A prime goal of psychological science is to understand how humans can flexibly adapt to rapidly changing contexts. The foundation of this cognitive flexibility rests on contextual adjustments of cognitive control, which can be tested using the list-wide proportion congruency effect (LWPC). Blocks with mostly incongruent (MI) trials show smaller conflict interference effects compared to blocks with mostly congruent (MC) trials. A critical debate is how proactive and reactive control processes drive contextual adjustments. In this preregistered study (N = 30), we address this conundrum, by using the theta rhythm as a key neural marker for cognitive control. In a confound-minimized Stroop paradigm with short alternating MC and MI blocks, we tested reaction times, error rates, and participants' individualized theta activity (2–7 Hz) in the scalp-recorded electroencephalogram. An LWPC effect was found for both, reaction times and error rates. Importantly, the results provided clear evidence for reactive control processes in the theta rhythm: Theta power was higher in rare incongruent compared with congruent trials in MC blocks, but there was no such modulation in MI blocks. However, regarding proactive control, there were no differences in sustained theta power between MC and MI blocks. A complementary analysis of the alpha activity (8–14 Hz) also revealed no evidence for sustained attentional resources in MI blocks. These findings suggest that contextual adjustments rely mainly on reactive control processes in the theta rhythm. Proactive control, in the present study, may be limited to a flexible attentional shift but does not seem to require sustained theta activity.