Figure 1. Illustration of two potential mechanisms for amplitude differences in average ERP.

The average amplitude difference between fencers and non-fencers may due to (A) true amplitude difference in single trials, or (B) the difference in single-trial latency variability.