The temporal extent over which time-frequency effects due to the test object could possibly extend into the delay period was computed as the time for wavelet power at each frequency to drop by a factor of e−2 relative to the test stimulus onset (Torrence and Compo, 1998). This is a conservative overestimate of the backward extent of test-object effects since it does take any latency of test-evoked neural signals into account. This period is plotted as a transparent gray region at the right of all time-frequency and frequency-band summary plots.
