Multiple factors influence the optimization of cognitive control, such as the likelihood of conflict associated with specific items. To investigate learning-guided, reactive adjustments of control, the item-specific proportion congruence (ISPC) manipulation is commonly used in which PC is manipulated within a 50% congruent list comprising mostly incongruent (MI) and mostly congruent (MC) items. The ISPC effect, a smaller Stroop effect for MI items compared with MC items, suggests a more focused/relaxed control setting is retrieved for MI/MC items. We investigated how performance-contingent reward modulates item-specific control, presenting reward as a pre-cue and feedback (Experiment 1) or only as feedback following responses (Experiment 2). First, participants completed a reward-free baseline phase where ISPC was manipulated, which yielded an ISPC effect. Then in a subsequent test phase MC or MI items were associated with reward for different groups. It was expected that reward would reinforce the more focused control setting for MI items when they were reward-associated, but would negate the more relaxed control setting for MC items when they were reward-associated. In Experiment 1, the ISPC effect persisted when rewards were associated with MI items, but disappeared when rewards were linked to MC items. In Experiment 2, however, preliminary results showed a persistent ISPC effect for both groups. While reward shifted reactive control and modulated the ISPC effect when presented in advance of the stimulus, it did not induce a shift in reactive control when only presented as feedback. This highlights the nuanced interplay between reward-based mechanisms and reactive control.