Using a conditioned suppression task, we investigated extinction and renewal of Pavlovian modulation in human sequential Feature Positive (FP) discrimination learning. In Experiment 1, in context a participants were first trained on two FP discriminations, X[right arrow]A+/A- and Y[right arrow]B+/B-. Extinction treatment was administered in the acquisition context a ("aaa" group) or in a new context b ("aba" group), and comprised X[right arrow]A- extinction and Y- control trials. Discriminative X[right arrow]A/A responding was lost in both groups when tested in the extinction context, but partially recovered in the "aba" and not in the "aaa" group when tested in the acquisition context, suggesting extinction and renewal of extinguished modulation. The same was observed for the Y[right arrow]B/B control pair, however, questioning whether the loss of discriminative X[right arrow]A/A responding represented genuine extinction of modulation. In Experiment 2, including only "aba" groups, participants were trained in context a on two FP discriminations, X[right arrow]A+/A- and Y[right arrow]B+/B-, after which the group "Extinction" was exposed to X[right arrow]A- extinction trials in context b, whereas the group "Control" was exposed to X- control trials; concurrently, both groups received further Y[right arrow]B+/B- training. In the group Control, differential Y[right arrow]B/B and X[right arrow]A/A responding were acquired and maintained throughout the experiment. In the group Extinction, while Y[right arrow]B/B responding was also maintained throughout, differential X[right arrow]A/A responding disappeared because of X[right arrow]A- extinction treatment when tested in the extinction context b, but partially reappeared when tested in the acquisition context a. This evidences "aba"-renewal of extinguished modulation. (Contains 5 figures and 1 table.)