Neural Substrates of Cognitive Motor Interference During Walking; Peripheral and Central Mechanisms

Abstract

Current gait control models suggest that independent locomotion depends on central and peripheral mechanisms. However, less information is available on the integration of these mechanisms for adaptive walking. In this cross-sectional study, we investigated gait control mechanisms in people with Parkinson's disease (PD) and healthy older (HO) adults: at self-selected walking speed (SSWS) and at fast walking speed (FWS). We measured effect of additional cognitive task (DT) and increased speed on prefrontal (PFC) and motor cortex (M1) activation, and Soleus H-reflex gain. Under DT-conditions we observed increased activation in PFC and M1. Whilst H-reflex gain decreased with additional cognitive load for both groups and speeds, H-reflex gain was lower in PD compared to HO while walking under ST condition at SSWS. Attentional load in PFC excites M1, which in turn increases inhibition on H-reflex activity during walking and reduces activity and sensitivity of peripheral reflex during the stance phase of gait. Importantly this effect on sensitivity was greater in HO. We have previously observed that the PFC copes with increased attentional load in young adults with no impact on peripheral reflexes and we suggest that gait instability in PD may in part be due to altered sensorimotor functioning reducing the sensitivity of peripheral reflexes.

Keywords: H-reflex; Parkinson disease; cognitive motor interference; fNIRS; gait control; motor cortex; prefrontal cortex.

FIGURE 1

Schematic representation of NIRS probe location.

FIGURE 2

H-reflex measurement setup. (1) The cables from the stimulator to the Tibial nerve in the popliteal fossa. (2) Reference electrode going to the lateral Malleolus. (3) The cables from the EMG-amplifier going to the Soleus. (4) The EMG-signal going to isolator which is connected with the filter. (5) The filtered EMG-signal is going to the CED-box. This box is connected with the pc. (6) The signal from the stimulator is also sent to the CED-box. (7) “Width-tool,” which sends the signal to the CED-box. With a pc the signals from the CED-box can be viewed in the “Signal Software.”

FIGURE 3

Groups results of the Soleus H-reflex gain. (A) Means and standard errors of linear regressions slopes at SSWS. (B) Means and standard errors of linear regressions slopes at FWS. PD, Parkinson disease; SSWS, self-selected walking speed; FWS, fast walking speed; ST, single task; DT, dual task.

FIGURE 4

Cortical activation while walking. Group data (means and standard errors) of task-related changes in oxy-Hb (left side) and deoxy-Hb (right side) concentrations while walking under different conditions, in (A) left prefrontal cortex, (B) left motor cortex, and (C) right motor cortex. oxy-Hb, oxygenated hemoglobin; deoxy-Hb, deoxygenated hemoglobin; M1, motor cortex; PFC, prefrontal cortex; PD, Parkinson disease; SSWS, self-selected walking speed; FWS, fast walking speed; ST, single task; DT, dual task.

FIGURE 5

Basic schematic illustration of the functional coupling between PFC and M1 during walking. (A) PFC activation during ST walking is devoted to inhibitory modulation of M1 activity via SMA/PM pathways. While PFC involvement in attentional control is tentative, M1 exerts excitatory modulation over SII via CST. (B) PFC activation during DT walking, albeit increased, is primarily devoted to attentional control. Thus, its inhibitory effect on M1 becomes less, and the later exerts a more substantial excitatory modulation over SII. Block arrows represent anatomical pathways. Red and blue arrows donate excitatory and inhibitory modulations, respectively. Sizes of blue and red arrows are relative to the proposed activation level. PFC, prefrontal cortex; M1, motor cortex; SMA, sensorimotor area; PM, premotor area; SII, spinal inhibitory interneurons; CST, corticospinal tract.