The precursors to motor learning, including speech motor learning are

1. motivation;
   
2. focused attention; and
   
3. pre-practice before entering the practice phase.

The clinician and parents may need to consider a behaviour management plan, implemented by a suitably qualified professional, for children who cannot focus or co-operate easily or who have motivation, attention, or compliance difficulties (Bitter, A15). It is important for parents (and us) to know that simply attending intervention sessions will have little or no impact on the speech of children with CAS unless they engage adequately in the motor learning and other aspects of treatment. It is also important that we recognise the limits of our professional expertise and not attempt to address issues, such as behaviour of concern (Chan et al., 2012) that are best managed by a professional counsellor or other professional, or indeed, by the child’s own family.

The conditions of practice for motor learning, including speech motor learning are

1. motivation;
   
2. goal and target setting (what will be practiced, and how many times);
   
3. instructions (how directions will be delivered);
   
4. modelling (e.g., simultaneous production/ immediate imitation/delayed imitation); and
   
5. the setting and with whom (e.g., where the practice take place and who will help).

Other factors may arise specific to a client. For example, the reinforcement (praise) used should not take up too much time, make too much noise, ‘interrupt’ or distract. It is usually necessary to guide parents in how to deliver reinforcement, providing explicit modelling and practice in sessions (with feedback to them). It is also necessary to choose and develop appealing activities for the child (and to an extent for the parents, too) that will facilitate and invite repeated opportunities for production of target behaviours or utterances.

The type of practice we aim for is repetitive practice, sometimes called motor drill. There must be sufficient trials (or ‘repeats’) of the target behaviour within a practice session for any motor learning to take place and for it to become habituated. Habituation is a step towards more automatic speech output processing.

There are four types of practice schedule, each with advantages and disadvantages. In the ‘real world’, we may not have much choice regarding practice distribution. We must decide, however, which targets to select and how many will be addressed concurrently or sequentially, and communicate this clearly to those concerned: the parents and any other helpers implementing practice away from the treatment room, and where indicated, the child him or herself. The options are: massed practice versus distributed practice; and random practice versus blocked practice.

It is essential during motor drill to give a child frequent information about his or her movement performance, building his or her ‘knowledge’ of what the speech motor apparatus is capable of, what it is doing ‘right now’, and what it did a moment before (just then). Interestingly, there are reports in the cognitive motor literature that adults derive most benefit from finely specified feedback. Conversely, if feedback to children is too specific, their performance can decrease. Skilled observations by the SLP/SLT allow the frequency of feedback to be tailored to suit, bearing in mind that it can distract some children and that, for some, saving any ‘reward’ until the end of a session is the most effective way to proceed.

During pre-practice the clinician models the utterance and provides detailed feedback on ‘movement performance’ to shape correct responses, and to prepare the child for the practice phase. This is called knowledge of performance (KP) feedback. In the practice phase the child should be able to adjust productions independently in the absence of both models and KP feedback. In the practice phase the clinician minimises the volume of modelling, really aiming to provide no models. Instead, knowledge of results (KR) feedback that diminishes over the course of treatment is provided. This KR feedback is delivered in response to about 80% of the child’s responses to start with either in a session or over several sessions, falling off to 10% as his or her capacity for self-monitoring, instating revisions and repairs and engaging in self-reinforcement builds. If the child is not doing well in a practice session with KR feedback only, then some KP feedback may be introduced to get him or her back on track.

There is usually a trade-off between rate and accuracy. A slower rate of production will, up to a point, increase accuracy. Varying the expected rate of production can be an effective technique to incorporate into motor drill, using speech, chanting (Melodic Intonation Therapy) and singing, because it encourages habituation of articulatory movement accuracy while working towards automaticity, a natural rate and natural prosody.

In the mid-1950s, Robert L. Milisen published an article about a multi-layered program for articulation therapy incorporating imitation and auditory and visual models (Milisen, 1954). Milisen’s method, called integral stimulation, has shaped the treatment of functional articulation disorders, the dysarthrias, and acquired apraxia of speech. It utilises hierarchical cueing procedures that begin with high levels of support via simultaneous production of slowly spoken simple utterances with visual and tactile cues. The cues are subtly, and expertly faded and amplified as required until, at the lowest level of support, they disappear completely and the client produces delayed repetition of increasingly complex stimulus items. Research by Rosenbeck, Lemme, Ahern, Harris and Wertz (1973) and Strand and Debertine (2000) shows that integral stimulation intervention in treatment of individuals with apraxia of speech is efficacious.

Although they may be unaware of its precise origins, the children’s version of integral stimulation is widely used by SLPs/SLTs who treat children’s speech and language difficulties. It involves a familiar procedure in which the clinician models an utterance and the child imitates it, while the clinician ensures that the child’s attention is as focused as possible on listening to the model while looking at the clinician’s face (watching the model, if you like).

Integral stimulation proceeds from bottom up, starting with simple phonetic segments and sequences and then short utterances; building in a hierarchy of difficulty to longer and more phonetically complex stimuli. Integral stimulation can be used alone when working with children with CAS, but it is thought to be more effectively applied in combination with tactile and gesture cues that shape the accuracy of articulatory gestures and prosodic cues (Strand, 1995; Strand, Stoeckel & Baas 2006), involving melodic intonation therapy techniques (Helfrich-Miller, 1983, 1984, 1994) or contrastive stress (Velleman, 2002). A prominent feature of the application of the integral-stimulation-combined-with-prosodic-cues approach with children with CAS is that syllable, word and sentence stress are emphasised early in therapy, that is, from the outset, and with young children if possible.

For non-verbal children with severe CAS, for whom the method described above is too difficult, Strand has developed and tested (Strand et al., 2006; see also Jakielski, Kostner & Webb, 2006) a variation of integral stimulation called Dynamic Temporal and Tactile Cueing (DTTC) for Speech Motor Learning. Incorporating the principles of motor learning (see above), it can be used with the non-verbal children who struggle unsuccessfully with the task of articulatory imitation and who seem unable to achieve even the remotest approximation for consonants or vowels. DTTC is an explicitly principled, modified version of the Eight-Step Continuum for Treatment of Acquired Apraxia of Speech (Rosenbeck et al., 1973), originally designed for adult clients with AOS. It allows for what Strand calls ‘a continuous shaping of the movement gesture’, to (1) improve motor planning and (2) program speech processing as speech and language acquisition progresses. The tiny steps and essential adjustments of the therapy dance within DTTC will have a familiar ring to many clinicians, and are as follows.

1. 
   

   Imitation

   
   

   In its implementation, DTTC begins with direct, immediate imitation of natural speech.

   
   
2. 
   

   Simultaneous production with prolonged vowels (most clinician support)

   
   

   If the child cannot imitate, the task is changed to the simplified, more ‘supported’ one of simultaneous production. At this easier level, the SLP/SLT says the utterance at normal volume with the child first, very slowly with the addition of touch cues and/or gesture cues as required. Slowing the utterance by sustaining the vowel ([si::::] rather than [ssssi], as explained in Chapter 6) helps the child, and at the same time lets the SLP/SLT run a visual check to see that the jaw and lip postures are correct (e.g., ensuring that there is no jaw slide and that there is acceptable facial symmetry).

   
   
3. 
   

   Reduction of vowel length

   
   

   As the simultaneous production phase of therapy advances the rate of stimuli production is increased (i.e., vowel length is reduced) allowing the child’s speech output to sound more natural.

   
   
4. 
   

   Gradual increase of rate to normal

   
   

   Practice continues at this level to the point where the child synchronises effortlessly with the therapist at normal rate, with normal movement gestures, and without silent posturing.

   
   
5. 
   

   Reduction of therapist’s vocal loudness, eventually miming

   
   

   Using delicate timing, the SLP/SLT is then in a position to reduce his or her vocal volume, eventually reaching a point where the clinician is producing a mime (mouthing the utterance) as the child says it aloud. Because of the intellectual closeness within the dyad, this can be a tricky point in therapy, and some children will dutifully follow exactly what the adult is doing so that the two are miming at each other! ‘Like a pair of goldfish’ as one parent commented. This is obviously not the goal, and children may need explicit instruction to keep their voice or voice box ‘turned on’ even though the adult’s is ‘off’. The gesture and touch cues may still be needed at this point and will almost certainly be necessary in the next step: the integral stimulation method proper.

   
   
6. 
   

   Direct imitation

   
   

   Ensuring that the child is secure and comfortable with moving to this harder level, the SLP/SLT instructs the child to watch the adult’s face (Look at me for help) while an auditory model is delivered. The child attempts to repeat the model and, if successful, does so many times. If unsuccessful, the therapist may backtrack to the simultaneous model or silent mouthing/miming level described above. Eventually all miming is faded, and the child directly imitates and ‘repeats’ targets numerous times before the final step – step 7 – is introduced.

   
   

   The key to successful implementation of integral stimulation is the clinician’s empathic, informed observations of and sensitivity to what the child is ‘giving’ by way of responses. The professional skill and flexibility involved in continually fine-tuning the hierarchy of stimuli and fine-tuning the amount of support provided to enable the child to imitate spontaneously, is critical. Auditory (including prosodic), visual, and tactile cues and the level of demand on the child are continually augmented and faded in each practice trial according to the child’s responses.

   
   

   The clinician’s alertness to the child’s responses is especially important with the CAS population, who have good and bad days with their speech-processing capacities. The SLP/SLT must be always be prepared to take the therapy ‘down a notch’ if required, and to explain to parents why this is happening.

   
   
7. 
   

   Introduction of a one-or two-second S-R delay (least support)

   
   

   Once the child is directly imitating the therapist’s model with normal rate, with prosody he or she can vary, and with appropriate articulatory gestures, the therapist inserts a new requirement. This is in the form of a one- to two-second delay before the child imitates, so that the child produces a slightly delayed response. To facilitate this for the children who find the delay difficult and want to ‘jump in’, miming while the child produces the delayed response can prove helpful.

   
   
8. 
   

   Spontaneous production

   
   

   Finally, the SLP/SLT elicits short and long spontaneous utterances, for example, by asking the child, ‘What is this called?’ using cloze tasks such as ‘Twinkle, twinkle ___ ___’, sentence completion such as ‘Mother elephant is very big, her baby is ___ ___’, ‘Three things I like about the beach are ___’, engaging in story telling (e.g., with wordless picture books), picture and object description, narrative and role play, and the like.

   
   

   Dr. Edythe Strand, who developed DTTC, is a consultant in the Department of Neurology, Division of Speech Pathology, at the Mayo Clinic in Rochester, Minnesota, and a Professor in the Mayo Medical School. Professor Strand’s primary research and clinical interests have been in neurologically mediated communication disorders, especially developmental and acquired AOS, dysarthria and neurogenic voice disorders. She has published articles and chapters regarding the clinical management of motor speech disorders in children, including treatment efficacy. Responding to Q45, she talks about DA, the Dynamic Evaluation of Motor Speech Skill (DEMSS) and DTTC.