The contralateral electromyographic (EMG) silent period (CSP) following transcranial magnetic stimulation (TMS) has been used as a measure of intracortical inhibition (ICI) and is particularly useful when the conditioning-test method is impractical. We have designed an automated method based on the cumulative sum (Cusum) statistic (Davey et al., 1986) to estimate the CSP duration. This was evaluated against the conventional manual method performed by two independent experts (MM1 and MM2) and against two automated methods based on statistical difference (AM1) and mean cumulative difference (AM2) (Nilsson et al., 1997; Garvey et al., 2001). With local ethics approval and consent, six adult volunteers were recruited. Surface EMGs were recorded from the right thenar during 20% maximum voluntary contraction. Trials of 20 magnetic stimuli were delivered using a Magstim 200 stimulator connected to a 9cm circular coil centred over the vertex. TMS intensity was selected by varying the TMS output in 5% steps so that a clear CSP could be identified at the lowest stimulus intensity. Analysis was performed offline using Signal Software (Cambridge Electronic Design). The mean CSP durations were: 23.3ms (± 6.4) by Cusum, 22.0ms (± 6.4) by MM1, 19.3ms (± 6.4) by MM2, 18.4ms (± 6.8) by AM1 and 15.3ms (± 6.7) by AM2. The intraclass correlation coefficient (ICC) between MM1 and MM2 was 0.88 (P<0.001). Of the 3 automated methods, the Cusum showed the strongest correlation with the MM1 (ICC 0.98, P<0.001) and MM2 (ICC 0.87, P<0.001). CSP durations were significantly different (repeated measures ANOVA with Bonferroni adjustment; P < 0.05) between Cusum and both expert raters (MM1; MM2). When the stimulus intensity was increased by 5% of maximum stimulator output (MSO), the ICC between all methods increased to 0.72 compared to 0.45 at the test intensity. At 5% MSO below test intensity, ICC between all methods decreased to 0.33 and there was no longer a significant correlation between MM1 and MM2. In conclusion, at higher intensity, both human raters and automated methods become more reliable while at lower intensity, when the ICI is barely detectable, all methods become less reliable. Overall, of the automated methods, the Cusum is best correlated with expert human raters and is a simple, graphical method of detecting CSP that can be easily automated.