Elsevier

Journal of Communication Disorders

Volume 87, September–October 2020, 106026
Journal of Communication Disorders

The influence of type of feedback during tablet-based delivery of intensive treatment for childhood apraxia of speech

https://doi.org/10.1016/j.jcomdis.2020.106026Get rights and content

Highlights

  • Tablet-delivered treatment was favorably received by children and families.

  • Only children provided with 100 % knowledge of performance showed significant immediate gains.

  • It is not yet clear which children are best suited to knowledge of results feedback.

  • Both feedback conditions produced significant gains in accuracy over time.

Abstract

Purpose

One of the key principles of motor learning supports using knowledge of results feedback (KR, i.e., whether a response was correct / incorrect only) during high intensity motor practice, rather than knowledge of performance (KP, i.e., whether and how a response was correct/incorrect). In the future, mobile technology equipped with automatic speech recognition (ASR) could provide KR feedback, enabling this practice to move outside the clinic, supplementing speech pathology sessions and reducing burden on already stretched speech-language pathology resources. Here, we employ a randomized controlled trial design to test the impact of KR vs KP feedback on children’s response to the Nuffield Dyspraxia Programme 3, delivered through an android tablet. At the time of testing, ASR was not feasible and so correctness of responses was decided by the treating clinician.

Method

Fourteen children with CAS, aged 4–10 years, participated in a parallel group design, matched for age and severity of CAS. Both groups attended a university clinic for 1-hr therapy sessions 4 days a week for 3 weeks. One group received high frequency feedback comprised of both KR and KP, in the style of traditional, face-to-face intensive intervention on all days. The other group received high frequency KR + KP feedback on 1 day per week and high frequency KR feedback on the other 3 days per week, simulating the service delivery model of one clinic session per week supported by tablet-based home practice.

Results

Both groups had significantly improved speech outcomes at 4-months post-treatment. Post-hoc comparisons suggested that only the KP group showed a significant change from pre- to immediately post-treatment but the group difference had dissipated by 1-month post-treatment. Heterogeneity in response to intervention within the groups suggests that other factors, not measured here, may be having a substantive influence on response to intervention and feedback type.

Conclusion

Mobile technology has the potential to increase motivation and engagement with therapy and to mitigate barriers associated with distance and access to speech pathology services. Further research is needed to explore the influence of type and frequency of feedback on motor learning, optimal timing for transitioning from KP to KR feedback, and how these parameters interact with task, child and context-related factors.

Introduction

Childhood apraxia of speech (CAS) is a disorder of speech motor control that causes substantial disruption to development of intelligible and natural sounding speech (ASHA, 2007). The speech of children with CAS is notable for substitutions and distortions of speech sounds and altered prosody. As a disorder of speech motor control, it is often recommended that CAS treatment apply principles of motor learning (PML), including high frequency of treatment sessions and high numbers of practice trials per session (Maas et al., 2008; Schmidt & Lee, 2011). However, parents often report difficulty accessing, attending and affording this level of clinical care and a willingness for alternative service delivery methods to alleviate these burdens (Ruggero, McCabe, Ballard, & Munro, 2012). It is here that theory-driven development of mobile technology applications can offer children with CAS access to engaging high intensity speech therapy that follows the best-practice PML. Apps can provide animated platforms in both exercise and game formats, to engage and motivate children to undertake the difficult and often tedious intensive practice. Incorporating state of the art automated speech recognition (ASR) also has potential, in the future, for delivering feedback on accuracy of speech attempts, leading to reduced burden on clinician time and the cost born by families. However, ASR, by nature, can only deliver feedback on correctness (i.e., knowledge of results, KR), and non-speech motor learning literature suggests that people benefit from knowledge of performance (KP) feedback when first acquiring new motor skills (Maas et al., 2008; Schmidt & Lee, 2011). It is not yet known how some of these principles of motor learning apply to children with CAS and whether some children respond differently to these principles than others. Maas and colleagues have reported that different children respond differently to principles of practice schedule (Maas & Farinella, 2012) and feedback frequency (Maas, Butalla, & Farinella, 2012). However, no study has specifically compared the effects of KP and KR on speech motor learning in children with CAS. It is important to know this before we embark on prescribing apps that may force specific principles or aspects of therapeutic method.

Here, we explore the impact of using KP versus KR feedback, (a) to test whether children respond differently to these two types of feedback in a controlled setting, and (b) to inform future development of speech therapy apps as ASR technology comes online. To isolate the effects of feedback type, independent of the performance of any ASR algorithm, clinicians supervised the app-delivered treatment and made perceptual judgments of children’s speech for feedback decisions.

There are different approaches to treatment for CAS currently used around the world. These include motor-based approaches, linguistic approaches and multi-modal communication approaches. In a systematic review of the evidence on treatment for CAS, Murray et al. (2014) identified three treatment protocols as having the strongest levels of evidence to support their use in a clinical setting to achieve positive treatment, maintenance and generalization effects. These included Dynamic Temporal and Tactile Cueing [DTTC] (Strand, Stoekel, & Baas, 2006), Rapid Syllable Transition Treatment [ReST] (Ballard, Robin, McCabe, & McDonald, 2010; Murray, McCabe, & Ballard, 2012), and Integrated Phonological Awareness Intervention (Moriarty & Gillon, 2006). There was suggestive evidence for ten other treatment approaches including the Nuffield Dyspraxia Programme – Third Edition (NDP3; Williams & Stephens, 2004), which is commonly used across Australia as best-practice (Gomez, McCabe, & Purcell, 2018). This review then led to the first and, currently, only randomized controlled trial (RCT) of treatment for CAS, comparing the NDP3 and ReST (Murray, McCabe, & Ballard, 2015). Results of the RCT indicated that both NDP3 and ReST treatments resulted in similar positive treatment outcomes, particularly for generalization to real words. The authors reported that NDP3 demonstrated greater immediate gains in speech accuracy and ReST treatment lead to better maintenance of treatment gains and generalization to untreated pseudo-words. However, a subsequent Cochrane review (Morgan, Murray, & Liégeois, 2018) offered a more conservative interpretation of these findings based on a re-analysis. This suggested no reliable difference existed between the two treatment groups on acquisition or maintenance of targets based on small absolute mean differences in accuracy scores between the groups and that both treatment protocols demonstrated a similar, moderate level of evidence (Morgan et al., 2018). Thus, the app-delivered intervention approach employed in this study is modeled on the NDP3 given the moderate level of evidence for positive treatment and maintenance effects as demonstrated by the RCT (Morgan et al., 2018; Murray et al., 2015). In addition, the NDP3 uses real words, which potentially can be analyzed with ASR algorithms built on databases of incidental speech; and its theoretical basis is in motor learning as a complex and hierarchical skill involving sounds, syllable shapes, words and prosody which are developed through repetition, elicitation and the provision of frequent, specific feedback on performance and results (Williams & Stephens, 2004, 2010). This is similar to the pre-practice phase of a motor learning approach.

Much of what we know about principles of motor learning (PML) has come from limb movement studies in non-disordered populations or investigations involving adults with acquired apraxia of speech (AOS) or dysarthria. Limb movement studies have demonstrated that greater long-term learning occurs when practice of motor targets is variable, randomized, and frequent, with delayed feedback provided on an intermittent schedule (see Maas et al., 2008 for a review). However, investigation into adult motor speech disorders revealed that some participants benefited more from low frequency feedback and others from high frequency feedback, with similar mixed results when exploring the effects of delayed versus immediate feedback (Austermann Hula, Robin, Maas, Ballard, & Schmidt, 2008). The type of feedback received also influences acquisition and retention effects. Non-speech motor learning literature has demonstrated that detailed feedback about the motor movement, designed to guide and shape subsequent movements towards a correct response (i.e., Knowledge of Performance, KP; e.g., “straighten your arm more quickly”) enhances acquisition but potentially inhibits maintenance of skill post-treatment. In contrast, feedback on the outcome or correctness of the completed movement (i.e., Knowledge of Results, KR; e.g., “That’s right” or “Not that time”) leads to greater maintenance of skill (Schmidt & Lee, 2011). However, KR is most effective when the learner has some internal representation of the target movement program and some ability to self-evaluate and self-correct (see Maas et al., 2008 for a review of non-speech and speech motor learning literature). As mentioned previously, the effects of KP and KR has not been directly compared in speech motor learning.

Few studies have explicitly investigated the influence of specific principles of motor learning in CAS. The principles that have been studied include (a) amount of practice, where providing ∼ 150 trials per session leads to greater treatment, generalization and maintenance effects than only 30–40 trials per session (Edeal & Gildersleeve-Neumann, 2011); (b) treatment intensity, where twice weekly treatment sessions led to significantly better outcomes than once per week treatment sessions (Namasivayam et al., 2015); (c) practice schedule (i.e., blocked versus random practice; Maas & Farinella, 2012), where findings were mixed across participants; (d) feedback frequency (i.e., low versus high frequency feedback; Maas et al., 2012) where findings were also mixed across participants (see Maas, Gildersleeve-Neumann, Jakielski, & Stoekel, 2014 for a review); and (e) distribution of practice (i.e., closely distributed at four times weekly for three weeks versus less closely distributed at twice weekly for six weeks; Thomas, McCabe, & Ballard, 2014).

In their RCT comparing treatment outcomes from the NDP3 and ReST treatments, Murray et al. (2015) suggested that inherent differences in the type and frequency of feedback provided to children may have influenced children’s responses to intervention. Although both groups made significant improvements with treatment, NDP3 treatment, which uses 100 % KP feedback, effected greater improvement on treated targets immediately post-treatment (i.e., greater acquisition) than ReST intervention, which utilizes 50 % KR feedback only (in line with PML). Conversely, the ReST group showed greater maintenance of treatment effects than the NDP3 group. The results from Murray et al. (2015) are consistent with previous work arguing that high frequency KP feedback confers an acquisition advantage, while low frequency KR feedback confers a maintenance advantage (e.g., Maas et al., 2008; Schmidt & Lee, 2011). Despite the more conservative interpretation offered by Morgan et al. (2018) concluding no reliable difference between the two treatments, exploration of the effects of feedback type and feedback frequency is warranted given that others have also reported equivocal effects for feedback parameters such as frequency when treating CAS (Maas et al., 2012). The current study was designed to specifically investigate the influence of the type of feedback received during speech production practice when therapy was delivered using mobile technology that has potential to provide KR feedback only via ASR. KP versus KR feedback has not previously been systematically compared; that is, the study of Murray et al. (2015) varied many components between their two treatment approaches such that the specific effect of feedback type could not be determined. To isolate the effect of feedback type, we maintained feedback frequency at 100 % for both experimental groups and administered the same treatment protocol to both groups.

Despite research consistently demonstrating that best practice intervention frequency for speech sound disorders, including CAS, is between 2 and 4 sessions per week (Murray, McCabe, & Ballard, 2014; Namasivayam et al., 2015; Sugden, Baker, Munro, Williams, & Trivette, 2018; Thomas et al., 2014), these intervention frequencies are uncommon in clinical practice (Gomez et al., 2018; Ruggero et al., 2012; Sugden, Baker, Munro, Williams, & Trivette, 2017). Parent involvement and home practice activities are routinely prescribed as a way to supplement face-to-face therapy sessions with a clinician (Lim, McCabe, & Purcell, 2017; Sugden, Baker, Munro, & Williams, 2016, 2017). Homework can also provide the frequent and regular practice of speech production targets that is needed for children to acquire new skills and habitualize these new movement skills, as well as different but related movement skills, into non-intervention contexts (Gordon-Brannan & Weiss, 2007; McLeod & Baker, 2017; Olswang & Bain, 2013). Effective home practice requires that the child be motivated to engage in their practice activities and that parents or carers can be available to supervise the practice sessions and provide feedback on the accuracy of the child’s productions. However, parents and children perceive speech practice as “work” (Thomas, McCabe, & Ballard, 2017; McAllister, McCormack, McLeod, & Harrison, 2011). In addition, research has demonstrated that treatment fidelity (i.e., adherence to the treatment protocol) and difficulty judging the accuracy of their child’s speech production attempts have been identified as barriers to the efficacy of parent-implemented approaches (Lim et al., 2017; Thomas et al., 2017; Thomas, McCabe, Ballard, & Bricker-Katz, 2018). It is here that computer-based or app-delivered home practice can be useful for strict protocol delivery as well as increasing a child’s engagement and motivation to participate in speech homework (Hair, Monroe, Ahmed, Ballard, & Gutierrez-Osuna, 2018; Nordness & Beukelman, 2010; Toki & Pange, 2010).

Computer software packages designed to act as a virtual speech-language pathologist (SLP) can be effective for a range of pediatric speech disorders (Chen et al., 2016; Furlong, Erickson, & Morris, 2017). Fewer than half of the programs in two recent review papers provided feedback to the user on speech attempts (Chen et al., 2016; Furlong et al., 2017). Program-driven feedback was mostly non-specific (e.g., a visual speech waveform or tracking number of trials completed). Explicit feedback on speech accuracy was experimenter/clinician controlled and judged. None of the included studies in either review included mobile technology.

The efficacy or effectiveness of therapy for pediatric speech sound disorders delivered via tablet and smartphone applications, however, has not been empirically tested (McKechnie, Ahmed, Gutierrez-Osuna, Monroe, McCabe & Ballard, 2018). This may be due in part to the risks in running time- and cost-intensive experimental trials in the fast turnover environment of the app market, compared with the relative low cost, both of producing and purchasing an app, and perceived low risk of the products themselves, given that apps are often frequently similar to printed materials already in use or form only a small part of an overall intervention approach (Edwards & Dukhovny, 2017). However, recent analysis of the quality and potential therapeutic benefit of mobile applications for children’s speech disorders found that less than 3% of more than 5000 identified apps met criteria that would warrant full evaluation (Furlong, Morris, Serry, & Erickson, 2018). Of that 3% (132 unique apps that were appraised), only 19 apps (14 %) were deemed to have therapeutic potential in the sense that engagement with the app offered a positive potential impact on the user’s speech sound disorder (Furlong et al., 2018). Positive potential impact was derived from the behavior change scale from the Perceived Impact section of the Mobile App Rating Scale (MARS; Stoyanov, Hides, Kavanagh, Zelenko, Tjondronegoro & Mani, 2015), where perceived impact was defined as potential to improve the user’s awareness, knowledge, attitudes, help-seeking behavior as well as the perceived behavioral change from use of the app (Stoyanov et al., 2015). Note that this analysis is not a valid replacement for well-controlled experimental studies of efficacy or effectiveness, given the high risk of bias.

Each app evaluated in Furlong et al. (2018) was rated by two reviewers on a 5-point likert scale from strongly disagree to strongly agree for the statement “the use of this app is likely to increase/decrease [insert target health behavior]”. Raters were experienced speech-language pathologists. Inter-rater reliability, calculated using two-way mixed Interclass Correlation, was reported only for the total MARS score which did not include the score for perceived impact. Nonetheless, satisfactory inter-rater reliability was obtained (ICC .72) (Furlong et al., 2018).

The majority of available computer- or app-based intervention tools offer digital stimulus presentation via engaging graphics and sound effects. They typically do not provide the child with explicit feedback on the accuracy of their productions (KP or KR) nor offer remote and/or automated assessment for the SLP to monitor. The lack of integrated, automated feedback is largely due to the challenges involved in developing ASR software which can provide decisions on speech production accuracy that are highly reliable with expert clinician judgements and delivered in a timely manner (see McKechnie et al., 2018 for a review). There has been limited research on computer-based or mobile technology approaches for CAS, perhaps due to the historical challenges in defining a relatively homogeneous group of children for testing and developing computerized approaches that treat the range of CAS features, not just segmental accuracy.

There is a mismatch between the need for children with CAS to receive intensive treatment and the reality of service delivery models in Australia and elsewhere. As a result, recent research evidence has emerged exploring the efficacy of ReST intervention delivered via alternative methods including tele-practice (Thomas, McCabe, Ballard, & Lincoln, 2016) and combined clinician- and parent-delivered methods (Thomas et al., 2017). However, there has currently been no well-controlled study published that investigates alternatives to fully clinician-delivered (in person or via video conferencing) speech therapy. In light of the fact that NDP3 is the most frequently used intervention for CAS in Australia (Gomez et al., 2018), it is timely to consider the impact of alternative service delivery methods on treatment efficacy for CAS using NDP3 intervention.

In light of the low number of commercially available apps deemed to have potential for therapeutic impact (Furlong et al., 2018), our group have developed Tabby Talks, which is a theory-driven, multi-tiered system for facilitating remote access to speech-language pathology services (Parnandi et al., 2013, 2015). Tabby Talks consists of three components: (1) Android platform application running on mobile tablets, (2) server-based learning management software (i.e., Moodle) running a speech analysis engine to evaluate children’s speech attempts offline for assessment of progress in therapy, and (3) a clinician interface allowing for the remote management and updating of clients and therapy exercises (see Table 1 and Fig. 1).

The clinician interface allows the clinician to create individual client profiles and input demographic information about each client; to create exercises containing customizable sets of stimuli to target specific speech production goals; and to enroll clients into exercises based on their individual assessment data. The Moodle server houses the speech analysis engine which evaluates the children’s speech offline and communicates data back to the clinician interface to allow for remote monitoring of progress through therapy (See Fig. 2A-B). The application running on the tablets requires each client user to log in with an individualized user-name and password. The app then displays the set of exercises into which that client has been enrolled and enables practice of speech targets in either a flash card/swipe through display or a simple memory game format. The user interface has icons for the client to touch, which will either play an audio model of the target word or start an audio recording to capture the client’s production attempt. The user interface tested here also has a non-ASR-based scoring system where the clinician can make a perceptual judgement of speech accuracy by selecting/awarding either a gold (for correct) or silver (for incorrect) star (see Fig. 3A–C). Additional details about the design and development of the Tabby Talks infrastructure have been published elsewhere (Parnandi et al., 2013, 2015).

One of the critical considerations, when exploring alternative service delivery options for CAS, is how these options will affect the structure of the treatment protocol and how different PMLs can be incorporated. Given the intention to incorporate ASR into Tabby Talks, and its restriction to KR feedback only, the first step in testing our program was to examine the impact of delivering primarily KR feedback (i.e., right / wrong decisions) vs KP feedback. While the PML approach advocates KR feedback during practice for best maintenance of treatment effects, a learner must first be trained in producing the target movement skills accurately through what is referred to as pre-practice. Pre-practice, unlike practice, is where the clinician/trainer provides detailed KP feedback to guide and shape performance so that the learner can experience the sensorimotor consequences of performing the targeted movement(s) correctly. Pre-practice serves to guide the learner in developing an internal reference of correctness that can be accessed later during practice, once KP is removed. This internal reference is needed to guide self-evaluation and self-correction. We propose that Tabby Talks may be best used in between the weekly in-clinic pre-practice sessions with the speech-language pathologist, to provide high intensity and frequent practice on speech behaviors that the child has begun to acquire.

This study aims to (a) explicitly investigate the influence of type of feedback (primarily KP vs primarily KR) on response to treatment for CAS using the Tabby Talks app; and (b) to determine the feasibility for such technology and software to provide an effective supplement to face-to-face intensive treatment. We compared intensive 4 days / week therapy with the clinician (100 % KP) with the more common model of once / week with the clinician (100 % KP) and 3 days / week with the app (100 % KR). Being an efficacy study, all sessions were conducted in the research clinic under supervision of the research team, so that parameters of the treatment protocol, including the frequency and type of feedback, could be carefully controlled. In this way, we simulated the once / week in-clinic model (i.e., the intervening 3 “at-home” sessions were completed in the clinic) and we also simulated the ASR-style KR that the app would provide during home practice sessions. Here, Tabby Talks was populated with stimuli from the NDP3 (with permission from the authors, Williams & Stephens, 2004).

All conditions other than feedback type were held constant across the two groups: children in both treatment conditions attended the clinic for all therapy sessions, all sessions were delivered by trained student speech-language pathologists under the supervision of experienced clinicians, all treatment stimuli were delivered via the Tabby Talks app, and the student clinicians delivered all feedback verbally.

This study explored the impact of using KP versus KR feedback, (a) to test whether children respond differently to these two types of feedback in a controlled setting, and (b) to inform future development of speech therapy apps as advancements in ASR technology progress towards full online integration into the apps. We also compared the two methods of treatment used here to our historical data for traditional paper-based delivery of NDP3 (Murray et al., 2015). Further, we invited participants to complete a questionnaire exploring satisfaction with the treatment process; motivation and engagement with therapy activities; app features, likes, and dislikes; ease of use; and interest in further treatment using the app. These results will be used to further inform future app development and are available as supplemental materials. We hypothesized that:

  • (i)

    Tablet-based delivery of NDP3 using high frequency KP feedback would obtain similar treatment outcomes to Murray et al.’s (2015) traditional paper-based delivery of NDP3.

  • (ii)

    Compared to participants in the high frequency KP group and the traditional paper-based NDP3 group, participants in the high frequency KR condition may demonstrate smaller treatment gains immediately post-treatment (i.e., evidence of slower acquisition and generalization) but greater maintenance at 1- and 4- months post-treatment (i.e., evidence of more robust learning).

  • (iii)

    The experimental groups would demonstrate at least similar long-term outcomes to Murray et al.’s (2015) traditional NDP3 delivery.

Section snippets

Method

This study was approved by the Human Research Ethics Committee at the University of Sydney (Protocol number 2013/703). All parents provided written informed consent for their child to participate and children older than 6 years of age provided written assent.

Results

To assess for treatment and generalization effects, first order autoregressive and unstructured linear mixed effects models were tested with and without the covariates of age and baseline speech disorder severity (i.e., PPC score for the Single Word Test of Polysyllables). In all cases, except for age for the treated items, both covariates were significant. For all dependent variables, the unstructured model including the covariate of severity provided the best fit, with residuals being

Discussion

This study compared two methods of feedback during tablet-delivered NDP3 treatment. This investigation is a necessary first step towards determining whether app-delivered right/wrong (KR) feedback during intensive at-home practice of new motor speech targets can effectively facilitate acquisition and maintenance of new segmental and suprasegmental speech patterns. Such technology has the potential to bridge the gap between optimal service delivery intensity in CAS and current service delivery

Conclusions

Mobile technology has the potential to increase the engagement and motivation of clients and to facilitate intensive practice of speech production targets (e.g., Hair et al., 2018). Combined with less frequent direct clinical contact via face-to-face sessions or telehealth, it can also mitigate barriers of distance and access to services for rural and remote families. With continued advancements in technology and the development and integration of accurate and reliable ASR software, mobile

CRediT authorship contribution statement

Jacqueline McKechnie: Software, Investigation, Data curation, Visualization, Supervision, Formal analysis, Writing - original draft, Writing - review & editing. Beena Ahmed: Conceptualization, Software, Resources, Writing - review & editing, Funding acquisition. Ricardo Gutierrez-Osuna: Conceptualization, Software, Resources, Writing - review & editing, Funding acquisition. Elizabeth Murray: Investigation, Supervision, Writing - review & editing. Patricia McCabe: Methodology, Investigation,

Acknowledgements

This research was made possible by NPRP Grant # 8-293-2-124 and 4-638-2-236 (Ahmed, Gutierrez, Ballard) from the Qatar National Research Fund (a member of the Qatar Foundation) and an Australian Postgraduate Award (McKechnie). During this project, Ballard was supported by an Australian Research Council Future FellowshipFT120100355. The statements made herein are solely the responsibility of the authors. The authors wish to thank the participants and families for their commitment to the

References (67)

  • L.M. Dunn et al.

    Peabody picture vocabulary test

    (2007)
  • D.M. Edeal et al.

    The importance of production frequency in therapy for childhood apraxia of speech

    American Journal of Speech-language Pathology

    (2011)
  • J. Edwards et al.

    Technology training in Speech-Language Pathology: A focus on tablets and apps

    Perspectives of the ASHA Special Interest Groups, SIG 10

    (2017)
  • L. Furlong et al.

    Mobile apps for treatment of speech disorders in children: An evidence-based analysis of quality and efficacy

    PloS One

    (2018)
  • R. Goldman et al.

    Goldman-Fristoe test of articulation

    (2000)
  • M. Gomez et al.

    Clinical management of childhood apraxia of speech: A survey of speech-language pathologists

  • M.E. Gordon-Brannan et al.

    Clinical management of articulatory and phonologic disorders

    (2007)
  • H. Gozzard et al.

    Single word test of polysyllables

    (2004)
  • H. Gozzard et al.

    Requests for clarification and children’s speech responses: Changing “pasghetti” to “spaghetti”

    Child Language Teaching and Therapy

    (2008)
  • A. Hair et al.

    Apraxia World: A speech therapy game for children with speech sound disorders

  • IBM Corp

    IBM SPSS statistics for windows (Version 24.0)

    (2016)
  • J. Iuzzini et al.

    Evaluation of a combined treatment approach for childhood apraxia of speech

    Clinical Linguistics & Phonetics

    (2010)
  • T.R. Kratochwill et al.

    Single-case designs technical documentation

    (2010)
  • J.M. Lim et al.

    Challenges and solutions in speech-language pathology service delivery across Australia and Canada

    European Journal for Person Centred Healthcare

    (2017)
  • E. Maas et al.

    Random versus blocked practice in treatment for childhood apraxia of speech

    Journal of Speech Language and Hearing Research

    (2012)
  • E. Maas et al.

    Feedback frequency in treatment for childhood apraxia of speech

    American Journal of Speech-language Pathology

    (2012)
  • E. Maas et al.

    Motor-based intervention protocols in treatment of childhood apraxia of speech

    Current Developmental Disorders Reports

    (2014)
  • E. Maas et al.

    Principles of motor learning in treatment of motor speech disorders

    American Journal of Speech-language Pathology

    (2008)
  • L. McAllister et al.

    Expectations and experiences of accessing and participating in services for childhood speech impairment

    International Journal of Speech-language Pathology

    (2011)
  • P. McCabe et al.

    Orthographically sensitive treatment for dysprosody in children with Childhood Apraxia of Speech using ReST intervention

    Developmental Neurorehabilitation

    (2014)
  • J. McKechnie et al.

    Automated speech analysis tools for children’s speech production: A systematic literature review

    International Journal of Speech-language Pathology

    (2018)
  • S. McLeod et al.

    Children’s speech: An evidence-based approach to assessment and intervention

    (2017)
  • H. Miller et al.

    Treatment of lexical stress, segmentation, and sound distortions in childhood apraxia of speech

  • View full text