Abstract SCHEDULING DISTRIBUTION AND MOTOR LEARNING GUIDED TREATMENT WITH CHILDHOOD APRAXIA OF SPEECH By Caitlin Webb April, 2011 Director: Laura Ball, PhD, CCC-SLP DEPARTMENT OF: Communication Sciences & Disorders The purpose of this study was to examine the use of motor learning guided (MLG) treatment with different treatment schedules in the treatment of participants with a diagnosis of childhood apraxia of speech (CAS). Five participants, chronological ages 4;6 to 5;11 years, received MLG treatment for diagnosed CAS in two different treatment schedules, mass and distributed. The mass schedule consisted of four weekly 60-minute treatment sessions for a total of 240 minutes of intervention. The distributed schedule consisted of 16 15-minute treatment sessions provided four days a week for a total of 240 minutes of intervention. With the mass treatment schedule, participants demonstrated an increase in performance accuracy by an average of 9.1%. With the mass treatment schedule, participants demonstrated an increase in probe accuracy by an average of 5%. With the distributed treatment schedule, participants demonstrated an increase in performance accuracy by an average of 21.4%. With the distributed treatment schedule, participants demonstrated an increase in probe accuracy by an average of 17%. Both treatment schedules produced positive outcomes with the distributed treatment schedule resulting in the highest improvement in speech production accuracy. The results of this study suggest that children with CAS may benefit from shorter and more frequent intervention sessions to yield motor learning of speech skills and to increase accuracy of speech production performance. Scheduling Distribution and Motor Learning Guided Treatment with Childhood Apraxia of Speech A THESIS Presented To The Faculty of the Department of Communication Sciences and Disorders East Carolina University In Partial Fulfillment Of the Requirements for the Degree Master of Science in Communication Sciences and Disorders by Caitlin L. Webb April 2011 © Copyright 2011 Caitlin Webb SCHEDULE DISTRIBUTION AND MOTOR LEARNING GUIDED TREATEMENT WITH CHILDHOOD APRAXIA OF SPEECH by CAITLIN WEBB APPROVED BY: DIRECTOR OF DISSERTATION/THESIS:_________________________________________ Laura Ball, Ph.D., CCC-SLP COMMITTEE MEMBER: _______________________________________________________ Martha Smith, Ph.D., CCC-SLP COMMITTEE MEMBER: _______________________________________________________ Kathleen Cox, Ph.D., CCC-SLP COMMITTEE MEMBER: _______________________________________________________ Paul Vos, Ph.D. CHAIR OF THE DEPARTMENT OF COMMUNICATION SCIENCES & DISORDERS ____________________________________ Gregg Givens, Ph.D., CCC-A DEAN OF THE GRADUATE SCHOOL_____________________________________________ Paul J. Gemperline, PhD ACKNOWLEDGEMENTS I would like to acknowledge the advice and guidance of my advisor, Dr. Laura Ball, and Dr. Martha Smith, whose encouragement, patience, and insightfulness allowed me to develop an understanding of the subject. I thank the members of my graduate committee, Dr. Kathleen Cox and Dr. Paul Vos, for their guidance and suggestions. My gratitude is given to Skye Lewis for the many hours she spent editing documents, supervising treatment schedules, providing support, and attending meetings, even when she had a thousand other things on her schedule. I would also like to thank Sarah Williamson and Rudi Carter for their help in providing therapy treatments and collecting data for the study. Finally, I would like to thank my family members for supporting and encouraging me to pursue this degree. My parents taught me the value in education and always encouraged me to reach for my goals. For their continued love, support, patience, and dog-sitting services, I will be eternally grateful. TABLE OF CONTENTS LIST OF TABLES……………………………………………………………………….. i LIST OF FIGURES……………………………………………………………………… ii CHAPTER 1: REVIEW OF LITERATURE……………………………………………. 1 Introduction………………………………………………………………………. 1 Etiology…………………………………………………………………………... 2 Diagnosis…………………………………………………………………………. 2 Treatment………………………………………………………………………… 3 Linguistic Approaches…………………………………………………… 4 Motor Learning Theory………………………………………………….. 6 Cueing…………………………………………………………… 7 Feedback………………………………………………………… 7 Practice Distribution…………………………………………….. 8 Mass Distribution………………………………………… 8 Distributed Distribution………………………………….. 9 Stimulus Distribution…………………………………………… 10 Motor Treatment Approaches……………………………………………. 11 Nonspeech Oral Motor Exercises………………………………... 11 PROMPT………………………………………………………… 12 Integral Stimulation……………………………………………… 12 Motor Learning Guided Treatment……………………………….. 13 Summary and Rationale…………………………………………………. 14 Research Question………………….…………………………………. 15 CHAPTER II: METHODS……………………………………………………………… 16 Participants………………………………………………………………………. 16 Stimuli……………………………………………………………………………. 19 Procedure………………………………………………………………………… 21 CHAPTER III: RESULTS & DISCUSSION……………………………………………. 26 Articulation Accuracy…………………………………………………………… 26 Probe Accuracy………………………………………………………………….. 34 Discussion……………………………………………………………………….. 42 MLG with Severe CAS………………………………………………….. 48 MLG with Moderate CAS……………………………………………….. 48 MLG with Mild CAS…………………………………………………….. 48 Potential Limitation……………………………………………………… 49 Clinical Implications…………………………………………………….. 49 REFERENCES………………………………………………………………………….. 51 APPENDIX A: IRB APPROVAL………………………………………………………. 57 APPENDIX B: STIMULI FOR PARTICIPANTS………………………………………. 67 LIST OF TABLES 1. Standardized Test Scores of Participants…………………………………………. 17 2. CAS Inclusion Parameters for Study……………………………………………... 20 3. Group Distribution and Treatment Schedule……………………………………... 22 4. Presentation of Stimuli by Treatment……………………………………………... 24 5. Accuracy of Treated Stimuli of Mass Schedule………………………………….. 27 6. Accuracy of Treated Stimuli of Distributed Schedule……………………………. 28 7. Baseline/Final Session Results of Treated Stimuli by Schedule………………….. 33 8. Accuracy of Probes During Mass Schedule………………………………………. 40 9. Accuracy of Probes During Distributed Schedule………………………………... 41 10. Baseline/Final Session Results of Probed Stimuli by Schedule…………………. 43 LIST OF FIGURES 1. Mass Schedule Mean Accuracy of Stimuli by Session…………………………..……... 29 2. Distributed Schedule Mean Accuracy of Stimuli by Session………………………..…. 30 3. Mass Schedule Baseline and Final Session Accuracy of Treated Stimuli……………... 31 4. Distributed Schedule Baseline and Final Session Accuracy of Treated Stimuli……….. 32 5. Change in Accuracy between Mass and Distributed Schedules……….....…………...... 35 6. Mean Change in Accuracy of Stimuli in Mass and Distributed Schedules………….…. 36 7. Mean Percent Change by CAS Severity Grouping……………………………………... 37 8. Mass Schedule Mean Accuracy of Probes by Session………………………………..… 38 9. Distributed Schedule Mean Accuracy of Probes by Session…………………..……….. 39 10. Mass Schedule Baseline and Final Session Accuracy of Probes…………………..…… 44 11. Distributed Schedule Baseline and Final Session Accuracy of Probes…………..…….. 45 12. Change in Probe Accuracy between Mass and Distributed Schedules…………..……... 46 13. Mean Change in Probe Accuracy in Mass and Distributed Schedules…………………. 47 I. Review of Literature Introduction Childhood apraxia of speech (CAS) is a motor speech disorder that occurs in a small percentage of the general population, between 1-10 in 1000. The majority of these cases are male (Souza, Payγ, & Costa, 2009; The Childhood Apraxia of Speech Association of North America, 2005). It has been suggested that there has been a substantial increase in occurrence during the past decade, but measurable population data is limited (American Speech-Language and Hearing Association [ASHA], 2007a). CAS has been a focus of research in the past few decades due to the controversy over its origin and characteristics and its increasing prevalence (Shriberg, Aram, & Kwiatkowski, 1997). This resulted in a call for research on CAS by ASHA to explore a variety of questions about the diagnosis and various treatment techniques (ASHA 2007a). Various diagnostic labels have been applied to children who have exhibited specific speech production behaviors, and these labels have historically been based on varying views of etiology (i.e., developmental apraxia of speech, developmental verbal dyspraxia, and the general term typically used for adults, apraxia of speech). In the past, developmental apraxia of speech identified deficits considered related to motor planning, whereas developmental verbal dyspraxia referred to phonological and language deficits co-occurring with motor deficits (Velleman, 2003). In 2007, the ASHA Ad Hoc Committee on Childhood Apraxia of Speech recommended the use of the label CAS and provided a formal definition. They described CAS as “a neurological childhood (pediatric) speech sound disorder in which the precision and consistency of movements underlying speech are impaired in the absence of neuromuscular deficits” (ASHA, 2007b). Etiology The etiology of idiopathic CAS remains in question; however, CAS has also been associated with known conditions. These include genetically based impairments, neurological disorders, developmental delays, prenatal or perinatal insults, and differences in the rate of development/quality of myelination (Cumley, Ball, Skinder, 2001). CAS can also co-occur with other neurological disorders such as Autism, Down Syndrome, and Fragile X (Flipsen & Gildersleeve-Neumann, 2009) and with other disorders such as otitis media, hypotonia, and sensory integration disorder (Teverovsky, Bickel, & Feldman, 2009), making it difficult to isolate a single etiology. Pre- and perinatal difficulties that have been connected to severe speech sound disorders (including CAS) include infections during pregnancy, preterm birth and low birth weight (Fox, Dodd, & Howard, 2002). However, there is a continued need for further investigation on the etiology of CAS and its co-existence with other disorders. Genetic research with the “KE” family, in which half of its members have orofacial apraxia, has identified a point of mutation on the FOXP2 gene. The FOXP2 gene was the first gene that researchers have been able to link to speech and language behavior. Located at chromosome 7q31, it has been implicated in “development of brain networks involved in orofacial learning, planning, and execution, particularly in motor sequences for speech, as well as conducting manual and other motor sequences” (Souza et al., 2009). Diagnosis Marrs (2010) reported that up to 75% of children diagnosed with CAS are actually misdiagnosed. There is an obvious need for consensus among clinicians with respect to appropriate diagnostic criteria and efficacious intervention. Although such hypotheses exist, there are currently no validated diagnostic criteria for CAS that differentiate it from other speech sound production disorders (ASHA, 2007a). Three main characteristics have been identified as being indicative of CAS. These include: errors on consonants and vowels that are characterized by their inconsistency when syllables or words are produced multiple times, expanded transition times between sounds/syllables, and inappropriate prosody (ASHA, 2007b). However, these characteristics are also commonly seen in other motor speech disorders (i.e., dysarthria) and may have contributed to an overdiagnosis of CAS (ASHA, 2007a). In an attempt to identify areas of research need, Crary (1993) discussed five areas hypothesized to be critical for differential diagnosis: the nonspeech motor skills, motor speech production, articulation and phonological skills, language performance, and an “other” category (e.g., attention and behavior). A comprehensive assessment for CAS includes speech intelligibility rating, receptive and expressive language skills, diadochokinetic performance, and speech consistency scores (ASHA 2007a; Velleman, 2003). Children with CAS may exhibit one or more of the following speech production errors: “non-diminished phonotactic (how sounds can be organized) errors beyond 3 years of age, regression, and variability in word usage and individual sounds” (ASHA, 2007a). Children with a motor speech disorder, often have co-occurring language deficits, often with higher receptive than expressive abilities (Crary, 1993). Treatment To be able to definitively report the effectiveness of CAS intervention, it is imperative for researchers to report “exactly what was treated (i.e. primary and secondary treatment targets) and how (i.e. nature, duration, and intensity), as well as what outcomes were measured and their results” (Morgan & Vogel, 2009). Treatment techniques, including various motor and phonological approaches, have been studied but almost all studies lack information on how intervention was delivered. Current practice suggests intensive services for CAS, particularly when the disorder is severe (ASHA, 2007; Hula et al., 2008; McCauley & Strand, 2008;). Indeed, it may be difficult to “dissociate motor and linguistic features of verbal dyspraxia because lexical, phonological, and articulation deficits co-occur” (Morgan & Vogel, 2009). Although, intensive services are needed, clinicians struggle to determine the appropriate amount (e.g., scheduling, collaboration, service delivery models, placements) (Cirrin et al., 2010). While providing intensive treatment can be challenging, some variables can be manipulated. These include quality and quantity of service, which includes the number of hours spent in therapy, child’s participation in therapy, proportion of adults to children during treatment (e.g., one-on-one, group), and the number of therapy sessions (Warren, Fey, & Yoder, 2007). However, because there are insufficient empirical data to support specific interventions, clinicians must rely on professional judgment and experience to determine the treatment approach. Treatment approaches reflect the broad scope of speech-language pathology, from linguistic to motoric, and some even employ non-speech oral movements. While each of these strategies approach CAS according to differently assumed etiology, they typically include similar aspects such as drill, remediation, self-monitoring, and feedback (Maas et al., 2008). Linguistic approaches. Clinicians who consider CAS a linguistic disorder often treat using a phonological, or linguistic-based approach (Velleman, 2003). The most common phonological intervention, Cycles Approach, aims to stimulate the entire phonological system (Hodson & Paden, 1991; Kamhi 2006; Velleman, 2006). Hodson (2006) states that targets are selected based on a normative perspective with consideration given to each child’s specific sound inventory since multiple sounds are developed at one time. This is conducted via traditional speech therapy (focused on motor production) in combination with a perceptual component (Kamhi, 2006). During each session, error sounds are targeted through review, auditory bombardment, word cards, production practice, and stimulability practice and probes (Garcia & Bauman-Waengler, 2009). Kamhi (2006) indicates that this approach has proven effective for children with CAS in some case studies, but additional evidence is needed for specific treatment and results on this particular population. Another phonologically based treatment, Phonotactic Therapy (Velleman, 2002), uses goal sets to focus on only one phonetic, phonotactic, and literacy goal at a time (Velleman, 2006). Phonotactic therapy addresses word structure by increasing syllable length in a step-by- step process using (1) modeling, adjacency, and fading; and (2) altering syllables by manipulating stress, harmony, and patterns. Phonotactic features affect speech movement patterns and the ability to vary the complexity of syllable structures (Velleman, 2006). This approach practices exaggerated vowels in syllables although vowel production is commonly impaired with CAS. This and the limited amount of phonotactic variability result in limited gains (Velleman, 2006). Prosodic approaches, which focus on production of varied stress and rhythm in speech, are another form of treatment used with CAS. The most common formalized prosodic therapy is melodic intonation therapy. Melodic intonation therapy was originally developed for and used with participants with apraxia of speech who also had aphasia. These participants have the acquired form of apraxia of speech, thus have had normal language in the past, unlike children with CAS who have a disorder in development (ASHA, 2007a). Also, participants with both apraxia of speech and aphasia have some level of language difficulty. As melodic intonation therapy implies, it is based on intonation, rhythm, and the use of gestures to help disordered individuals stay in time with their speech (McCauley & Strand, 2008). Square (1994) notes that this treatment assumes that a major component of Apraxia of Speech is related to the disruption of “the oscillating rhythmic substrate that underlies speech and the entire motor circuit” and this therapy aims to make a person’s speech more cyclic in nature. However, this approach can be difficult because as speakers, we automatically segment speech, making it difficult to work on problematic vowel productions. While this approach is considered to be linguistic based, it does have a motor component, with aspects that include arm swinging/leg tapping, finger counting, and vibro-tactile stimulation (Square, 1994). Motor learning theory. An intensive motor approach is considered the optimal treatment program (ASHA, 2007) for CAS. Two primary goals for motor approaches are to increase automatic speech oral motor flexibility (Velleman, 2006). These treatments are time-intensive, drill based, structured, and have an emphasis on self-monitoring (Stein, Harvey, & Macko, 2009). They are based on the assumption that motor programs are “road maps” that help speakers achieve speech targets that are perceived as normal (Square, 1994). Schmidt & Lee (2005) define motor learning as “a set of processes that produces a relatively permanent acquired capability for movement that is not directly observable”. Pre-practice activities are completed prior to each therapy session to motivate the person and gain selective attention to the task through setting goals, emphasizing the importance of the tasks, and ensuring the client understands the instructions (Caruso & Strand, 1999; Schmidt & Lee, 2005). Practice is not begun until the participant is engaged in the specific motor activity. It is important that repetitive practice is paired with focused, selective attention to ensure the person learns the movements and remembers them for future productions (Caruso & Strand, 1999). To increase awareness of learned behaviors, motor learning theory manipulates various aspects of learning, such as cueing, feedback, and distribution of practice (Schmidt & Lee, 2005). Cueing. Cueing is used to increase motor learning and can be provided through tactile, auditory, visual, or verbal modes (McCauley & Strand, 2008). Verbal imitation is a commonly used cue for motor learning but is often used in conjunction with other modalities (Caruso & Strand, 1999). However, research has shown that using imitation alone with children with CAS does not yield the improvement seen with typically developing peers (Moriarty & Gillon, 2006). Feedback. Feedback is another method used to increase motor learning. Feedback is defined in a variety of ways based on the type, monitoring, timing, and/or frequency provided. Feedback type refers to the differences between giving knowledge of the results (e.g., correct or incorrect) versus knowledge of performance (e.g., “I heard you say…”) and feedback frequency refers to how often feedback should be given, frequently or infrequently (Maas et al., 2008). Moriarty and Gillon (2006) noted that children with CAS showed the greatest improvement when given information about phonological and phonetic makeup of the targets (i.e., knowledge of their performance), rather than simply correct or incorrect. Similar to feedback frequency, feedback timing considers the effects of providing immediate feedback after each utterance versus delayed feedback after a set or variable number of utterances. Monitoring refers to the use of intrinsic (self-monitoring) or extrinisic (clinician-provided) feedback (Maas et al., 2008). Intrinsic, when the person determines whether their own productions are correct or incorrect, and extrinsic feedback, when the person is given feedback (i.e., accuracy of performance) by the clinician, are both considered to potentially change the acquisition of skills because they may shape current behaviors (Sheppard, 2008). However, research suggests that when doing motor activities, either speech or nonspeech, excess feedback provided too quickly interrupts a person’s self-monitoring abilities and inhibits the ability to gauge success (Swinnen, Schmidt, Nicholson, & Shapiro, 1990). Practice distribution. Distribution of practice is based on the time spent in actual practice compared to the time spent at rest. Motor learning theory provides the principle that the most effective distribution of practice can enhance performance, as well as the retention of skills (Caruso & Strand, 1999). Practice can be distributed by two categories: distributed, defined as many sessions that last a short period of time; and mass, defined as the same amount of practice but in fewer but longer sessions (Caruso & Strand, 1999). In motor learning studies not involving speech tasks, performance (accuracy within a session) and learning (retention over time) were greater when longer rest periods (i.e., mass practice) were provided (Schmidt & Lee, 2005). It is expected that the most effective distribution of practice can enhance performance during speech treatment sessions as well as the retention rate of those skills; however, a lack of evidence is available to examine the effects of motor learning for speech production using these different schedules (Caruso & Strand, 1999). Mass practice schedule. Use of a mass practice schedule has been found to be the most effective distribution of practice in some speech studies (Hall, Jordan, & Robin, 1993). Evidence suggests that when utilizing a phonological approach, such as language and literacy training, a mass practice schedule produces the highest results (Kamhi, 2006). In terms of CAS, high intensity services are considered to produce the best results among clinicians; however, little evidence supports this claim and there is lack of a true definition of what characterizes high intensity. When a university language-literacy camp for children aged four to five years old was conducted and each child received approximately two hours a day of intensive therapy, minimal progress was noted in the children diagnosed with CAS (Edwards & McDonald, 2009). Distributed practice schedule. Motor learning evidence consistently suggests that distributed practice results in greater learning than mass practice, and that it helps with both immediate performance and retention for various motor tasks, lasting up to nine months. This is opposed to mass practice, which typically shows a loss of performance gains shortly after therapy ends (Maas et al., 2008). In a study done using Motor Learning Guided Treatment (MLGT) with swallowing and feeding disorders, it was determined that distributing practice was most efficient, gave more opportunities for feeding, and allowed optimal brain plasticity (Sheppard, 2008). In 2000, Shea, Lai, Black & Park looked at memory consolidation between massed and distributed practice based on the theory that “two associations are of equal strength but of a different age, a new repetition has a greater value to the older one” (McGeogh, 1943). In two experiments, Shea et al. (2000) examined a person’s balance and performance on a numeric keyboard task to show generalization. Results for both experiments indicated that memory consolidation and increased participant performance occurred through distributed practice (Shea et al., 2000). Motor learning evidence for speech is limited but many clinicians base their distribution of practice on the text of Caruso and Strand (1999). They recommend distributed practice as most efficacious for serial motor learning tasks and producing the greatest gain in performance and learning because it is a continuous skill (i.e., has no beginning or end and can be stopped at any moment such as swimming) (Caruso & Strand, 1999). Another possible reason for benefit of distributed practice is the fact that children, especially younger ones, can maintain attention to a task for a maximum of 30 minutes, limiting the amount of quality intervention per session (ASHA, 2007a). In a study on the effects of treatment scheduling on phonemic awareness skills, distributed practice was found to be more effective for maintenance of knowledge (i.e., learning), but this was not proven true for immediate knowledge (i.e., performance) and over the course of the school semester, the advantage appeared to be minimal (Ukrainetz et al., 2009). A motor programming approach requires the clinician to use a more distributed practice schedule in order to give the client the most effective prognosis. It is important to note that research has not determined whether “impaired motor systems are sensitive to the same principle of learning as intact motor systems” (Maas et al., 2008). Some researchers say that there is no difference between the amount of gains for the two types of practice. One study looked specifically at the difference between a three-week summer camp and a typical speech therapy period of 12 months for children with cleft palate who had articulation disorders. While both treatment schedules showed significant decreases in the children’s misarticulations, there were no significant differences between the two groups (Pamplona, Ysunza, Patino, Ramirez, Drucker, & Mazon, 2004). Another study looked at a traditional Lee Silverman Voice Treatment program, a voice treatment for people with Parkinson’s, which involves a massed practice schedule and compared it to an extended version of the program. In this study, distributed practice did not enhance voice abilities in comparison to the traditional program (Maas et al., 2008). Stimulus distribution. Distribution of stimuli is another consideration of motor learning and can be organized in a blocked or random pattern, or a combination of the two (Schmidt & Lee, 2005). A blocked distribution in motor learning is completed when one task, with multiple trials, is completed before another stimulus is practiced. Caruso and Strand (1999) report that in speech therapy this leads to better performance as one target sound/word is practiced multiple times. Blocked distribution is used when targeting new goals and for children who have limited verbalizations (Gildersleeve-Neumann, 2007). During the earliest stages of learning, evidence suggests that this form of distribution provides the most success (Caruso & Strand, 1999). Random distribution, on the other hand, is used when multiple stimuli are being used and the same task is never repeated in consecutive trials (Schmidt & Lee, 2005). The order of the presentation of the stimuli is randomized throughout the session and it is suggested that this distribution results in better retention of skills for motor learning (Caruso & Strand, 1999). However, random practice requires increased motor planning and cognitive involvement of the individual (Gildersleeve-Neumann, 2007). Both forms of stimulus distribution require further investigation before a definitive recommendation can be made for speech tasks (Caruso & Strand, 1999). Schmidt and Lee (2005) suggest that using a combined randomized-block schedule (i.e., a combination of random and blocked distribution) allows the clinician to combine many of the positive features from both forms of stimulus distribution. Motor treatment approaches. Common motor approaches used for CAS treatment include (a) non-speech oral motor exercises, (b) Prompts for Restructuring Oral Muscular Phonetic Targets (Hayden, 2009), and (c) motor learning (e.g., integral stimulation, Nuffield Centre Dyspraxia Programme, and Motor Learning Guided Treatment) (Caruso & Strand, 1999; Williams, McLeod, & McCauley, 2010). Nonspeech oral motor exercises. Although commonly used, nonspeech oral motor exercises are typically associated with therapy of dysarthria. McCauley & Strand (2008) stated that non-speech oral motor exercises could be used to help increase coordination and strength of the muscles involved in speech. To achieve this, clinicians use sensory stimulation (i.e., pressing under jaw, practicing sucking, biting, blowing, and moving tongue) and manipulation to increase strength, stability, range of motion, and respiratory support (McCauley & Strand, 2008). McCauley & Strand (2008) note that this approach is most often used in conjunction with articulation practice, but not always. It is essential to note that CAS is not a disorder of muscle weakness and there is a lack of evidence on the effectiveness of oral motor therapies for CAS (Kamhi, 2006). It has also been noted that “no speech sound requires the tongue tip to be elevated towards the nose and that no sound is produced by puffing out the cheeks, …oral movements that are irrelevant to speech movements will not be effective as speech therapy techniques” (Lof, 2009). PROMPT. Hayden (2009) described PROMPT as an approach in which the clinician manually guides the various speech structures (e.g., tongue, lips, cheeks) in an attempt to manipulate muscles and structures of speech, ranging from the mouth to the larynx. It uses tactile and kinesthetic cues and also addresses the posture of the head, neck, and trunk (Hayden, 2009). Adults with acquired AOS used the PROMPT system, and experienced a gain in skills at the end of therapy but lacked retention of those skills over time (Square, Chumpelik, Morningstar, & Adams, 1986). Integral Stimulation. Integral stimulation is another motor approach that has been used to treat CAS (Caruso & Strand, 1999). Integral Stimulation is geared towards matching the cognitive motor learning to the current skill level (Gildersleeve-Neumann, 2007). It aims to achieve an adequate speech signal by focusing on oral movement patterns. Integral Stimulation uses auditory, visual, and tactile cues to enhance the success of the concept, “listen to me, watch me, and do as I do” (McCauley & Strand, 2008). A hierarchical approach is followed, moving from simple to complex productions; subsequently, as more complex productions are achieved, the clinician support (e.g., feedback, frequency) is gradually decreased. For example, at stage one, a child watches and listens while simultaneously producing the stimulus with the clinician. Once simultaneous production accuracy is mastered, the clinician next models the stimulus and the child repeats the model while the clinician silently mouths it, and gradually fades the “mouthing” cue. In considering motor learning theory and the need for repetitive practice, Integral Stimulation uses multimodal cues, which may detract from the number of potential productions in a session. Also, integral stimulation does not take into account the time required for the articulatory system to “reset” to produce consistently accurate productions. This delay is needed to allow time for sensory and motor signals to be converted and transported prior to another action can being accurately produced (Wolpert, Ghahramani, & Flanagan 2001). Motor Learning Guided Treatment. Considering that CAS is a motor speech disorder, it is most appropriate to use a treatment that engages the oral motor system. Motor Learning Guided Treatment (MLGT) includes components from Integral Stimulation and adds features from motor learning theory. MLGT focuses on motor learning as a “set of internal processes associated with practice or experience leading to relatively permanent changes in the capability for movement” (Schmidt & Lee, 2005). The basis of this approach considers four main motor learning principles: background information (e.g., motivation, family involvement), conditions of practice, rate, and feedback (Maas et al., 2008). MLGT applies a hierarchical approach in which practice starts with randomized stimuli. Once sufficient practice (i.e., person is motivated and expectations are clear) has been completed, stimuli are introduced in a random-block distribution. Stimuli are selected based on similar error patterns to target behaviors. A delay follows each attempt to allow time for internal processing. The amount (high or low frequency) of feedback given is also restricted (Fountain, Lasker, & Stierwalt, 2007). MLGT has been used in both the treatment of speech and non- speech tasks. Although MLGT has been used with adults diagnosed with apraxia of speech, it has not been used with children diagnosed with CAS. While MLGT is believed to be beneficial in treating motor speech disorders, including hypothetically CAS, there are few studies published examining its effectiveness in disorders other than AOS (Maas et al, 2008). Summary and Rationale Based on the current literature, specific aspects of motor learning have proven to be more successful with remediating CAS. Imitation should be combined with other cueing strategies, including verbal, visual, & auditory cues, to increase speech (Moriarty & Gillon, 2006). This cueing system should be paired with providing feedback on knowledge of performance based on Moriarty & Gillon’s (2006) findings that children with CAS experience greater improvement when information is given about the phonological and phonetic makeup of their productions and not just correct/incorrect. Another aspect of motor learning that has proven successful is using a combined randomized-block schedule. Using a combination of a random and blocked stimuli distribution allows the clinician to provide the positive features from both distribution types, allowing enough practice for a new skill to be learned and increasing overall skill abilities (Schmidt & Lee, 2005). While certain aspects of motor learning have been recognized as being the most beneficial with children with CAS, there continues to be a lack of evidence on the effectiveness of various practice schedules using the MLGT with children diagnosed with CAS (ASHA, 2007a). Research is needed to definitively report the distribution of therapy that is used with children with CAS and what outcomes were measured while using the distribution schedules (Morgan & Vogel, 2009). Motor learning studies involving nonspeech tasks suggest that a mass schedule will provide greater performance and learning (Schmidt & Lee, 2005); however, there is a lack of evidence to examine if these findings can also apply to speech production tasks and with the growing population of children with CAS (Caruso & Strand, 1999). This project addresses one aspect of treatment for children with CAS. The various structures of practice, distributed and massed, using MLGT with children with CAS were studied and analyzed. The performance and learning rates of these sessions will be examined. As a benefit, this research provides information on intervention techniques used to support children with CAS and to identify differences in various structures of practice. This project enhances the knowledge regarding CAS and its clinical intervention. It also aims to provide an effective intervention strategy for clinicians by determining the most successful distribution of practice with children with CAS. Research Question When treatment time is held constant (e.g., 300 minutes) among typically used treatment schedules [distributed: 1 or 2 times a week sessions and massed: 5 times a week sessions], do children with CAS show greater response to one schedule? II. Methods Participants Five children chronological ages ranging from 4;6 to 5;11 years were recruited from the clients of East Carolina University Speech-Language and Hearing Clinic and the surrounding area through advertisement. Approval was obtained from the ECU Institutional Review Board and consent forms (Appendix A) were signed prior to testing. Participants were excluded if they had any known neurological/organic conditions or scored below average receptive language testing. Inclusion criteria for participation was based on: (a) passing an audiometric screening; (b) corrected visual acuity to interact sufficiently with stimuli; (c) native American English (range of cultural/racial backgrounds) to reduce potential variability resulting from production of non-English words; and (d) characteristics of CAS as defined in the ASHA Technical Report (ASHA, 2007a). To determine whether participants met the CAS characteristic inclusion criteria, a series of assessments were completed and results are displayed in Table 1. A hearing screening was administered at 25 dB HL for 1000 Hz, 2000 Hz, and 4000 Hz, adjusted from the recommended 20 dB HL for children due to the participants being tested in a noisy clinic room, in order to confirm probable normal hearing (ASHA, 1997). An oral mechanism screening was administered to rule out craniofacial abnormalities that could interfere with task performance. The Primary Test of Nonverbal Intelligence (PTONI) (Ehrler & McGhee, 2008) was administered to screen for cognitive ability necessary to complete the tasks in the study and determine inclusion criteria. The Test of Auditory Comprehension of Language, 3rd Edition (TACL-3) (Carrow-Woolfolk, 1998) was administered to assess receptive language abilities and establish inclusion factors. The Clinical Assessment of Articulation and Phonology (CAAP) Table 1 Standardized Test Scores of Participants 1 2 3 4 5 Age 5;11 5;3 5;9 4;11 4;6 Gender Male Male Female Male Male Hearing & Vision Pass Pass Pass Pass Pass PTONI 132 (Very Superior) 77 (Low) 99 (Average) 103 (Average) 113 (Above Average) TACL-3 115 (Above Average) 115 (Above Average) 126 (Superior) 106 (Average) 102 (Average) CAAP 41 48 12 20 46 KSPT SS Oral Simple Complex 108 17 <3 63 <3 <2 108 <17 64 106 <49 55 106 <49 9 DDK # per sec /p?/ /t?/ /k?/ Patticake 2.67 2.67 2.33 1 2.7 3.3 3 1.3 2.7 3 3.3 1.3 3 3 3.33 .83 3 2.67 2.3 1.3 (Secord, Donohue, & Johnson, 2002) measured the child’s speech production skills of single words. Results of this assessment also were used to determine the participant’s phonetic inventory and single word intelligibility and were the basis of stimulus creation. The Kaufman Speech Praxis Test for Children (Kaufman, 1995) provided information for a vowel inventory, consistency of vowel production, complexity of syllables produced, phoneme prolongations, and accuracy of nonsense word repetition. Diadochokinesis (DDK) tasks were completed to determine syllable repetitions (e.g., maximum repetition rate, alternating repetition rate, and diadochokinesis). Participants were between the ages of 4;6 and 5;11 at the beginning of the study. Participants 1, 2, and 3 had finished Kindergarten prior to participating in the study and the remaining two participants were enrolled in preschool. Each participant completed all testing during a single 1-hour session. Scores on the PTONI ranged from 77-132. Participant 2, who scored a standard score of 77 on the PTONI, was the only participant who did not fall under the “average” to “very superior” range. Due to the discrepancy between his receptive language score on the TACL-3 (SS- 115, above average) and his score on the PTONI, he was included in the study. His low score on the PTONI was considered to be a reflection of attention and not of actual ability and that his receptive language score was more likely a closer measure of performance. Receptive language was in the average to superior range for all participants, with standard scores ranging from 102-126. Standard scores on the CAAP ranged from <55-75, indicating profound to moderate speech delays, with the participants’ total number of errors on mono- & multisyllabic words and cluster words ranging between 12-48. Standard scores on the KSPT included: 63-108 for oral tasks, <49- <3 for simple tasks, and 64- <2 for complex tasks. As tasks increased in motoric difficulty on the KSPT, scores decreased. DDK tasks were slow and inaccurate for all participants (Mean of p?= 2.67-3, t?= 2.67-3.3, k?= 2.3-3.33) compared with normative scores of p? (4.00-4.03), t? (4.07-4.14), and k? (4.56-4.58) (Flecther, 1972). For inclusion in this study, children were diagnosed with CAS if they exhibited one or more features from categories including nonspeech motor behavior, motor speech behavior, speech sound and structure, and prosody (Table 2). Nonspeech motor behaviors included impaired oral volitional movement, and/or articulatory groping obtained from DDK and KSPT tasks. Motor speech behaviors included impaired prolongation of fricatives, difficulty with monosyllabic and trisyllabic DDK sequences, and an inability to do nonword and multisyllabic word repetitions. Speech sound and structure features were obtained from the KSPT and CAAP and included vowel errors, inconsistency in speech errors, errors in production order of sounds, morphemes, and words, articulatory regression, improved performance on automatic productions in comparison to volitional productions. Prosody errors obtained from the CAAP and KSPT included syllable segregation and excessive-equal stress patterns. Stimuli An error analysis was completed based on speech sound errors made on the CAAP and KSPT. This error analysis was used to identify errors as consistent or inconsistent, and then further distinguished by place within syllables (initial, medial, and final). Phonemic combinations that the participant never produced accurately in syllables/words were not included. Targets were selected such that the participant (1) produced a maximum of one phoneme in error at baseline, and (2) had inconsistent error production in that word position. Once 80-100 stimuli were created for each child, a random number generator (Stat Trek, 2011), was used to select 60 stimuli for use in practice. From the group of 60, 20 stimuli were randomly selected for use as untreated probes. These probes were used at the beginning of the session to Table 2 CAS Inclusion Parameters for Study S1 S2 S3 S4 S5 Nonspeech Motor Behavior (1 or more) Impaired oral volitional movement x x x Lingual/mandibular articulatory groping x x x x Motor Speech Behavior (1 or more) Impaired prolongation of fricatives x x x x Impaired production of monosyllabic & trisyllabic DDK sequences x x x Impaired nonword repetition x x x Impaired multisyllabic word repetition x x x x x Speech Sound & Structure (1 or more) Vowel errors x x x x x Inconsistent speech errors x x x x x Errors in production order Sounds x x x x x Morphemes x x x x x Words x Prosody (1 or more) Syllable segregation (Staccato Speech) x xx x x Excessive-equal stress x x x x x Note: Items from this checklist were selected from the literature and in particular the ASHA Technical Report (2007) estimate generalization of skills to untreated probes. The remaining 40 stimuli were randomly assigned to two groups of 20 stimuli for practice (see Appendix B). Stimuli were of varied complexity, from simple 1-word productions (e.g., key, pig, cheek, lemon) to sentences with more complex movement patterns (e.g., “I am five”, “That toy is too expensive”), based on the participant’s abilities. Procedure The five participants were randomly assigned to two groups. Treatment schedule order was counterbalanced for each group to reduce order effects in performance and ensure each group completed both schedules. Table 3 illustrates group assignments. Group 1 included two participants (Ss 2, 3); this group was assigned to the Mass schedule for treatment first, during which they participated once a week for 60-minutes for 240 minutes of total treatment. During these 60-minutes sessions, the 20 stimuli were randomized for practice and after completing all stimuli they were re-randomized 3-5 times for practice. When all four sessions of the Mass schedule was completed, Group 1 began the Distributed schedule, during which they received 4 weeks of intervention four times a week (M-TH) for 15-minutes (16 sessions) for 240 minutes of total treatment. Group 2 consisted of participants 1, 4, and 5 and was assigned the Distributed schedule initially, followed by the Mass schedule. During each session, participants produced the 20 probe items to establish an initial baseline and evaluate potential generalization of treatment to untreated targets. Twenty treated stimuli randomized prior to each session. During data collection, participants were in quiet clinical treatment rooms as free from distraction as possible. They were seated in child-sized chairs at a small table with stimuli Table 3 Group Distribution and Treatment Schedule Group Schedule 1 Schedule 2 1 Participants: 2 & 3 Mass (1x/week for 60 min) Distributed (4x/week for 15 min) 2 Participants: 1, 4, & 5 Distributed (4x/week for 15 min) Mass (1x/week for 60 min) presented via a PowerPoint presentation using words and clip-art pictures. The stimulus words and sentences were individualized for each participant and no two sets of targets were the same. Each session was structured in a similar manner, varying only in amount of practice time per session. The researcher started the session by presenting each probe given a stimulus picture and no verbal model to the participant. The participant named the word or sentence and each response was judged as correct or incorrect. After completion of all 20 probes, the 20 treated stimuli were introduced. The researcher began treatment by presenting a stimulus with a verbal model and stimulus picture. The participant imitated the stimulus utterance without assistance, followed by a delay interval of 3 seconds. After the delay, the researcher continued therapy by cueing (pointing) the participant to produce the same stimulus item again but provided no additional verbal cue. The participant then produced the stimulus utterance a second time, without feedback. This point-no verbal cue procedure was 3 additional times, with a 3 second delay interval between each production. Feedback was provided once all 5 productions were complete. This procedure was followed for all 20 treated stimuli and each production judged as correct or incorrect. As further illustrated in Table 4, each schedule is broken down into the number of stimuli produced and the procedure for all stimuli with imitation of verbal model beginning followed by a 3 second delay & 4 productions with no verbal model and delay between each production prior to feedback being given. The number of stimulus sets practiced varied depending on the treatment schedule (mass vs. distributed). During the mass schedule, the 20 stimuli were practiced five times, for a total of 100 utterances produced in a 60-minute session. In the distributed schedule, the 20 stimuli were practiced one time, for a total of 20 utterances practiced in a 15-minute session. Table 4 Presentation of Stimuli by Treatment Schedule Distribution of Stimuli Mass 100x – M-(3 sec delay)-Imitation1-(3)-I2-(3)-I3-(3)-I4-(3)-I5-Feedback Distributed 20x – M-(3 sec delay)-Imitation1-(3)-I2-(3)-I3-(3)-I4-(3)-I5-Feedback Note: M= Model and I= Imitation. Each stimulus word/sentence began with a model and then participant was required to imitate stimuli 5 times. Each session was digitally audio and video recorded using for later scoring of responses and analysis. III. Results and Discussion Descriptive statistics were used to analyze data, as the goal of this exploratory research was to evaluate whether altering treatment distribution could effectively alter the speech of children with CAS. The following analyses involved direct comparisons among participants on learning and generalization of stimuli. Articulation Accuracy Each participant’s productions of treated targets were used to calculate overall accuracy. For mass practice, accuracy was determined by calculating the mean performance of all repetitions, approximately 500 total for each participant, of the treated targets (i.e., Accuracy = Accurate Productions/Total Productions). For distributed practice, accuracy was determined by calculating the mean performance of all utterances, approximately 100 total for each participant, (i.e., Accuracy = Accurate Productions/Total Productions). These data were used to investigate and compare change and consistency across treatment for all participants. Data from the mass schedule and distributed schedule for all participants is shown in Tables 5 & 6. Figures 1 & 2 illustrate the pattern of change for both schedules. Each participant’s production accuracy increased from the baseline to the final session in both the mass and distributed schedules. Percent of change was calculated by subtracting the final session accuracy from the baseline accuracy for both schedules (i.e., Change in Accuracy = Percent Accuracy in Final Session – Percent Accuracy in Baseline). Participant 3 showed the greatest percentage of change with the mass schedule versus the distributed schedule. As illustrated by Figures 3 & 4, the remaining four participants experienced greater improvement with the distributed schedule. Data from each schedule, mass and distributed, for all participants is shown in Table 7. Table 5 Accuracy of Treated Stimuli of Mass Schedule Participant Session 1 Session 2 Session 3 Session 4 1 60.2 51.6 69.8 64.6 2 45.5 39.5 38.2 57.5 3 N/A 57.1 59.39 75.17 4 73.2 84.8 77.2 77 5 60.8 63.8 89.8 68 Table 6 Accuracy of Treated Stimuli of Distributed Schedule Participant 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 41 47 50 62 50 45.3 54 56 48 49 52 58 62 62 61 64 2 40 47 68 44 47 57 59 47 43 39 49 36 40 45 36 72 3 65 50 48 52 59 52 58 62 66 60 61 60 63 77 83 78 4 80 92 94 87 64 81 82 82 85 96 87 95 83 79 84 89 5 45 47 40 40 43 29 53 57 51 43 76 45 45 79 43 75 Figure 1. Mass Schedule Mean Accuracy of Stimuli by Session 0 20 40 60 80 100 1 2 3 4 P erc en t C orr ec t Session 1 2 3 4 5 Figure 2. Distributed Schedule Mean Accuracy of Stimuli by Session 0 20 40 60 80 100 1 3 5 7 9 11 13 15 Pe rc en t Corr ec t Session 1 2 3 4 5 Figure 3. Mass Schedule Baseline and Final Session Accuracy of Treated Stimuli 0 20 40 60 80 100 1 2 3 4 5 Pe rc en t Corr ec t Participant Baseline Final Session Figure 4. Distributed Schedule Baseline and Final Session Accuracy of Treated Stimuli 0 20 40 60 80 100 1 2 3 4 5 Pe rc en t Corr ec t Participant Baseline Final Session Table 7 Baseline/Final Session Results of Treated Stimuli by Schedule Mass Schedule Distributed Schedule Participant Baseline Final % Change Baseline Final % Change 1 60.2 64.6 +4.4 41 64 +23 2 45.5 57.5 +12 40 72 +32 3 57.1 75.17 +18.1 65 78 +13 4 73.2 77 +3.8 80 89 +9 5 60.8 68 +7.2 45 75 +30 Mean Change: +9.1 +21.4 As illustrated in Figure 5, no participant experienced more than 18.1% improvement with mass schedule from baseline to the final session (n=4). Participants experienced greater levels of improvement from baseline to the final session (n=16) during the distributed schedule, with the highest percent of increase at 32%. Overall, the amount of change seen from the mass schedule ranged from 3.8-18.1% (M=9.1%) versus the distributed schedule range of 9.0-32% (M=21.4%) (Figure 6). Mean accuracy of participants by severity of CAS (i.e., severe, moderate, mild) can be seen in Figure 7. Percent change data from each schedule, mass and distributed, for all participants is shown in Table 7. The mass schedule yielded a significant difference between baseline and final session accuracy (t(4) = 3.40, p = .03, 95% CI [-16.53, -1.67]). The distributed schedule also resulted in a significant difference between baseline and final session accuracy (t(4) = 4.71, p = .009, 95% CI [-34.02, -8.78]). While significant differences between baseline and final session accuracy were identified, interpretation of these statistical findings should be guarded due to the small sample size. Probe Accuracy Each participant produced 20 probes, or untreated words/phrases, at the beginning of each session. These probes were used to determine impact of treatment on untreated stimuli. For both schedules, accuracy was determined by calculating the mean performance accuracy (i.e. Accuracy = Accurate Productions/Total Productions). Figures 8 & 9 illustrate the trends toward improvement for the mass and distributed practice schedules. Data from each session for the mass and distributed schedules for all participants is shown in Tables 8 & 9. With the exception of Participant 4, all participants experienced some improvement during each treatment schedule (i.e., mass & distributed). No increase in accuracy was observed Figure 5. Change in Accuracy between Mass and Distributed Schedules 0 5 10 15 20 25 30 35 1 2 3 4 5 Pe rc en t Ch ang e Participant Mass Schedule Distributed Schedule Figure 6. Overall Mean Change in Accuracy of Stimuli in Mass and Distributed Schedules 0 5 10 15 20 25 Mass Distributed M ea n Pe rc en t Ch ang e Treatment Schedule Figure 7. Mean Percent Change by CAS Severity Grouping 0 5 10 15 20 25 30 Mass Distributed Pe rc en t Ch ang e Severe CAS Moderate CAS Mild CAS Figure 8. Mass Schedule Mean Accuracy of Probes by Session 0 20 40 60 80 100 1 2 3 4 Pe rc en t Corr ec t Session 1 2 3 4 5 Figure 9. Distributed Schedule Mean Accuracy of Probes by Session 0 20 40 60 80 100 1 3 5 7 9 11 13 15 Pe rc en t Corr ec t Session 1 2 3 4 5 Table 8 Accuracy of Probes during Mass Schedule Participant Session 1 Session 2 Session 3 Session 4 1 55 45 60 60 2 65 45 50 65 3 N/A 55 60 70 4 80 75 75 80 5 70 70 70 75 Table 9 Accuracy of Probes during Distributed Schedule Participant 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 45 50 45 55 35 50 30 50 45 45 55 50 40 60 60 55 2 50 45 60 60 60 65 65 45 65 60 70 30 35 70 45 60 3 65 65 60 70 80 45 75 70 80 55 65 80 75 85 85 90 4 80 65 85 80 50 70 85 80 80 80 80 70 80 80 85 80 5 45 55 75 60 60 55 45 60 55 55 60 60 60 80 70 85 on probes during the mass schedule for Participants 2 and 4, nor during the distributed schedule for Participant 4. However, during the distributed schedule, Participant 2 demonstrated a 10% gain on probe accuracy. Data from each schedule, mass and distributed, for all participants is shown in Table 10. These findings are illustrated in Figures 10 & 11. Percent of change in accuracy on probes was calculated by subtracting the final session accuracy from baseline accuracy for both schedules (i.e., Change in Accuracy on Probes = Percent Accuracy in Final Session – Percent Accuracy at Baseline). As shown in Figure 12, the range of accuracy on probes varied among participants. Probe accuracy during the mass schedule ranged from a 0%-15% improvement (M=5%). Probe accuracy during the distributed schedule ranged from 0%-40% improvement (M= 17%), as shown in Figure 13. Data from each schedule, mass and distributed, for all participants is shown in Table 10. The smallest degree of change as measured by accuracy on probes was achieved during the mass schedule. While overall a greater increase in accuracy on probes was seen during the distributed schedule, the percent of change are smaller than those of the treated stimuli. No significant difference between baseline and final session on probes was found with the mass schedule (t(4) = 1.83, p = 1.4, 95% CI [-12.60, 2.60]). The distributed schedule also did not demonstrate a significant difference between baseline and final session probes (t(4) = 1.94, p = 1.3, 95% CI [-36.51, 6.51]). It is noted that these statistical findings are interpreted guardedly due to the small sample size in the study. Discussion All participants included in the study were diagnosed with CAS of diverse degrees of severity, from mild to severe. Additionally, research has shown that children with CAS make limited progress in treatment. Nonetheless, the participants demonstrated improvement on Table 10 Baseline/Final Session Results of Probed Stimuli by Schedule Mass Schedule Distributed Schedule Participant Baseline Final % Change Baseline Final % Change 1 55 60 +5 45 55 +10 2 65 65 0 50 60 +10 3 55 70 +15 65 90 +25 4 80 80 0 80 80 0 5 70 75 +5 45 85 +40 Mean Change: +5 +17 Figure 10. Mass Schedule Baseline and Final Session Accuracy of Probes 0 20 40 60 80 100 1 2 3 4 5 Pe rc en t Corr ec t Participant Baseline Final Session Figure 11. Distributed Schedule Baseline and Final Session Accuracy of Probes 0 20 40 60 80 100 1 2 3 4 5 Pe rc en t Corr ec t Participant Baseline Final Session Figure 12. Change in Probe Accuracy between Mass and Distributed Schedules 0 10 20 30 40 50 1 2 3 4 5 Pe rc en t Ch ang e Participant Mass Schedule Distributed Schedule Figure 13. Mean Change in Probe Accuracy between Mass and Distributed Schedules 0 5 10 15 20 25 Mass Distributed M ea n Pe rc en t Ch ang e Treatment Schedule treated targets during both treatment schedules (i.e., mass & distributed). However, participants demonstrated a greater response to distributed treatment than mass treatment. MLG Treatment with Severe CAS. Participants 1, 2, and 5 were all diagnosed with severe CAS. During the distributed schedule, these participant’s accuracy levels increased to a greater degree than those participants with less severe CAS. These three participants all experienced more than a 20% change in overall accuracy. Accuracy on probes varied widely within this severe group during the distributed schedule. Participant 5 experienced the greatest change in accuracy at 40%, while participants 1 & 2 experienced only a 10% increase. The high increase experience by Participant 5 may have been related to increased and more focused attention during the shorter sessions associated with the distributed practice, allowing him to better monitor his own speech productions. During the mass schedule, Participants 1 & 5 experienced no more than a 5% change in overall accuracy while Participant 2 experienced no change at all. Overall accuracy of probes during the mass schedule was identical to accuracy on treated stimuli, with participants 1 & 5 experiencing a 5% increase and Participant 2 showing no change. MLG Treatment with Moderate CAS. Participant 4 was diagnosed with moderate CAS. Participant 4 demonstrated increased accuracy of treated stimuli during both treatments schedules. However, during the distributed schedule, a 9% increase in accuracy was obtained. Whereas, during the mass schedule, accuracy was limited to a 3.8% increase. This participant experienced no change in accuracy on probes during either treatment. MLG Treatment with Mild CAS. Participant 3 was diagnosed with mild CAS. An increase in accuracy on treated stimuli during both treatments was attained. During the distributed schedule, an improvement of 13% accuracy was achieved. During the mass schedule, an improvement of 18.1% was achieved. Unlike the other participants, greater improvement on treated stimuli was observed during the mass schedule. Although it’s difficult to provide a rationale for why this participant was different, it was noted that she had difficulty removing herself from a camp environment to participate in the distributed sessions, which may have impacted her overall progress during that schedule. Similarly, she was the only female and least severe; therefore, this may reflect gender and/or severity based differences in response to treatment. Increased accuracy on probes was seen for both treatments. However, greater improvement was observed during the distributed schedule, with a 25% increase compared to a 15% increase during the mass schedule. Potential Limitation of the Study. Overall the participants had significantly greater difficulty maintaining attention during the mass schedule due to the lengthy sessions involved. Distractors, such as tokens & prizes, were used in order to maintain motivation throughout these sessions. Attention fluctuated and made a clear impact on accuracy of treated stimuli. For participant’s short attention span consistent with their ages, the distributed schedule may have reduced the requirement for prolonged attention. Clinical Implications. Based on the findings of this study, additional research is warranted to further evaluate the response of children with CAS to MLG interventions. Future research should focus on examining more participants and distributing gender so that comparisons between male and females regarding response to treatment can be completed. Additional analyses may also be completed to evaluate CAS response to MLGT based on age, severity, length of treatment and sessions, and type of stimuli. Anecdotally, clinicians often report difficulty providing drill-based practice to young children. However, this research illustrated a successful approach to providing drill-based treatment in an enjoyable environment. Participants with limited attention showed greater benefit from the distributed treatment schedule than the mass treatment schedule. While the participants produced 120 utterances per session, the distributed sessions lasted only 15 minutes at a time. In a typical clinical setting this may be difficult to schedule, as traditional therapy sessions last between 30 minutes to 1 hour (i.e., a mass treatment schedule). However, by having shorter, more intense sessions (i.e., distributed treatment schedule), the child can engage for a larger portion of the session and perhaps experience progress towards performance and learning at a faster rate than in a traditional session. REFERENCES American Speech-Language-Hearing Association. (1997). Guidelines for Audiologic Screening [Guidelines]. Retrieved from www.asha.org/docs/html/GL1997-0199.html American Speech-Language-Hearing Association. (2007b). Childhood apraxia of speech. Position Statement, 1-3. Retrieved from www.asha.org/docs/html/PS2007-00277.html American Speech-Language-Hearing Association. (2007a). Childhood apraxia of speech. Technical Report, 1-74. Retrieved from www.asha.org/docs/html/TR2007-00278.html Bowen, C. (1998). Developmental phonological disorders: Information for families. Retrieved from http://www.speech-language-therapy.com/parentinfo.html. Carrow-Woolfolk, E. (1998). Test of Auditory Comprehension of Language. Austin, TX: PRO- ED Inc. Caruso, A. J., & Strand, E. A. (1999). Clinical management of motor speech disorders in children. New York, NY: Thieme. Cirrin, F.M., Schooling, T.L., Nelson, N.W., Diehl, S.L., Flynn, P.F., Staskowski, … & Adamczyk, D.F. (2010, July). Evidence-based review: Effects of different service delivery models on communication outcomes for elementary school-age children. Language, Speech, and Hearing Services in Schools, 44, 233-264. Crary, M. A. (1993). Developmental motor speech disorders: Phonologic disability, dysarthria and apraxia. Sand Diego, CA: Singular Publishing Group. Cumley, G., Ball, L., & Skinder-Meredith, A. (2001). Developmental apraxia of speech. Augmentative Communication News, 14(2), 1-4. Edwards, C., & McDonald, L. (2009, November). University language-literacy camp for preschoolers. Presentation at annual meeting of the American Speech-Language-Hearing Association, New Orleans, LA. Ehrler, D.J., & McGhee, R.L. (2008). Primary Test of Nonverbal Intelligence. Austin, TX: PRO- ED, Inc. Flipsen Jr., P. & Gildersleeve-Neumann, C. (2009). Childhood apraxia of speech: Some basics of assessment and treatment. Presentation at annual meeting of the American Speech- Language-Hearing Association, New Orleans, LA. Fountain, S., Lasker, J., & Stierwalt, J.A.G. (2007, November). Improving Speech Production in Profound Apraxia Using MLG and AAC. Presentation at the annual meeting of the American Speech-Language-Hearing Association, Boston, MA. Fox, A.V., Dodd, B., & Howard, D. (2002). Risk factors for speech disorders in children. International Journal of Language & Communication Disorders, 37(2), 117-131. Garcia, D. & Bauman-Waengler, J. (2009, November). The changing picture of Childhood Apraxia of Speech. Presentation at annual meeting of the American Speech- Language-Hearing Association, New Orleans, LA. Gildersleeve-Neumann, C. (2007, November 06). Treatment for Childhood Apraxia of Speech : A Description of Integral Stimulation and Motor Learning. The ASHA Leader. Hall, P. K., Jordan, L. S., & Robin, D. A. (1993). Developmental apraxia of speech: Theory and clinical practice. Austin, TX: Pro-Ed Inc. Hayden, D. (2009). Diagnosis and treatment: Childhood motor speech disorders and the PROMPT system. Retrieved from http://www.apraxia-kids.org Hodson, B.W. (2006). Identifying phonological patterns and projecting remediation cycles: Expediting intelligibility gains of a 7 year old Australian child. Advances in Speech- Language Pathology, 8(3), 257-264. Hodson, B.W., & Paden, E.P. (1991). Targeting intelligible speech: A phonological approach to remediation, 2nd ed. Austin, TX: Pro-Ed Inc. Hula, S., Robin, D.A., Maas, E., Ballard, K.J., & Schmidt, R.A. (2008). Effects of feedback frequency and timing on acquisition, retention, and transfer of speech skills in acquired apraxia of speech. Journal of Speech, Language, and Hearing Research, 51(5), 1088-1113. Kamhi, A. G. (2006). Clinical forum. Treatment decisions for children with speech-sound disorders. Language, Speech, & Hearing Services in Schools, 37(4), 271. Kaufman, N. (1995). Kaufman Speech Praxis Test for Children. Detroit, MI: Wayne State University Press. Lof, G.L. (2009, November). Nonspeech oral motor exercises: An update on the controversy. Presentation at annual meeting of American Speech-Language-Hearing Association, New Orleans, LA. Maas, E., Robin, D. A., Hula, S., Freedman, S. E., Wulf, G., Ballard, K. J., & Schmidt, R. A. (2008). Principles of motor learning in treatment of motor speech disorders. American Journal of Speech-Language Pathology, 17(3), 277-298. Marrs, I. (2011). Seventy five percent of apraxia cases wrongly diagnosed. Retrieved from http://www.speechlanguagefeeding.com. McCauley, R. J., & Strand, E. A. (2008). Treatment of childhood apraxia of speech: Clinical decision making in the use of nonspeech oral motor exercises. Seminars in Speech & Language, 29(4), 284-293. McGeogh, J.A. (1943). The Psychology of Human Learning. New York, NY: Longmands Green. Morgan, A. T., & Vogel, A. P. (2009). A cochrane review of treatment for childhood apraxia of speech. European Journal of Physical & Rehabilitation Medicine, 45(1), 1-110. Moriarty, B.C., & Gillon, G.T. (2006). Phonological awareness intervention for children with childhood apraxia of speech. International Journal of Language & Communication Disorders, 41(6), 713-734. Pamplona, C., Ysunza, A., Patino, C., Ramirez, E., Drucker, M., & Mazon, J. J. (2005). Speech summer camp for treating articulation disorders in cleft palate patients. International Journal of Pediatric Otorhinolaryngology, 69(3), 351-359. Raymer, A. M., Beeson, P., Holland, A., Kendall, D., Maher, L. M., Martin, N., … & Gonzalez Rothi, L. (2008). Translational research in aphasia: From neuroscience to neurorehabilitation. Journal of Speech, Language & Hearing Research, 51(1), S259-75. Schmidt, R. A., & Lee, T. D. (2005). Motor control and learning: A behavioral emphasis. Champaign, IL: Human Kinetics Publishers. Secord, W., Donohue, J., & Johnson, C. (2002) Clinical Assessment of Articulation and Phonology. Greenville, SC: Super Duper Publications. Shea, C. H., Lai, Q., Black, C., & Park, J. H. (2000). Spacing practice sessions across days benefits the learning of motor skills. Human Movement Science, 19(5), 737-760. Sheppard, J.J. (2008). Using motor learning approaches for treating swallowing and feeding disorders: A review. Language, Speech, and Hearing Services in Schools, 39, 227-236. Shriberg, L.D., Aram, D.M., & Kwiatkowski, J. (1997). Developmental apraxia of speech: I. Descriptive and theoretical perspectives. Journal of Speech, Language, and Hearing Research, 40, 273-285. Souza, T.N.U., Payγ, L.M.C., & Costa, R.C.C. (2009). Childhood speech apraxia in focus: Theoretical perspectives and present tendencies. Prσ-Fono Revista de Atualizaηγo Cientνfica, 21(1), 75-80. Square, P. A. (1994). Treatment approaches for developmental apraxia of speech. Clinics in Communication Disorders, 4(3), 151-161. Square, P.A., Chumpelik, D.A., Morningstar, D., & Adams, S. (1986). Efficacy of the PROMPT System of therapy for the treatment of acquired apraxia of speech: A follow-up investigation. Clinical Aphasiology, 16, 221-226. Stat Trek (2011). Random Number Generator. Retrieved from http://stattrek.com. Stein, J., Harvey, R.L., & Macko, R.F. (2009). Stroke recovery and rehabilitation. New York, NY: Demos Medical Publishing. Strong, C.J., Mayer, M., & Mayer, M. (1998). The Strong Narrative Assessment Procedure. Eau Claire, WI: Thinking Publications. Swinnen, S.P., Schmidt, R.A., Nicholson, D.E., & Shapiro, D.C. (1990). Information feedback for skill acquisition: Instantaneous knowledge of results degrades learning. Journal of Experimental Psychology: Learning, Memory, & Cognition, 16(4), 706-716. Teverovsky, E.G., Bickel, J.O., & Feldman, H.M. (2009). Functional characteristics of children diagnosed with childhood apraxia of speech. Disability and Rehabilitation, 31(2), 94-102. The Childhood Apraxia of Speech Association of North America (2005). What is childhood apraxia of speech? Retrieved from http://www.apraxia-kids.org. Ukrainetz, T.A., Ross, C.L., & Harm, H.M. (2009). An investigation of treatment scheduling for phonemic awareness with kindergarteners who are at risk for reading difficulties. Language, Speech, and Hearing Services in Schools, 40, 86-100. Velleman, S.L. (2002). Phonotactic therapy. Seminars in Speech and Language, 23(1), 43-56. Velleman, S.L. (2003). Resource guide for childhood apraxia of speech. Clifton Park, NY: Delmar/Thomson Learning. Velleman, S.L. (2006, November). Childhood apraxia of speech: Assessment/treatment for the school-aged child. Presentation at annual meeting of American Speech-Language-Hearing Association, Miami, FL. Velleman, S.L., & Strand, K. (1994) Developmental verbal dyspraxia. In J.E. Bernthal & N.W. Bankson (eds.) Child phonology: Characteristics, assessment, and intervention with special populations (p. 110-139). New York, NY: Thieme. Warren, S. F., Fey, M. E., & Yoder, P. J. (2007). Differential treatment intensity research: A missing link to creating optimally effective communication interventions. Mental Retardation and Developmental Disabilities Research Reviews, 13, 70–77. Williams, A.L., McLeod, S., & McCauley, R. (2010). Interventions for speech sound disorders in children. Baltimore, MD: Brookes Publishing. Wolpert, D.M., Ghahramani, Z., & Flanagan, J.R. (2001). Perspectives and problems in motor learning. TRENDS in Cognitive Science, 5(11) 487-494. Appendix A: IRB Approval Appendix B: Stimuli for Participants Table B1 Participant 1 Stimuli List Untreated Stimuli Mass Schedule Distributed Schedule 1. Toys 2. Wand 3. Geek 4. Base 5. Cap 6. Cone 7. Fight 8. Guard 9. Rack 10. Hive 11. Peas 12. Lip 13. Guess 14. Rib 15. Line 16. Key 17. Pen 18. Wedding 19. Lamb 20. Gas 1. Good 2. Dog 3. Lock 4. Hedge 5. Guide 6. Top 7. Lemon 8. Ride 9. Ramp 10. Wood 11. Loop 12. Coin 13. Duck 14. Kiss 15. Penny 16. Day 17. Guy 18. Move 19. Path 20. Pony 1. Boat 2. Lid 3. Tea 4. Feed 5. Vein 6. Cough 7. Toe 8. Teeth 9. Roof 10. Rip 11. Mud 12. Net 13. Bake 14. Beet 15. Tongue 16. Go 17. Lime 18. Zit 19. Taxi 20. Comb Table B2 Participant 2 Stimuli List Untreated Stimuli Mass Schedule Distributed Schedule 1. Deep 2. Hit 3. Badge 4. Peak 5. Beg 6. Bone 7. Bag 8. Toad 9. Funnel 10. Move 11. Fell 12. Hay 13. Yawn 14. Mitt 15. Dot 16. Hug 17. Black 18. Two 19. Hide 20. Lap 1. Pup 2. Duck 3. Hoop 4. Beak 5. Look 6. Dog 7. Tent 8. Deck 9. Up 10. Lie 11. Tail 12. Balloon 13. Photo 14. Foot 15. Edge 16. Peel 17. Good 18. Light 19. Tool 20. Wag Table B3 Participant 3 Stimuli List Untreated Stimuli Mass Schedule Distributed Schedule 1. We went to the lake 2. Pack a lunch 3. The baby is asleep 4. Seal the envelope 5. The bunny has a pink nose 6. I’d like soup and a sandwich 7. I need a map 8. The fan is going fast 9. Sit on my lap 10. The ham is in the oven 11. Look at those kids 12. I swim in the ocean 13. Don’t peek 14. She was too sick to go 15. I put seventeen violets in the vase 16. I want to talk to you 17. The boy is nine 18. I saw a snake 19. Have some pie 20. Hop like a bunny 1. I want to vacation at the beach 2. Don’t believe the gossip 3. A pentagon has five sides 4. My zip code is 27832 5. The panda is black and white 6. Lock the van 7. The ketchup is on the table 8. The beach is fun 9. Let’s have pizza 10. Get the phone 11. They went to a magjc show 12. Caitlin made a mistake 13. We found a sunken ship 14. She has a white dove 15. That was a big wave 16. We won the basketball game 17. I’m good at addition 18. I take a vitamin at night 19. A diamond is a shape 20. I want a bun 1. My spine is on my back 2. Jen had a headache 3. Pillow fights at night 4. I like school 5. I have a cousin in Maine 6. A daisy has white petals 7. Let’s play in the sun 8. Bees like pollen 9. Ben had a vanilla shake 10. I need a nap 11. That toy is too expensive 12. Put on the baseball cap 13. Let’s have a picnic 14. An animal was in the cave 15. I am five 16. I don’t want the olive 17. Bees make hone in a hive 18. The moon is full tonight 19. Beauty is only skin deep 20. Give the baby a bottle Table B4 Participant 4 Stimuli List Untreated Stimuli Mass Schedule Distributed Schedule 1. Base 2. House 3. Lace 4. Check 5. Pies 6. Cane 7. Type 8. Dive 9. Chew 10. Peas 11. Hive 12. Face 13. Bone 14. Five 15. Cheek 16. Big 17. Pave 18. Lion 19. Kneel 20. Yes 1. Knee 2. Toss 3. Net 4. Sheep 5. Pitch 6. Bit 7. Mitt 8. Fish 9. Vet 10. Cello 11. Leave 12. Dig 13. Seal 14. Night 15. Goose 16. Cone 17. Move 18. Sun 19. Chat 20. Nut 1. Balloon 2. Dove 3. Keys 4. Eyes 5. Shop 6. Lies 7. Wave 8. Knock 9. Bus 10. Map 11. Shoes 12. Knife 13. Choke 14. Cup 15. Mop 16. Gill 17. Love 18. Cape 19. Neck 20. Toys Table B5 Participant 5 Stimuli List Untreated Stimuli Mass Schedule Distributed Schedule 1. Puff 2. China 3. Game 4. Wood 5. Hoof 6. Duck 7. Back 8. Pull 9. Wave 10. Man 11. Bag 12. Kit 13. Day 14. Kiss 15. Bat 16. Type 17. Duke 18. Choke 19. Chat 20. Feet 1. Comb 2. Tide 3. Cheat 4. Pool 5. Moon 6. Bike 7. Cheetah 8. Say 9. Bee 10. Key 11. Fat 12. Mood 13. Go 14. Paid 15. Map 16. See 17. Pave 18. Seat 19. Loop 20. Add 1. Sack 2. Fight 3. Teach 4. Pad 5. Tail 6. Match 7. Gate 8. Boot 9. Mall 10. Watch 11. Tape 12. Same 13. Cheek 14. Knife 15. Jeep 16. Guy 17. Hood 18. Mad 19. Beak 20. Cat