KINESIOLOGICAL ELECTROMYOGRAPHY - Biomechanics and Motor Control of Human Movement - page 268

Figure 10.14 Results of a cross-correlation, Rxy , between two adjacent electrodes in
Figure 10.13. Here, R xy = 0 . 6, indicating a common pick-up of R xy = 0 . 36, or 36%
cross-talk. (Reprinted from the Journal of Electromyography and Kinesiology ,Vol.4,
Winter, D. A., A. J. Fuglevand, and S. Archer. “Crosstalk in Surface Electromyography:
Theoretical and Practical Estimates,” pp. 15 - 26, 1994, with permission of Elsevier.)

correlation is 0.6 at zero phase shift. Because the electrode pairs were placed
parallel with the muscle fibers, each pair was the same distance from the
motor point; thus, any common m.u.a.p.'s will be virtually in phase. With the
peak R xy = 0 . 6, the net correlation, R xy =

0 . 36, tells us the two EMG had
36% of their signals in common. R xy drops off sharply as the distance between
electrode pairs is increased (Winter et al., 1994). At 2.5 cm, R xy = 0 . 48 (23%

cross-talk); at 5.0 cm, R xy = 0 . 14

(2% cross-talk). Reducing the size of the electrodes reduces the pick-up zone
of the electrode, and thus the cross-talk. The motor units at a greater distance
from the electrodes contribute most to the cross-talk, and the m.u.a.p.'s from
those units are longer duration (lower frequency). Thus, by processing the
EMG through a differentiator, the higher-frequency (closer) m.u.a.p.'s are
emphasized, while the lower-frequency (further) m.u.a.p.'s are attenuated.

0 . 24 (6% cross-talk); and at 7.5 cm, R xy =

10.2.6 Recommendations for Surface Electromyogram Reporting and
Electrode Placement Procedures
Over the past 30 years, the availability of EMG recording equipment and
the number of laboratories using such equipment has exploded dramatically.
Most labs have developed their own protocols regarding the details needed for
reporting results and for selection of the electrode sites over the muscles. The