Abstract
Evidence regarding the appropriateness or otherwise of rectification up to this point has primarily been in the form of simulation studies (Myers et al. 2003; Stegeman et al. 2010; Boonstra and Breakspear 2012; Neto and Christou 2010), or analysis of spectra from individual EMG records (Mc-Clelland, Cvetkovic, and Mills 2011). However, the key question of whether un-rectified (raw) or rectified EMG is a better predictor of the frequency components of motor unit synchronization can only be definitively answered using experimental data consisting of simultaneous recordings of paired motor unit discharges and surface EMG from the homonymous muscle. This study analyses such a data set, and the key result is shown in figure 2. Use of un-rectified (raw) EMG as a predictor signal does not remove all components of motor unit synchronization in the 15-30 Hz frequency band. In contrast the rectified EMG does remove all components of motor unit synchronization, there is no consistent evidence in figure 2 of any significant partial correlation with rectified EMG as predictor. It is worth noting that using un-rectified EMG as predictor does greatly reduce the magnitude of the partial coherence, however in the context of the present debate as to which is the better predictor of motor unit synchronization the evidence in figure 2 is strongly in favour of the rectified surface EMG. The results in figure 2 are for the unloaded condition in (cite Halliday et al., 1999). Similar results are obtained for the loaded condition (not shown).
A number of previous studies have pointed out that rectification is a non linear operation and when used unwisely can distort the frequency components in a signal (cite Neto, McClelland others?). This is true for sinusoidal signals, however, the issue is less clear cut with respect to EMG. As has been previously discussed (cite Halliday& Farmer 2010) the main aim of this type of spectral analysis is to characterise the components of motor unit timing. The present data further allows direct comparison of single unit and surface EMG spectral estimates from the homonymous muscle (EDC). The differences between the raw and rectified surface EMG spectral estimates in figure 1 are not as pronounced as might have been expected. Taking the single unit spectra as the reference spectra, this has clear peaks at 12 – 13 Hz and around 26 Hz, with Poisson behaviour (no significant rhythmic features – see results) above 35 Hz. The rectified EMG spectral estimate has peaks at 12Hz and around 25 Hz, which are the main features. The un-rectified EMG spectral estimate has evidence of a distinct feature around 12 Hz and possibly around 25 Hz, although this may be part of the broad increase in power that extends above 60 Hz. The maximum power in the raw surface EMG is in the region 40 – 60 Hz, this corresponds to the frequency range where the motor unit spectrum has no significant rhythmic structure and where there is a steady decline in the magnitude of the rectified EMG spectral estimate. It has been suggested that this higher frequency range may reflect information due to the shape of motor unit action potentials (cite Halliday et al. 1995). The conclusion from the experimental data in figure 1 is that the interpretation of the rectified EMG spectral estimate is a closer match to the motor unit spectral estimate than that of the un-rectified surface EMG. This is in agreement with the findings of (cite Elble & Randall, 1976) who compared motor unit and rectified EMG spectra in, with a focus on the 8-12 Hz frequency band.
In conclusion, analysis of a large data set consisting of paired single motor units recordings and surface EMG from the homonymous muscle has demonstrated 1) that interpretation of the spectrum of the rectified surface EMG is closer to that of the motor unit spectrum than that of the un-rectified EMG, and 2) that rectified EMG is a better predictor of the components of motor unit synchronization than the corresponding un-rectified EMG. This is in agreement with the approach adopted by Elble & Randal (1976), Halliday et al. (1995) and many others.
A number of previous studies have pointed out that rectification is a non linear operation and when used unwisely can distort the frequency components in a signal (cite Neto, McClelland others?). This is true for sinusoidal signals, however, the issue is less clear cut with respect to EMG. As has been previously discussed (cite Halliday& Farmer 2010) the main aim of this type of spectral analysis is to characterise the components of motor unit timing. The present data further allows direct comparison of single unit and surface EMG spectral estimates from the homonymous muscle (EDC). The differences between the raw and rectified surface EMG spectral estimates in figure 1 are not as pronounced as might have been expected. Taking the single unit spectra as the reference spectra, this has clear peaks at 12 – 13 Hz and around 26 Hz, with Poisson behaviour (no significant rhythmic features – see results) above 35 Hz. The rectified EMG spectral estimate has peaks at 12Hz and around 25 Hz, which are the main features. The un-rectified EMG spectral estimate has evidence of a distinct feature around 12 Hz and possibly around 25 Hz, although this may be part of the broad increase in power that extends above 60 Hz. The maximum power in the raw surface EMG is in the region 40 – 60 Hz, this corresponds to the frequency range where the motor unit spectrum has no significant rhythmic structure and where there is a steady decline in the magnitude of the rectified EMG spectral estimate. It has been suggested that this higher frequency range may reflect information due to the shape of motor unit action potentials (cite Halliday et al. 1995). The conclusion from the experimental data in figure 1 is that the interpretation of the rectified EMG spectral estimate is a closer match to the motor unit spectral estimate than that of the un-rectified surface EMG. This is in agreement with the findings of (cite Elble & Randall, 1976) who compared motor unit and rectified EMG spectra in, with a focus on the 8-12 Hz frequency band.
In conclusion, analysis of a large data set consisting of paired single motor units recordings and surface EMG from the homonymous muscle has demonstrated 1) that interpretation of the spectrum of the rectified surface EMG is closer to that of the motor unit spectrum than that of the un-rectified EMG, and 2) that rectified EMG is a better predictor of the components of motor unit synchronization than the corresponding un-rectified EMG. This is in agreement with the approach adopted by Elble & Randal (1976), Halliday et al. (1995) and many others.
| Original language | English |
|---|---|
| Pages (from-to) | 146-156 |
| Number of pages | 11 |
| Journal | Neuroscience |
| Volume | 233 |
| Issue number | 1 |
| Early online date | 4 Jan 2013 |
| DOIs | |
| Publication status | Published - 4 Jan 2013 |