Figure 1.

Illustration of the single-trial measures and their subsequent effects on the grand-averaged event-related potentials (ERPs). First column (Baseline, green): example of baseline trials with P300 elicited in four trials and absent in one trial. Second column, % Trls no ERP: each individual P300 remains unchanged, however the proportion of P300-absent trials increases. This is measured as a percent change in the number of P300-absent trials to total number of trials and results in a reduced classic grand-averaged P300 amplitude (bottom row). Third column, Amplitude Mean: each individual P300 is smaller in amplitude, number of P300-absent trials remains the same as baseline. Amplitude Mean is calculated as the mean of the estimated P300 amplitudes using only the four trials with an elicited P300. The middle P300-absent trial is not included in the calculation. The change in Amplitude Mean also results in a reduced classic grand-averaged P300 amplitude (bottom row). Fourth column, Latency Standard Deviation (SD): number of elicited P300s and their amplitudes are unchanged. Only variation is the latency of each P300 peak such that the latency variance (jitter) is increased (both slower and faster), but the average P300 latency remains constant, resulting in a smaller (and broader) averaged P300 amplitude (bottom row). Fifth column, Latency Mean: the latency of each P300 peak is consistently slower, resulting in a slower averaged ERP latency and no change to the ERP amplitude (bottom row). Green arrows used to help visually clarify shifts from baseline P300s (first column, green).