Response conflicts and their sequential adaptations are typically investigated for conflicts between an instructed and a competing automatic response tendency, triggered by different stimulus features. In the present study, comprising six Experiments, we investigate conflicts between a participant’s and a co-actor’s competing response goals and their sequential adaptations in a novel paradigm. Specifically, the participant and a co-actor (another human or the computer) move a target relay-like in two steps from the bottom center to the top left or right corner of the computer-screen. In a first block, participants must always follow the co-actor’s goal choice (follower training; Experiments 1, 2). In the critical second block, participants are free to follow either’s goal choice. Experiments 3-6 explored modifications of the basic paradigm (order of blocks; leader training; humanized computer co-actor). In all six Experiments, participants showed congruency effects whereas conflict adaptation effects were almost exclusively found for second blocks (mostly free-choice) and only if participants cooperated with a real or imagined human co-actor. Thus, the ‘relay paradigm’ allows to examine socially triggered response conflicts between co-actors in different roles (follower, leader, decider) and settings (lab or online) and offers new insights into the functionality of conflict adaption.