The list-wide proportion congruency effect describes how the congruency effect varies with the frequency of incongruent trials in a list. That is, the congruency effect is larger in mostly congruent (MC) lists than in mostly incongruent (MI) lists. It has been shown that adaptation to these lists does not change symmetrically when transitioning between the lists: moving from MC to MI rapidly decreases the congruency effect, whereas moving from MI to MC results in little or no increase (Abrahamse et al., 2013). We conducted two studies to replicate this asymmetric list shift effect, using a within-participant design in which all participants experienced both transitions (MC-to-MI and MI-to-MC) within a Stroop paradigm. Although (marginally) significant asymmetric list shift effects were observed, further inspection of the data suggested that practice effects for frequent trial types could explain the findings (see Schmidt, 2016). A third study included frequency-unbiased (i.e., 50% congruent) items embedded in the MC and MI lists, which should not be affected by practice effects.The results showed no significant adaptation between lists for unbiased items and thus no asymmetry, but again selective practice effects for biased items. Taken together, the results suggest that the asymmetric list shift effect is likely to arise from practice effects for frequent trial types rather than from a carryover of task shielding when switching from MI to MC lists.