Antagonistic Priming Reveals an Anticipatory Mode of Cognitive Control Under Time Pressure

Automatic reactions can endanger survival in some critical situations, like running toward anexit engulfed in flames, fighting a river current, or steering a car away from a cliff during an icyskid. Across four Stroop-derived experiments conducted between 2017 and 2022 with a WEIRDundergraduate sample (total N = 192), we examined whether people can preempt such counterproductiveimpulses via a fast and anticipatory mode of cognitive control. The Antagonistic-Priming paradigm used in the present series of experiments presented a prime word that activatesResponse 1, followed -after a variable stimulus-onset asynchrony (SOA)- by a colourpatch that usually requires the opposite Response 2. At list onset, participants either did or didnot receive anticipatory information (AI) stating that prime and target would be incongruent on80% of trials. With a 300-ms SOA, a binary response set (red vs. green), and the AI present,participants were able to reverse the classic Stroop pattern: they were faster and more accurateon incongruent trials than on congruent trials (Experiment 1; replicated in Experiment 4 with a450-ms SOA and black–white responses). Eliminating the SOA (Experiment 2) or expandingthe response set to four colours (Experiment 3) abolished the reversal. These results provide evidenceof a proactive, instruction-based control mechanism that readies the antagonist responsebefore conflict arises, refining list-wide proportion-congruent accounts and informing hazard-responsetraining.