Abstract
Two methods are presented for the data analysis of signals derived by subtracting conditioned from control reflex data in full-wave rectified and averaged electromyographic (EMG) signals. The first method uses the ratios of the mean amplitudes and standard errors of the mean (S.E.M.s) (i.e. Student's t values) of a series of data points in such a difference signal, and deals with the detection of latencies of reflex components which are susceptible to a conditioning stimulus or experimental situation. The second method applies a modified cumulative sum (CUSUM) technique to full-wave rectified difference signals. This modified CUSUM technique determines the magnitude of the effects of the conditioning situation above that expected due to chance fluctuations, taking into account the effect of reflex modulations on such chance fluctuations in a post-stimulus period. The modified CUSUM technique proved particularly useful when various subtle but consistent, opposing changes occurred sequentially in the conditioned series thus yielding a complicated pattern of effects in a difference signal, with a low signal-to-noise ratio.
Original language | English |
---|---|
Pages (from-to) | 117-125 |
Number of pages | 9 |
Journal | Journal of Neuroscience Methods |
Volume | 58 |
Issue number | 1-2 |
DOIs | |
Publication status | Published - 1995 |
Keywords
- (Human)
- Conditioning
- CUSUM
- Electromyography
- Reflex analysis
- Significance testing