SIGNAL PROCESSING - Biomechanics and Motor Control of Human Movement - page 25

250

200

L4

150

100

C7

50

0

0 10 20 30 40 50 60

% of STRIDE

70 80 90 100

Figure 2.7 Ensemble-averaged profiles over the stride period of EMG signals of
paraspinal muscles at C 7 and L 4 levels for one of the subjects. Note the 2 nd harmonic
peaks occurring during the weight acceptance periods of the left and right feet to
balance the trunk and head. The C 7 amplitude is lower than the L 4 amplitude because
the inertial load above C 7 is considerably lower than that above L 4 . More important
is the timing of C 7 so that it is ahead of L 4 , indicating that the head is balanced first,
ahead of the trunk. (Reproduced by permission from Gait and Posture )

later) of L4 and C7 muscle profiles over the stride period for one subject). A
cross-correlation of these two signals showed that C7 was in advance of L4
by about 70 ms. For all 10 young adults in this study, all signals at C7, T2,
T4, T6, T8, T10, T12, and L2 were separately cross correlated with the L4
profile. The phase shift of these signals is presented in Figure 2.8. The ear-
lier turn on of the more superior paraspinal muscles indicates a “top-down”
anticipatory strategy to stabilize the head first, then the cervical level, the tho-
racic level, and finally the lumbar level. This strategy resulted in a dramatic
decrease in the A/P head acceleration over the stride period compared to the
A/P acceleration of the pelvis. In a subsequent study on fit and healthy elderly
(Wieman, 1991) the head/hip acceleration (%) in the elderly (41.9%) was sig-
nificantly higher (p < .02) compared with that of young adults (22.7%); this
indicated that the elderly had lost this “top-down” anticipatory strategy, and
the paraspinal EMG profiles bore this out.

2.1.8.3 Measurement of Synergistic and Coactivation EMG Profiles.
There is considerable information in EMG profiles regarding the action
of agonist/antagonist muscle groups during any given activity. Recently,
cross-correlation techniques have been used to quantify coactivation patterns
(agonist/antagonist active at the same time) and non-coactivation patterns
(agonist/antagonist having synergistic out of phase patterns): Nelson-Wong