A deficit- or process-oriented approach to intervention (Wilson, 2004; Sugden, 2007; Blank et al., 2012; Smits-Engelsman et al., 2012) also known as the General Abilities Approach (Pless and Carlsson, 2000) is based on maturational, neurodevelopmental theory and on the assumption that underlying sensorimotor processes need to be remedied and improvements lead to a better performance of motor skills. An example of a bottom-up approach is Sensory Integration therapy (SIT) (Ayers, 1972). SIT uses sensory stimulation in order to enhance motor development and cognitive abilities but does not include integration of primary reflexes in order to develop motor milestones such as rolling, belly crawling and crawling on hands and knees (Niklasson, 2013a; Lin et al., 2014).

A functional skills/task-oriented approach to intervention (Pless and Carlsson, 2000; Wilson, 2004; Thelen and Smith, 2006; Sugden, 2007; Blank et al., 2012; Smits-Engelsman et al., 2012) is a top-down approach, which implies that the child or adult is taught how to handle a task, a special skill or a strategy in its environmental context. The most promising and also the most recommended method is Cognitive Orientation to Daily Occupational Performance (CO-OP). Coming from the field of occupational therapy, CO-OP is mainly focused on cognitive strategies for the acquisition of age-specific functional skills.

Among skills/task-oriented approaches a perspective on motor development provided by the framework of dynamical systems theory (DST) has gained much attention and has been influential for developmental psychology research for more than 25 years, mainly due to its connection to physical science (Hollenstein, 2011). Originated in mathematics DST emphasizes i.e., “changes over time,” “emergence,” “non-linear,” and “self-organization” (Thelen and Smith, 2006). Using the framework of DST a dynamic systems perspective on motor development (Goldfield and Wolff, 2004) holds that motor behavior is an emerging result of the interaction between different subsystems. Without any special entity directing it, the system is self-organizing with changes appearing to be non-linear. An example could be the transition from crawling to walking. According to the dynamic systems perspective this transition is due to a dynamic interaction between the child's ability, the context of the environment and the task in question. An important force for a developmental change is the child's motivation to explore (Thelen, 2000; Thelen and Smith, 2006). According to the dynamic perspective the nervous system is a part of an embodied system with the brain acting as a medium “informationally coupled” to the environment yielding patterns for control and coordination mainly through visual information (Goldfield and Wolff, 2004). Shumway-Cook and Woollacott (1995) meant that one limitation of this perspective could be the anticipation that the nervous system is supposed to play a rather unimportant role.

From our point of view this remark is well founded since our studies (Niklasson et al., 2009, 2015) have indicated that the persistence of aberrant primary reflexes together with a vestibular dysfunction are hindrances for the development of motor milestones. Therefore, we tentatively suggest that motivation is not enough. The child must primarily be able to handle gravity (Adolph and Franchak, 2017) in order to develop a dynamic interaction between perception and movement.

In accordance with substantial research (e.g., Rasmussen and Gillberg, 2000), which has shown that individuals do not grow out of motor problems, DSM-5 (American Psychiatric Association., 2013; Purcell et al., 2015) recognizes DCD as persistent throughout the life span. However, standardized tests for adults are lacking. Sometimes MABC-2 and BOT-2 are used but their psychometric properties when used to assess adults are still unclear (Barnett, 2014). So far, a self-rating questionnaire such as the Adult Developmental Co-ordination Disorders/Dyspraxia Checklist (ADC) (Kirby et al., 2010) is most commonly used. ADC is a comprehensive instrument, which was developed and tested on individuals in the age range of 17 to 42 years. It consists of 40 items, 10 rates childhood performance while 30 rates current adult performance. The items cover both motor behavior, for example movement skills and handwriting, and behavior beyond the motor capacity such as social skills and general organization. Despite the growing awareness and knowledge about DCD in adults during recent years, evidence based research is still scarce (Missiuna et al., 2012; Tal-Saban et al., 2012) and the need to develop appropriate assessment tools for adults is urgent (Barnett, 2014).

Our recent study (Niklasson et al., 2015), which used the method Retraining for Balance in a group of adults, indicated promising results. The method follows hierarchical principles (Wiest, 2012) and includes both assessment tools and guidelines for intervention regarding persistent primary reflexes and vestibular dysfunction. The main logic behind RB is that un-integrated primary reflexes together with vestibular dysfunction cause an arrested sensorimotor development, which will be released only through a proper and methodical suppression of the reflexes. This is why the method can be used and also apparently works in all age groups. There is a strong connection between certain primary reflexes and the vestibular system. One example of special interest is the Moro reflex, which is expected to be gradually suppressed during the first 6 months of life. Nonetheless it was frequently found among participants in both the child- and the adult-group in our study (Niklasson et al., 2015). Another study by Konicarova and Bob (2012) showed that a persisting Moro reflex was closely linked to Attention Deficit Hyperactivity Disorder (ADHD) in school children aged 8 to 11 years.

In the newborn the Moro reflex is unconditioned and released through a sharp alteration of the infant's position in space. The first form of anxiety is then the fear of falling, closely connected to the excitement of the vestibular part of the eighth cranial nerve. The other part of this cranial nerve, which is connected to the auditory system, is stimulated by sudden loud noises. Since the newborn is rather insensitive to noise the reaction and fear for loud and/or sudden noises, a Startle reflex develops somewhat later and is conditioned. The common denominator for both anxiety and fear is an irritation of at least one of the branches of the eighth cranial nerve (Feldenkrais, 1988). Although it has been argued that The Moro reflex and the Startle reflex are different entities (e.g., Rousseau et al., 2017) they are thought to be part of the same developmental chain (Goddard Blythe, 2014) and therefore we find it likely that they are two sides of the same coin. In a recent study, Poli and Angrilli (2015) demonstrated that a strong Startle reflex was associated with higher anxiety levels. We speculate that retention of the Moro reflex and a too easily elicited Startle reflex among both adolescents and adults with sensorimotor problems and/or anxiety is more common than expected. Therefore, we suggest that the hitherto neglected roles of persistent primary reflexes and of vestibular dysfunction ought to be parts of future DCD diagnosing. Retraining for Balance constitute such a diagnosing instrument.

Through a combination of theoretical knowledge (e.g., Ayers, 1972; Thelen, 2000) and clinical practice (e.g., Blythe, 2009) we came to understand the importance of mixing the inhibition of aberrant primary reflexes with vestibular stimulation and thereby creating assessments and interventions grounded in fundamental neurological development. In a previous naturalistic study (Niklasson et al., 2009), comprised of 232 children and adolescents diagnosed with sensorimotor disorder, results showed that the integration of persisting primary reflexes together with vestibular stimulation according to the method Retraining for Balance (RB) enhanced sensorimotor development. At the time we could not find any reports on equal interventions with adults. Therefore, we decided to study a group of adults who had voluntary signed up to be assessed and treated for SMD and compare their results with a group of older children with an average age of 12 years (Niklasson et al., 2015).

Training according to RB (Niklasson et al., 2007) is a process, which takes on average 3 years. During that time participants are doing exercises for about 15 min/day at home together with either their parents or their spouse. The method is a blend of five different approaches and the perspective includes the following notions:

The deficit- or process-oriented approach (Wilson, 2004; Sugden, 2007; Blank et al., 2012; Smits-Engelsman et al., 2012) or the General Abilities Approach (GA) (Pless and Carlsson, 2000; Niklasson et al., 2009), which implies that deficits in underlying neurological structures needs to be remediated. A logic of RB as well as of GA is that persistent primary reflexes will become a hindrance for development of postural reactions and of gross motor milestones such as rolling, belly-crawling and creeping on hands and knees which in turn will affect the development of more complex motor skills.

In our recent study (Niklasson et al., 2015), prior to their first visit, the participants were interviewed on telephone by one of two therapists. After the conversation a decision was made as whether or not the individual should come for assessment. This verbal history taking is similar to the Adult Developmental Co-ordination Disorders/Dyspraxia Checklist (ADC) (Kirby et al., 2010) and to the Movement ABC-2 checklist or the revised version of the Developmental Coordination Disorder Questionnaire (DCD-Q-R) (Wilson et al., 2009) for children. Regarding the children, Conners' Teacher Rating Scale (Conners, 1990; Janols and von Knorring, 1991; Niklasson et al., 2009) and Conners' Parent Symptom Questionnaire (Conners, 1973; Niklasson et al., 2009) were used before and after intervention. Adult participants, as well as the children's parents, were also asked to complete the questionnaire Reasons for training (Bergström et al., 1999) in order to indicate additional problems. Besides sensorimotor difficulties both groups reported attention problems, generalized anxiety, reading problems and sensitivity to stress. The similarity between the groups might be an indication that comorbidities do not wane with age and hints toward the importance of an early identification of sensorimotor problems in order to minimize additional problems later in life (Missiuna et al., 2015).

In contrast to the instruments Movement ABC-2 (Henderson et al., 2007) and Bruininks-Oseretsky Test of Motor Profiency-2 (Bruininks and Bruininks, 2005), which mainly assess the child's abilities on the functional level, the method Retraining for Balance uses instruments, which primarily evaluate motor patterns below the functional level (Niklasson and Niklasson, 2007a,b). The RB perspective is partly an answer to Wann (2007) who rhetorically wrote that although DCD is a heterogeneous diagnosis “there must be common perceptuomotor subsystems that are poorly developed and refined.” Previously, Sigmundsson (2003) stressed the importance of finding links or fundamental mechanisms between behavior and basic neural processing and argued for a process-oriented approach. We are well aware of the complexity of diagnosing developmental disorders (e.g., Dewey and Bottos, 2004; Flouris et al., 2005; Pennington, 2009; Zwicker et al., 2009) but when it comes to the visibility of sensory and motor difficulties and DCD one logic of RB has been that the lowest common denominator is the interaction between the vestibular system and primary reflexes (Niklasson et al., 2009, 2015). Both vestibular function and primary reflexes belong under the concept neurological soft signs (Ayd, 2000; American Psychiatric Association., 2013).

In our previous study (Niklasson et al., 2015) the concept sensorimotor disorder was tentatively suggested to label a diagnosis of DCD plus vestibular problems. Vestibular dysfunction has so far been overlooked within the framework of DCD but our studies (Niklasson et al., 2009, 2010, 2015) indicate the need for reconsideration. Recently the importance of a well-functioning vestibular system has been stressed in quite a few articles, connecting it to mental health (e.g., Gurvich et al., 2013), cognitive development (e.g., Weiner-Vacher et al., 2013) and a recent editorial (Besnard et al., 2015) declared that “The sky's the limit for new ideas and developments in vestibular therapy (both pharmacological and physical devises).”

The instruments used by the method Retraining for Balance are as follows:

RB-P. Retraining for Balance–Physiological Test (RB-P) (Niklasson and Niklasson, 2007a; Niklasson et al., 2015) is a battery of 41 different subtests aiming at detecting aberrant primary reflexes (e.g., Capute et al., 1978; Blythe, 2009; Goddard Blythe, 2009) and determine the relation between primary reflexes, postural reactions, gross motor milestones, and sports-related gross motor skills (Capute and Accardo, 1991). Of special interest are primary reflexes (McPhillips et al., 2000; Zafeiriou, 2004) connected to the vestibular system i.e., the asymmetrical tonic neck reflex, the tonic labyrinthine reflex, the symmetrical tonic neck reflex and the Moro reflex. The primary reflexes, except the symmetrical tonic neck and the Moro reflexes are assessed in prone, supine, table position and standing position. There are different approaches to how to examine and evaluate neurological soft signs including primary reflexes in children (e.g., Touwen and Prechtl, 1970; Prechtl, 1977; Touwen, 1979; Vaivre-Douret et al., 2016) and adults (e.g., Rodnitzky, 1988). However, to our knowledge only RB uses vestibular related primary reflexes, postural reactions and gross motor skills in combination. A prerequisite for locomotion is the development of posture, which in turn is dependent on the integration of primary reflexes (Zafeiriou, 2004). During the first year of life, starting with the lifting of the head while in prone position the child is normally expected to defy the gravitational force via rolling, belly crawling, creeping on hands and knees and then further to the ability to walk. This developmental succession is of importance for RB. As shown in our previous study (Niklasson et al., 2015) both the younger group and the group of adults diverged from expected ability levels at first assessment. Of special interest is the ability to belly crawl, which we hold as an important transition between activities in prone position and the four feet position. Our experience is in line with both Adolph (2008) who wrote that only about 50% of the infants who later will creep on hands and knees also belly crawl and with Holt (1991) who showed that a skipping of belly crawl might lead to bottom shuffling. Not only will bottom shufflers walk later, they will also miss out valuable time in 4 feet position which should enable them to develop increasing strength in fingers, hands, neck and trunk and the ability to develop a cross pattern. A well-developed cross pattern is a prerequisite for the acquisition of sports-related gross motor skills. Our recent study (Niklasson et al., 2015) showed that even these skills were below expected levels in both groups before intervention.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.