Article Text
Statistics from Altmetric.com
A view of apraxia
A previous exposition of apraxia in Practical Neurology1 offered an account of underlying impairment and accompanying clinical assessment based on a distinction between ideational, ideomotor and limb-kinetic apraxia. This conceptualisation stems from 19th-century models of higher cortical motor, language and visual function, revived by Geschwind2 3 in the 1960s. Liepmann in 19004 had posited a hierarchical model of action control over three discrete levels. At the top, movement formulae (‘visual engrams’ of the action) provided overall targets, which activate innervatory patterns to stimulate the appropriate muscles, coordinated contraction of muscles then led to execution of the action. (1) Ideational apraxia arose when movement formulae were either impaired or disconnected from the innervatory patterns. (2) Ideomotor apraxia resulted from disruption to innervatory patterns. (3) Limb-kinetic apraxia, assumed to be a disruption to the smooth neural transmission of the motor commands, was, even for Liepmann, not a full apraxia. It existed between apraxia and paresis only within a broader view of dysfunction.
Liepmann’s model proved valuable in studying apraxia. However, despite surviving in several neurology textbooks, the ideational–ideomotor dichotomy has now been replaced as a way of understanding and classifying apraxias,5–8 similar to the disbandment of Broca’s versus Wernicke’s aphasia, or the dichotomy of associative versus apperceptive agnosia9 (although these distinctions, too, still persist in some areas of clinical practice).
A radical rethink was needed, owing to issues with the underlying theoretical model of apraxia and difficulties in distinguishing between ideational and ideomotor clinically. Goldenberg7 made the plea to ‘relegate the dichotomy of ideational and ideomotor to the history of neuropsychology’ (p332). Limb-kinetic apraxia should not be thought a pure apraxia, but rather was a primary sensory-motor dysfunction. Patients with ‘limb-kinetic apraxia’ have intact conceptual planning, but without the contextual variability of apraxia; their clumsiness varies with fine motor coordination complexity rather than with psychomotor complexity. It is confined to distal control and is typically unilateral.10
There are several problems with the ideational–ideomotor (dis)connection model. What constituted ‘visual engrams’ and ‘innervatory patterns’, and the relationship between them, was at the very least highly underspecified, at worst misguided. The same applied to how ‘engrams’ and ‘innervation’ might relate to actual brain/central nervous system function. Clinically, the model forced clinicians to categorise a patient’s problem as either ideational or ideomotor apraxia. As these categories were derived from a theoretical rather than an experimental model, many patients had clinical problems that did not conform to the classification. In addition, the classification rested on the interpretation of observed behaviours (eg, altered proximity, body part as object, wrong grasp) that could stem from many different underlying causes.
Another view of apraxia
An alternative perspective is to see action planning and control arising from highly interactive, integrated neurophysiological and neuropsychological networks, as opposed to discrete hierarchical formulations. Alternative accounts have highlighted control networks distributed across the left hemisphere,11–15 on the one hand stressing interaction with visual and tactile-kinaesthetic perception and feedback, and on the other with subcortical, especially extrapyramidal, elements of motor control in initiation, timing, rhythm, force of movements.
At their core, these alternative accounts see action planning and control as comprising and depending on multiple underlying processes. If one could identify these building blocks of control and how they interact, then one could analyse apraxic behaviour in terms of which underlying factor(s) is at fault. This would enable the diagnosis of what someone can or cannot do in terms of preservation–impairment of elements of action semantics and spatiotemporal control (see below), rather than assigning some general category label that is not specific to the person’s problem. It would enable targeting of rehabilitation directly at the root causes rather than attempting to deliver interventions based on broad categories that may or may not apply to the individual and be directed at derailments with multiple underlying causes.
The quest was therefore to uncover constituents of action planning and control that operated independently from each other within the interactive networks, but that were also compatible with and transparently linked to underlying neurophysiological processes/networks.
The starting point for building alternative models were systematic observations and the testing out of apraxic behaviour to search for components of action that could break down independently of each other.16–23 Such studies have highlighted, for instance, people who can name objects (and so have no visual agnosia or aphasia) but who cannot state their use and/or demonstrate their application. Others, despite recognising usage, cannot pantomime the action but can faultlessly perform when they handle the object. Another may describe an object and correctly conjecture how it might be employed, but cannot discern its exact function. Others attempt to use it but show wrongly configured grasp, apply to the wrong target of the action (eg, wrong body part, other object) or execute inappropriately (eg, stirring tea with a dunking motion). Part actions (eg, finger, elbow, shoulder extension-flexion) may be possible in isolation, but when performance requires control of multiple degrees of freedom simultaneously across these joints (eg, to reach, pick up a glass, bring it to the lips) movements may be incomplete, out of phase, with trial and error attempts and so forth.
Clinically, too, one would systematically evaluate different modes of elicitation (eg, verbal, visual, tactile), contexts of action (eg, clinic, naturalistic, with–without handling objects) and types of gesture (eg, transitive–intransitive, meaningful–meaningless) to identify which aspect(s) of action semantics and planning/control is/are impaired, aiming to identify the precise source of difficulty. For example, the source of breakdown in stirring tea may relate to not recognising what a spoon is for (despite correct naming), not establishing the association cup–spoon, unable to grasp and manipulate the spoon appropriately, and showing contrasting ability in miming stirring spontaneously compared with imitation, either with or without the tea and spoon present, and with and without touching them. Elucidating the problem might involve contrasting stirring tea with stirring paint (into which one would not dunk a biscuit) or stirring with using the spoon for sugar (using a different grasp and plane, and so on). There are several validated bedside and clinical screening tests that facilitate this evaluation.24 25
There is ongoing debate about precisely which of the underlying components to action control one should seek, but by and large the processes distil into elements around (1) action semantics/conceptual components of action (knowledge of function, manipulation knowledge and mechanical problem solving8 22 23 26) and (2) kinematic, spatiotemporal constituents of control (coordination of the degrees of freedom across multiple joints and the amplitude, trajectory, orientation/plane, spatial proximity/dimensions and relative phasing/timing of subparts of actions8 13–15 27).
The dissociations between conceptual components of action and spatiotemporal constituents of control in turn link to neurophysiological networks (which interact but are dissociable) providing a neuroanatomical and neurophysiological basis for this approach.11–13 15 22 Dorsal–dorsal, dorsal–ventral and ventral neuroanatomical streams have been identified that link to the knowing ‘where’ and ‘how’ (dorsal) and ‘what’ (ventral) of action control, with further subdivision of the dorsal stream into ‘grasp’ and ‘use’ systems.11 15
This more individual approach has more than theoretical relevance. It identifies the root cause(s) of the person’s difficulty with action, and thus facilitates a more targeted rationale for intervention.28 It also opens up the possibility to explore different characteristic profiles of praxic breakdown across neurological disorders.29
References
Footnotes
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed. This paper was reviewed by Alexander Leff, London, UK.
Correction notice This paper has been amended since it was published Online First. Owing to a scripting error, some of the publisher names in the references were replaced with ‘BMJ Publishing Group’. This only affected the full text version, not the PDF. We have since corrected these errors and the correct publishers have been inserted into the references.
Linked Articles
- Editors’ commentary
Other content recommended for you
- The clinical assessment of apraxia
- Biparietal variant of Alzheimer's disease: a rare presentation of a common disease
- Brain activation during ideomotor praxis: imitation and movements executed by verbal command
- Apraxia, agnosias, and higher visual function abnormalities
- Utility of testing for apraxia and associated features in dementia
- Pure apraxic agraphia with abnormal writing stroke sequences: report of a Japanese patient with a left superior parietal haemorrhage
- Lateralised motor control: hemispheric damage and the loss of deftness
- Natural history and survival of 14 patients with corticobasal degeneration confirmed at postmortem examination
- Corticobasal syndrome: a practical guide
- Dyspraxia in a patient with corticobasal degeneration: the role of visual and tactile inputs to action