highly significantp = 0.001
Follow-up of this (occasion × trial type) interaction showed no effect of trial type or block at baseline ( p values >0.20) but a highly significant main effect of trial type [ F (1,14) = 17.54, p = 0.001, ηρ 2 = 0.56], showing much faster RTs during pattern trials than random trials over the entire experiment (random trials: mean = 0.33 s, SEM = 0.39; pattern trials: mean = 0.298 s, SEM = 0.034) at follow-up in this group (Table 3 ).