Identification of connectivity in human motor control: exciting the afferent pathways

Motor control involves various parts of the central nervous system (CNS) and requires the exchange of information between neural populations in the CNS. Information exchange in facilitated by the formation of functional networks between populations of was localized at the sensorimotor area contralateral to the position perturbation. Position-cortical coherence has an important advantage over CMC measured during an unperturbed task. Significant CMC is detected in only 40% of a normal healthy populations. Significant PCC was detected in all healthy subjects. In addition, because CMC is affected by both afferent and efferent pathways it is not possible to relate changes in CMC to specific pathway connectivity. Position-cortical coherence primarily reflects afferent pathway connectivity. The muscle stretch evoked potential represents the time course of cortical activation in response to transient joint movement. Peaks in the StrEP at different latencies allow the separation between the arrival of sensory feedback at the cortex and subsequent processing of this information. In normal young subjects (n Æ 22), the StrEP was characterized by an early peak within 60ms after movement onset localized at the contralateral primary motor cortex and a complex of late peaks between 60 and 300ms after movement onset over the vertex. Stretch evoked potential waveforms and features were consistent across different tasks and sessions.Also in (subacute) stroke survivors afferent pathway connectivity can be assessed by PCC or the StrEP. All (subacute) stroke survivors presented PCC (chapter 4) and a StrEP (chapter 6), even those with very poor motor function who were unable to performan isotonic wrist flexion task. Abnormal and heterogeneous StrEP waveforms were seen in subacute stroke subjects, but no significant difference was found between StrEP features of subjects with good and poor function. However, presence of PCC did differ between subacute subjects with good and poor motor function. Future research should show whether PCC could aid in giving stroke survivors a more detailed prognosis of their potential motor function recovery or allow applying rehabilitation therapy targeted to critical time windows. vineurons, relying on the synchronization of neural oscillations. This thesis presents the development and evaluation of techniques that quantify the corticomuscular connectivity (i.e. connectivity between cortex and muscles) in motor control. Intramuscular coherence (IMC) quantifies the common (supra-spinal) drive to different parts of a single muscle. While IMC analyses does not require complex measurement techniques, IMC analysis is an attractive technique to apply during functional tasks like walking. However, the applicability of IMC analysis is limited due lowreliability and agreement of IMC variables between sessions (chapter 2). The smallest real difference indicated that large differences in IMC variables are needed to detect changes in common drive to muscles between sessions. Corticomuscular coherence (CMC) is the coherence between cortical activity, recorded by EEG or MEG, and muscle activity, recorded by EMG. This is a widely applied measure of corticomuscular connectivity. The phase of the complex corticomuscular coherency is often interpreted as a transmission delay between cortex and muscles. We showed that phase analysis for the estimation of transmission delay gives unreliable results in closed loop systems (chapter 3). As evidence is accumulating that CMC arises in a closed loop system, phase analysis of corticomuscular coherency is not a valid measure of the transmission delay between cortex and muscles. Based on techniques from the field of system identification, two measures of pathway connectivity were presented (chapter 4 and 6). External mechanical perturbations were applied to ‘open’ the closed loop motor control system. Two connectivity measures were derived from the application of joint position perturbations during a static motor task: position-cortical coherence (PCC) and muscle stretch evoked potentials (StrEPs). Because the mechanical perturbations ‘enter’ the motor control system at the beginning of the afferent sensory pathways, integrity of the afferent sensory pathways is required and sufficient for the detection of PCC and StrEP. Indeed, in the normal subject group all subjects presented PCC as well as a StrEP, consistent with a normal integrity of the afferent sensory pathways. Position-cortical coherence quantifies the correlation between a joint position perturbation and cortical activity, recorded by EEG, in the frequency domain. Presence of significant PCC at a specific frequency indicates that the cortical activity is synchronized to the position perturbation at that frequency. In normal young subjects (n Æ 22) significant PCCwas localized at the sensorimotor area contralateral to the position perturbation. Position-cortical coherence has an important advantage over CMC measured during an unperturbed task. Significant CMC is detected in only 40% of a normal healthy populations. Significant PCC was detected in all healthy subjects. In addition, because CMC is affected by both afferent and efferent pathways it is not possible to relate changes in CMC to specific pathway connectivity. Position-cortical coherence primarily reflects afferent pathway connectivity. The muscle stretch evoked potential represents the time course of cortical activation in response to transient joint movement. Peaks in the StrEP at different latencies allow the separation between the arrival of sensory feedback at the cortex and subsequent processing of this information. In normal young subjects (n Æ 22), the StrEP was characterized by an early peak within 60ms after movement onset localized at the contralateral primary motor cortex and a complex of late peaks between 60 and 300ms after movement onset over the vertex. Stretch evoked potential waveforms and features were consistent across different tasks and sessions. Also in (subacute) stroke survivors afferent pathway connectivity can be assessed by PCC or the StrEP. All (subacute) stroke survivors presented PCC (chapter 4) and a StrEP (chapter 6), even those with very poor motor function who were unable to performan isotonic wrist flexion task. Abnormal and heterogeneous StrEP waveforms were seen in subacute stroke subjects, but no significant difference was found between StrEP features of subjects with good and poor function. However, presence of PCC did differ between subacute subjects with good and poor motor function. Future research should show whether PCC could aid in giving stroke survivors a more detailed prognosis of their potential motor function recovery or allow applying rehabilitation therapy targeted to critical time windows.