Subjects of
Research

Introduction

Curriculum
vitae

Theory of
CDT

Clinical
Applications

Publication
List

Neurotherapy
Course

Download articles

Videos

Therapy Place
in Tartu

Abstract. The key strategies on which the discovery of the functional organization of the central nervous system (CNS) under physiologic and pathophysiologic conditions have been based included (1) our measurements of phase and frequency coordination between the firings of a- and g-motoneurons and secondary muscle spindle afferents in the human spinal cord, (2) knowledge on CNS reorganization derived upon the improvement of the functions of the lesioned CNS in our patients in the short-term memory and the long-term memory (reorganization), and (3) the dynamic pattern approach for re-learning rhythmic coordinated behavior. The theory of self-organization and pattern formation in nonequilibrium systems is explicitly related to our measurements of the natural firing patterns of sets of identified single neurons in the human spinal premotor network and re-learned coordinated movements following spinal cord and brain lesions. Therapy induced cell proliferation and, maybe, neurogenesis seem to contribute to the host of structural changes during the process of re-learning of the lesioned CNS. So far, coordinated functions like movements could substantially be improved in every of the more than 100 patients with a CNS lesion by applying coordination dynamic therapy. As suggested by the data of our patients on re-learning, the human CNS seems to have a second integrative strategy for learning, re-learning, storing and recalling, which makes an essential contribution to the functional plasticity following a CNS lesion.

A method has been developed by us for the simultaneous recording with wire electrodes of extracellular action potentials from single human afferent and efferent nerve fibres of undamaged sacral nerve roots. A classification scheme of the nerve fibres in the human peripheral nervous system (PNS) could be set up in which the individual classes of nerve fibres are characterized by group conduction velocities and group nerve fibre diameters. Natural impulse patterns of several identified single afferent and efferent nerve fibres (motoneuron axons) were extracted from multi-unit impulse patterns, and human CNS functions could be analyzed under physiologic and pathophysiologic conditions. With our discovery of premotor spinal oscillators it became possible to judge upon CNS neuronal network organization based on the firing patterns of these spinal oscillators and their driving afferents. Since motoneurons fire occasionally for low activation and oscillatory for high activation, the coherent organization of subnetworks to generate macroscopic function is very complex and, for the time being, may be best described by the theory of coordination dynamics. Since oscillatory firing has also been observed by us in single motor unit firing patterns measured electromyographically, it seems possible to follow up therapeutic intervention in patients with spinal cord and brain lesions not only based on the activity levels and phases of motor programs during locomotion but also based on the physiologic and pathophysiologic firing patterns and recruitment of spinal oscillators. The improvement of the coordination dynamics of the CNS can be partly measured directly by rhythmicity upon the patient performing rhythmic movements coordinated up to milliseconds. Since rhythmic, dynamic, coordinated, stereotyped movements are mainly located in the spinal cord and only little supraspinal drive is necessary to initiate, maintain, and terminate them, rhythmic, dynamic, coordinated movements were used in therapy to enforce reorganization of the lesioned CNS by improving the self-organization and relative coordination of spinal oscillators (and their interactions with occasionally firing motoneurons) which became pathologic in their firing following CNS lesion. Paraparetic, tetraparetic spinal cord and brain-lesioned patients re-learned running and other movements by an oscillator formation and coordination dynamic therapy. Our development in neurorehabilitation is in accordance with those of theoretical and computational neurosciences which deal with the self-organization of neuronal networks. In particular, jumping on a springboard ‘in-phase’ and in ‘anti-phase’ to re-learn phase relations of oscillator coupling can be understood in the framework of the Haken-Kelso-Bunz coordination dynamic model. By introducing broken symmetry, intention, learning and spasticity in the landscape of the potential function of the integrated CNS activity, the change in self-organization becomes understandable. Movement patterns re-learned by oscillator formation and coordination dynamic therapy evolve from reorganization and regeneration of the lesioned CNS by cooperative and competitive interplay between intrinsic coordination dynamics, extrinsic therapy related inputs with physiologic re-afferent input, including intention, motivation, supervised learning, interpersonal coordination, and genetic constraints including neurogenesis.

The theory of reorganizing the lesioned human CNS, based on measurements in humans (brain-dead individuals (HT1-6) and patients with and without CNS lesion) of self-organization of the human spinal cord neuronal networks, is used to re-learn lost somatic, autonomic functions and higher mental functions in patients with CNS lesions, and will be shown to be in accordance with the data from animal research if comparable; some differences will be made clear. With the development of a special coordination dynamic therapy device to train simultaneously the coordinated movements of arms, hands and fingers, and legs, feet, and the trunk an essential further step has been done towards efficient reorganization of lesioned CNS.

With our tools and methods available to repair the lesioned CNS by reorganizing (relearning), it became also possible to diagnose instability and deterioration of integrated functions of the CNS in patients with severe lesions or minor defficiencies in the organization of the CNS, such as scoliosis. The essential improvement of higher integrative functions in patients with severe brain lesions by coordination dynamic therapy opens up the possibility to improve higher mental functions in individuals with a severe CNS lesion or a physiologically functioning CNS.

Key-words: Human neurophysiology - Integrative CNS functions - Single nerve-fibre action potentials - Natural impulse patterns - Self-organization - Spinal oscillators - Phase and frequency coordination - Coordination dynamics - CNS lesion - Coordination dynamic therapy - Re-learning - Transfer of learning - Co-movements - Spasticity release - Neuro-rehabilitation - Rehabilitation

The human CNS possesses billions of neurons each one of which having connections to an average of approximately 4000 other neurons. New concepts and tools are needed if the inherent complexity of the most complex system of all, the human CNS and its relation to behavior and thinking, is to be understood and repaired in the case of a lesion. Presently, there is a huge void between the knowledge what a single neuron does (which we know a lot of) and what many of them do when they cooperate. The understanding of the principles of organization among large numbers of neurons is of a paramount importance, as this organization lies at the root of the understanding of ourselves, of the world we live in, of how we touch, see, hear, plan, act and think and how we re-learn behavior and thinking in the case of a CNS lesion. Such fundamental behavioral functions depend on temporally coherent functional units distributed throughout different regions of the CNS, and standard methods have not the potential to elucidate them. For example, responses to sensory stimuli or activities in relation to motor acts are commonly averaged over successive trials in order to improve the signal-to-noise ratio. This averaging procedure destroys any temporal structure in the activation pattern that is not precisely locked to the stimulus or the motor response. Thus, temporal codes were either ignored or remained undiscovered with the commonly applied methods of single unit analysis [145]. In our human neurophysiologic studies [103-140,186-189,193] simultaneous natural firing patterns of several identified single afferent and efferent neurons were therefore recorded and analyzed, but not averaged. In this way we could discover the self-organization of premotor spinal oscillators (functional unit of a motoneuron and interneurons) and phase and frequency coordination between the firings of oscillatory and not oscillatory firing motoneurons and afferents.

By recording single nerve-fibre action potentials from nerve roots it is possible to analyze simultaneously afferent and efferent impulse patterns and investigate the coupling changes of self-organized premotor spinal oscillators to generate, under physiologic conditions and following a CNS lesion, rhythmic and non-rhythmic movements and autonomic functions like continence. These data partly enable the understanding of the integrated functions of the human spinal cord and supraspinal centres; moreover, they allow to substantially improve locomotor and other functions in patients with CNS lesions. With the improvement of the treatment applied within the first 7 hours following spinal cord lesion, including methylprednisolone [14] administration, more patients with spinal cord lesions are paretic and a reorganization of the integrative functions of the CNS is possible and needed. A first paradigm shift in the understanding of the functioning of the CNS concerning the self-organization of neuronal networks gives more reorganization possibilities by the so-called neuronal network plasticity. The second paradigm shift, namely that neurogenesis may be induced in the human adult spinal cord [186] and adult brain [31], opens further possibilities for repair, regeneration and reorganization of the CNS.

In the last 100 years, many authors have come to suggest that it is the rhythmic firing of human neuronal networks which is responsible for rhythmic movements (trembling, tremor) of the body (trunk) and legs, arms and fingers (for references, see [60,168]). Descartes tackled tremor as early as in 1649 [20]. The conclusions drawn by R. Jung from measurements on tremor and clonus in 1941 [60] were similar to the findings concerning the self-organization of premotor spinal oscillators in recent papers [118,120,127]. Only, R. Jung did not differentiate between different motoneuron types and analyzed tremor by only using mechanical and electromyographic recordings.

In 1939 and 1950 [52-54], E.v. Holst expressed his disagreement with the common regard of the CNS beeing only a reflex apparatus producing motor output. His ‘relative coordination’ of different rhythms [52] of the CNS in different species including man is very similar to recent findings of the relative coordination of human spinal oscillators [107,118,131]. It was the opinion of the Sherrington school that all reflexory, excitatory and inhibitory influences onto motor output are due to direct interactions at the motoneuron pool itself (reflex theory) [19]. The opinion of R. Jung was that the bottom-level coordination mechanism, at which all impulses run together, is the ‘Schaltzellenapparat’ (neuronal network apparatus) of the spinal cord [60].

Our current research on spinal oscillators and rhythm coupling supports the rhythm theory of R. Jung and E.v. Holst that assumes that the neuronal networks with their rhythmic properties driving the motoneurons are the bottom level basic mechanism for coordination, and essentially contribute to the coordination dynamics of arm and leg movements. At least to understand movements, the coordination dynamics of neuronal networks of the spinal cord must be understood and integrated in theories on human movements, since neuronal networks of the spinal cord show remarkable functions and plasticity, as has been shown by electromyography by the improvement of the motor program in patients with a complete rostral spinal cord lesion [177]. Brain-dead humans may be able to perform coordinated leg movements when kept at an intensive care unit for several weeks, anencephalic [91] and healthy newborn babies (Fig. 43) can step automatically, and spinal oscillators self-organize themselves and may fire rather coordinatedly in patients with a complete spinal cord lesion.

In the last 15 years the understanding of the functioning of the CNS has changed, moving away from the rigid reflex and neuronal response chain theories towards the concept of dynamic self-organization of neuronal networks [3,63], and this has direct implications for neurorehabilitation because of the increased network plasticity including large scale plasticity [143]. The regulation of neural stem cells and neurogenesis in the intact and damaged adult mammalian and human brain, and probably spinal cord, offers further possibilities with respect to the regeneration and reorganization of the injured human CNS [4,45,166,167].

This review is designed to build up a scientific basis to reorganize the lesioned human CNS by re-learning, so that motor, autonomic and higher mental functions of patients can be improved on the same basis. Since it seems impossible to review the functioning of the whole human CNS, the present review concentrates on oscillators and self-organization in neuronal networks. To get more hard data on the functioning of the human nervous system, new precise human neuroelectrophysiologic methods are needed, like single motor unit electromyography [97,147], the tungsten electrode method [59,67,158] and the single-nerve fibre action potential recording method [103-140].

The case reports are used here to document the theory, to derive human data from lesion studies and use them in theory and to explore possibilities in neurorehabilitation when extrapolating from theory to clinical settings (in similarity to pre-experiments in animal research), since full clinical studies for reorganizing the human CNS, requiring up to 5 years time and more are time, energy and money consuming. In the long term, the clinical trials have to be backed by full clinical studies. But in all patients who obtained coordination dynamic therapy from the author (G.S.) for at least 3 months, the functioning of the CNS could substantially be improved.

This review starts with the new development in human electrophysiology with which it is possible to analyze partly the organization of neuronal networks of the human CNS. As a first step, the firing of motoneurons in the occasional and oscillatory firing mode in relation to the skin and muscle spindle afferent activity is demonstrated; the recordings were obtained from brain-dead individuals (HTs). The partial loss of phase and frequency coordination following CNS lesion was derived from recordings from paraplegics made during surgery for implantation of electrical bladder stimulators. Single motor unit firing was partly verified electromyographically from leg muscle recordings. The partial loss of phase and frequency coordination prompted the development of coordination dynamic therapy to restore the impaired coordination dynamics. The case reports show that the lesioned CNS can be repaired. With the knowledge derived from the case reports and other patients, the theory has been summarized and updated.

To improve motor, vegetative and higher mental functions in patients with CNS lesion, it is appropriate to observe neuronal network organization in the human lower sacral range, where motor and vegetative (volitional and automatic) functions are generated. Self-organized premotor spinal oscillators have been shown to have two driving phases per oscillation cycle [126] when primarily the somatic nervous system is activated (pin-pricking of sacral dermatoms to induce escape reaction), but the oscillators have three driving phases [126] when additionally the parasympathetic division is activated (stimulation of urinary bladder filling (vegetative) afferents by bladder catheter pulling). The complexity of neuronal network organization becomes obvious when in addition to somatic also vegetative functions are activated. It is further believed that the vegetative functions are the door to the higher mental functions.

Due to the 4 new developments in human neurophysiology mentioned, it is possible to reorganize the lesioned or functionally impaired human central nervous system. The four new repair-related concepts are:

1. The CNS is viewed as a neuronal network which organizes itself. The organization can be changed by re-learning.
2. The self-organization is based on a relative (specifically changing) phase- and frequency coordination of rhythmically firing subneuronal networks and single neurons.
3. Neurogenesis and functional cell proliferation is induced and controlled by learning. Methods for re-learning basic CNS functions use especially rhythmic, dynamic, coordinated movements.
4. It seems from the success in re-learning movements, vegetative and higher mental functions in patients with CNS lesion that the human CNS has a second integrative strategy to learn, re-learn, store and recall network states.

The lesioned human CNS can be repaired by re-learning of partially lost phase and frequency coordination through coordinated rhythmic movements. The severely lesioned CNS can only efficiently be repaired if integrative, coordinated functions are re-learned. The re-learning of relative phase and frequency coordination of the lesioned CNS can be achieved by:

1. Using special coordination dynamic therapy devices which offer exact phase and frequency coordination up to a few milliseconds for re-learning.
2. The training of automatisms, postures and old learned movements which are only little impaired in their functioning by the lesion.

Rather than asking what is the best method to re-organize the lesioned CNS we should ask what method is most efficient in re-organizing the lesioned CNS by re-learning. The increase of the rate of re-learning is determined by 4 factors:

1. The exactness of the coordination of the performed movements during the therapy, to functionally reconnect disconnected network parts to recouple arms or legs that cannot be moved.
2. The increase of the integrativity of the coordination dynamic therapy, which increases the number and complexity of simultaneously exercised phase and frequency coordinations and makes it possible to re-learn integrative functions such as the higher mental functions.
3. The enhancement of the movement induced re-afferent input to strengthen the physiologic self-organization of the lesioned CNS and its communication with the environment.
4. The increase of the intensity of the therapy to force the ‘adaptive machine’ CNS to adapt.

The theory of self-organization and pattern formation in nonequilibrium systems [63,172-178,191] builds upon the concepts of synergetics [190]. Patterns of coordination are viewed in terms of their nonlinear dynamics. The patterns of coordination are characterized by low dimensional collective variables or order parameters whose dynamics are function-specific. Observable patterns of coordination are mapped onto attractors (see below) of the order parameter dynamics. Biological boundary conditions act as parameters on the collective dynamics. Several coordinative patterns can coexist under the same condition (multistability). Loss of stability leads to switching of patterns and gives rise to nonequilibrium phase transitions. Fluctuation and differential stability govern the switching dynamics among multiple coordinative patterns. If a certain coordinative pattern has a very high stability, it may seem as if this pattern is ‘hard wired’ and is generated by a pattern generator. If the coordination dynamics is not specified by the constraints to a particular pattern (spontaneous pattern formation), it is called intrinsic dynamics. The pattern in that concept that emerges is a direct consequence of cooperative and competitive interactions between the intrinsic dynamics, the intensional pertubation (intentional impulse patterns) and the extrinsic dynamics (movement induced afferent input).

If x is a characteristic collective variable, describing the dynamic pattern, then x = x (t) = x_t, where t is time and x obeys the dynamical law dx /dt = f_intr (x_t, parameters, noise) (the right side is called vector field). Special solutions to this equation are called attractors if they are asymptotically stable; all neighboring solutions converge in time to the attractor solution. Attractors play a key role in the modeling process because the behavior of the collective variable in time may be mapped onto attractors. The attractor basin is defined as the set of all initial points from which trajectories converge to a given attractor.

In learning a bimanual coordination task by synchronizing finger movements to a visually specified phasing relation, the coordination dynamics is captured through a collective variable, the relative phase, j. With the equation for the order parameter dj /dt = - ¶V /¶j and the potential V = -a cos (j) - b cos (2j) (Fig. 54), the behavior of the system can be visualized if j is identified with the coordinate of a particle that moves in an overdamped fashion in the potential, V.

In a theoretical analysis of the patterns of interlimb co-ordination in the gaits of quadrupedal locomotion, the collective variables are given by three relative phases that describe the co-ordination patterns of arms and legs. Gaits were classified by their symmetry properties, which can be expressed as invariances under groups of transformations [174].

In the severely lesioned CNS only many collective variables (control parameters) can describe the coordination dynamics, and multistability may be large among attractor states. To increase the differential stability of those attractor states of subnetworks which give rise to physiologic movements (Fig. 107), supervised learning using the special coordination dynamic therapy device may be necessary.

The key concepts to repair the lesioned human CNS are to discover the functional organization of the human CNS under physiologic and pathophysiologic conditions, and to re-organize the lesioned CNS for physiologic functioning. Self-organization and pattern formation (collective effects of firings of many neurons) was explored in a part of the CNS, namely the spinal cord, by measuring (1) phase and frequency coordination between a - and g -motoneurons and secondary muscle spindle afferents in normal individuals and in patients with CNS lesions [107,115,116-121,123,128-131], and by (2) extracting knowledge on CNS organization from the lesioned CNS of patients [137-140,186-189,193]. By improving physiologic functioning of the lesioned CNS by a coordination dynamic therapy [137-140,186-189] in patients in the short-term memory and in the long-term memory (reorganization) the progress of re-learning was analyzed. Because of measured phase and frequency coordination in the process of neuronal network organization in man, the theory of coordination dynamics is used for re-learning motor, vegetative, and higher mental functions. The re-learning (repair) is seen as a change of the existing inner coordination dynamics tendencies after the lesion (with no or only pathologic functioning of arms, legs and trunk) to achieve CNS coordination dynamics which will generate again physiologic movements, vegetative and higher mental functions (Fig. 108). The change of the coordination dynamics is achieved by the coordination dynamic therapy which uses the strategies (1) of accurate movements coordinated up to a few milliseconds by supervised instrumented learning to reconnect functionally disconnected neuronal network parts (Fig. 104), (2) of increasing the integrativity of coordinated movements or behaviors to repair integrative CNS functions like higher mental functions, (3) of enhancing movement induced afferent input to stabilize physiologic network states of the lesioned CNS and to destabilize pathologic network states like spasticity by offering more physiologic afferent input to the lesioned neuronal networks for physiologic self-organization, and by supplying - through instrumented, coordinated movements - physiologic regulation (motor control) to the networks by using the receptors of the periphery, especially the secondary muscle spindle afferents, and (4) of going to the limits of exercising by increasing the intensity of the therapy to force the ‘adaptive machine’ CNS to adapt.

The self-organization of neuronal network parts and its evolution with time (coordination dynamics) is realized by interrelating (of firings of motoneurons) the existing inner coordination dynamics tendencies of the network (tendencies of phase and frequency coordination), the intentional impulse patterns and the afferent input from the periphery including movement induced re-afferent input. Network parts are functionally connected or interlaced by network communication including regulatory processes with other network parts and the periphery.

The link between neuronal activities and movements resides in collective effects (pattern formation) at the microscopic level that create macroscopic order and disorder. Coordination can be observed and measured (1) between the moving of arms and legs (Figs. 50,103F), (2) in the motor programs of muscle activation in electromyographic (EMG) recordings [131], and was measured (3) as relative coordinated firings between premotor spinal oscillators (assembly level) and single neurons (Figs. 31-39).

Following CNS lesion, coordination is partly lost at all levels of observation (macroscopic level (Fig. 103F), assembly level (Fig. 47C), single neuron level (Figs. 12,13), motor control level (regulation loops, clonus)). The intrinsic dynamics of the neuronal networks, as measured by phase and frequency coordination of firings of neurons, shows poor coordination. By exercising on the special coordination dynamic therapy device, i.e. by instrumented supervised motor and motor control learning, phase and frequency coordination can be re-learned (Fig. 108) and the re-learned coordination measured (Figs. 50,103F). Since the neuronal networks of the CNS communicate via receptors with the environment, the coordination upon performing coordinated movements reaches the neuronal networks via regulatory loops (for example, external oscillator loops or g-loops). The relationship between the coordination dynamics of arms and legs and the coordination dynamics of premotor spinal oscillators partly induced by secondary muscle spindle (joint and other) afferent activity constitutes a relationship between macroscopic and assembly level. Therefore, the networks of g-loops or external loops of premotor spinal oscillators establish with other regulatory loops a connection between the macroscopic and the neuronal assembly level.

The macroscopic competition between a physiologic attractor state (e.g., running) and a pathologic attractor state (e.g., extensor spasticity) may be seen at the premotor network level as the competition for different coordinations among the impulse patterns of a- and g-motoneurons, interneurons and muscle spindle and other afferents.

Relative coordination does not mean approximate coordination, but means specific changing coordinations among single neurons or assemblies. The coordination dynamic therapy improves the precision of relative coordination in CNS neuronal networks, i.e. the exactness of evolving changes with the time of coordinated firing among neurons.

Following CNS lesion, the coordination dynamics substantially changes. If, in a severe CNS lesion, therapy is started too late, the impaired neuronal networks adapt by themselves in an uncontrolled way. The CNS networks will then nearly always generate also pathologic neuronal networks states, which give rise to different kinds of spasticity, unphysiologic posture and unphysiologic hand and leg positioning (Fig. 78).

In re-learning of old learned functions two kinds of changes of the existing coordination tendencies may occur in the cooperative and competitive interplay between the pathologic inner coordination tedencies and the pattern formation tendencies induced by the afferent impulse patterns from the periphery, induced by the therapy. Firstly, when the re-learning involves a shift of preexisting attractive states, mainly cooperative mechanisms becone involved, and parametric changes in the coordination dynamics correspond to adjustment of actual phase relations between the spinal oscillators and neurons within the same coordination strategy in the premotor network. Secondly, when the necessary re-learning requirements differ strongly from the existing coordination tendencies, competitive mechanisms may induce loss of stability (seen as enhanced variability in the progress of the re-learned task) and/or attraction of the re-learned movement to an underlying coordination tendency. Examples of the latter are the slipping during leg movement into extensor spasticity or slipping, when giving the hand, into the grasp automatism (Fig. 78E,F,G).

Whereas analysts of motor behavior consider motor control to be distinct from motor learning, motor control and motor learning is used when performing coordination dynamic therapy. When the patient tries to adapt his motor behavior to the coordination dynamic therapy devices, he uses the motor control (materialized by afferent activities of muscle spindle, skin, joint and other receptors) for re-learning. But when the hand partly slips from the lever and the patient tries to grip additionally the lever at the appropriate moment when the fingers starts to slip, then he exercises motor learning in addition to motor control learning.

So far, improvement of the organization of the lesioned CNS has been measured indirectly by the improvement of movements. E.g., to measure the changes of the coordination dynamics during re-learning, walking or running speed was used (Fig. 78H). But by measuring the coordination dynamics (rhythmicity) of arms and legs during exercising on the special coordination dynamic therapy device, it is possible to measure more directly the changes of the organization of the CNS with ongoing therapy (Figs. 50,103F).

Parametric changes of coordination dynamics, namely mainly the cooperative interplay, and dramatic changes, namely the more competitive interplay between the intrinsic dynamics and the to-be-re-learned coordination dynamics, were observed in patients. When coordination dynamic therapy was started soon after the CNS lesion, mostly smooth curves of improvement of movements were observed (parametric changes, see Figs. 67, 78H). When the therapy was started long after severe CNS lesions, often instabilities in the movement progress were observed (dramatic changes, see Figs. 71F, 93, 99). It seems therefore that when coordination dynamic therapy was started long after a severe lesion, false established coordination dynamics had essentially to be reorganized and a mainly competitive interplay took place between the existing pathologic and the to-be-re-learned physiologic coordination dynamic tendencies. The competitive interplay with the instabilities occurring in the network organization reflected itself in the instability of progress, which means that the performance of a movement (e.g., walking) can transiently become worse.

Kelso, Zanone and Schöner [175,176] evaluated changes of the intrinsic coordination dynamics tendencies during learning of coordination tasks in non-lesioned CNS by measuring the dynamics of the relative phase during the performance of bimanual coordination tasks, including the measurement of non-equilibrium phase transitions from anti-phase to in-phase movements when increasing the frequency of performance. By analyzing different bimanual coordination tasks, they tried to analyze the transfer of learning from a learned new coordination task to another not trained coordination task.

During coordination dynamic therapy of the lesioned human CNS, has so far been only partly possible to measure the changing momentary intrinsic coordination tedencies during re-learning at the different levels of description (level of movements, assembly level (premotor spinal oscillator level), single neuron level). The improvement of the coordinated movement of an arm, finger or leg can be measured. The improvement of the average coordination dynamics when exercising on the special coordination dynamic therapy device can be used to quantify the improvement of the inner coordination tendencies on the macrsoscopic level (Figs. 50,103F). By performing electromyography, the coordinated firing of premotor spinal oscillators can be measured. Therefore, re-learned coordination can also partly be measured additionally on the assembly level and single neuron level.

With respect to re-learning of movements after CNS lesion using coordination dynamic therapy or other re-learning methods the question is important what is being re-learned by improving the coordination dynamics in the patient. What is transfered for the improvement of the general coordination tendencies, if a certain movement is exercised and re-learned. In patients following CNS lesion phase and frequency coordination among premotor spinal oscillators was shown to get partly lost. The oscillators have partly lost their rhythmic properties, a- and g-motoneurons changed their recruitment in the occasional firing mode, control of spinal oscillators was partly lost, and the phase relations between spinal oscillators, g-motoneurons and secondary muscle spindle afferents were partly lost. It seems therefore that the coordinated firing of the neurons in the lesioned CNS get partly lost with respect to time and space (loss of physiologic CNS organization) as is communication with the environment (partial loss of motor control). When the specific property of the lesioned CNS ‘timed firing in space’ of its neurons (phase relations between the firings) is partly lost, then the timing of firing of the neurons has to be re-learned.

If the CNS has only been affected by a minor lesion, then the CNS may be able to repair itself (re-learning by itself). But if a patient suffered a severe CNS lesion, the CNS cannot repair the basic functional structure by itself, as experienced from improving functions in patients with severe CNS lesions. Instrumented supervised re-learning can offer the lesioned neuronal networks dynamic physiologic sets of phase relations, evolving with time, for re-learning. It seems therefore that the phase relations between the firings of the different neurons have to be re-learned for different (movement) patterns for re-learning the general improved functioning of the CNS, because descending impulse patterns and afferent impulse patterns from the periphery contribute to the ‘timed firing’ of neurons. How the relative coordination evolves over time via relative phase and frequency coordination between the component neurons can directly be seen in Figures 31 to 39 for the premotor spinal neuronal network. The coordination dynamics is being re-learned by re-learning the appropriate phase and frequency coordinations among the firings of the neurons, evolving over time. The collective effects of many firing neurons resides in the collective effect of relative coordination of many phase relations. The self-organization of the assembly ‘premotor spinal oscillator’ is achieved by the timed firing of the component neurons activated by descending impulse patterns (volition or intension) and/or afferent impulse patterns from the periphery.

Dynamic neuronal network organization does not only mean evolution of the network organization over time (flow of a vector field) but additionally means that also movements of arms and legs are performed with high positive and negative accelerations, so that the CNS is activated as integratively as possible. Also, those phase relations are activated and re-learned which are only activated when fast changing movements are performed. Such dynamic activation of the system will also give rise to higher order rhythm couplings and will also activate strongly the fast systems in the neuronal networks. With respect to the premotor spinal network this means that also many (dynamically responding) a₁-motoneurons and primary muscle spindle afferents are activated.

During the process of re-organization of the CNS, when performing coordination dynamic therapy, the timing of firing of neurons (phase entrainment between neurons or between different neurons in an assembly), premotor spinal oscillators (phase entrainment between assemblies), of different neuronal network parts with similar and dissimilar functions, and the timing of firing of the different CNS parts is re-learned. The phase relations with an exactness of up to a few milliseconds offered by supervised re-learning (Fig. 34) seem to be crucial for re-connecting of network parts (Fig. 104) to improve the integrative functions of the CNS.

The question could be put forward, what is being re-learned for walking when the patient is exercising on the special coordination dynamic therapy device, because the movements seem to be quite different. It seems that the overlap in the intrinsic coordination dynamics tendencies, i.e. the overlapping of the activated performed phase relations of the two movement patterns over time is re-learned for walking. The overlap of the activated coordination dynamics tendencies is transfered from the task of turning levers and pedals to the task of walking. This view is supported by the experience that about 1000 turns on the special coordination dynamic therapy device will make walking and running easier in normal individuals and in patients with CNS lesion.

When connections between neuronal network parts are destroyed by a lesion or are impaired, e.g. by demyelinisation, mutual phase relations between neurons will have changed dramatically. The re-learning of the collective effects of timed firing of neurons (pattern formation) includes therefore also the building up of new connections between network parts. A host of structural changes, including neurogenesis, cell proliferation and changes of membrane properties, will make contributions to the adjustment of the timed firing of neurons in the different parts of the CNS.

Competitive and cooperative mechanisms can only be re-learned (and measured) in relation to existing coordination tendencies, before, during and after re-learning, making the individual learner rather than the group or the species the significant unit [176]. The instantaneous coordination dynamics can be measured when exercising on the special coordination dynamic therapy device (Figs. 50,103F).

Interesting and important in the process of re-learning by the coordination dynamic therapy is that so far, no unphysiologic movement patterns have ever been produced if the supported and/or supervised exercised movements were physiologic. No drifting into obscure patterns has ocurred. Apart from some instabilities, the pathologic movements always gradually or suddenly changed in the direction of physiologic movements. It seems therefore as if the re-learning was supervised to go in the direction of physiologic movements. Maybe, the same mechanism is at work which is responsible for the primary organization of the CNS during ontogenesis. A genetic support in the control of the process of re-learning seems possible. The influence learning methods have on neurogenesis and cell proliferation in animals [169-171] supports that view. The observation that also improvement in the physiologic functioning of the CNS can be achieved in children with Down’s syndrome (trisomy 21) if coordination dynamic therapy or conventional therapy is applied seems to contradict genetic guarding in the process of re-learning. However, different participating genes may cooperate in a different way during ontogenesis and repair, and ontogenesis may be rather pathologic because of the genetic defects, whereas the therapy induced genetically guarded repair mechanisms are mainly physiologic.

The understanding of the interrelationship between learning and re-learning methods and the host of structural changes taking place in response to them, including neurogenesis, cell proliferation, neurite growth, membrane property changes and synapse efficiacy changes, will not only bring further insight into the functioning of the human CNS, but will also give a further basis for improving the efficiacy of learning and re-learning methods. From the peripheral nervous system in animals it is known that only those motoneurons survive during development, which make functional contact in the periphery [102].

Based on early pharmacotherapy [14,83] in spinal cord and brain lesions to reduce the self-destruction of the CNS, on new measurements in the PNS providing more knowledge on the self-organization and coordination dynamics of neuronal networks in the CNS, on regenerative capacity by neurogenesis and/or axonal regeneration [4,545,51,75,157,166,167], and on the demonstrated higher plasticity following rhythmic, dynamic, coordinated (and other) movements in a quantified intensive therapy, it could be shown that substantial recovery is possible in patients with spinal cord and brain lesions.

Even if new drugs are able to substantially enhance the regenerative capacity of the human CNS, it should be borne in mind that most likely, the growth of neurites is unspecific, which means that a reorganization (tackled in this vertical review) of the regenerating CNS is still needed to probably transform mass contractions into useful movements.

Researchers from theoretical and computational neurosciences have been attacked for not measuring on human CNS. Firstly, this argument holds for many researchers working on the CNS. Secondly, comparative measurements in animals and humans are needed so that knowledge obtained from animal studies can be transposed to humans and that human medicine can benefit from animal research. A cooperation of researchers from several fields is needed, including those from the clinical field. Actually, for the time being the major problem with respect to neurorehabilitation seems to be that new knowledge on neuronal network plasticity and regenerative capacity is to be brought to the patient. Theoretical modeling without including hard human data is an empty exercise, a mere theory. So is collecting facts in the absence of any understanding of the basic operating modes of the human CNS [63].

It has been shown that somatic, autonomic and higher mental functions of the lesioned CNS in patients can be repaired using the knowledge of four new developments in neurosciences, namely (I) the concept of self-organization of neuronal networks, (II) the concept of rhythmic firing of subneuronal networks and of the rhythm coupling (relative coordination of phase and frequency) of these rhythmically firing networks, (III) the concept of regeneration including neurogenesis in adult patients with CNS lesions, and (IV) the concept of integrative learning, re-learning after CNS lesion, storing and recalling, which follows from the case reports and further practical experience with patients with CNS lesions. With the newly developed oscillator formation and coordination dynamic therapy using also oscillator-theory-based equipment, the reorganization was so efficient and its extent so large that in some patients the goal of the neurorehabilitation ‘Cure rather than care’ has been reached. With the strategy in mind to reorganize the CNS as integratively as possible by means of special coordination dynamic therapy devices, the higher mental functions improved with the motor functions in severe as well as in minor CNS lesions, leading to the conclusion that higher mental functions may substantially be enhanced by integrative coordination dynamic therapy.

The spirit generated in the neuronal networks of the CNS may get released from the genetically determined network structure. There is no more the question whether the spirit can ever understand the functioning of the neuronal networks of the CNS which generated it; rather, we should ask if the spirit can understand how it is generated in the networks; then it may develop further, with sufficient motivation and appropriate learning methods, to become released from the genetically pre-determined neuronal network.