” and says “pit” and then reads the sentence with the absence of a normal prosody and the teacher wonders what went wrong. Well, the answer is that coarticulation was not taught. My therapy for final consonant deletion first builds up the syllable at the end of a word. When a vowel is added then the blending of words occurs and a natural prosody is taught. It is a fact that the effects of coarticulation cross word boundaries. One “run “ or sequence of development could be /dIdI/ then /pHdIdI/ then /pHdIdIn/ then /aipHdIdIn/ then /aipHdIdInǝ / then /aipHdIdInǝ kΛ p / then /aipHdIdInǝ kΛpǝ ti/ (I put it in a cup of tea). While working on fixing the process of final consonant deletion, unstressed syllables are being practiced and learned. The child is learning that the movement of the tongue is important especially the tip of the tongue to the alveolar ridge. This in turn supports the suppression of the process of gliding through medial consonants i.e. /pHjIjIn/ vs. /pHdIdIn/ (put it in). Anecdotally, when I first entered the field as a professional one of my first children to see had no consonants in his speech. He substituted a glottal stop for consonants in the string of syllables. Using the technique I just described the child became a normal speaker within 5 months.
In a recent therapy session I saw a child who presented with t/k and omission of middle syllables. The child produced /sI?ǝ r/ for sister and træ?ǝ r/ instead of cracker. The following discussion talks about the variety of techniques used to demonstrate to another professional (Special Ed. teacher) who was trying to teach the child to read. When teaching the child to produce the /st/ blend in “sister” it was first necessary to simplify the word into two syllables. Each syllable was then established as an entity in the child’s production using verbal-visual cues. First /sIs/ was built up with the final /s/ prolonged. Next /dǝ r/ was established in rapid syllables /dǝrdǝrdǝr/. Finally using a verbal-visual cue and a repeated model of /sIs/ /dǝr/ the word was built up in a sequence of two syllables. The verbal-visual cue was “When I put up my left hand you say /sIs/ and when I put up my right hand you say /dǝr/”. The child needed to learn the timing and not be impulsive. /sIsdǝr/ was established at 100% in 5/5 productions. It was then made into a longer series of syllables by adding /mais/ then /maisIs/ then /maisIsdǝr/ then /maisIsdǝr/ + is verbing (My sister is + verbing).
The following is a technique where the therapist controls the child’s productions at the syllabic level. The context and syllabic environment are also controlled. The child’s production of /træ?ǝr/ is dealt with in the following manner: First elicit /ek/, then /ekǝ/ and then /ekǝkǝ/ are produced. The /kræ/ syllable is introduced and /ǝkrækǝ/ is produced. Each syllabic utterance is practiced a number of times. The process of fronting of velars could be further addressed by presenting /ekǝkǝp/ then /mekǝkǝp/ and finally /aimekǝkǝp/. The therapist can now introduce a third syllable with /k/: /kǝkǝkǝ/. This will eventually be built up into two different words, their parts are already blended and the child has already experienced normal speed, movement, and prosody. Additional parts are added and the vowels are manipulated. This then becomes /aikıkǝkæn/ or /aimekakUki/. The string becomes more complex and meaning changes as different phonemes are placed into the sequence. Initially phonemes that are not competitive or the child’s error sound are placed into the sequence. All phonemes have features: place, manner, and voicing. Voicing being the least important though if one of the child’s problems was devoicing of final consonants then the feature of voicing would become meaningful. Phrases such as /aipIkǝkon/, /aimekǝkΛpǝk∪fi/ are then constructed and practiced. Later on in therapy after /k/ has been established at 100% in phrases that do not contain competitive phonemes or his error sound then /t/ or /d/ is introduced. /t/, /d/ contain interfering features that can instigate the error phonological process to appear if used too early in the therapy process. In the above phrase that contained the sequence /kΛpǝ/ (cup a ) this is used instead of /kΛpΛv/ (cup of) because in running speech that is how it is said. First, have the child produce /tekǝ/ then /aitekǝ/ then /aitekǝtako/ (I take a taco) noting the pattern of movement t-k-t-k. Another complete phrase could be /aigatǝgotǝkæmp/ (I got to go to camp/. At a high level of proficiency, the child is capable of producing complex syllabic sequences with age
appropriate content with 100% intelligibility. This is the end result of teaching coarticulated strings of syllables in a controlled manner. The single word level is never employed and meaning is layered onto the syllabic string and intelligibility is programmed into the child’s speech.
This article is a companion piece to an article that I had the pleasure of being published in Advance in June, 2008.
This is a subject that is most crucial when one decides what to base their therapy decisions on. Is the input (target selection) more important than type of therapy? The type of therapy and there are more than 20 types as reported by Baker in 2006 (though not one was The Siegelman Method: based on Coarticulation Therapy) are conflicting in their approach and in what theory of phonology they are based upon. There are two distinctions that need to be addressed when one answers this question. The first is: when doing therapy, do you carry it out using single words as your main level or do you use strings of syllables thus bypassing the word level. The second is: what theory of phonology do you employ, an abstract one (Generative or Optimality Phonology) or Articulatory Phonology which supports the phenomenon of Coarticulation (Kent and Minifie, 1977, Kozhevnikow and Chistovich, 1966).
Gierut (2005) made a statement concerning target selection vs. type of therapy after comparing three types of therapy, minimal pairs, cycles, and a language approach, and found that there were no differences in their success rates (or lack thereof). Gierut stated that the type of therapy chosen was not the important issue but that target selection (what is the input) is the important variable. This is a comment that I heard stated in an inservice that I attended at the beginning of the school year in Anchorage, Alaska. I challenged it then and I am challenging it now. I consider it quite an outrageous statement to make. Gierut then goes on in a number of other articles based on Optimality Phonology that the input is the crucial variable and that success comes from a reranking of constraints in therapy done with traditional techniques. Gierut stated in 2005 that her therapy does not employ sounds in isolation or connected speech but rather in single words (at times nonsense words are employed).
It is my contention that The Siegelman Method (Coarticulation Therapy as supported by the literature in Coarticulation and Articulatory Phonology) which strings together syllables into a predetermined phrase is a much more efficient therapy technique than any of the therapies that have come out of Optimality Phonology. Having said that, I do accept the knowledge base that Optimality Phonology puts forth with some reservations (differential responses and low percentages of accuracy). In a DVD that I put together and which can be purchased at www.coarticulationruns.com, it shows how Coarticulation Therapy can accommodate complexity (Barlow,2001), learnability theory, spr, (Morrisett et al, 2006 ) and implicationally related error patterns (Dinnsen and O’Conner, 2001).
Connected speech is a string of syllables that overlap in their motor movement as well as acoustically. Articulatory Phonology states that the basic unit of phonology is the articulatory gesture and not an abstract concept such as the feature, segment (Browman and Goldstein, 1992). The Siegelman Method teaches the principles of Coarticulation from the beginning of therapy. It passes through the two important stages of vCv, and vCvCv. The Siegelman Method (2008, 2009) posits that the benefits of Coarticulation are as important as the phoneme that is targeted or the type of error pattern that is selected (fronting, stopping, syllable deletion, etc.) The previous mentioned types of error patterns are not viewed as phonological processes (based on Generative Phonology which I reject) but as what gesture pattern needs to be taught and fixed in connected speech. In The Siegelman Method the “flow of movement” is taught. Articulatory Phonology (Browman and Goldstein, 1992) combines phonetics and phonology. This is a very important statement as in all previous theories of phonology, phonetics and phonology are seen as separate. In The Siegelman Method, the “movement has meaning”. The levels of therapy, vCv and vCvCv should be considered as input because Coarticulation in and of itself is the prime reason for success. No other therapy uses Coarticulation in this manner. It is the movement of going into and out of the consonant from the adjacent vowel that is being taught from the very of therapy which is very different than going from the single word into a sentence. This sets the foundation from which all success grows out of.
When the type of technique is chosen to fix an unintelligible child’s phonological system, it is important that it be based on a phonological theory rather than one of the many other areas that one might choose such as nonspeech oral motor exercises or traditional therapy. But in our field, which theory of phonology one selects seems to be unimportant. The field evolved from traditional therapy into a diverse array of phonological theories. There is Generative Phonology, Autosegmental Phonology, Optimality Phonology, and Articulatory Phonology. There are others but I do have time and space constraints so I won’t mention them. There have also been paradigm shifts: traditional therapy into abstract phonologies (Generative/Optimality) or as I learned it from traditional therapy and a static sequence of segments into Coarticulation and later on supported by Articulatory Phonology. It should matter to the individual therapist which theory they base their therapy on but from what I have observed personally in my years in the field and from conferences that I have attended, it truly does not enter into the decision making process.
There seems to be an unstated rule in this field that you are not suppose to criticize someone else’s therapy but I find it difficult not to comment on other specific techniques. One example that stands out is: in a recent position that I held in Peoria, Az., the lead therapist in that district allowed another SLP to conduct /r/ therapy with the following technique, that therapist stated “the best way to do /r/ therapy is to have the client lay down on the floor as this allows the tongue to go into the /r/ position”. There are different levels of expertise when it comes to doing therapy but allowing this type of therapy is part of the reason why public school speech therapy is failing. In another commentary, Gierut et al. (2001) stated that in the technique of Cycles which teaches /s/ clusters before singleton /s/, that to begin with /s/ clusters (an adjunct cluster) predisposes a child to form erroneous hypotheses about syllable structure and that the child is provided with input that does not conform to the universal structure of syllables (see the sonority sequencing principle). It appears that Therapists will always use what they think works before basing decisions on the literature.
The Siegelman Method eschews using single words in therapy and instead builds up strings of syllables (vC, vCv, and vCvCv) into a phrase. Once the vCvCv is mastered and phrases are produced at a high degree of accuracy, it is at this point in therapy that vCCv strings are introduced. It is also at the vCvCv stage that contrasting gestures are introduced. The concept of the Adjacent Sequential Syllable (Siegelman, Powerpoint Presentation, 2011) is employed in my technique. To put it simply, it means what comes next, and to have a reason for its use. The next syllable could be a duplicate, a contrasting gesture, or something based on maximal differences, or a major class difference or on sonority. The use of single words as the main level of therapy is a major reason for the failure or excruciatingly slow progress of these therapies (my opinion). There are therapies that use syllables and then go to single words thus losing the benefits of the flow of movement of Coarticulation Therapy. There are therapies that use minimal pairs (single words) and auditory bombardment as there are several therapies that have come out of Optimality Phonology that use constraints or complexity but also employ single words to effect a change in the child’s phonological system. The one point of commonality of all these disparate therapies is that they all buy into the use of single words. Two examples of how The Siegelman Method (Coarticulation Therapy) can change this use of single words can be seen in the following:
To get back to the main focus of this article, that the choice of therapy is crucial to the success of therapy and thus improving the intelligibility of the client in the fastest and most efficient manner. My choice is strings of syllables rather than single words. My choice of phonological theory is Articulatory Phonology rather than Generative or Optimality Phonology. And my choice of therapy is The Siegelman Method: Coarticulation Therapy based on the literature of Coarticulation and Articulatory Phonology. I have now made the case for the retraction of Gierut’s previous statement.
Barlow, J. A. (2001). Recent Advances in Phonological Theory and Treatment. Language, Speech, and Hearing Services in Schools, 32, 225-228
Browman, C.P., Goldstein, L., et al. (1984) Articulatory synthesis from underlying dynamics. Journal of the Acoustic Society of America,75, 522. Articulatory Phonology from Wikinfo
Browman, C.P., and Goldstein, L. (1992), Articulatory Phonology: An Overview. Phonetica, 49, (3-4), 155-80
Dinnsen, D.A. and O’Connor, K.M. (2001). Implicationally Related Error Patterns and the Selection of Treatment Targets. Language,Speech, and Hearing Services in Schools,32,257-270
Gierut, J. A. and Champion, A. H. (2001). Syllable Onsets 2: Three-Element Clusters in Phonological Treatment. Journal of Speech, Language, and Hearing Research, 44, 88-104
Gierut, J. A. (2005). Phonological Intervention the How or the What? In Journal of Speech-Language Pathology, 6, 8-17
Kent, R.D., and Minifie, F.D. (1977), Coarticulation in Recent Speech Production Models. Journal of Phonetics, 5, 115-133
Kozhevnikov, V.& Chistovich, A. (1966), Speech: Articulation and Perception, U.S. Department of Commerce.
Morrisette, M. L., Farris, A. W., and Gierut, J. A. (2006), Applications of learnability theory to clinical phonology. Advances in Speech-Language Pathology, 8(3): 207-219.
Siegelman, S. (2008), Coarticulation Therapy: A New Paradigm. Advance, Vol.18, Issue 25, page 9.
Siegelman, S. (2009), Coarticulation Therapy: Tips, Runs, and Opinions. Advance, April 6th, in the Clinicians in the Classroom column
Siegelman, S. (2011), The Siegelman Method: Coarticulation Therapy - A Phonological Therapy. Presented at a Conference produced by Northern Speech Services, April, 7-8, 2011.
My therapy technique, The Siegelman Method: Coarticulation Therapy (Siegelman, 2008, 2009) based on the literature in Coarticulation, and Articulatory Phonology (Goldstein and Fowler, 2003) has been used to treat phonologically based disorders in children (though anecdotally I have used it to treat childhood apraxia, and literal paraphasia in adults after having had a stroke). Now the question is , how can this technique and its knowledge base be employed to address the issue of disordered prosody in children with autism. To make the target more specific we could specify children with high-functioning autism (Peppe et al, 2007).
First, What is Prosody?
According to the Queen Margaret University College webpage, prosody is not what you say but how you say it. Prosody is the rhythm, stress, and intonation of speech and its features can be found in the suprasegmental patterns of speech (Wikipedia). Prosody can function on a number of different levels (Shriberg et al, 2001). The different levels of prosody are: grammatical, pragmatic, and affective prosody with each one carrying a different form of information. The emotional states of excitement and calm present themselves in prosody and can be measured by speech rate in syllables per second (Kehrein,2002). The prosody of speech is more than just stringing words together in a sentence. A narrative reflects discourse rather than sentence effects (Tseng, C., et al, 2005). The prosody and rhythm of fluent speech is affected by cross phrase prosodic patterns and syllable duration patterns. There are also tonal and timing patterns which will affect the prosody of speech (Diehl et al, 2009)
How do deficits in prosody present themselves in the speech of children with autism?
McCann et al, in 2005, posited that deficits in the prosody of children with autism are due to a deficit in Theory of Mind skills. In this theory, the disordered prosody of autistic children can be attributable to a deficit with inferring the mental states of others. Children on the autistic spectrum just do not know that prosody contains information necessary to deliver the speaker’s intention.
Disordered prosody reveals itself in the social communication of children with autism and can be found in the pragmatic and affective aspects of prosody (McCann et al, 2005, and Shriberg et al, 2001). These effects changes little over time and can last throughout their life time. Shriberg and his colleagues, in 2001, found that a core feature of this disordered prosody is evident in the suprasegmentals of speech with deficits in pitch, and aberrant stress patterns, along with a monotonic intonation. McCann et al, (2005), added that this speech is characterized by poor inflection, and excessive or misassigned stress.
The literature on disordered prosody in the speech of children with autism has a number of areas of varying levels of disagreement. In 2007, Peppe et al, wrote that prosodic development in children with high functioning autism appeared to have a delay in prosodic development or are deviant in other aspects of prosody. In 2009, Diehl and Paul wrote the there is no current data on the typical developmental sequence of the acquisition of prosodic ability. In the area of articulation or phonological deficits in the speech of autistic children, McCann et al, (2005), wrote that phonology may or may not be impaired, while in 2001, Shriberg et al wrote that there is a high prevalence of residual articulation errors in the speech of children with high functioning autism. It seems that there is a great deal of individuality in the behavioral characteristics of children with autism as there is in the literature that is being written in this field. Treatment for Disordered Prosody in Children with Autism:
Hargrove et al, in 2009 wrote that a positive change in disordered prosody can be achieved with treatment. Shriberg et al, in 2001, placed the focal point of disordered prosody within the utterance rather than within multisyllabic single words. This is a very important statement to make because it takes the therapy out of single words and into phrases and needs to be taken into consideration when one chooses what type of therapy they will use. Shriberg et al, go on to write that they support traditional activities to increase the speaker’s metalinguistic knowledge of appropriate suprasegmental targets. Diehl and Paul wrote in 2009, that there is a dearth of literature on techniques to improve disordered prosody in children with autism. They then wrote that techniques to improve disordered prosody should be both temporally and spatially sensitive. This last statement leads into the next and most important segment of this paper.
How Articulatory Phonology Affects Prosody: Articulatory Phonology (Browman and Goldstein,1986, 2000) puts forth that the articulatory gesture is the basic unit of phonology. This phonological theory supports Coarticulation and the overlapping of gestures (movement) and is the theoretical foundation of The Siegelman Method (Siegelman, 2008, 2009). I will now employ this phonological theory to support the use of my therapy technique as the method of choice to effect a change in the disordered prosody of a child with autism.
Articulatory Phonology began with explaining the phonetic-phonology interface (as part of one system instead of being viewed separately) is now being used to explain the phonetic-prosody interface (Riggs and Byrd, 2008). The theoretical tenets of Articulatory Phonology which were incorporated into the task dynamic model (Saltzman et al, 2008) have also been extended to explain or provide a framework for investigating the higher order dynamics of prosodic phrasing, syllable structure, and lexical stress.
The view that phonological units are seen as being inherently dynamic (moving through space over time) is now also being used to view prosodic boundaries as having a temporal component (Riggs and Byrd, 2008). This movement in time is seen in a lengthening effect that can extend over three syllables prior to the phrase boundary. Saltzman stated that there are prosodic gestures that are active at the phrase boundary which work to slow the rate of utterance. Beckman, (1989), stated that Articulatory Phonology’s model of the timing of articulatory gestures could be used to explain the prosodic organization of an utterance into words, larger phonological units and other suprasegmental structures. Understanding the phonetics-prosody interface helps with explaining the effects that phrases have on articulation, prosody, and speech production (Byrd and Choi, 2010). Two of the main authors in this area of research, Byrd and Saltzman, wrote in 1998 and 2003, that speech motor control is dependent upon prosodic structure, that prosodic gestures adjacent to phrase boundaries are capable of being temporally stretched or shortened (slowing as it nears a boundary and speeding up as the boundary recedes in time), and that this warping of the temporal fabric of an utterance represents a coming together of the fields of prosody and speech dynamics. How can therapy based on Articulatory Phonology (The Siegelman Method: Coarticulation Therapy) be successful in effecting a change in the disordered prosody of children with autism?
The Siegelman Method: Coarticulation Therapy – A Phonological Therapy has been used to effectively change a child’s disordered phonology for over 30 years via the manipulation of syllables built up into a predetermined phrase through coarticulation and overlapping gestures. This technique eschews the single word (see the June, 2011 issue of Advance). The review of the literature that has been presented, naturally extends the tenets on Articulatory Phonology into the realm of prosody. This is possible because this is the way that speech is produced. Using the natural rhythm in a phrase and by manipulating the stress placed on overlapping syllables (gestures) in a given time frame in a functional phrase, one should be able to change the disordered prosodic patterns in the speech of a child on the spectrum with high functioning autism. One last note, the string of syllables needs to be in a vCvCv….structure and contain approximately 5 words.
I used this technique during the spring of 2010 with an 8th grade middle school student who presented with high functioning autism, gliding of /r/, gliding through /d/ as a medial syllable, and has a monotone speech pattern. The goals were to improve the student’s prosody while also improving on his production of /r/. Each goal was worked on during the same session.
Thank you, Steven Siegelman M.S. CCC-SLP
Beckman, M.E., (1989). Timing Models for Prosody and Cross-word Coarticulation in Connected Speech, Proceeding HTL’89 Proceedings of the Workshop on Speech and Natural Language.
Browman, C. P., and Goldstein, L.M. (2000). Competing Constraints on Intergestural Coordination and Self-Organization of Phonological Structures. Bulletin do la Communication Parlee p. 25-34
Browman, C. P., and Goldstein L. M. (1986). Towards an Articulatory Phonology. Phonology Yearbook 3, 219-252
Byrd, D. (2003). Frontiers and Challenges in Articulatory Phonology.
Byrd, D., and Choi, S.,(2010). At the juncture of prosody, phonology, and phonetics – The interaction of phrasal and syllable structure in shaping the timing of consonant gestures, Papers in Laboratory Phonology 10. Mouton de Gruyter.
Byrd, D., and Saltzman, E., (1998). Intragestural Dynamics of Multiple Prosodic Boundaries, Journal of Phonetics, 26, 173-199.
Byrd, D., and Saltzman, E., (2003). The Elastic Phrase: Modeling the Dynamics of Boundary-Adjacent Lengthening. Journal of Phonetics, 31, 149-180.
Diehl, J.J., and Paul, R. (2009). The Assessment and Treatment of Prosodic disorders and Neurological Theories of Prosody, International Journal of Speech Language Pathology, 11, 287-292.
Goldstein, L., and Fowler, C. A., (2003). Articulatory Phonology: A phonology for public language use. In Phonetics and Phonology in Language Comprehension and Production: Differences and Similarities, (159-207), ed. Antje S. Meyer and Niels O. Schiller. Mouton de Gruyter.
Hargrove, P., Anderson, A., and Jones, J. (2009). A Critical Review of Interventions Targeting Prosody, International Journal of Speech-Language Pathology, 11, 298-304.
Kehrein, R., (2001). The Prosody of Authentic Emotions, in Speech Prosody, An International Conference.
McCann, J., Peppe, S., and Gibbon, F.E. (2005). Prosody and Its Relationship to Language in School-Aged Children with High-Functioning Autism. Prosody and Language in Autism.
Peppe, S., McCann, J., Gibbon, F., O’Hare, A., and Rutherford, M., (2007). Receptive and Expressive Prosodic Ability in Children with High-Functioning Autism, Journal of Speech Language Hearing Research, 50, 1015-1028.
Riggs, D., and Byrd, D. (2008). The Scope of Phrasal Lengthening in Articulation: Prosody and Prominence, Laboratory Phonology 11.
Saltzman, E. Inter-Unit Timing in Speech Production: Pattern, Stability, and Hierarchy.
Saltzman, E., Nam, H., Krivokapic, J., and Goldstein, L., (2008). A Task-Dynamic Toolkit for Modeling the Effects of Prosodic Structure on Articulation.
Shriberg, L.D., Paul, R., McSweeny, J.L., Klin, A., Cohen, D.J., and Volkmar, F.R., (2001). Speech and Prosody Characteristics of Adolescents and Adults With High-Functioning Autism and Asperger Syndrome, Journal of Speech, Language, and Hearing Research, 44, 1097-1115.
Siegelman, S., (2008). Coarticulation Therapy: A New Paradigm, Advance, Vol.18, Issue 25, page 9.
Siegelman, S., (2009). Coarticulation Therapy: Tips, Runs, and Opinions. Advance, April 6th, in the Clinicians in the Classroom column.
Tseng, C., Pin, S., Lee, Y.M., Wang, H., and Chen, Y. (2005). Fluent Speech Prosody: Framework and Modeling, Speech Communication, 46, 284-309.
I need to preface this essay by first stating that I retired from work in 2012. I therefore am not accountable to anyone nor do I have to jump through any more hoops. I no longer have a supervisor editing or censoring what I say. Having said that I will now proceed to set out my take on the field.
I have been in the field for 33 years. Split evenly between the public school system and home health (primarily geriatrics). My time spent in the public schools was divided into two time periods separated by 17 years in home health. I will be targeting my comments specifically to the school system and the SLPs who ply their trade in this forum. From my perspective the majority of therapists in the public school arena are well intentioned people who want to help others though the quality of their therapy in the area of phonological (articulation) therapy is woefully lacking. The fault lies not only with the individual therapists but with the PhDs (academics) in the university system and the antiquated, outmoded, and rigid thinking with the professionals who run ASHA itself.
I read an article in June of 2012 in the online edition of the NY Times. The article was under the heading Psy Blog – Understanding Your Mind The Dunning-Kruger Effect: Why The Incompetent Don’t Know They’re Incompetent. The article stated that the poorest performers are the least aware of their own incompetence secondary to failing to learn from their mistakes and paying no attention to negative feedback. In my view this describes the typical SLP who session after session, year after year, stays the course and uses the same failing technique of single word therapy while accepting results that are less than stellar. I have spoken with therapists about this problem. Most just ignore it and refuse to deal with this issue. It is just not a high priority for them. I have heard remarks such as: My colleagues do single word therapy, I was taught single word therapy in graduate school, and This is what I know and this is what I do. They accept the status quo as the way things are done. They are impervious to change. These therapists do an injustice to those who are open to change.
A number of decades ago I was given to read The Structure of Scientific Revolutions by Thomas S. Kuhn (1962). Kuhn put forth a rationale for how fields change through paradigm shifts. This leads to changes in and acceptance of new information that determines how a field conducts itself. New therapies come into vogue based on this shift and old ones are relegated to obsolescence. This transformation is important for our field. But the question arises as to which new information should be a priority. It’s possible to remain in the abstract and go from traditional therapy → generative phonology → optimality phonology. Or one could go from traditional therapy → coarticulation which is how the sound and movement blend into the context. If one were to stay in the abstract then therapy would still be conducted by traditional means but if one accepts the tenets of coarticulation then therapy changes and single words are no longer the vehicle of change but syllabic sequences (vcvcv) takes center stage. What has always mystified me is that even after years of the acceptance of a paradigm shift (pick one) there are still therapists who defend their use of traditional therapy. Kuhn said that proponents of old techniques are resistant to change. It is my opinion that these therapists either need to retrain or else leave the field.
The failure of traditional therapy to consistently provide a high degree of success has led practitioners in the field to look for other ways to conduct therapy. Over time This has led to the adoption of some truly bizarre techniques. You only have to look at /r/ therapy to see this effect (lying on the floor to assist the backward movement of the tongue, placing sticks in the mouth to forcibly move the tongue into position).
The constant search for what works only detracts from what should be a high degree of professionalism. One has to wonder why current practicing SLPs do not reject single word therapy in the light of years of failure, a lack of a supportive theory of therapy and a shift in the paradigm. The answer is – this is all they know and to reject it and change means that they would have to reject years of work and dedication to a faulty premise, and the idea that they actually did functional therapy.
Did the rejection of traditional therapy and the paradigm shift to generative phonology and eventually to Optimality Phonology for target selection make this a better field? To me the obvious answer is no. During this time period via the work done at Haskins Laboratory amongst other places the phenomena of coarticulation was revealed. This then morphed into Articulatory Phonology. The Siegelman Method, Coarticulation Therapy, came out of this stream of knowledge. Kuhn stated that the new paradigm should be neater, more suitable, simpler, and or more elegant. The Siegelman Method is exactly that. It is both simple and elegant in its format and execution while obtaining consistently excellent results.
Public school speech therapy is rife with mediocrity and ultimately failure in the area of phonological therapy. This in an area that can be anywhere from 30%-80% of a caseload is unacceptable. Therapists, Special Ed. teachers, and their departments are never held accountable for their continual failure. Therefore, there is no urgent need to change in order to protect their jobs. Phonological Therapy in the schools is governed by a ruled system mired in thinking that is associated with the tenets of Traditional Articulation Therapy and behavioral rules. The idea that you can’t begin therapy on /r/ until after the first grade because this phoneme is a later developing sound is ridiculous or children with an articulation problem are not allowed in therapy because the category of need is redefined by saying that the child’s problem does not interfere with progress in the curriculum. Discrete positions in a single word are by default the accepted norm for conducting articulation therapy. It makes no difference if the therapy is traditional therapy, cycles, word shells, minimal pairs or something that came out of Optimality Phonology with its various therapy techniques. It is my contention that most therapists experience very little success with fixing a child’s sound system in connected speech because single word therapy is ultimately artificial. Children experience little to no progress through no fault of their own. I have seen colleagues who have exited children because of the lack of progress made over the years and have met SLPs who would give their articulation cases to their SLPAs. I have worked with SLPs who outright refused to do articulation therapy (my first position in the Washington Elementary School System in Phoenix) in a special program put together to fix the unintelligible kindergartens in that district. These two therapists instead insisted on doing the language therapy. I have come to the conclusion that SLPs who become experts in language therapy or go into augmentative communication do so in order not to have to do articulation therapy (a little cynical humor). I have had supervisors specifically ask not to have to observe articulation therapy because they either consider it too boring, it is all about the games, or therapy looks all the same (some variation of single word therapy). After all, how often can one observe articulation bingo? Analogous to this are reading specialists and/or special education teachers who are teaching oral reading skills using similar techniques to SLPs who are teaching phonological principles of sounds in the initial, medial, and final positions of single words. Not being able to get the child to orally read using the appropriate prosody and fluency is similar to the SLP who is able to get the child to produce the sound in single words but not able to have the child use the sound appropriately in connected speech. This is because they do not specifically teach coarticulation skills and ignore this fact their whole career. Both fields use similar techniques to develop fluency in connected speech. I witnessed a knowledgeable special ed. teacher using oral motors exercises to teach a student who also was in speech with a phonological disorder how to pronounce a word that begin with his “sound of problem”. The teacher was ultimately unsuccessful in her attempt In the fall of 2011, I had a temporary position in Phoenix and was asked to take data on the production of phonemes in either the initial or final position of single words for a group of preschoolers during free play for a SLPA who was doing traditional therapy. To my way of thinking it was a meaningless task to chart progress that is tied to a form of therapy that I consider artificial and ignorant of the way one actually speaks. It would be inappropriate to hold the SLPA for 100% of the fault for conducting ineffectual therapy. The SLPA was enrolled in a Speech Path program at a local university. I would instead question the teaching that she is receiving at this institution. The real problem here is that this is typical of the quality of therapy taught in graduate programs.
What does one say to a parent at an IEP meeting when they ask point blank, why their child hasn’t made any progress in their years of therapy? During my last 5 years in the field I re-entered the school system after spending the previous 17 years in home health.
I did this in order to travel with the skill. I went from Phoenix to the U.S. Virgin Islands to Anchorage, AK. to Lodi, Ca. and then to West Covina, Ca. and back to Phoenix. I conducted my share of IEP meetings. The inevitable evolution of the IEP meeting into a contentious minefield filled with advocates and administrators where one had to be wary of every statement made was eye opening and informative. Usually an administrator sits in on the meeting. From my prospective they do this not to support you but to protect the school district’s interests. The main mission of the IEP meeting is to help the student while protecting the district. This means that the parent does not receive 100% of the information that the school district uses to make a decision. Instead the district gives the parent just enough information to get the least expensive and least painful outcome to the district. I have worked in districts in my travels (California, Arizona, and Alaska) where I was explicitly told not to discuss methodology with the parent when explaining their child’s failure is actually a failure in therapy technique.
Now back to the original question of how to respond to the parent who wants to know why their child hasn’t made any progress. Districts would want you to be less than 100% forthcoming, to withhold information and to misdirect your answer. Methodology is what it’s all about. The parent needs to be told that the reason her child hasn’t made progress is because the therapy employed is junk and has nothing to do with spontaneous connected speech. I have had administrators say that the reason you don’t discuss methodology with parents at an IEP meeting is that they could then say why hasn’t their child received this type of therapy in the past and if they could get this therapy in the future if I were to leave that district. I have made it a point to specifically tell parents the reason for the failure of their child to make progress is the type of therapy given to their child in the past. I then explain to the parents the difference between my therapy (strings of syllables built up into a phrase) vs. single word therapy with phonemes in the initial, medial, and final positions. I explain why this type of therapy (single word) is ineffectual and then go on to demonstrate in specific cases the remarkable speed of change in their child’s phonological system that is possible.
The field makes heroes out of people who have come up with therapies that have ultimately failed to hold up over time. I won’t even get into the Monster Study in the field of stuttering. Several years ago I attended the ASHA School Conference in Las Vegas and saw presentations on cycles and other single word techniques and found them to be lacking in vision and based on theories that are no longer cutting edge. Even Gierut’s criticisms of cycles have not stopped people from using this technique. Correcting the production of a phoneme in the spoken language of a child via a phonological therapy should be a skill every therapist should have acquired sometime early in their career. But this is not so. Connected speech is a skill that needs to be taught via coarticulation. “Single Word Therapies” are antagonistic to this idea. The great majority of SLPs in the field take it for granted that the way to conduct phonological therapy is to do it on the single word level. Even when the target selection method changes as it did when Optamility Theory came into vogue, the therapy method remained single word therapy. It doesn’t matter if you base your target on complexity or is feature based the therapy is still single word. This is a major failure of the field. Single word therapy is the main reason for children making little to no real progress year after year. When therapy is carried out in a manner that does not take into account how connected speech is motorically formed and produced then is it any wonder why progress is missing. The only therapy that takes this into account is The Siegelman Method (Coarticulation Therapy).
The literature of the last 40 years in the area of phonological therapy has been focused on the single word. It has been described, probed, torn apart and put back together. But has this accumulative knowledge had a positive effect on what the typical SLP does for therapy. I have always thought that there is a disconnect between the body of knowledge in Speech and the conduct of therapy. The therapies that have come out of this have for the most part failed. It is time to change from the single word to coarticulation as the method of choice to conduct phonological therapy. Whether it is choice of therapy technique or target selection, The Siegelman Method, is by far the most effective technique. It is based in the literature and easily reproducible by others. All one needs to do is reject single word therapy in all its forms. SLPs need to make a paradigm shift in their therapy.
In the future when writers compare the relative success of several varieties of therapies, The Siegelman Therapy, Coarticulation Therapy, should be amongst the mix. If it could be set up, I would enjoy the idea of a contest pitting my therapy (conducted by myself) against the experts with their own therapies as to whose is the most effective and efficient in changing the phonological behavior of an unintelligible five year old kindergarten child.