“…Another property of GAMM that makes it particularly valuable for EEG research is its ability to account for nonlinearity of time series data. In addition, by including random-effect variables, the modeling enabled us to appropriately deal with structural variation across participants and images in EEG amplitude (for other ERP research that applied GAMM, see De Cat, Klepousniotou, & Baayen, 2015;Kryuchkova, Tucker, Wurm, & Baayen, 2012;Meulman et al, 2015;Tremblay & Newman, 2015). Not unexpectedly, a large proportion of the variance in the data is accounted for by the random-effect variables, Participant and Image.…”