Seminar 2 –
Assessment
Peter Wilson:
RMIT University, Melbourne, Australia
Models of movement assessment: What are we doing and where are we going?
This paper examines different models for assessing movement clumsiness in children (or Developmental Coordination Disorder—DCD). I first outline some of the key diagnostic issues that bear on how the disorder might best be conceptualised and assessed. For instance, the term “DCD” itself may mean different things to different people, creating some uncertainty about the nature of the movement dysfunction. The term “coordination” may suggest to some that underlying movement dynamics are somehow “disordered” in these children, while to others it may simply denote that the integrity of movement is not preserved at a functional level. As well, there is continued debate about what the core “symptoms” of DCD are, what cut-off points should demarcate typical from atypical skill development, whether DCD sub-types exist (each with their own markers and treatment approaches), and whether the causal basis of “pure DCD” is the same as that for DCD seen in comorbid conditions like ADHD.
Not surprisingly, approaches to assessment and treatment are varied, each drawing on distinct theoretical assumptions about the aetiology of the condition and its developmental course. I discuss the importance of embedding assessment within a developmental framework. This means more than paying homage to sometimes outmoded theory in the clinical literature, but actively weighing up the strengths and weaknesses of current theory against logical and empirical criteria. A guiding theory should, thus: permit multi-level assessment; acknowledge the interplay between mechanism and process; specify learning mechanisms and constraints; map the level(s) of construct assessment to treatment, and be pragmatic & useful.
In this paper, I define five main approaches to motor assessment according to their broad conceptual origin: Normative Functional Skill Approach, General Abilities Approach, Neurodevelopmental Theory, Dynamical Systems Theory, and the Cognitive Neuroscientific Approach. Each conceptual framework is shown to support assessment and treatment methods with varying degrees of conceptual and psychometric integrity. The Normative Functional Skill Approach supports the major screening devices for DCD and cognitive (or top-down) approaches to intervention. It has origins in the normative period of developmental psychology, drawing on the maturational frameworks of McGraw, Gesell, and Piaget. The goal here was to chart a timetable of development defined by the acquisition of motor milestones, age-directed change over a normal time course, and a descriptive level of analysis as distinct from mechanistic. Consistent with this approach, the diagnostic criteria for DCD also specify deviation from normative standards.
The approach is still the mainstay for diagnosing DCD and has informed the more commonly used descriptive tests for children like the Movement Assessment Battery for Children (MABC) and the Bruininks-Oseretsky Test of Motor Proficiency (BOTMP). Unfortunately, specialist movement scales for infants are not well developed; rather, assessment is generally deferred to tests of general intellectual development where no clear distinction between cognitive and motor function is drawn (e.g., the Bayley Scales and the McCarthy Scales). While the MABC (nee TOMI) and BOTMP have served researchers and clinicians well over a number of years, neither adopt a lifespan perspective on motor development, nor reflect more recent advances in motor control and motor development.
The General Abilities Approach is borne of early theorising in physical and occupational therapy, based on the working assumption that sensory-integrative and sensorimotor functions support later motor and intellectual development. Movement clumsiness is thought to arise from underlying difficulties in the organization of these base functions. Sensory Integration (SI) theory is prototypical of this approach. Evolved from a clinical framework, SI theory has spawned diagnostic tests with poor validity and that provide little insight into the workings of the perceptual-motor system (e.g., the Sensory Integration of Praxis Tests—SIPT). Moreover, evaluation data for SI training and traditional perceptual-motor approaches is not encouraging, further questioning the theoretical basis of the approach.
In the case of Neurodevelopmental Theory, a medical model and neuromaturational norms have traditionally been used to understand signs of abnormal motor development, and has guided the selection of assessment tools and their interpretation. Unfortunately, no clear theoretical basis for understanding DCD in its own right has been widely adopted in the medical profession. Abnormal motor signs (such as the persistence of primitive reflexes and poor postural control) were used as markers of other medical syndromes like cerebral palsy and muscular dystrophy. More generally, early sensorimotor function and motor milestones were regarded as indices of general neural integrity. The composition of neurodevelopmental tests does not conform well to any distinct and current model of development. They are composed of an eclectic blend of neurological, cognitive, and intellectual tests that vary from one study to another.
More specifically, traditional tests of soft neurological signs (e.g., Touwen, 1979) remain open to clinical interpretation and have, until recently, been unable to reliably pinpoint malfunctioning sites within the brain. With the advent of more sophisticated neuroimaging technologies and converging research methodologies there is now hope that particular soft signs might be used as specific markers of brain system development, some of which may provide clues about the neural loci of DCD. The Dynamical Systems Approach supports promising trends in biomechanical or kinematic analysis of movement, ecological task analysis, and task-specific intervention. Kinematic approaches provide important insights into the changing coordination dynamics that occur with age, and putative biomechanical and neuromotor bases for these changes. I argue that task-specific analysis (or Ecological Task Analysis) provides currently the most pure version of a dynamical systems approach. Here, task, performer, and environmental factors all feature in the assessment process, with individual variation embraced. The clinician is, thus, better equipped to identify those control parameters that are particularly important to the development of skill at a task-specific level.
Finally, the Cognitive Neuroscientific Approach is a cross-disciplinary
framework that seeks to understand the development of motor skills in terms
of unfolding brain-behaviour interactions. This approach supports a brain-systems
oriented approach to assessment (and treatment), but only as an adjunct to
a conventional movement skill exam. From this perspective, deficits in the
internal modelling of movement and in timing control are two hypotheses that
might explain points of departure from normal development. While training
studies have shown preliminary support for the internal modelling account,
future work needs to isolate what components of the treatment package are
exerting effect. Taken together, a multi-level approach to assessment and
treatment is recommended for children with DCD. The use of multiple and converging
measures will circumvent some issues with diagnosis and promote a fuller
appreciation of motor development at different levels of function—behavioural,
neurocognitive, and emotional.
A hybrid assessment model that embraces both dynamic systems and cognitive
neuroscientific principles will provide the impetus for this change. I suggest
that multi-level assessment might involve three sequential steps: (1) functional
performance assessed against (age-relevant) movement goals, (2) movement
strategies that reflect the interaction of contextual, biomechanical, and
task factors, and (3) impairment factors (i.e., the integrity of supportive
neurocognitive and affective systems). Levels 2 and 3 might only be implemented
if deficiencies are observed in a previous step. The hope is that assessment
of multiple levels of function will ultimately map more seamlessly to intervention.
In short, the assessment strategy should be theoretically-principled, construct-valid,
and flexible enough to cater for individual differences in presentation,
learning style and progression.