apraxia, evidence-based interventionstatic.crowdwisdomhq.com/asha/7402 article.pdffrom clinics in...

1

Acquired Apraxia of Speech: Evidence-Based Intervention

Please note: ASHA CEUs may not be earned more than once for the same program. If you completed thisproduct when it was initially published in the Division newsletter, you are not eligible to take it again. Call the

Action Center at 800-498-2071 to check your eligibility.

Contents

CEU Introduction: Empirical Support for Treatment of Acquired Apraxia of Speech ................................. 3

PROMPT Treatment Method and Apraxia of Speech ....................................................................................... 5

Sound Production Treatment for Apraxia of Speech ........................................................................................ 11

Principles of Motor Learning and Treatment for AOS ..................................................................................... 17

Oral Motor Training and Treatment for Apraxia of Speech .............................................................................. 23

Treatment Research on Augmentative and Alternative Communication for Adults With Apraxia of Speech ........ 27

2


3


Note: Julie Wambaugh was issue editor for Division 2 at the time the content of this self-study was first published inNeurophysiology and Neurogenic Speech and Language Disorders. For current information on ASHA’s special interest divisions, visitthe division pages on the ASHA Web site (http://www.asha.org/about/membership-certification/divs/) or call the ASHAAction Center at 1-800-498-2071.

Julie Wambaugh, Issue Editor

Introduction: Empirical Support for Treatment of Acquired Apraxia ofSpeech

This issue of the Division 2 Newsletter is devotedto the discussion and review of evidence supportingthe use of various treatments for acquired apraxia ofspeech (AOS). The field of AOS treatment research isrelatively new, in that the first published reports oftreatments for AOS appeared in the early to mid-1970s (Keith & Aronson, 1975; Rosenbek, Lemme,Ahern, Harris, & Wertz, 1973). Since that time, avariety of treatment methods have been discussed (seeMcNeil, Doyle, & Wambaugh, 2000, for a review).Investigators have been slowly, but consistently,making progress in the evaluation and refinement ofAOS treatments. Researchers have moved fromreporting uncontrolled case studies to employingexperimental methods to assess treatment effects. Inconsidering Robey’s (2001) discussion of the typicalstages involved in the development of evidence-basedtreatments, the field of AOS treatment research couldbe described as being in the early stages of evaluating“efficacy.”

Although numerous approaches have beenproposed for the treatment of AOS, we will focus ontwo treatment methods that have received significantlymore research attention than others: Prompts forRestructuring Oral Muscular Phonetic Targets

(PROMPT; Square, Chumpelik, & Adams, 1985) andSound Production Treatment (SPT; Wambaugh,Doyle, Kalinyak, & West, 1996). In the first article ofthe issue, Arpita Bose and Dr. Paula Square review thebasic principles and methods underlying PROMPT.They then present the evidence supporting the use ofPROMPT along with a discussion of the need foradditional research. Similarly, in the second article, Ireview the procedural aspects of SPT and summarizethe research supporting its use.

The third and fourth articles of the issue moveaway from discussions of specific treatment protocolsto discussions of aspects of treatment that deserveserious consideration when applying almost any AOStreatment. Dr. Kirrie Ballard discusses principles ofmotor learning that appear to be pertinent to thetreatment of AOS, such as structure of the practicesession, type and frequency of feedback, and transferbetween part and whole task components. Shereviews not only the sparse literature in speech-language pathology, but also the literature in limb motorlearning. Dr. Linda Shuster then focuses on the issue ofutilizing oral motor training in the treatment of AOS.She reviews the literature that examines the relationshipbetween non-speech oral abilities and speech abilities

4


to determine if evidence supports the use of non-speech tasks in the treatment of AOS.

In the final article, Dr. Margaret Rogers presentsconvincing arguments urging clinicians and researchersto consider augmentative and alternative communica-tion approaches for individuals with moderate orsevere AOS. Additionally, Dr. Rogers addresses theextremely important issue of adequately defining anddocumenting AOS in study participants. Beyond theAAC/AOS literature, the entire existing AOS treat-ment literature is replete with inadequate participantdescriptions. Consequently, evaluation of existingtreatment data is difficult and generalization of findingsis almost impossible. As correctly noted by Dr.Rogers, this issue continues to be a major obstructionin the development of evidence-based AOS treat-ments.

The topics discussed in this issue represent only asampling of the available AOS treatments and con-cerns that clinicians must consider in their managementof AOS. The road to providing clinicians with thor-oughly tested AOS treatments is long. Althoughprogress has been made in demonstrating positivetreatment effects with selected approaches, there is anobvious need for systematic, sustained evaluation oftreatments across individuals with varying severities ofAOS.

ReferencesKeith, R. L., & Aronson, A. E. (1975). Singing as therapy for

apraxia of speech and aphasia: Report of a case. Brain andLanguage, 2, 483-488.

McNeil, M. R., Doyle, P. J., & Wambaugh, J. L. (2000). Apraxiaof speech: A treatable disorder of motor planning andprogramming. In S. E. Nadeau, L. J. Gonzalez Rothi, & B.Crosson (Eds.), Aphasia and language: Theory to practice (pp.221-266). New York: Guilford Press.

Robey, R. R. (2001). Treatment effectiveness, treatment effi-cacy, and clinical trials. ASHA Special Interest Division 2,Neurophysiology and Neurogenic Communication DisordersNewsletter, 11(1), 6-9.

Rosenbek, J. C., Lemme, M. L., Ahern, M. B., Harris, E. H., &Wertz, R. T. (1973). A treatment for apraxia of speech inadults. Journal of Speech and Hearing Disorders, 38, 462-472.

Square, P. A., Chumpelik, D., & Adams, S. (1985). Efficacy of thePROMPT system of therapy for the treatment of acquiredapraxia of speech. In R. Brookshire (Ed.), Clinical AphasiologyConference Proceedings (pp. 319-320). Minneapolis, MN:BRK.

Wambaugh, J. L., Doyle, P. J., Kalinyak, M. M., & West, J. E.(1996). A minimal contrast treatment for apraxia of speech.In M. Lemme & D. Robin (Eds.), Clinical aphasiology (vol.24, pp. 97-108). Austin, TX: Pro-Ed.

5


PROMPT Treatment Method and Apraxia of SpeechArpita Bose and Paula A. SquareUniversity of TorontoToronto, Canada

STAGE VISEQUENCED MOVEMENTS

Co-articulated Multiple PlanesTiming

STAGE VTONGUE CONTROL

(anterior - posterior)

STAGE IVFACIAL RETRACTION & ROUNDING

(horizontal plane)

STAGE IIIJAW

(vertical plane)

STAGE ITONE

STAGE IIPHONATION

Voice onset time(+ voice - voice)

Segmentalproduction(timing &duration)

BREATH CONTROL

STAGE VIIPROSODY

Figure 1. The Motor Speech Hierarchy (Reprintedfrom Clinics in Communication Disorders, Vol. 4, D.Hayden and P. A. Square, “Motor Speech TreatmentHierarchy: A Systems Approach,” pp. 162-164, 1994,with permission from Elsevier Science.)

IntroductionAcquired apraxia of speech (AOS) is a neurogenic speech disorder that is characterized by speech sound

errors, slowed rate of speech production, and disrupted prosody (Square, Martin, & Bose, 2001). The speechdifficulties observed with AOS are considered to arise from disturbances in the specification of the temporal andspatial aspects of speech sound production (McNeil, Doyle, & Wambaugh, 2000). That is, AOS is thought toreflect a disrupted speech motorsystem.

Basic Principles andMethod

Facilitation of the speech motorsystem of individuals with apraxia ofspeech (AOS) can be provided in avariety of ways, including tactile andkinesthetic support (see Square,Martin, & Bose, 2001) or melodicand rhythmic methods (Sparks &Deck, 1986; Square, Roy, & Martin,1997). Prompts for RestructuringOral Muscular Phonetic Targets(PROMPT) is a tactile-kinesthetic-based treatment method that employscomponents of both articulatory/kinematic and rhythmic/rate controlstrategies in the treatment of AOS.This method was originally devel-oped by Chumpelik (1984) for thetreatment of children with develop-mental motor speech disorders andhas subsequently been modified andeffectively employed with adults withmotor speech disorders (Square-Storer & Hayden, 1989). The oral-facial cues/prompts are targeted toguide the client through the temporal and spatialaspects of speech movements using a combination ofproprioceptive, pressure, and kinesthetic cues. Thesecues are designed to heighten sensory input regardingthe place of articulatory contact, extent of mandibularopening, presence and extent of labial rounding and

6


retraction, voicing, muscle tension, timing of segments,manner of articulation, and/or coarticulation. As amethod, PROMPT helps the client to gain voluntarycontrol of the motor speech system and links thenecessary motor movements that help create holisticmotor-phoneme maps to linguistic equivalents for usein communication (Hayden, 1999). That is, PROMPTis considered to facilitate the translation of articulatory–kinematic parameters of speech production to thecorresponding phonemes and, further, to the resultantwords and phrases intended for production.

Hayden and Square (1994) developed the MotorSpeech Hierarchy that guides the PROMPT systemfor both assessment and treatment of motor speechdisorders. This hierarchy focuses on the training anduse of appropriate degrees of freedom and boundariesfor speech movements in the overall speech systemand speech subsystems (jaw, labial-facial, lingual) in ahierarchical fashion.

The first two stages of the hierarchy focus onneuromuscular pretuning for speech and are usuallynormal in individuals with AOS, but usually abnormal inindividuals with dysarthria. Stage I focuses on posturalsupport for speech and emphasizes the attainment oftrunk, neck, and head control and the suppression ofabnormal oral-motor reflexes. Stage II focuses onphonatory control that will support speech for at least2-3 seconds.

Stages III, IV, and V of the Motor SpeechTreatment Hierarchy assess the three supralaryngealarticulatory subsystems: jaw, lips, and tongue. Thesearticulatory subsystems may be affected individually orin combination with each other in various degrees ofseverity. Stage III focuses on the control of jawmovements that are appropriate in speech. Horizontaland anterior-posterior jaw sliding are physicallyinhibited by the clinician. Maximal jaw opening for“normalized” speech is established, in that the degreeof opening of the jaw is physically restricted by theclinician so that the opening does not exceed thatrequired for /A/ in connected speech. Control over jawgradation is then established using appropriate wordsthat contain vowels and consonants that requireconsistent use of degrees of jaw opening. At Stage IVof the hierarchy, symmetry of movement on both sidesof the face and coordinated movement of the upperand lower lips are encouraged through physical

manipulation in speech. Labial refinements occur atStage IV and are integrated with jaw control todevelop more complex movement synergies betweenthe two subsystems. Once jaw and labial parametersof action are normalized and integrated, Stage V isintroduced. In Stage V, anterior and posterior tongueaction, as well as control of tongue height and locationand area of contraction along the tongue body, areestablished and then integrated with jaw and labialmovements.

Stage VI of the hierarchy focuses on increasingprecise control of parameters of the supralaryngealarticulators (e.g., direction or extent of articulatormovement) within longer sequences of speech move-ments. At this level, prompts that indicate key elementsof movement at points of natural stress may be used(e.g., slight pressure on the lips when there is astressed word). Finally, in Stage VII, temporal aspectsof speech production, such as appropriate rates andintonation contours, are practiced. Given that speech isa very complex action, the clinician may need to workon more than one stage at any one time. On the otherhand, movement impairment could be at only one stage(e.g., lip and facial muscles); in that case, the clinicianmight work at a single stage. In the PROMPT method,the clinician can work at one or more stages dependingupon the needs of an individual client andhis/her impairments.

The stimuli employed in each stage of PROMPTare dependent upon the individual’s speech motorproblems and specific goals. Consequently, PROMPTcan be used to cue speech segments at the phoneme,word, or phrase level (Square-Storer & Hayden,1989) or to facilitate only certain parameters ofmovements, such as jaw opening or the temporal flowof multi-syllabic utterances (see Square et al., 2001,for discussion). Hayden (1999) detailed three types ofprompt cues depending upon the type of supportrequired by the speech motor system. First, “param-eter prompts” are larger organizing and supportingpostures that provide support for control of speechmovements in specific subsystems. For example, if theclient presents with an overall retracted facial posture(i.e., facial muscles are contracting and presents with aperpetual smile), then the parameter prompt wouldhelp release the muscle contraction and allow a morerelaxed and rounded facial position from which move-

7


ment for speech can begin. Second, “complexprompts” are more intricate cues that signal the place,length, and intensity of contractions and provide asmuch information as possible about the motor-pho-neme map (i.e., they are based on the motor templatesof phonemes). For example, the complex prompt forthe vowel /i/ would include input to the anterior portionof the mylohyoid muscle resulting in contraction of thefront of the tongue as well as input to the orbicularisoris muscle resulting in contraction to create a re-tracted facial position. Third, “surface prompts”provide the least amount of tactile-kinesthetic input andprovide information about timing and transition ofvarying planes of movement, particularly in multi-syllabic utterances. For example, the surface promptsfor both /m/ and /i/ would be used to help with thetiming involved in the transition from one sound to theother in the word (e.g., “me”). (For detailed discussionon different types of prompts see Square et al., 2001).

Clinical Studies/Empirical EvidenceEven though PROMPT has been used clinically

for a number of years with AOS speakers, there arefew experimental studies that have investigated re-sponses to PROMPT treatment. Square and col-leagues (Square, Chumpelik, & Adams, 1985; Square,Chumpelik, Morningstar, & Adams, 1986; Square-Storer & Hayden, 1989) have reported the positiveeffects of PROMPT in 4 adults with AOS and coexist-ing aphasia. In these studies, PROMPT treatment wasused to train subjects to produce minimally contrastivephonemes, polysyllabic words, and short phrases.Positive results were reported for all target behaviorsfor the subjects, with varying degree of generalization.Four subjects demonstrated better production abilitiesin trained phonemes and words, while 3 of the 4subjects showed improvement in functional phrases. Inaddition, the subjects also demonstrated improvementin speech intelligibility as measured by the Assessmentof Intelligibility of Dysarthric Speech (AIDS, Yorkston& Beukelman, 1981). There was some generalizationto untrained items, but performance was variable. Theresults of these investigations were encouraging,particularly because positive treatment effects weredemonstrated even for speakers with very severeAOS. Unfortunately, these preliminary studies utilized avery limited number of training utterances and subjects

and employed an uncontrolled simultaneous treatmentdesign to demonstrate effects and to comparePROMPT with other treatment methods (e.g., integralstimulation).

Freed, Marshall, and Frazier (1997) examinedthe effects of PROMPT treatment in an adult withsevere AOS and aphasia using a multiple-baselineacross behaviors design. The treatment items consistedof a core vocabulary of 30 words and short phrasesthat were functionally and personally relevant to theparticipant. Treatment was employed sequentially forsix separate lists, each containing five items, over aperiod of 41 weeks. Results revealed robust acquisi-tion of all 30 items with strong maintenance effects.There was some variability in performance across lists.Although there was no formal measure of generaliza-tion, the subject was encouraged to practice thevocabulary in appropriate social contexts. It wasreported that the subject could successfully use thelearned vocabulary spontaneously in appropriatecontexts. In addition to the positive effects during theacquisition phase of treatment, this investigationhighlighted the long-term maintenance of the treatmentgains and the importance of utilizing the trained utter-ances in socially appropriate contexts.

In a more recent investigation Bose, Square,Schlosser, and van Lieshout (2001) examined theeffectiveness of PROMPT treatment on the acquisitionand generalization of precision and automaticity ofspeech movements in an individual with AOS andBroca’s aphasia. Using a single subject multiple probedesign across behaviors, treatment effects and gener-alization were examined for 30 functionally and per-sonally relevant short phrases (15 training and 15generalization) grouped in five items each in threelinguistically different forms of sentences: imperatives,active declaratives, and interrogatives. The speakerparticipated in one hour treatment sessions three timesa week over a period of 7 weeks. Results indicatedimproved speech precision and sequencing of speechmovements for trained and untrained sentences forimperative and active declarative forms only. Therewas no effect of treatment on the interrogatives. Theresults of this study indicated that for this individualwith AOS and aphasia, PROMPT treatment was moreeffective for linguistically simpler utterances, thushighlighting the fact that careful consideration should be

8


given while constructing treatment stimuli for suchclients. In addition, results indicated that PROMPTcould be used to improve articulatory ease and profi-ciency for individuals with moderate AOS and aphasia.

EvaluationOverall, clinical observations and studies examin-

ing effects in 6 participants suggest that PROMPT canbe an effective treatment method for speech motorimpairments in AOS. Studies demonstrate that speak-ers with even severe AOS can acquire functionalvocabularies and use them in social contexts andsuggest that, with extensive practice, treatment gainsmay be maintained over a long period of time. Forclients with less severe impairments, articulation canimprove even with short durations of treatment.Despite the usefulness of this method, the practicingclinician should note the following:

• First, extensive clinical training and practice isrequired to master PROMPT before it can be usedeffectively. Delivering appropriate PROMPTs forimproving articulation requires practice and pa-tience on the part of clinicians.

• Second, intensive practice is required from theclient in order to maintain gains and make theeffects generalizable to different situations.

• Third, even though the PROMPT system is basedon the Motor Speech Hierarchy, none of the clini-cal studies to date has followed the hierarchy todevelop treatment goals. This is in part due to thefact that speech (words and utterances) is pro-duced by the complex integration of various sub-systems (articulators), and it is difficult to con-struct stimuli that focus on just one subsystem ofspeech (e.g., jaw control), yet maintain functionaland personal relevance for the client. As recom-mended by Hayden and Square (1994), more thanone subsystem can be worked on simultaneouslyif required.

• Fourth, PROMPT involves frequent touching ofclients’ face and articulators by the clinician. Someclients may show high tactile sensitivity. Bothclients and clinicians may not be fully comfortablewith such intervention.

As demonstrated in the preceding paragraphs,there is limited information on effectiveness andefficacy of treatment using the PROMPT system forclients with AOS. This is attributable, in part, to thefact that this method to treat adults with AOS isrelatively new in the whole scheme of treatment

approaches and still requires development and refine-ment. Nevertheless, the existing clinical case studiesand single subject studies do demonstrate the positiveeffects of the PROMPT system.

The current state of knowledge warrants thatadditional efficacy research be undertaken. Suchresearch should be designed with the following consid-erations:

• The effectiveness of the PROMPT treatment shouldbe investigated both in isolation and in combina-tion with other treatment methods for AOS;

• The relative effectiveness and efficiency of PROMPTshould be compared to other treatment methods;

• Well-controlled designs should be employed toassess the specificity of the treatment andgeneralizability to a broad variety of situationsand contexts;

• Sample size should be increased and should in-clude a variety of participants, using carefullyselected inclusion and exclusion criteria; and

• Research should seek to identify the optimumschedule of treatment delivery to enhance learn-ing and maintenance of motor skills.

ReferencesBose, A., Square, P. A., Schlosser, R., & van Lieshout, P. (2001).

Effects of PROMPT Therapy on speech motor function ina person with aphasia and apraxia of speech. Aphasiology,15, 767-785.

Chumpelik, D. A. (1984). The PROMPT system of therapy.Seminars in Speech and Language, 5,139-156.

Freed, D. B., Marshall, R. C., & Frazier, K. E. (1997). Long termeffectiveness of PROMPT treatment in a severely aprac-tic-aphasic speaker. Aphasio-logy, 11, 365-372.

Hayden, D., & Square, P. A. (1994). Motor speech treatmenthierarchy: A systems approach. Clinics in CommunicationDisorders, 4, 162-174.

Hayden, D. (1999). PROMPT manual level 1 and 2. Santa Fe, NM:PROMPT Institute.

McNeil, M. R., Doyle, P. J., & Wambaugh, J. L. (2000). Apraxiaof speech: A treatable disorder of motor planning andprogramming. In S. E. Nadeau, L. J. Gonzalez Rothi, & B.Crosson (Eds.), Aphasia and language: Theory to practice (pp.221-266). New York: Guilford Press.

Sparks, R., & Deck, J. (1986). Melodic intonation therapy. In R.Chapey (Ed), Language intervention strategies in adultaphasia (pp. 320-332). Baltimore, MD: Williams & Wilkins.

Square, P. A., Chumpelik, D., & Adams, S. (1985). Efficacy of thePROMPT system of therapy for the treatment of acquiredapraxia of speech. In R. Brookshire (Ed.), Clinical aphasiology(Vol. 15, pp. 319-320). Minneapolis, MN: BRK.

9


Square, P. A., Chumpelik, D., Morningstar, D., & Adams, S.(1986). Efficacy of the PROMPT system of therapy for thetreatment of acquired apraxia of speech: a follow upinvestigation. In R. Brookshire (Ed.). Clinical aphasiology(Vol. 16, pp. 221-226). Minneapolis, MN: BRK.

Square, P. A., Martin, R. E., & Bose, A. (2001). The nature andtreatment of neuromotor speech disorders in aphasia. InR. Chapey (Ed), Language intervention strategies in aphasiaand related neurogenic communication disorders (pp. 847-884). Baltimore, MD: Lippincott, Williams & Wilkins.

Square, P. A., Roy, E. R., & Martin, R. E. (1997). Apraxia of speech:another form of praxis disruption. In L. J. G. Rothi & K. M.Heilman (Eds.), Apraxia: The neuropsychology of action (pp.173-206). Erlbaum, UK: Psychology Press

Square-Storer, P. A., & Hayden, D. (1989). PROMPT treatment.In P. A. Square-Storer (Ed.), Acquired apraxia of speech inaphasic adults (pp. 190-219). UK: Lawrence Erlbaum.

Yorkston, K., & Beukelman, D. (1981). Assessment of intelligi-bility of dysarthric speech. Tigard, Oregon: C.C. Publica-tions.

10


11


Sound Production Treatment for Apraxia of SpeechJulie L. WambaughUniversity of Utah and VA Salt Lake City Healthcare SystemSalt Lake City

Acquired apraxia of speech (AOS) is a neuro-genic speech disorder that is considered to reflect anunderlying disruption of the ability to specify speechmotor programs (McNeil, Robin, & Schmidt, 1997;McNeil, Doyle, & Wambaugh, 2000). The sounderrors that typify AOS are errors of spatial targetingand timing of the articulators (Square, Martin, & Bose,2001). Consequently, treatments for AOS frequentlyfocus on facilitating or improving the articulatory-kinematic aspects of speech production (McNeil et al.,2000). Sound production treatment (SPT) is one suchtreatment that has been systematically investigated interms of its effects on consonant production in speak-ers with AOS.

ProceduresSPT was designed to promote correct produc-

tion of specific sounds targeted for intervention.Treatment combines modeling, repetition, minimal paircontrast, integral stimulation, articulatory placementcueing, and feedback. With the exception of minimalpair contrast, these techniques are traditional articula-tion training techniques that have been shown to havepositive effects in speakers with AOS (Wertz,LaPointe, & Rosenbek, 1984; Rosenbek, Lemme,Ahern, Harris, & Wertz, 1973). Although the use ofminimal pair contrast is more commonly associatedwith facilitating phonological development than withpromoting correct articulation, it has been incorporatedinto SPT for several reasons. Despite observations thatthe majority of AOS sound errors appear to bemotoric in nature (McNeil et al., 1997; McNeil et al.,2000), it is possible that some sound errors producedby speakers with AOS and aphasia may be phonologi-cal in nature. Minimal contrast pairs may providefacilitory effects to sound selection through homonymyavoidance (i.e., sound differentiation is necessary in theproduction of minimal contrast pairs in order that theword pairs do not sound the same). Additionally,minimal contrast practice involving the target soundand the replacing sound provides potentially important

practice in detailing the articulatory parameters (i.e.,movement specifications) needed to differentiate thesounds.

The techniques that comprise SPT are applied ina response-contingent hierarchy. That is, subsequentsteps of the hierarchy are utilized only upon an incor-rect response. The basic steps of the hierarchy are asfollows:

1. The clinician provides a verbal model of the targetsound/word/phrase/minimal pair and requeststhat the speaker repeat the target. If the response iscorrect, the clinician provides positive verbal feed-back, requests another repetition of the target andthen presents the next item. This procedure isfollowed in all of the following steps when a correctresponse is provided. Feedback at this step (andsubsequent steps, with the exception of Step 4)takes the form of knowledge of results (Schmidt &Lee, 1999).

2. If the response is incorrect, the clinician providesappropriate feedback, presents a written letter cor-responding to the target sound (i.e., “this is thesound you are working on”) and repeats Step 1.

3. If the response is incorrect, the clinician providesfeedback and asks the speaker to attempt the targetutterance with integral stimulation (i.e., watch me,listen to me, say it with me). A maximum number ofattempts is specified (usually three to five attempts).

4. If the response is incorrect, the clinician providesarticulatory placement cues specific to the sounderror (cues may be verbal, visual, tactile).

5. If the response is incorrect, the target sound may beattempted following a model in a simplified con-text, depending upon the original target (i.e., ifworking at the word level of production, the soundmay be attempted in isolation; if working at thephrase level of production, the sound may be at-tempted in a word level production).

6. If the response is incorrect, the next target is at-tempted.

Eight to ten exemplars (e.g., words, phrases)containing the target sound are usually used for treat-ment. Each exemplar is submitted to the hierarchy inrandom order until all exemplars have been attempted

12


(this constitutes one trial). As many trials as possibleare then conducted over a period of 45 minutes to onehour. Typically, four to eight trials can be completed inone session, with the number of trials increasing assound accuracy increases. The sounds selected fortreatment are specific to individual needs. Minimalcontrast pairs are devised to contrast the target soundwith the sound(s) that most frequently replace(s) thetarget sound (i.e., the perceived sound substitution). Incases in which errors are clearly distortions, minimalcontrast is not utilized. Real words are utilized for boththe target and minimal contrast stimuli as frequently aspossible. Although most of the research with SPT hasinvolved sequential treatment of sounds (see below),more than one sound may be treated concurrently.

Evidence for SPT EffectivenessThe empirical support for use of SPT rests on a

series of single-case experimental analyses that havefocused on explicating its acquisition, response gener-alization, stimulus generalization, and maintenanceeffects (Table 1, on page 13). Acquisition refers tocorrect production of trained behaviors in contextssimilar to treatment. Response generalization relates toincreases in untrained behaviors as a result of training asimilar behavior. Response generalization effects ofSPT have been measured to: (a) untrained exemplarsof trained behaviors and (b) untrained, but relatedbehaviors. Stimulus generalization concerns use oftrained behaviors in untrained contexts. Maintenancerefers to the durability of changes in behaviors aftercessation of treatment and has been measured duringtraining of subsequent behaviors and following comple-tion of all treatment. In all of the following SPT investi-gations, probe data served as the indicators of treat-ment effects, with probes being conducted at least oneday following a given treatment session.

Wambaugh, Kalinyak-Fliszar, West, and Doyle(1998) investigated the effects of SPT with 3 speakerswith chronic aphasia and moderate to severe AOS.They used a multiple baseline design across behaviorsand subjects and treated three sounds sequentially witheach speaker (e.g., Speaker 1 received treatment firstfor /z/, then for /S /, and finally for /dZ/). Increases incorrect production of trained items were demonstratedfor each speaker with all sounds, although criterionlevels of performance were not reached for one sound

for two of the speakers (in both cases, the sound was /S/). Response generalization to untrained exemplars oftrained sounds was strong, with correct productionsparalleling those of trained sounds. Response generali-zation to untrained sounds was minimal. Stimulusgeneralization, production of trained items in untrainedphrases, was positive for one speaker (for all sounds)and negligible for the other speakers. Maintenanceeffects were measured during training of subsequentbehaviors and at 6 weeks following completion of alltreatment. Maintenance data revealed that for 2speakers, correct productions remained high for thosesounds that had reached criterion levels of perfor-mance during treatment. For the third speaker, perfor-mance remained high for one sound, decreased byapproximately 20% for a second sound, and returnedto baseline levels for the sound that had not reachedtreatment criterion levels.

Wambaugh and Cort (1998) further examinedresponse generalization effects of SPT with anotherspeaker who presented with moderate AOS andBroca’s aphasia. They speculated that responsegeneralization effects across sounds would be evidentwhen error patterns were similar across those sounds.The speaker in their investigation demonstrated a singleerror pattern of devoicing across stops and affricates.The investigators hypothesized that improvement inproduction of voicing of one sound might result inimprovement of voicing of untreated sounds. Treatmentwas applied first to /b/, then to /dZ/, and finally to /d/and /g/ (simultaneously) in the context of a multiplebaseline design across behaviors. Additionally, produc-tions of /T/ were measured for experimental controlpurposes. The speaker’s erroneous productions of /T/involved a spatial targeting error (i.e., perceived /f/substitution) and therefore, were expected to beresistant to generalization effects. During treatment of /b/, modest improvements in productions of /dZ/, /d/,and /g/ were observed (i.e., improvements in accuracyof 20% to 45%). Productions of /b/ reached highlevels of accuracy during treatment. When treatmentwas extended to /dZ/, additional improvements in /d/and /g/ were observed and /dZ/ reached criterionlevels of performance. Treatment was extendedsimultaneously to /d/ and /g/ and relatively little im-provement was observed with either sound. This mayhave been due to the fact that performance levels were

13


Table 1. Summary of Investigations of Sound Production Treatment

AC: Accompanying ConditionsMBD: Multiple Baseline Designacq: acquisitionrg(ue): response generalization to untrained exemplarsrg(us): response generalization to untrained soundssg: stimulus generalizationm: maintenance

Study Purpose of Study Design N AOS &Aphasia

Dependent Measures Findings

Wambaugh,Doyle,Kalinyak, &West, (1996)

To examine response &stimulus generalizationeffects of tx.; to examinerelationship betweenprobe & tx. data;Participant also reportedin 1998 study –additional data reportedin 1996 investigation

MBDacrossbehaviors

1 mod-severeAOS;Broca’saphasia

% correct productionof target sounds intrained & untrainedwords & in differentstimulus contexts;follow-up at 6 weekspost-tx; measured inprobes

acq-positiverg(ue)-positiverg(us)-negligiblesg-positivem-positive

WambaughKalinyak-Fliszar,West, &Doyle,(1998)

To examine acquisition,response gen., stim. gen.& maintenance effects oftx. To provide replica-tions of tx. effects.

MBDacrossbehaviors& subjects

3 mod-severeAOS;Broca’saphasia

see above acq-positiverg(ue)-positiverg(us)-negligiblesg-positive for only 1participantm-positive

WambaughWest, &Doyle,(1998)

To examine the effects oftraining several soundswithin a sound classsimultaneously. Toexamine the effects oftreatment on soundaccuracy and rate ofproduction.

MBDacrossbehaviors

1 mild-mod.AOS;Broca’saphasia

% correct sounds intrained and untrainedsentences; follow-up at6 wks. post-tx.sentence durationmeasures; measured inprobes

acq-positiverg(ue)-positiverg(us)-positive

Wambaugh& Cort(1998)

To examine the responsegeneralization effects oftraining on sounds forwhich error patterns weresimilar. To providereplications of tx. effects.

MBDacrossbehaviors

1 mod.AOS;Broca’saphasia

% correct sounds intrained and untrainedwords; measured inprobes

acq-positiverg(ue)-positiverg(us)-partial

Wambaugh,Martinez,MacNeil, &Rogers,(1999)

To examineovergeneralizationTo provide replicationsof tx.

MBDacrossbehaviors

1 mod.-severeAOS;Broca’saphasia;dysarth.


acq-positiverg(ue)-positiverg(us)-negligibleovergeneral-ization evident

Wambaugh& Martinez(submitted)

To attempt to mini-mizeovergeneral-izationeffects with the speakerfrom Wambaugh et al.(1999)

MBDacrossbehaviors

1 mod.-severeAOS;Broca’saphasia;dysarth.


acq-positiverg(ue)-positiverg(us)-negligibleovergeneral-ization negligible

Wambaugh(2000, 2001)

To examine and promotestimulus generalization

MBDacrossbehaviors

2 1-mod.-severe1-mild-moderate

% correct sounds inwords and phrases;measured in probes

acq-positiverg (ue)-positiverg(us)-negligible &partialsg-positive to somecondition, facilitatedby additional tx.

14


high prior to the initiation of treatment with thesesounds or to difficulty on the participant’s part indealing with two sounds concurrently. Across soundgeneralization to /T/ did not occur. As in theWambaugh et al. (1998) study, response generalizationto untrained exemplars of trained sounds was strong.Maintenance effects were also strong.

Generalization of treatment effects across soundswas examined in the preceding studies from theperspective of assessing positive effects of treatment.Generalization effects may also be negative, as in thecase of overgeneralization (i.e., incorrect overuse oftrained behaviors). Wambaugh, Martinez, McNeil, andRogers (1999) conducted an error analysis of re-sponses in order to try to explain problematic mainte-nance effects in another experimental analysis of SPT.The participant, who demonstrated moderately severeAOS and Broca’s aphasia, received SPT appliedsequentially to /p/, /k/, and /S/ in the context of amultiple baseline design across behaviors. The speakerachieved high levels of correct production of /p/ duringtreatment, which then decreased during /k/ treatment.Similarly, he attained high levels of accuracy of pro-duction of /k/ during treatment and then decreaseddramatically in productions of /k/ during /S/ training.The error analysis revealed that this speakerovergeneralized productions of the sound currentlyunder treatment to all other sounds under study.Retraining of the previously learned sounds in conjunc-tion with each other returned productions to highlevels. Productions were maintained at high levels at 6weeks following treatment. Anecdotally, varyingdegrees of overgeneralization had been observed withother speakers who received SPT, but not to thedegree observed in the investigation by Wambaughand colleagues (1999).

The speaker who participated in the precedinginvestigation is currently participating in another studyinvolving SPT (Wambaugh & Martinez, submitted). Heis receiving SPT applied to three groups of threesounds each, with the sounds in each group beingtreated concurrently. Preliminary findings indicate thatovergeneralization appears to be impeded by themultiple sound training. Only minor, transient instancesof overgeneralization have been observed to date. Avariation of SPT was applied to sound groups withanother AOS speaker in an earlier investigation

(Wambaugh, West, & Doyle, 1998) and that speakeralso demonstrated positive acquisition effects for alltrained sounds with minimal overgeneralization. Thus, itappears that treating more than one sound in parallelwith SPT may be an effective, and perhaps moreefficient, method of sound remediation, for somespeakers.

The focus of the preceding investigations was onacquisition and response generalization effects and noton stimulus generalization. Although Wambaugh,Kalinyak-Fliszar, West, and Doyle (1998) measuredstimulus generalization to phrases, no efforts weremade to promote generalization when it was found tobe lacking for two of the three participants. Wambaugh(2000; 2001) examined stimulus generalization effectsacross several elicitation contexts with two speakerswith AOS and aphasia. Multiple baseline designsacross behaviors and contexts were used to examinegeneralization from word and phrase level training tosentence completion, multisyllabic words, and longerutterances. When generalization was found to belacking in a particular context, treatment was sequen-tially modified and extended to that context andadditional treatment gains were achieved.

ConclusionsSignificant progress has been made toward

providing objective evidence to support the use ofSPT in the treatment of AOS. It is clear that SPT canbe expected to result in improved production of targetsounds and that training relatively few exemplars of asound will result in improved productions of untrainedexemplars. It appears that significant generalizationacross sounds is unlikely to occur unless an underlyingpattern of errors exists across sounds. Much moreresearch is needed to specify and refine aspects ofSPT prior to the consideration of larger scale research.For example, systematic replications should be con-ducted to examine effects with: (a) patients of differingseverities; (b) modifications of the treatment applica-tion; (c) extended lengths of maintenance measures;and (d) different stimulus generalization measures.Issues of social validation and levels of outcomemeasures also require study. Despite the dauntingamount of research still required to demonstrate theeffectiveness of SPT, clinicians are justified in consid-ering SPT for treatment with patients with moderate

15


AOS.Dr. Julie Wambaugh is an associate professor in theDepartment of Communication Disorders at theUniversity of Utah and a researcher with the VASalt Lake City Healthcare System.

ReferencesMcNeil, M. R., Doyle, P. J., & Wambaugh, J. L. (2000). Apraxia

of speech: A treatable disorder of motor planning andprogramming. In S. E. Nadeau, L. J. Gonzalez Rothi & B.Crosson (Eds.), Aphasia and language: Theory to practice (pp.221-266). New York: Guilford Press.

McNeil, M. R., Robin, D. A., & Schmidt, R. A. (1997). Apraxia ofspeech: Definition, differentiation, and treatment. In M.R. McNeil (Ed.), Clinical management of sensorimotor speechdisorders (pp. 311-344). New York: Thieme.

Rosenbek, J. C., Lemme, M. L., Ahern, M. B., Harris, E. H., &Wertz, R. T. (1973). A treatment for apraxia of speech inadults. Journal of Speech and Hearing Disorders, 38, 462-472.

Schmidt, R. A., & Lee, T. D. (1999). Motor control and learning: Abehavioral emphasis. Champaign, IL: Human Kinetics.

Square, P. A., Martin, R. E., & Bose, A. (2001). The nature andtreatment of neuromotor speech disorders in aphasia. InR. Chapey (Ed.), Language intervention strategies in aphasiaand related neurogenic communication disorders (4th ed., pp.847-884). Baltimore, MD: Lippin-cott Williams & Wilkins.

Wambaugh, J. L. (2001). Stimulus generalization effects of soundproduction treatment. Manuscript in preparation.

Wambaugh, J. L. (2000, January). Stimulus generalization effectsof treatment for articulatory errors in apraxia of speech. Postersession presented at the biennial Motor Speech Confer-ence, San Antonio, TX.

Wambaugh, J. L., & Cort, R. C. (1998, January). Treatment forAOS: Perceptual and VOT changes in sound production.Poster session presented at the biennial Motor SpeechConference, Tuscon, AZ.

Wambaugh, J. L., Doyle, P. J., Kalinyak, M. M., & West, J. E.(1996). A minimal contrast treatment for apraxia of speech.In M. Lemme & D. A. Robin (Eds.), Clinical aphasiology (Vol.24, pp. 97-108). Austin, TX: Pro-Ed.

Wambaugh, J. L., & Martinez, A. L. (submitted). Acquisition andstimulus generalization effects of sound production treatmentfor apraxia of speech: A follow-up investigation. Paper submit-ted for presentation at the biennial Motor Speech Confer-ence, Williams-burg, VA.

Wambaugh, J. L., Martinez, A. L., McNeil, M. R., & Rogers, M.(1999). Sound production treatment for apraxia of speech:Overgenerali-zation and maintenance effects. Aphasiology,13, 821-837.

Wambaugh, J. L., Kalinyak-Fliszar, M. M., West, J. E., & Doyle,P. J. (1998). Effects of treatment for sound errors in apraxiaof speech. Journal of Speech, Language, and Hearing Re-search, 41, 725-743.

Wambaugh, J. L., West, J. E., & Doyle, P. J. (1998). Treatment forapraxia of speech: Effects of targeting sound groups.Aphasiology, 12, 731-743.

Wertz, R. T., LaPointe, L. L., & Rosenbek, J. C. (1984). Apraxia ofspeech in adults: The disorder and its management. Orlando,FL: Grune & Stratton.

16


17


Principles of Motor Learning and Treatment for AOSKirrie J. BallardUniversity of Iowa, Iowa City

In this country, the prevailing view is that apraxiaof speech (AOS) represents an impairment in speechmotor control. The predominant treatment approachesfor AOS focus on affecting the organization of theneuromotor system for relearning of speech motorskills. Relatively few studies investigating treatmentefficacy for AOS have been published (Wambaugh &Doyle, 1994). Several researchers in this field, inaddressing the paucity of treatment efficacy literature,have suggested appealing to the much deeper literaturein the field of limb motor control and learning (e.g.,Duffy, 1995; McNeil, Robin, & Schmidt, 1997).Studies of limb motor control have explored theconditions that influence acquisition (i.e., performanceduring the training session or phase), retention (i.e.,maintenance post-training), and transfer (i.e., generali-zation) of motor skills in the limbs and much of thework has a foundation in the behavioral tradition.

While we typically measure and report acquisitionin clinical settings, the findings summarized belowdemonstrate that acquisition is not necessarily a goodindicator of how behaviors will be retained long term(e.g., Schmidt & Bjork, 1992). Schmidt and Lee(1999), in a comprehensive review of the literature inlimb motor learning in normal adults, have identifiedseveral principles that influence how flexible and skillfulcontrol of motoric behaviors is acquired and retained.These principles pertain to (a) the structure andamount of practice provided, (b) the nature andfrequency of the feedback that a person receivesduring practice, and (c) the part-whole relationshipbetween tasks. These principles do not dictate thetreatment protocol applied, but rather how the proto-col is applied. Below, studies in limb motor learningwill be presented and, where available, publishedapplications of these principles to neurological impair-ments of limb and speech motor systems will bediscussed.

Principles of Practice StructurePrinciples of practice structure address (a) the

order of trials within a practice session (i.e., random orblocked order), and (b) the amount of practice re-quired for learning (i.e., high or low number of trials).Order of trials during practice: In blocked practice,all trials for a given task are grouped together within apractice session. In random practice, trials on severaltasks are presented in random order. The taskspresented may involve different actions or variants of asingle action. Schmidt and Bjork (1992) noted thatpracticing tasks in random order introduces difficulty,or “high contextual interference” (Battig, 1979), to thepractice session as it prevents one from developing astable set for a task. In effect, it requires greaterattention, forces retrieval and organization of a differentresponse on every trial, and results in increasedprocessing of the relationships governing variations of agiven task. Thus, the random context may more closelysimulate a natural context and so result in betterretention, and also transfer to novel contexts.

Several studies have demonstrated that randompractice order facilitates long-term learning (e.g.,Goode & Magill, 1986; Shea & Morgan, 1979). Sheaand Morgan had subjects practice three similar armmovement tasks. One group practiced these in randomorder and the other group in blocked order. The groupthat underwent random practice demonstrated greaterretention at 10 days post-training and more positivetransfer. This effect has been repeated for similar taskconditions and has been demonstrated for learningtiming of movements, perceptual anticipation, controlof force, ability to detect movement error, and coordi-nation of bimanual movements (see Schmidt & Lee,1999, for a review). In summary, blocked practicetends to promote acquisition of the skill during thetraining period, but random practice leads to greaterretention of skill post-treatment and more likelihood oftransfer to novel contexts or related tasks.

A small number of studies have examined thegeneralizability of these findings to individuals withmotoric impairments. Hanlon (1996) examined thelearning of different functional actions in individuals

18


with a hemiparetic upper limb following cerebrovascu-lar accident (CVA). Random practice resulted ingreater retention than blocked practice. Adams andPage (2000) and Knock, Ballard, Robin, and Schmidt(2000) have extended this to the speech motor system.Adams and Page trained 40 adults (20-41 yrs) whodid not have AOS to produce an utterance at a slowedrate of speech. All subjects reached a similar level ofskill during the acquisition phase, but retention scoreswere higher when multiple tasks were presented andbehaviors were practiced in random order. Knock andcolleagues used a phonetic placement approach (e.g.,Van Riper & Irwin, 1958) with two individuals withacquired apraxia of speech to compare random andblocked practice orders. They trained simple conso-nant-vowel (CV) syllables or CVC words. Mostspeech targets were acquired at a similar rate underboth conditions. However, retention at 4 weeks post-treatment was higher for the behaviors trained inrandom order. The participants did not transfer theeffects of treatment to novel speech behaviors. Oneindividual demonstrated generalization to a novelcontext, and this was best maintained for the behaviorspracticed in random order.Number of trials during practice: In the motorlearning literature, high amounts of practice often referto 50 or more repetitions of a given action per trainingsession. Learning is clearly facilitated by a high numberof practice trials. This has been reiterated in a series ofrecent studies in the physical therapy literature examin-ing constraint-induced therapy and intensive practice(Kunkel et al., 1999; Liepert, Bauder, Miltner, Taub,& Weiller, 2000; Miltner, Bauder, Sommer, Dettmers,& Taub, 1999; Taub et al., 1993). Here, individualswith a hemiparetic limb subsequent to CVA undergo 6to 8 hours of therapy a day for a 2-week period.Therapy involves constraining the intact limb so that theindividual is forced to use only the impaired limb.Improvements in function have been far greater thanwhen receiving the same total time of therapy spreadover sessions of 15 minutes a day. However, therelative contributions of the intensive practice versusthe constraint procedure have not been well differenti-ated. One study has adapted intensive constraint-induced therapy to language therapy for aphasia withsimilar findings (Pulvermüller et al., 2001).

Importantly, the studies above provided intensive

practice but presented a range of tasks over each day.Shea and Kohl (1990) observed that high numbers ofpractice trials may interact with others principles ofmotor learning. They trained individuals to generatespecific levels of force. One group received “constant”practice at a single “criterion” force level (100 trials). Asecond group received “varied” practice where theypracticed the criterion task (100 trials) and force levelssurrounding the criterion (100 trials each). A thirdgroup received constant practice but produced thesame total number of trials as the varied group. Acqui-sition performance was lowest and retention highest forthe variable practice group. The criterion groupreceiving the higher number of trials performed with thehighest accuracy during practice, but demonstratedpoorest retention. In the context of practice on a singletask, or task level, it is possible to provide too manytrials so that long-term learning is degraded.

Principles of Feedback StructureAnother group of principles that influence motor

learning relates to the type, amount, and timing offeedback during practice.Type of feedback: Schmidt and Lee (1999) discusstwo primary types of feedback: knowledge of perfor-mance (KP) and knowledge of results (KR). KPfeedback provides information about how an action isperformed (e.g., “Bring your tongue further forward”),while KR feedback merely indicates whether or not aresponse is acceptable. KP can be provided in theform of verbal comments or visual feedback onkinematics (i.e., position, timing, velocity, and coordi-nation) or kinetics (i.e., force) of movement throughinstrumentation. Kinematic feedback is frequently usedto reveal characteristics of a movement that theindividual is unable to perceive unaided, in which caseit can facilitate learning. However, when the goal of thetask is clear, kinematic KP feedback may provide noadditional benefit over the instructor simply providingKR (Swinnen, Walter, Lee, & Serrien, 1993). Kineticfeedback has been used in sports to train sprinters tominimize starting times in races, for example. Broker,Gregor, and Schmidt (1993) demonstrated that thisform of feedback can have long-lasting effects on skilllevel.

In treatments for AOS, the influence of kinematicfeedback has been tested using electropalatography

19


(Howard & Varley, 1995), and electromagneticarticulography (EMA; Katz, Bharadwaj, & Carsten,1999). Katz and colleagues presented a treatmentstudy with a 63-year-old individual with Broca’saphasia and AOS. They used EMA to provide visualfeedback on tongue tip position during silent articulatormovement, movement during humming, a nonwordtask, and a real word task designed to improveplacement for /s/ and /S/. Five treatment sessions wereprovided. Improvements were observed for trainedbehaviors, but were greater for the nonspeech andnonword than the speech task. Retention examined 10weeks post-treatment was difficult to interpret. Provid-ing kinematic feedback to individuals with AOS hasintuitive appeal and further research is well-justified.Use of kinetic feedback has received more attention indysarthria and dysphagia treatments (e.g., Dworkin,1991; Stierwalt & Robin, 1996) than in AOS.

Schmidt and Lee (1999) argue that KP feedbackis beneficial to performance during practice, but ismost beneficial to retention when the informationprovided is not obtainable from other inherent sources.It has also been suggested that KP is most useful in theearly stages of training, but KR should become thepredominant type of feedback to encourage accurateself-evaluation of performance and independence inperformance of actions. The use of KR feedback,which has received a good deal of attention, clearly isimportant in the learning pro-cess. Some aspects that havebeen studied include the fre-quency and the timing of KRfeedback.Frequency of KR feedback:Schmidt and colleagues haveperformed a series of studiesinvestigating the effect of KRfeedback frequency (seeSchmidt & Lee, 1999, for areview). When the number oftrials was held constant and theultimate goal was to producethe actions without KR, highfrequency feedback (i.e., on100% of trials) resulted in morerapid acquisition, but poorerretention of trained limb motor

behaviors than low frequency feedback (i.e., 30-60%of trials; e.g., Wulf, Schmidt, & Deubel, 1993). In thenormal speech motor system, Adams and Page (2000)have also reported greater retention when feedback ispresented at a low frequency. Schmidt and colleagues(Salmoni, Schmidt, & Walter, 1984; Schmidt, 1991;Schmidt & Bjork, 1992) argued that higher feedbackfrequency, particularly during later stages of acquisi-tion, may interfere with self-evaluation processes andencourage dependence on external sources of feed-back. Consequently, treatment effects will not transferto the more naturalistic no-KR condition typically usedin retention tests. It may also over-shape a behaviorresulting in artificially reduced variability and thus lossof flexibility with contextual changes. Low frequencyfeedback guides an individual’s understanding ofacceptable versus unacceptable responses while alsodeveloping independence (Schmidt & Lee, 1999).Timing of KR feedback: Schmidt and Lee (1999)argued that a brief delay of 3-4 seconds between anindividual’s response and the provision of KR feed-back, and between feedback and presentation of thenext stimulus, facilitates both acquisition and retentionof trained limb actions. When feedback is providedimmediately, it may interfere with self-evaluation. Oncefeedback on a response is provided, a brief delayallows an individual to analyze the previous movement,compare this with the received feedback, and then

1Similar effects reported in a study/studies of speech motor learning.2KR refers to knowledge of results.

Table 1. Influence of select principles of motor learning on acquisitionand retention of limb motor skills.

Principles that Facilitate

Acquisition

Principles that Facilitate

Long-Term Retention

Structure of Practice

Order of trials Blocked order Random order1

Number of trials High (> 50) High (> 50) 1

Structure of KR2 Feedback

Frequency of feedback

High frequency

(100% of trials)

Low frequency1

(30-60% of trials)

Timing of feedback Delayed 3-4 sec Delayed 3-4 sec

20


plan the next movement (e.g., Swinnen, Schmidt,Nicholson, & Shapiro, 1990; also see Salmoni et al.,1984, for a review).

Transfer Between Part and WholeTask Components

Another category of principles addresses thepart-whole relationship between tasks. It is the ten-dency in speech therapy, and other fields of motorlearning, to work on less complex tasks that are easilystimulable, prior to targeting more complex tasks (e.g.,Dworkin, 1991, Nuffield Centre DyspraxiaProgramme, 1985). The easier tasks may be referredto as “lead-up activities” (Schmidt & Lee, 1999), andit is assumed that these represent a part of the whole,more complex, task. Schmidt and Lee noted that theaction in the lead-up task may be altered when per-formed in the context of the more complex wholeaction, and so minimal or no transfer may occur. Thisgeneral hypothesis is supported by work in anotherfield of study—treatment of linguistically-based com-munication disorders (syntax, Ballard & Thompson,1999; Thompson, Ballard, & Shapiro, 1998; seman-tics, Plaut, 1996, Kiran, 2001; phonology, Gierut,1998). These reports have demonstrated that training askill results in transfer to theoretically related skills thatare of lesser complexity, but not to related skills ofgreater complexity or to unrelated skills. In the speechmotor system, one could argue that a given movementpattern for a speech skill is more complex than asimilar nonspeech movement pattern. Thus, trainingnonspeech movement patterns should not result inimproved motor control for related, but more complex,speech movement patterns. Studies by Katz et al.(1999) and Howard and Varley (1995) with AOSseem to support this proposal. Furthermore, Powell,Elbert, and Dinnsen (1991) studied childhood phono-logical disorders and reported that greater transfer tountreated phonological targets is observed when moredifficult, or non-stimulable, behaviors are targeted firstin treatment, rather than stimulable targets (stimulablehere refers to sounds that a child can produce whengiven an auditory-visual model). While it may seemunintuitive to begin therapy at a more complex level,where the client will experience greater difficulty,research suggests that this promotes transfer.

ConclusionsWhile there is considerable support for using

specific principles of motor learning to train novelmovement skills in the limbs, the evidence for thespeech system is sparse. The few studies availablesuggest that retention is enhanced by random order ofpractice (Adams & Page, 2000; Knock et al., 2000),low frequency KR feedback (Adams & Page, 2000),and a high number of trials during each practicesession. These conditions may render the learning taskmore difficult and so acquisition may be less rapid. Ifclinicians choose to apply principles of motor learningto treatment of AOS, they are cautioned that theeffects have been studied primarily in different motorsystems and the one study of AOS cited here (Knocket al., 2000) included only two subjects. Currently,there is insufficient data to state confidently that thespeech motor system will respond to these principles ina similar way to the limb system.Dr. Kirrie Ballard is an assistant professor in theDepartment of Speech Pathology and Audiology atthe University of Iowa.

ReferencesAdams, S. G., & Page, A. D. (2000). Effects of selected practice and

feedback variables on speech motor learning. Journal ofMedical Speech-Language Pathology, 8, 215-220.

Ballard, K. J., & Thompson, C. K. (1999). Treatment and gener-alization of complex sentence production in agramma-tism. Journal of Speech, Language, and Hearing Research, 42,690-707.

Battig, W. F. (1979). The flexibility of human memory. In L. S.Cermak & F. I. M. Craik (Eds.), Levels of processing inhuman memory (pp. 23-44). Hillsdale, NJ: Erlbaum.

Broker, J. P., Gregor, R. J., & Schmidt, R. A. (1993) . Extrinsicfeedback and the learning of kinetic patterns in cycling.Journal of Applied Biomechanics, 9, 111-123.

Duffy, J. R. (1995) Motor speech disorders: Substrates, differentialdiagnosis and management. St. Louis, MO: Mosby.

Dworkin, J. P. (1991). Motor speech disorders: A treatment guide.St. Louis, MO: Mosby.

Gierut, J. A. (1998). Treatment efficacy: Functional phonologicaldisorders in children. Journal of Speech, Language, andHearing Research, 41, S85-S100.

Goode, S., & Magill, R. A. (1986). The contextual interferenceeffects in learning three badminton serves. Research Quar-terly for Exercise and Sport, 57, 308-314.

Hanlon, R. E. (1996). Motor learning following unilateral stroke.Archives of Physical Medicine and Rehabilitation, 77, 811-815.

21


Howard, S., & Varley, R. (1995). Using electropalatography totreat severe acquired apraxia of speech. European Journalof Disorders of Communication, 30, 246-255.

Katz, W. F., Bharadwaj, S. V., & Carstens, B. (1999). Electro-magnetic articulography treatment for an adult withBroca’s aphasia and apraxia of speech. Journal of Speech,Language, and Hearing Research, 42, 1355-1366.

Kiran, S. (2001). Effect of exemplar typicality on naming treatmentin fluent aphasia. Unpublished doctoral dissertation. North-western University, Evanston, IL.

Knock, T. R., Ballard, K. J., Robin, D. A., & Schmidt, R. (2000).Influence of order of stimulus presentation on speechmotor learning: A principled approach to treatment forapraxia of speech. Aphasiology, 14, 653-668.

Kunkel, A., Kopp, B., Müller, G., Villringer, K., Villringer, A.,Taub, E., & Flor, H. (1999). Constraint-induced move-ment therapy for motor recovery in chronic strokepatients. Archives of Physical Medicine and Rehabilitation, 80,624-628.

Liepert, J., Bauder, H., Miltner, W., Taub, E., & Weiller, C. (2000).Treatment-induced cortical reorganization after stroke inhumans. Stroke, 31, 1210-1216.

McNeil, M. R., Robin, D. A., & Schmidt, R. A. (1997) Apraxia ofspeech: Definition, differentation, and treatment. In M. R.McNeil (Ed.), Clinical management of sensorimotor speechdisorders (pp. 311-344). New York: Thieme.

Miltner, W. H. R., Bauder, H., Sommer, M., Dettmers, C. & Taub,E. (1999). Effects of constraint-induced movement therapyon patients with chronic motor deficits after stroke: Areplication. Stroke, 30, 586-592.

Nuffield Speech Therapists (1985). Nuffield Centre DyspraxiaProgramme. Royal National Throat, Nose, and Ear Hos-pital Hearing and Speech Centre, London, UK.

Plaut, D. C. (1996). Relearning after damage in connectionistnetworks: Toward a theory of rehabilitation. Brain andLanguage, 52, 25-82.

Powell, T. W., Elbert, M., & Dinnsen, D. A. (1991). Stimulabilityas a factor in the phonologic generalization ofmisarticulating preschool children. Journal of Speech andHearing Research, 34, 1318-1328.

Pulvermüller, F., Neininger, B., Elbert, T., Mohr, B., Rockstroh,B., Koebbel, P., & Taub, E. (2001). Constraint-inducedtherapy of chronic aphasia after stroke. Stroke, 32, 1621-1626.

Salmoni, A. W., Schmidt, R. A., & Walter, C. B. (1984). Knowl-edge of results and motor learning: A review and criticalreappraisal. Psychological Bulletin, 95, 355-386.

Schmidt, R. A. (1991). Frequent augmented feedback can de-grade learning: evidence and interpretations. In J. Requin& G. E. Stelmach (Eds.), Tutorials in motor neuroscience (pp.59-75). Dordrecht: Kluwer Academic Publishers.

Schmidt, R. A., & Bjork, R. A. (1992). New conceptualizations ofpractice: Common principles in three paradigms suggestnew concepts for training. Psychological Science, 3, 207-217.

Schmidt, R. A., & Lee, T. D. (1999). Motor control and learning: Abehavioral emphasis (3rd ed.). Champaign, IL: HumanKinetics.

Shea, C. H., & Kohl, R. M. (1990). Specificity and variability ofpractice. Research Quarterly for Exercise and Sport, 61, 169-177.

Shea J. B., & Morgan, R. L. (1979). Contextual interference effectson the acquisition, retention, and transfer of a motor skill.Journal of Experimental Psychology: Human Learning andMemory, 5, 179-187.

Stierwalt, J., & Robin, D. A. (February, 1996). Treatment ofstrength in a person with TBI. Paper presented at theConference on Motor Speech, Amelia Island, Florida.

Swinnen, S., Schmidt, R. A., Nicholson, D. E., & Shapiro, D. C.(1990). Information feedback for skill acquisition: Instan-taneous knowledge of results degrades learning. Journalof Experimental Psychology: Learning, Memory, and Cogni-tion, 16, 706-716.

Swinnen, S. P., Walter, C. B., Lee, T. D., & Serrien, D. J. (1993).Acquiring bimanual skills: Contrasting forms of informa-tion feedback for interlimb decoupling. Journal of Experi-mental Psychology: Learning, Memory, and Cognition, 19,1328-1344.

Taub, E., Miller, N. E., Novack, T. A., Cook, E. W., Fleming, W.C., Nepomuceno, C. S., Connell, J. S., & Crago, J. E. (1993).Technique to improve chronic motor deficit after stroke.Archives of Physical Medicine and Rehabilitation, 74, 347-354.

Thompson, C. K., Ballard, K. J., & Shapiro, L. P. (1998). The roleof syntactic complexity in training wh-movement struc-tures in agrammatic aphasia: Optimal order for promot-ing generalization. Journal of the International Neuropsy-chological Society, 4, 661-674.

Van Riper, C., & Irwin, J. (1958). Voice and articulation. EnglewoodCliffs: Prentice Hall.

Wambaugh, J. L., & Doyle, P. J. (1994). Treatment for acquiredapraxia of speech: A review of efficacy reports. ClinicalAphasiology, 22, 231-243.

Wulf, G., Schmidt, R. A., & Deubel, H. (1993). Reduced feedbackfrequency enhances generalized motor program learn-ing but not parameterization learning. Journal of Experi-mental Psychology: Learning, Memory, and Cognition, 19,1134-1150.

22


23


Oral Motor Training and Treatment for Apraxia of SpeechLinda I. ShusterWest Virginia University, Morgantown

The relationship between speech abilities andnon-speech oral abilities is unclear and has beendebated for many years (e.g., Folkins et al., 1995;Weismer, 1997; Weismer & Liss, 1991). Because thesame anatomical structures are involved in both speechand non-speech movements, it is not surprising thatinvestigators and clinicians have speculated that there isa relationship between the two. However, studies ofnormal speakers, as well as individuals with neuromo-tor speech problems, have yielded inconclusive results.

Describing the nature of the relationship betweenapraxia of speech (AOS) and non-speech oral abilitiesis particularly controversial, because there is notuniversal agreement on the underlying basis for AOS.While most investigators and clinicians consider AOSto be a disorder of speech motor control (Brookshire,1973; Duffy, 1995; McNeil, Robin, & Schmidt,1997), others, such as Dogil and Mayer (1998) haveproposed that the deficit in AOS is at a linguistic level.If AOS is a linguistic problem, and not a disorder ofspeech motor control, then treatment of AOS, intheory, should not directly target speech productionprocesses or non-speech oral movements.

Several studies have investigated non-speech oral abilities in AOS (Clark & Robin,1998; Hageman, Robin, Moon, & Folkins, 1994;Robin, Bean, & Folkins, 1989; McNeil,Weismer, Adams, & Mulligan, 1990). In some ofthese studies, the subjects with AOS displayedproblems with non-speech movements, while inothers they did not.

The EvidenceRobin and colleagues (1989) used a strain gauge

system attached to the lips to study coordination of thelips and peak articulatory velocity of the lower lip inspeakers with AOS. They used speech tasks as wellas a non-speech task; however, only peak velocitywas measured for the non-speech task. The investiga-tors hypothesized that individuals with AOS mightdemonstrate lower peak velocities (slower move-ments) for lower lip movements, perhaps due to a

weakness in that musculature. However, they foundthat the highest peak velocities generated by all of thespeakers with apraxia were higher than the highestpeak velocity generated by the normal control subject,which did not support their hypothesis.

McNeil and colleagues (1990) were interested inthe effect of neurogenic communication disorders onthe ability to control the speech apparatus. Theycompared speakers with AOS to three other groups ofsubjects: individuals with conduction aphasia, individu-als with ataxic dysarthria, and a group of normalcontrols. Based on both the normative data on speechmotor control and on the typical clinical assessments ofnon-speech oral competencies, they chose to study theability of the subjects to exert fine force as well as theability to find and hold a static position with the follow-ing articulators: the upper and lower lips, the tongue,and the jaw. McNeil and colleagues found that partici-pants with AOS demonstrated significantly greaterinstability in orofacial isometric force and static positionthan did their normal subjects. However, their partici-pants with dysarthria also displayed similar instability,and the data from their participants with conductionaphasia were equivocal. Therefore, these measures didnot clearly differentiate the three diagnostic groups.

Hageman and colleagues (1994) comparedindividuals with AOS to normal speakers on theperformance of predictable and non-predictable oralmotor tracking tasks. The articulators that they investi-gated were the lower lip, jaw, and the larynx (i.e.,fundamental frequency of sustained phonation). Thepredictable targets were sinusoidal waveforms withfrequencies of 0.3, 0.6, and 0.9 Hz, while the non-predictable target was a complex waveform composedof 10 frequencies of equal amplitude. The 10 frequen-cies ranged from 0.1 to 1.0 Hz in 0.1 Hz steps. Theseinvestigators found that the subjects with AOS per-formed similarly to the normals on the unpredictabletracking task; however, they performed more poorlythan the normals on the predictable tasks. In particular,the movements of the subjects with AOS tended to lagbehind the target on the predictable tasks. Hageman

24


and colleagues proposed three possible interpretationsfor these results: (a) individuals with AOS are unable todevelop a motor plan; (b) individuals with AOS areable to develop a plan, but the plan is poorly definedor inaccurate; and (c) a plan is developed, but theindividual with AOS is unable to access that plan.However, the authors acknowledge that: “It is appar-ent from these data that there are no easy answers tohelp us understand the mechanisms underlying apraxiaof speech” (p. 227).

Clark and Robin (1998) found that individualswith AOS demonstrated variability in certain aspects ofmotor programming, in particular generalized motorprograms (GMPs) and parameterization of thoseprograms. A GMP is a “fundamental movementpattern” (Wulf & Schmidt, 1996) that underlies avariety of different motor behaviors. Parameterizationis used to tailor the GMP for a particular action, forexample, through the timing or the amplitude of themovement. Using non-speech tasks, Clark and Robinfound that 2 of their 4 subjects with AOS demon-strated poor parameterization accuracy and unim-paired generalized motor program (GMP) accuracy,while the other two demonstrated the opposite pattern:good parameterization, but impaired GMP accuracy.However, the investigators also studied individuals withconduction aphasia (CA), and some of these subjectsalso demonstrated some difficulties with parameteriza-tion, although they displayed generally accurate GMPs.The authors suggested that their results may be evi-dence for different subtypes of AOS. That is, onesubtype may have difficulty in developing or executinga GMP while another subtype may have intact GMPs,but may be unable to accurately parameterize thoseGMPs.

SummaryIn summary, a relationship between non-speech

oral abilities and AOS has not been demonstratedconsistently. While some studies have found speakerswith AOS to have an impairment of non-speech oralabilities, others found no difference between theirnormal subjects and those with AOS on non-speechmeasures. In addition, there was often no clear differ-entiation between different neurogenic communicationdisorder populations on these behaviors. For example,McNeil and colleagues’ (1990) subjects with ataxic

dysarthria displayed fine force and isometric positioninstability similar to their subjects with AOS. In Clarkand Robin’s study (1998), 2 of the participants withconduction aphasia demonstrated a pattern of impair-ment that was similar to 2 of the subjects with AOS(i.e., accurate GMPs, but problems in parameteriza-tion). As a result, no non-speech oral impairments havebeen identified that definitively characterize this disor-der. More importantly, there is very little informationregarding the relationship between impaired non-speech oral skills and the speech impairment of AOS.Clark and Robin, for example, noted the need foradditional studies that would investigate parameteriza-tion accuracy and the intactness of GMPs for bothspeech and non-speech tasks.

So what are the implications of these findings forthe treatment of AOS? Regarding AOS treatment,Brookshire (1973) stated: “we see a concern foradministering treatment which is directed toward thepatient’s ‘key’ areas of deficit and not to ancillaryareas…” (p. 192-193). It is obvious that we needfurther evidence regarding which non-speech oralmovement deficits in AOS are key. Importantly, wehave been unable to locate any published studies onthe use of non-speech oral training for AOS. However,despite the lack of data to support such an approach,perusal of some of the publications targeted towardspeech-language pathologists reveals that severalindividuals are offering workshops on oral motortherapy, and adults with AOS are included in the list ofindividuals who might benefit from this treatment. Inaddition, there are books and therapy materials in printthat describe the use of non-speech oral motor trainingfor AOS. This training includes tasks such as passivemovement of the patient’s articulators by the clinician,sensory stimulation of the articulators, and the repeatedproduction of various non-speech movements such astongue protrusion and lip retraction.

Given the constraints on the number of treatmentsessions imposed by many third-party payers, it iscritical that clinicians make optimal use of therapy time.In light of the data that are currently available, it seemspremature for clinicians to use clinical time to treatnon-speech oral behaviors in the absence of efficacydata when there are treatment studies documenting theeffectiveness of various approaches which use speechtasks (e.g., see other papers in this issue). Future

25


investigations should continue to explore non-speechoral behaviors and their relationship to speech in orderto determine which of these behaviors are key factorsin the underlying basis for AOS. Once these behaviorsare identified, investigations should be designed toexplore treatments that will improve the communicativeeffectiveness of individuals with AOS.Linda I. Shuster, PhD, CCC-SLP is an AssociateProfessor in the Department of Speech Pathologyand Audiology at West Virginia University.

ReferencesBrookshire, R. H. (1973). An introduction to aphasia. Minne-

apolis: BRK Publishers.

Clark, H. M., & Robin, D. A. (1998). Generalized motorprogramme and parameterization accuracy in apraxiaof speech and conduction aphasia. Aphasiology, 12, 699-713.

Dogil, G., & Mayer, J. (1998). Selective phonological impair-ment: A case of apraxia of speech. Phonology, 15, 143-188.

Duffy, J. R. (1995). Motor speech disorders: Substrates, differen-tial diagnosis, and management. Baltimore: Mosby.

Folkins, J. W., Moon, J. B., Luschei, E. S., Robin, D. A., Tye-Murray, N., & Moll, K. L. (1995). What can non-speechtasks tell us about speech motor disabilities? Journal ofPhonetics, 23, 139-147.

Hageman, C. F., Robin, D. A., Moon, J. B., & Folkins, J. W.(1994). Oral motor tracking in normal and apraxicspeakers. Clinical Aphasiology, 22, 219-229.

McNeil, M. R., Robin, D. A., & Schmidt, R. A. (1997). Apraxiaof speech: Definition, differentiation, and treatment. InM. R. McNeil (Ed.), Clinical management of sensorimotorspeech disorders (pp. 311-344). New York: Thieme.

McNeil, M. R., Weismer, G., Adams, S., & Mulligan, M. (1990).Oral structure nonspeech motor control in normal,dysarthric, aphasic, and apraxic speakers: Isometricforce and static position control. Journal of Speech andHearing Research, 33, 255-268.

Robin, D. A., Bean, C., & Folkins, J. W. (1989). Lip movementin apraxia of speech. Journal of Speech and HearingResearch, 32, 512-523.

Weismer, G. (1997, March 12). Assessment of oromotor, nonspeechgestures in speech-language pathology: A critical review(Telerounds 33). Tucson, AZ; National Center forNeurogenic Communication Disorders.

Weismer, G., & Liss, J. M. (1991). Reductionism is a dead-endin speech research: Perspectives on a new direction. InC. A. Moore, K. M. Yorkston, & D. R. Beukelman (Eds.),Dysarthria and apraxia of speech: Perspectives on manage-ment (pp. 15-27). Baltimore: Paul H. Brookes Publish-ing Co.

Wulf, G., & Schmidt, R. A. (1996). Average KR degradesparameter learning. Journal of Motor Behavior, 28, 371-381.

26


27


Treatment Research on Augmentative and AlternativeCommunication for Adults With Apraxia of Speech

Margaret A. RogersUniversity of Washington, Seattle

The typical constellation of symptoms amongindividuals with apraxia of speech (AOS) makes themexcellent candidates for augmentative and alternativecommunication (AAC). When present in relativeisolation, AOS is a modality-specific disorder disrupt-ing speech motor programming, sparing communica-tion through other language modalities. Even whenAOS is concomitant with aphasia, as is typically thecase, many of the affected individuals have only mild tomoderate impairments of auditory comprehension,reading, and writing, thus enabling these modalities toaugment spoken communication. Unfortunately, thereis a paucity of research in the application of AAC withthis population. The neglect of this area is regrettablebecause most individuals with moderate to severeAOS may have the communication and cognitivecompetencies to allow them to use AAC to communi-cate more effectively than with their impaired speech.While there is insufficient evidence to confirm thisproposal, the likelihood of its veracity mandates thatAAC be considered a central component of clinicalpractice with this population.

This article is intended to review the extantresearch on the use of AAC among individuals withAOS, to present a rationale concerning why AACshould be a central component of clinical practice withthis population, and, lastly, to make suggestionsregarding methods that, if applied, would greatlyenhance the value of future research regarding theeffectiveness and social validity of using AAC ap-proaches with individuals with AOS. Accordingly, thisarticle is intended to encourage clinicians to provideAAC options to their patients with AOS and toincrease the awareness of the research communityconcerning the need to investigate the effectiveness andsocial validity of AAC approaches for individuals withAOS.

The Four EsResearch regarding the benefits of treatment may

concern the efficacy, effectiveness, effects (or out-

come), and/or efficiency of interventions (Kendall &Norton-Ford, 1982). As discussed by Robey (2001),these four terms are important to differentiate whenconsidering the contribution of specific studies to thetreatment research literature. Accordingly, a briefreview is provided because an important conclusion ofthis literature review is that studies providing directevidence of the benefits of AAC for individuals withAOS are almost entirely lacking.

Efficacy and effectiveness both refer to thedemonstration of behavioral changes as a direct resultof treatment. The term efficacy indicates the possibilityof a treatment’s benefit to individuals in a definedpopulation when the treatment is applied under idealconditions, whereas effectiveness refers to possiblebenefits when the treatment is applied under averageconditions, such as during typical service deliveryconditions. Efficiency refers to the demonstration ofcomparative effectiveness of two or more interventionsand typically entails a comparison of two treatments interms of time or number of sessions required to meetcriterion, number of trials, treatment cost, or effort(Schlosser, 1999). Treatment effects or outcomesrefer to the demonstration of behavioral changes overthe time period associated with the application of atreatment, though they may not be directly caused bythe treatment. For example, treatment outcome studiesmay demonstrate treatment effects by comparing pre-and post-treatment measures, but, unlike studiesdesigned to demonstrate efficacy or effectiveness, theydo not include probe measures that sample targetbehaviors repeatedly throughout the course of treat-ment. Accordingly, treatment outcome studies do notcontrol for the potential contribution of extraneousfactors that can affect the results and subsequentlyconfound internal validity. Thus, outcome researchdesigns, such as case studies, do not provide directevidence for the efficacy, effectiveness, or efficiency ofa treatment. Rather they simply indicate that changeshave or have not occurred between two points in time.Case studies do offer important data, especially in theearly phases of collecting information about treatment

28


effects (Robey & Schultz, 1998). The strength of theinferences that can be drawn from a treatment studydepends on the level of experimental control, thedegree to which the participants and treatment meth-ods are adequately described, and the ways in whichthe consequences of treatment are measured(Chambless & Hollon, 1998 as cited by ANCDS adhoc Practice Guidelines Coordinating Committee,2001). For example, studies that include periodicmeasurement during the treatment phase and includemeasures of maintenance and generalization offerstronger evidence than those that evaluate the conse-quences of treatment based solely upon theparticipant’s end-of-treatment performance on the setof trained items. Lastly, studies that also obtain infor-mation concerning patient and/or caregiver satisfactionwith the treatment in terms of its influence on theparticipant’s quality of life contribute an importantdimension to treatment research.

Social ValidityThe nature of the evidence provided by treatment

research is not exhaustively described by the four Es.The perceived value of changes achieved throughtreatment is another important variable to consider(Kazdin, 1977). The social validity of a treatmentconcerns both a social perspective (members ofsociety) and a personal perspective (patients, family,caregivers). To evaluate the social validity of anintervention, ideally both the individuals with AOS andthose with whom they communicate would providefeedback concerning whether they value the changesthat have occurred. The communication skills selectedfor intervention and the effect on communication mustbe valued by the individuals with the communicationdisorder as well as those with whom they communicateand must be viewed as beneficial to the goal of inter-acting more effectively with others (Calculator, 1988).

Apraxia of SpeechOne of the phrases frequently used in defining the

four Es is “individuals in a defined population.” Thisphrase points to one of the most persistent obstacles inevaluating the AOS treatment literature. None of thestudies cited in Table 1 (on page 30) provide specificinformation concerning how the determination of AOSwas made nor even a definition of AOS as the basis

for making the clinical diagnosis. In fact, articles wereselected to be included in Table 1 on the basis of avery broad set of inclusionary criteria that included anyof the following descriptors: Broca’s aphasia, non-fluent aphasia, agrammatic speech, verbal apraxia,apraxia of speech, telegraphic speech, repeatedattempts to produce target words, groping and fre-quent or predominant production of phonemicparaphasias. No exclusionary criteria were used.Clearly, some of these descriptors could equally applyto individuals with Transcortical Motor aphasia (with-out AOS), to Broca’s aphasia (without AOS), or toConduction aphasia (without AOS). However, it wasdeemed necessary to cast a wide net for this prelimi-nary analysis of the AAC/AOS literature despite thefact that the validity of the inferences that can be drawnare certainly compromised by the lack of nosologicspecificity. To enable generalization and application oftreatment research to “individuals in a defined popula-tion,” it is necessary to define and describe the re-search participants with a fair amount of precision anddetail. Research pertaining to AOS has always sufferedfrom a lack of generally accepted diagnostic criteria,and this continues to be one of the most significantmethodological issues impeding treatment research inthis area.

Because the population of individuals with AOSis heterogeneous, in addition to reporting the diagnosticcriteria, it is also important that researchers describethe participants’ strengths and additional impairments.For example, it would be very important to report thepresence of non-linguistic sequelae such as limbapraxia, hemiparesis, visual field cuts, initiation difficul-ties, and reduced cognitive flexibility. Because AACinvolves using multiple modalities to communicate,impairments that can affect pointing, gesturing, panto-miming, carrying devices, and using the visual modalityfor reading, writing, and drawing would greatly influ-ence both the choice of AAC options and the outcomeof intervention. Individuals with initiation difficulties andreduced cognitive flexibility present particular chal-lenges as often they do not exhibit generalization andrequire much prompting to use and switch amongdifferent AAC options. Almost all individuals with AOShave a concomitant aphasia. Accordingly, their abilityto read, write, and comprehend spoken language aretypically impaired, however, often to a lesser degree

29


than their ability to speak. Thus, while AOS is amodality-specific disorder, ostensibly disrupting thestages of speech production from phonetic encodingthrough motor programming, individuals with thisdisorder can exhibit an array of other problems(McNeil, Robin, & Schmidt, 1997; Rogers & Storkel,1999; Wertz, LaPointe, & Rosenbek, 1984).

Augmentative and AlternativeCommunication

In the introductory chapter of Beukelman andMirenda’s (1998) seminal book on AAC, they providethree quotes that define communication, AAC, and thepopulation that may benefit from AAC. These quotesare presented below to provide a definitional basis forusing these concepts. The first quote is from theNational Joint Committee for the CommunicativeNeeds of Persons with Severe Disabilities (1992) anddefines communication as:

Any act by which one person gives to or receivesfrom another person information about thatperson’s needs, desires, perceptions, knowledge,or affective states. Communication may be inten-tional or unintentional, may involve conven-tional or unconventional signals, may take lin-guistic or nonlinguistic forms, may occur throughspoken or other modes. (p. 2)

The second quote is from ASHA (1989) anddefines AAC as:

An area of clinical practice that attempts to com-pensate (either temporarily or permanently) forthe impairment and disability patterns of indi-viduals with severe expressive communicationdisorders (i.e., the severely speech-language andwriting impaired). (p. 107)

It is important to emphasize that AAC involves“other modes” of communication (Beukelman &Mirenda, 1998), should always be multimodal innature, and should utilize “the individual’s full communi-cation capabilities, including any residual speech orvocalizations, gestures, signs, and aided communica-tion” (ASHA, 1991, p. 10).

The third quote is also from ASHA (1991) andspecifies the population of individuals that may benefitfrom AAC.

Individuals with severe communication disor-ders are those who may benefit from AAC—those for whom gestural, speech, and/or writtencommunication is temporarily or permanently

inadequate to meet all of their communicationneeds. For those individuals, hearing impair-ment is not the primary cause for the communi-cation impairment. Although some individualsmay be able to produce a limited amount ofspeech, it is inadequate to meet their variedcommunication needs. (p. 10)

Individuals with AOS have great potential tobenefit from AAC as their speech impairment istypically moderate to severe and inadequate to meettheir communication needs. The degree to whichindividuals have concomitant impairments disruptingtheir ability to use language in other modalities willaffect the utility of some AAC options, but should notpreclude these individuals from using AAC altogether.The presumed interaction among such factors as theuser’s impairment profile, communication needs,personality and affective state, the availability ofcaregiver support for communication, and the type ofAAC intervention is important to consider whendesigning treatment studies and planning intervention.

Comprehensive CommunicatorsGiven the heterogeneity of the population of

individuals with AOS, different intervention approacheswill be required to meet their needs and to accommo-date the unique array of strengths and impairments ofeach individual. Despite the heterogeneity, manyindividuals with AOS can use orthographic displays ofsingle words and even short phrases. Many can makegood use of gesture and pantomime. Often theseindividuals can either point to or write the initial seg-ments of words that they cannot produce verbally.Many also have the cognitive capabilities to allow useof complex systems, such as computer-based commu-nication devices. Communication notebooks requiringthe ability to locate items by category and/or alphabeti-cally are often the cornerstone of AAC approacheswith these individuals. Based on the classificationscheme developed by Garrett and Beukelman (1992)for individuals with severe aphasia, the majority ofindividuals with AOS would be classified as compre-hensive communicators. Comprehensive communi-cators retain a variety of communication skills, such asthe ones just listed. They want to communicate withmore than one partner and in more than one setting.They initiate interaction and demonstrate some prag-matic competence in discourse (Garrett & Beukelman,

30


Table 1. Publications concerning the application of augmentative and alternative communication to individu-als with probable apraxia of speech.

AOS = apraxia of speech

Citation Treatment Design[type of evidence](# of participantswith AOS)

Conclusion

Skelly, Schinsky,Smith,& Fust, 1974

paired Amer-Ind training with speech case study[effects](n=3)

Amer-Ind gestures facilitate speechproduction

Rabidoux , Florance,& McCauslin, 1980

Handi Voice case study[effects](n=1)

aided communication facilitatescommunication success

Dowden, Marshall,& Tompkins, 1981

Amer-Ind training case study[effects](n=2)

gestures may be acquired; mayfacilitate speech; gestures as AACmay improve functionalcommunication

Lane & Samples, 1981 Blissymbol training case study[effects](n=4)

individuals with less severe aphasialearned better than individuals withmore severe aphasia

Kearns, Simmons,& Sisterhen, 1982

Amer-Ind training vsAmeri-Ind training paired with speechtraining

Multiple baselineAcross behaviors[efficacy](n=1)

Unimodal training results inacquisition and maint-enance ofAmer-Ind gestures, but faciliation ofverbal responses resulted only withextensive multimodality training

Coelho & Duffy, 1985 manual sign training case study[effects](n=1)

successful acquisition of signs, butnot used by participant in hisinteractions at home

Beukelman, Yorkston,& Dowden, 1985

multiple communication systems:multimodality communication, notebooksto augment auditory comprehension andexpression; voice output communicationaid

case study[effects](n=1)

AAC applied with communicationneeds assessment continuallymonitored throughout interventionproved successful in facilitatingreturn to work

Garrett, Beukelman, &Low-Morrow, 1989

multiple communication systems: speech,writing, drawing, word dictionarytopically organized, alphabet card,printed conversational control phrases


training multiple systems andidentification of most efficientsystem in context resulted inimproved conversational initiation,increased conversational turns andfewer communication breakdowns

Bellaire, Georges, &Thompson, 1991

communication notebook training single-subject multiplebaseline design[effectiveness](n=2)

post-treatment, participants able torequest basic items andcommunicate personal information,generalization required extensivetime and message-specific trainingin target context

Yanak & Light, 1991 Amer-Ind, word board, andcommunication notebook training


participant learned to communicateby pointing to stored messages andgeneralized to actual socialsituations

King & Hux, 1995 intervention using Talking WordProcessing Software


spoken computer output enhancedindependent written language skills

31


1998). AAC interventions for comprehensive commu-nicators are complex in that these individuals can use awide variety of techniques such as communicationnotebooks, limited natural speech, gestures, alphabetcards, writing, drawing, maps and other schemata,customized or electronic word dictionaries, andelectronic communication devices with voice output.Training on all useable techniques is warranted. Fur-thermore, a critical aspect of AAC intervention forcomprehensive communicators is the need for guidedpractice concerning when to use which technique andto learn to switch modalities or techniques based onthe situation or when a communication breakdown isencountered (Garrett & Beukelman, 1998). Trainingalso needs to occur in natural communication contextswith a variety of partners to facilitate generalization. AsGarrett and Beukelman (1998) indicate, failure toteach and train in naturalistic situations is one of themost common reasons comprehensive communicatorsfail to use their AAC systems effectively.

Intervention StudiesTable 1 displays a list of studies that concern the

application of AAC to individuals with AOS based onthe set of descriptors mentioned above. The listcontains 12 peer-reviewed articles mostly published intwo journals: Augmentative and Alternative Com-munication and Aphasiology (also Clinical Aphasio-logy or Proceedings of the Clinical AphasiologyConference). There are five studies that includedcommunication boards or notebooks as a key compo-nent of intervention (Bailey, 1983; Bellaire, Georges,& Thompson, 1991; Beukelman, Yorkston, &Dowden, 1985; Garrett, Beukelman, & Low-Morrow,1989; Yanak & Light, 1991). There are five studiesthat report training of manual gestural systems: manualsign (Coelho & Duffy, 1985) and Amer-Ind (Dowden,Marshall, & Tompkins, 1981; Kearns, Simmons, &Sisterhen, 1982; Skelly, Schinsky, Smith, & Fust,1974; Yanak & Light, 1991). Two articles reporttraining voice output communication aids (Beukelmanet al., 1985; Rabidoux, Florance, & McCauslin,1980).

While none of the 12 studies directly addressefficiency, all of them make reference to the extensivetraining and in some cases, assessment procedures,required to incur acquisition and/or generalization of

the AAC method. For example, the study by Bellaireand colleagues (1991) used a single-subject multiplebaseline design to investigate the use of picture com-munication boards by two individuals with severeBroca’s aphasia. Their investigation demonstrated thattraining resulted in both participants being able to usethe board for making requests (e.g., coffee andcookies) and personal responses (e.g., name and prioroccupation), but that generalization outside of thetreatment room required extensive training (greaterthan 23 sessions). Furthermore, attempts to promotegeneralization using role-playing techniques in thetreatment room did not result in the participantsspontaneously using the communication board duringcoffee hour at the nursing home (the context chosen toassess generalization). Generalization training had tooccur in the natural settings in which the users wouldcommunicate. Training during coffee hour involvedproviding verbal cues to use the board, modeling useof the board, and physical assists when necessary.Neither participant exhibited response generalization tountrained items, but both demonstrated successfulacquisition and maintenance of trained items represent-ing various objects and actions. This study also indi-cated that training and generalization of social re-sponses may be more complicated than trainingcommunication for the purposes of transaction, such asrequesting specific information or objects.

All but two of the studies, Bellaire and colleagues(1991) and Kearns and colleagues (1982), consist ofcomprehensive case reports. Thus, most of the rel-evant literature provides information concerningtreatment effects and lacks the experimental controlnecessary to provide stronger evidence of treatmentefficacy or effectiveness. Nonetheless, all of these casestudies provide detailed descriptions of the partici-pants’ communication needs and abilities upon whichassessment, AAC development, and training strategieswere based and offer a rich foundation for futureresearch. These pioneering reports are exceedinglyuseful for dissemination of information concerningclinical decision-making and for generating testablehypotheses concerning AAC intervention with individu-als with AOS. However, without experimental control,the impact of these studies is limited.

The study by Yanak and Light (1992) is anexample of a case study that provides detailed de-

32


scriptions of the participant’s communication needsand abilities upon which assessment, AAC develop-ment, and training strategies were based. The 38-year-old individual described in this study had suffered acerebrovascular accident 18 months prior to the onsetof the intervention program. He exhibited severeexpressive deficits with mild auditory and readingcomprehension deficits and limited ability to writeprimarily due to word-finding difficulties. He desired tobe able to communicate with unfamiliar partners, to liveindependently, and to achieve his goal of reemploy-ment. The intervention program consisted of client-based individual sessions designed to teach AACtechniques (Amer-Ind, word board and communica-tion notebook use) and community-based activitiesdesigned to facilitate use and generalization of thesetechniques. The authors describe the process ofvocabulary selection for the communication notebookas entailing the analysis of the demands and opportuni-ties in specific environments (ecological inventories)allowing the user to either validate or discard vocabu-lary items suggested by the clinician (Sigafoos & York,1991). Three types of activities were described tofacilitate interactions: group sessions with two otherclients, simulations of community interactions role-played by the clinician, and actual community-basedinteractions with a variety of unfamiliar partners thatwas initiated 7 weeks into the 12-week program.During the group sessions that continued throughoutthe intervention program, the client was expected topresent problems encountered in daily living andpropose solutions to be considered by the group.Modification of the communication notebook wasincorporated as an ongoing process. This studyprovides useful information concerning the integrationof simulation and interactions outside the clinicalsetting, as a means of both guiding vocabulary selec-tion and promoting generalization. One component thatthe authors concluded would have improved theoutcome of the study was if they had incorporatedpartner or “facilitator” training. The client’s familytended to “pre-empt” (Calculator, 1988) or dominatethe conversational interactions, thereby limiting theamount of time and opportunities the client had toformulate his messages.

Several of the studies indirectly address the socialvalidity of the AAC intervention. For example, the

participant in Bailey’s (1983) study of Blissymboltraining requested that the symbols be removed fromhis communication board and indicated a preferencefor the single word vocabulary. While an indirectmeasure of social validity, this case report indicatesthat the user did not value the Blissymbols. Theparticipant in the Coelho and Duffy (1985) studydemonstrated good acquisition of the manual signs butdid not use them at home. It is difficult to knowwhether this was due to inadequate generalizationtraining, an inability to use the signs without prompting,or because the user did not value the communicationapproach. The uncertainty regarding why generaliza-tion does not always or easily occur underscores theneed for direct measures of social validity to determinewhether user dissatisfaction is a contributing factor.The primary dependent variable in treatment researchis typically the target behavior that is to be changed bythe intervention (e.g., acquisition of an AAC system,increased use of AAC in requesting or commenting,increased initiation or number of conversational turnswith AAC use). It is perhaps equally as important todetermine whether participants value the consequencesof the intervention. Accordingly, social validity mea-sures ought to be included as a dependent variable intreatment research, because the perceived value of anintervention cannot necessarily be inferred from otheroutcome measures.

SummaryThe following statements summarize the informa-

tion that can be gleaned from these 12 studies con-cerning AAC intervention for individuals with AOS:

1. Many of the participants demonstrated acquisitionof the targeted behaviors in controlled settings;

2. Many of the participants exhibited meaningful andsubstantial improvements with respect to commu-nication over the course of treatment;

3. Extensive training was required, especially forgeneralization;

4. Studies addressing the role of partner support wereconspicuously lacking (but see Garrett &Beukelman, 1995; Kagan, Black, Duchan, Simmons-Mackie, & Square, 2001);

5. Measures of social validity were absent;

6. Acceptance of AAC by participants was not directlyaddressed by any of the studies, but was recog-

33


nized as an important factor by many of the inves-tigators;

7. The absence of AOS criteria in this body of literaturelimits the generalizability of the conclusions;

8. The majority of studies only measured treatmentoutcomes and, without experimental control, offerlimited evidence concerning the value of AAC forthis population.

In conclusion, the synthesis of these studiesindicates future success for the application of AAC toindividuals with AOS, but also reveals specific areaswhere additional research effort is required.

In January 2001, reimbursement for AAC wasapproved for Medicare recipients, making the acquisi-tion and training of AAC devices more accessible for alarger segment of the population and opening doors forimproved utilization of AAC generally. AAC interven-tion is time and resource intensive and, accordingly, thebenefits of these approaches need to be demonstrated.Most of the case studies reviewed indicate that sub-stantial time is required to promote successful use ofAAC and necessarily involves the following steps:

1. Conduct assessment of abilities and communica-tion needs;

2. Conduct ecological inventories for vocabulary se-lection;

3. Train use of the selected tools and strategies;

4. Provide opportunities for simulated and actualcommunity-based interactions using AAC; and

5. Provide facilitator training to family and othercommunication partners.

While the new Medicare policy of funding AACdevices clearly represents significant progress, withoutcomplementary funding for therapy to provide thenecessary training, the battle is only half won. Further-more, without improved treatment research to supportthe efficacy of AAC practices, continued funding maybe at risk. It is imperative that future investigationsinclude experimental control, because without that levelof methodological rigor, the scientific contribution isinsufficient. For many individuals with AOS, it is likelythat AAC offers the greatest potential for independentand wide-ranging communicative interactions. Without

question, research in this area is needed given thepaucity of extant studies and the clear potential forpositive results. More importantly, the opportunity formany individuals with AOS to substantially improvetheir communicative interactions may be diminished ifthe potential utility of AAC is not explored in muchgreater depth with this population.Dr. Margaret Rogers is an Associate Professor ofSpeech and Hearing Sciences at the University ofWashington in Seattle, WA.

ReferencesANCDS ad hoc Practice Guidelines Coordinating Commit-

tee. Golper, L. (Chair), Wertz, R. T., Frattali, C., Yorkston,K., Myers, P., Katz, R., Beeson, P., Kennedy, M. T.,Bayles, K., & Wambaugh, J. Evidence-based practiceguidelines for the management of communication disordersin neurologically impaired individuals: An introduction. Apublication of the Academy of Neurologic Communi-cation Disorders and Sciences (ANCDS), http://www.duq.edu/ancds, September, 2001.

ASHA (1989). Competencies for speech-language patholo-gists providing services in augmentative communica-tion. Asha, 31, 107-110.

ASHA (1991). Report: Augmentative and alternative com-munication. Asha, 33 (Suppl. 5), 9-12.

Bailey, S. (1983). Blissymbols and aphasia therapy: A casestudy. In C. Code & D. Muller (Eds.), Aphasia therapy(pp. 178-186). London: Edward Arnold.

Bellaire, K. J., Georges, J. B., & Thompson, C. K. (1991).Establishing functional communication board use fornonverbal aphasic subjects. In T. E. Prescott (Ed.),Clinical aphasiology (Vol. 19, pp. 219-227). Austin, TX:Pro-Ed.

Beukelman, D. R., & Mirenda, P. (1998). Augmentative andalternative communication: Management of severe commu-nication disorders in children and adults. Baltimore: PaulBrookes Publishing Co.

Beukelman, D. R., Yorkston, K. M., & Dowden, P. A. (1985).Communication augmentation: A casebook of clinical man-agement. Boston: College-Hill Press.

Calculator, S. N. (1988). Promoting the acquisition and gen-eralization of conversational skills by individuals withsevere disabilities. Augmentative and Alternative Com-munication, 4 (2), 94-103.

Chambless, D. L., & Hollon, S. D. (1998). Defining empiricallysupported therapies. Journal of Consulting and ClinicalPsychology, 66, 7–18.

Coelho, C. A., & Duffy, R. J. (1985). Communicative use ofsigns in aphasia: Is acquisition enough? Aphasiology, 4,1-8.

34


Dowden, P. A., Marshall, R. C., & Tompkins, C. A. (1981).Amer-Ind sign as a communicative facilitator for apha-sic and apraxic patients. In R. H. Brookshire (Ed.),Proceedings of the clinical aphasiology conference (pp. 131-140). Minneapolis, MN: BRK Publishers.

Garrett, K. L., & Beukelman, D. R. (1992). Augmentativecommunication approaches for persons with severeaphasia. In K. M. Yorkston (Ed.), Augmentative commu-nication in the medical setting (pp. 245-337). Tucson, AZ:Communication Skill Builders.

Garrett, K. L., & Beukelman, D. R. (1995). Changes in theinteraction patterns of individuals with severe aphasiagiven three types of partner support. In M. Lemme(Ed.), Clinical aphasiology (vol. 23, pp. 237-251). Austin,TX: Pro-Ed.

Garrett, K. L., & Beukelman, D. R. (1998). Adults with severeaphasia. In D. R. Beukelman & P. Mirenda (Eds.),Augmentative and alternative communication: Manage-ment of severe communication disorders in children andadults (2nd ed., pp. 465-500). Baltimore: Paul BrookesPublishing Co.

Garrett, K. L., Beukelman, D. R., & Low-Morrow, D. (1989).A comprehensive augmentative communication sys-tem for an adult with Broca’s aphasia. Augmentative andAlternative Communication, 5, 55-61.

Kagen, A., Black, S., Duchan, J. F., Simmons-Mackie, N., &Square, P. (2001). Training volunteers as conversationpartners using “Supported Conversation for Adultswith Aphasia” (SCA): A controlled trial. Journal of Speech,Language, and Hearing Research, 44, 624-636.

Kazdin, A. E. (1977). Assessing the clinical and applied impor-tance of behavior change through social validation.Behavior Modification, 1, 427-452.

Kearns, K. P., Simmons, N. N., & Sisterhen, C. (1982). Ges-tural sign (Amer-Ind) as a facilitator of verbalization inpatients with aphasia. In R. H. Brookshire (Ed.), Clinicalaphasiology (vol. 14, pp. 183-191). Minneapolis, MN:BRK Publishers.

Kendall, P., & Norton-Ford, J. (1982). Therapy outcomesresearch methods. In P. Kendall & J. Butcher (Eds.),Handbook of research methods in clinical psychology (pp.429-460). New York: John Wiley and Sons.

King, J. M., & Hux, K. (1995). Intervention using TalkingWord Processing Software: An aphasia case study.Augmentative and Alternative Communication, 11, 187-192.

Lane, V. W., & Samples, J. M. (1981). Facilitating communica-tion skills in adult aphasics: Application of Blissymbolsin a group setting. Journal of Communication Disorders,14, 157-167.

McNeil, M. R., Robin, D. A., & Schmidt, R. A. ( 1997). Apraxiaof speech: Definition, differentiation, and treatment. InM. R. McNeil (Ed.), Clinical management of sensorimotorspeech disorders (pp. 349-386). New York: Thieme Medi-cal Publishers.

National Joint Committee for the Communication Needs ofPersons with Severe Disabilities. (1992). Guidelines formeeting the communication needs of persons withsevere disabilities. Asha, 34 (Suppl. 7), 1-8.

Rabidoux, P., Florance, C. L., & McCauslin, L. (1980). The useof a Handi-Voice in the treatment of a severely apractic,non-verbal patient. In R. H. Brookshire (Ed.), Proceed-ings of the clinical aphasiology conference (pp. 131-140).Minneapolis, MN: BRK Publishers.

Robey, R. R. (2001). From clinical practice to clinical outcomeresearch and back again. ASHA Division 2: Neurophysi-ology and Neurogenic Speech and Language Disorders, 11(1), 3-18.

Robey, R. R., & Schultz, M. C. (1998). A model for conductingclinical-outcome research: An adaptation of the stan-dard protocol for use in aphasiology. Aphasiology, 12,787-810.

Rogers, M. A., & Storkel, H. L. (1999). Planning speech onesyllable at a time: The reduced buffer capacity hypoth-esis in apraxia of speech. Aphasiology, 13, 793-805.

Schlosser, R. W. (1999). Comparative efficacy of interven-tions in augmentative and alternative communication.Augmentative and Alternative Communication, 15, 56-68.

Skelly, M., Schinsky, L., Smith, R. W., & Fust, R. S. (1974).American Indian sign (Amer-Ind) as a facilitator ofverbalization for the oral verbal apraxic. Journal ofSpeech and Hearing Disorders, 39, 445-455.

Sigafoos, J., & York, J. (1991). Using ecological inventories topromote functional communication. In J. Reichle, J.York, & J. Sigafoos (Eds.), Implementing augmentativeand alternative communication (pp. 61-70). Baltimore:Paul H. Brookes Publishing Co., Inc.

Wertz, R. T., LaPointe, L. L., & Rosenbek, J. C. (1984). Apraxiaof speech in adults: The disorder and its management.Orlando, FL: Grune & Stratton, Inc.

Yanak, K., & Light, J. (1991). Use of augmentative communi-cation with persons with Broca’s aphasia: A case study.National Student Speech-Language-Hearing AssociationJournal, 19, 27-34.

apraxia, evidence-based interventionstatic.crowdwisdomhq.com/asha/7402 article.pdffrom clinics in...

Documents