A randomized control trial comparing ultrasound visual feedback and ReST (Rapid Syllable Transition Training) to improve speech in childhood apraxia of speech (CAS).

All children will receive treatment 2 days per week for 6 weeks for a total of 12 sessions. Each session will be 1 hour in duration. ReST Treatment Protocol Stimuli: ReST uses multi-syllable nonword stimuli that include varied consonants and vowels (C, V). Three C sounds are selected from each child’s impaired sounds, plus 3 Vs. All possible CV units for the 6 sounds are randomly combined to form 40 strings, half with strong-weak (SW) stress and half weak-strong (WS) stress (e.g. KAdiku, biTUga), presented orthographically. A random selection of 50% of the stimuli will be treated and 50% will be tested for treatment generalization. Children unable to read the stimuli are given spoken models for imitation until independent production emerges (Ballard et al, 2010). No statistical difference in treatment outcome has been noted between imitation and reading (McCabe et al, 2014). To test ecological validity, 20 real words (same length and sounds as treated items, e.g. toboggan) are tested with naming in a 100-utterance speech sample. Individual treatment sessions conducted by one of a pool of SPs will include 10 minute pre-practice and 50 minute practice in accordance with principles of motor learning (PML) in ReST (Ballard et al, 2010). Pre-practice: Using 10 randomly selected stimuli, the clinician defines parameters of a correct response and guides the child through attempts at reading stimuli aloud or imitating a clinician model, with cues on articulatory accuracy and stress, with 100% feedback on performance. When the child achieves 5 correct productions with the guided cueing, or 10 minutes has elapsed, the session moves on to practice. Practice: 100 trials of reading aloud or imitating production of randomly selected treatment stimuli. Feedback on correctness is provided for 50% of randomly selected trials. Children randomised to ReST will receive 1 hour treatment per day for 2 days per week for 6 weeks, a total dose of 1200 practice trials at half the intensity of the completed RCT (Murray, McCabe & Ballard, 2012)(ANZCTR: ACTRN12612000744853). The completed trial delivered treatment for 1 hour sessions, 4 times per week for 3 weeks. Ultrasound Treatment Protocol Ultrasound biofeedback involves training articulatory patterns using real-time visual displays of the tongue. Based on our previous work in the ultrasound biofeedback program, both in CAS and in Residual Speech Sound Errors, each child will have 4 target sequences available for treatment (CV, VC, or CC), although each session will be structured to address only two of those sequences; thus the target sequences will be cycled through in pairs (target sequence 1 and 2 treated for two sessions, followed by target sequence 3 and 4 treated for two sessions). Individual treatment sessions will be conducted by SPs for 60 minutes. Within the session, 10 minutes of pre-practice with the ultrasound will be provided to allow the clinician to instruct the client in the visual patterns associated with the two desired sequences. This will be followed by 50 minutes of structured practice, broken down into four 12-minute time blocks. These time blocks will alternate, allowing practice with the ultrasound in Time Block A, followed by practice without the ultrasound (i.e., generalization) in Time Block B, ultrasound in Time Block C, and finishing with no ultrasound training in Time Block D. A timer will be used to ensure adherence to this structure. Practice items will begin with the target sequence in isolation (CV, VC, or in the case of a CC target this will be paired with a vowel e.g. CCV), but complexity will increase based on performance (from syllables to monosyllabic words, multisyllabic words, phrases, and sentences). Six trials will be attempted in a block, and advancement to more complex items with the same target sequence will be contingent upon achieving at least 5 of 6 trials correct in the block (i.e., “step up”). Achieving 1 or 0 correct items will mean that the next practice attempt on that target sequence will “step down” to a simpler level of linguistic complexity. If the participant does not successfully “step up” to advance to more complex stimuli, then the next practice block will commence on the other treatment target sequence. Hence, practice will involve alternating between the two treatment targets throughout the session as a way to incorporate practice variability. Within each block, the amount of feedback (knowledge of performance and knowledge of results) will be dependent upon the level of complexity that the child is attempting. Higher frequency feedback with more knowledge of performance (KP) will be provided in blocks targeting simpler linguistic items, and lower frequency feedback with less KP will be used in blocks addressing more complex items. For example: Syllable level: 4/6 trials receive KP+KR (knowledge of results) feedback Monosyllabic word level: 3/6 trials receive KP+KR feedback, 1 receives KR feedback only Multisyllabic word level: 2/6 trials receive KP+KR feedback, 2 receive KR feedback only Phrase level: 1/6 trials receive KP+KR feedback, 2 trials receive KR feedback only Sentence level: 1/6 trials receive KP+KR feedback, 1 trial receive KR feedback only. Participants' parent/carers in both groups will sign a consent form with child participants providing written assent. The Participant Information Statement (PIS) details 2 weekly treatment sessions are required and 3 follow-up sessions. Researchers will monitor participant attendance and reschedule missed appointments for the near future. No other strategies to monitor adherence is required. In our Australian studies to date, only 2 participants of 54 recruited did not finish, due to parental inability to manage transport (Murray et al., 2012 & Thomas, McCabe & Ballard, 2014). Attrition will be addressed through intention to treat analysis. We are the only group in Australia studying CAS intervention. We receive >50 referrals/year to our Sydney treatment research lab. Hence recruitment of the required participants is feasible and strengthened by our well-established networks with peak bodies, community SP clinics, and CAS communities.