Brain activations during motor imagery of locomotor-related tasks: a PET study

Abstract

Positron emission tomography (PET) was used to study the involvement of supraspinal structures in human locomotion. Six right-handed adults were scanned in four conditions while imagining locomotor-related tasks in the first person perspective: Standing (S), Initiating gait (IG), Walking (W) and Walking with obstacles (WO). When these conditions were compared to a rest (control) condition to identify the neural structures involved in the imagination of locomotor-related tasks, the results revealed a common pattern of activations, which included the dorsal premotor cortex and precuneus bilaterally, the left dorsolateral prefrontal cortex, the left inferior parietal lobule, and the right posterior cingulate cortex. Additional areas involving the pre-supplementary motor area (pre-SMA), the precentral gyrus, were activated during conditions that required the imagery of locomotor movements. Further subtractions between the different locomotor conditions were then carried out to determine the cerebral regions associated with the simulation of increasingly complex locomotor functions. These analyses revealed increases in rCBF activity in the left cuneus and left caudate when the W condition was compared to the IG condition, suggesting that the basal ganglia plays a role in locomotor movements that are automatic in nature. Finally, subtraction of the W from the WO condition yielded increases in activity in the precuneus bilaterally, the left SMA, the right parietal inferior cortex and the left parahippocampal gyrus. Altogether, the present findings suggest that higher brain centers become progressively engaged when demands of locomotor tasks require increasing cognitive and sensory information processing.

Figure 1

Experimental design: schematic representation of the behavioral and physiological assessments, as well as the sequence of scanning conditions. Subjects were administered two motor imagery tests before scanning. Then before each scan, the subjects were shown thrice a video illustrating the task they had to imagine as if in the first‐person perspective, and were asked to rate the vividness of the mental images after each viewing, and again after each scan. EMG and ECG were recorded for 1 min just before and during each scan. The subjects were scanned during two non‐locomotor conditions (rest and imagining standing, which were presented in a counterbalanced order) as well as during three experimental imagery conditions: initiating gait, walking, and walking with obstacles; the latter conditions were presented in a pseudorandom order. The experimental conditions were repeated twice (2 sets), for a total of eight scans.

Figure 2

A: Percent frequency of the scores from the modified VMIQ corresponding to the vividness of motor images (scale 1 = perfectly clear; 5 = no image) when imagining one's self or someone else performing motor tasks with the lower limbs prior to the scanning session. B: Frequency distribution of scores representing the vividness of motor images (on a 5‐point scale) when subjects imagined the experimental conditions after viewing the video just before the scans (before) and during the scans. C: Mean (SD) perceived distance when imagining walking at “slow” and “fast” paces during the three time periods. Note the steeper slope in the fast pace compared to the slow pace condition.

Figure 3

Mean (SD) heart rate changes relative to baseline during rest and during each imagined conditions: S: standing; IG: gait initiation; W: walking; WO: walking with obstacles.

Figure 4

Merged PET‐MRI sections illustrating increases of regional cerebral blood flow associated with the imagining of locomotor‐related tasks. The images were averaged over the 6 subjects and represent: (A) Standing, (B) Initiating gait, (C) Walking, (D) Walking with obstacles minus the control condition Rest. Each subtraction yielded focal changes in blood flow shown as t‐statistic images. The range is coded by the color scale (L = left; R = right).

Figure 5

Merged PET‐MRI sections illustrating increases of regional cerebral blood flow (rCBF) changes associated with the imagining of locomotor‐related tasks. The images were averaged over the 6 subjects and represent the three imagined locomotor conditions: walking, initiating gait and walking with obstacles) minus the control condition rest. Conventions are as in Figure 4. Areas of activation in (A) the leg area in the precentral gyrus and (B) the arm area in the postcentral gyrus.

Figure 6

Merged PET‐MRI sections illustrating increases of regional cerebral blood flow (rCBF) changes associated with the imagining of walking. The images were averaged over the 6 subjects and represent walking minus the initiating gait condition. Conventions are as in Figure 4. Areas of activation are the left caudate nucleus and the left cuneus

Figure 7

Merged PET‐MRI sections illustrating increases of regional cerebral blood flow (rCBF) changes associated with the imagining of walking with obstacles. The images were averaged over the 6 subjects and represent walking with obstacles minus the walking condition. Conventions are as in Figure 4. Areas of activation are the left SMA and the left parahippocampal area.