Table 1.
Demographics of the Stroke Group.
Figure 1.
(A) The arm and hand positions on the gripping apparatus. Participants executed unimanual isometric grip to match their forces (white bar) as accurately as possible to the target force (black bar) at 5%, 25% and 50% of MVC. (B) Typical force trials of an age-matched control and stroke participant at 25% submaximal target force for 20 s.
Figure 2.
The coefficient of variation (CV) at the three force levels.
(A) Differences in CV of force for stroke (paretic hand) and control (non-dominant hand) participants. The stroke participants demonstrated greater variability than the control participants (p<0.05). (B) Differences in CV of force for the paretic and non-paretic arm of the stroke participants. The paretic arm exhibited more variability than the non-paretic arm participants (p<0.05). Error bars represent standard error.
Figure 3.
Normalized power spectrum of the force output from 0–12 Hz.
The majority of the power was concentrated below 1
Figure 4.
Representative data for a stroke and a control participant.
(A) Forces (duration = 2 s). (B) Power spectrum below 1 Hz (Y axis) across time (X axis). The warmer colors (i.e., red) indicate greater power and cool colors (i.e., blue) indicate reduced power. On the left side, the stroke participant demonstrates high power in 0–0.5 Hz and low power in 0.5–1 Hz. In contrast, the control participant (on the right side) demonstrates high power in both 0–0.5 Hz and 0.5–1 Hz. (C) The top panel shows the force band-passed between 0.5–1 Hz. At higher frequencies (>0.5 Hz), the control participant exhibits higher power (amplitude) than the stroke participant. The bottom panel shows the force output low pass filtered at 0.5 Hz. At lower frequencies (<0.5 Hz), the stroke participant exhibits greater power (amplitude) than the control participant.
Figure 5.
The modulation of power below 1 Hz for stroke and control groups.
(A) The stroke group demonstrated significantly greater relative power at 0.14, 0.21 and 0.28 Hz compared with the control group (p<0.05). However, this relationship reversed at higher frequencies. Specifically, the stroke group had significantly reduced power at 0.49 Hz, 0.56, and 0.77 Hz compared with the control group (p<0.05). (B) The average power in frequency bands 0.07–0.35 Hz and 0.49–0.77 Hz for stroke and control participants. The stroke group shows greater power within 0.07–0.35 Hz (p<0.05) and reduced power within 0.49–0.77 Hz (p<0.05).
Figure 6.
The modulation of power below 1 Hz within stroke and control groups.
(A) For the stroke group, the paretic hand demonstrated significantly greater power at 0.21 and 0.28 Hz relative to the non-paretic hand (p<0.05). Further, the paretic hand showed significantly reduced power at 0.56 Hz, 0.7 Hz, 0.77 Hz and 0.84 Hz compared with the non-paretic hand (p<0.05). (B) The power in frequency bands 0.07–0.35 Hz and 0.49–0.77 Hz within stroke group. For the stroke group, the paretic hand shows greater power within 0.07–0.35 Hz (p<0.05) and reduced power within 0.49–0.77 Hz (p<0.05) relative to the non-paretic hand. (C) For the control group, the dominant hand’s power density below 1 Hz did not vary significantly from the non-dominant hand. (D) Similarly, the power in frequency bands 0.07–0.35 Hz and 0.49–0.77 Hz does not differ between the dominant and the non-dominant hand for the control group.
Figure 7.
The prediction of CV of force from power below 1 Hz.
(A) For stroke and control group. A backward multiple linear regression model was applied to predict CV of force (dependent variable) from the absolute power contributions at frequencies below 1 Hz (independent variable). The model predicted (R2 = 0.82) the CV with specific contributions from frequency at 0.28 Hz (part r = 0.86) and 0.63 Hz (part r = −0.40). (B) Within stroke group. A backward multiple linear regression model was applied to predict the difference in CV of force (dependent variable) between hands in stroke participants from the differences in the power below 1 Hz (independent variable). The model predicted (R2 = 0.80) the difference in CV of force with specific contributions from frequency at 0.35 Hz (part r = 0.82) and 0.63 Hz (part r = −0.32).