Feedback and Feedfoward control modes:
On the basis of all motor control theories are the Feedback and Feedfoward control modes, used by the CNS on the control of musculoskeletal system.
The feedforward model is capable of achieving a causal relationship between inputs to the system (the distance of an object; the coordinates of the arm position) and it is important in motor learning situations (e.g., during an arm movement, it could predicts the next state – position and velocity – through the current state and motor command). Despite the feedfoward components not being directly related to sensorial information, it can be influenced by feedback signals [18
The feedback components depend on sensorial information (the position of the arm), whereas the feedfoward components are based on system dynamics knowledge (which should be the adequate position to reach an object) . The feedback control, somehow acts as an inverse model. The inverse model inverts the system. First, it is provided the motor command, causing a desired change in the state (elbow extension). Then, the inverse model act as a controller mechanism, providing the necessary motor commands and achieving the desired state transition [19
Mechanical impedance and viscoelasticity - key properties of the CNS:
Both theories of motor control are based on the ability of the CNS easily adapt to sudden changes of the environment. This ability depends on some key properties of the CNS, such as the mechanical impedance (a basic concept on the equilibrium point theory) and the viscoelasticity (basic concept on both theories).
The mechanical impedance is an important dynamic relation between small forces and position variations and it is a basic concept on the Equilibrium Point Theory. The mechanical impedance of the neuromuscular system determines the reaction forces on the hand in response to perturbations from the manipulated object and choosing an adequate mechanical impedance may be one of the ways the CNS controls the behavior of the complete system (hand+object) [20
]. The mechanical impedance could be improved trough viscoelastic changes by feedfoward motor commands way, specifically in cases on rapid changes in the magnitude or nature of external forces [21
]. The viscoelasticity is a property of biological materials that are both solid and fluid-like, such as tendons and ligaments. These materials possess time-dependent stress-strain relations, that change as the loading speed changes [22
Based on these key properties of the CNS, some authors [5
] suggested the integration of the Equilibrium Point Theory and Internal Model Theory in movement control. They defended that the CNS relies on the viscoelasticity property when the Internal Models are imperfect or the environment is unstable [23
]. In this new model, the viscoelasticity depends on feedback controller, as while the internal models are the result of forward controller. On the following graph (Figure 1) it is presented a model created to explain how the motor control could be programmed, based on new integral model.
At least two different models are referred on the literature to explain the neural regulation of agonist-antagonist muscle activation at a spinal level: the Common Drive Model and the Disynaptic Reciprocal Ia Inhibition.
Common drive model :
The nervous system does not control the firing rates of motor units individually, to generate the muscle synergy [5
], instead the CNS programmed the excitation of the motoneuron pool [25
]. A motoneuron pool is a group of motor neurons with common targets and afferent inputs [26
]. In this muscle activation mode, a flexor muscle and a extensor muscle around one specific joint are controlled as if they were one muscle. The way a group of agonist-antagonist muscles are activated depends on the function to be performed but also depends on the spine origin proximity of their nerve roots. This model defends that the “flex” and “extend” commands channels in the CNS are reciprocally organised and its main characteristic is that the inhibition of the antagonist muscle happens prior to excitation of the agonist muscle – co-contraction phenomena [26
]. The co-contraction is important in two specific situations both depending on the environment conditions: (i) during states of uncertainty or (ii) when it was required a compensatory force correction [25
] (e.g., if they occurs destabilizing forces during a upper limb dynamic task etc. [11
] (Figure 2).
The disynaptic reciprocal Ia inhibition:
This model was proposed [14
] to explain the reciprocal pattern of muscles activation on voluntary movements [27
]. The brain controls agonist α-motoneurones and Ia inhibitory interneurones, which have monosynaptic projections to motoneurones of the antagonists in parallel [14
]. Despite this mechanism involving only a single interneuron it has a special characteristic - their reciprocal organisation [27
]. This organisation allows that during a simultaneous activation of two antagonist muscles, the muscular contractions generate less force. During this process occurs a “double action of the reciprocal inhibitory action” and the a-motoneurones supplying the antagonistic muscles is simultaneously depressed [14
] (Figure 3).
The main difference between these two models is related to the processing mechanism: the Disynaptic Reciprocal Ia inhibition depends on a single interneuron with monosynaptic projections to motoneurones of the antagonists and the Common Drive Model defends that exist different motoneurons (one for the agonist and other for the antagonist muscles) linked by common targets and afferent inputs. Therefore, it remains unclear how is the co-contraction programmed by the CNS and further research should clarify these mechanisms related to specific tasks (e.g. postural or dynamics) and specific environment conditions (e.g. presence of perturbing forces) [2