In the task cueing procedure, participants perform two tasks in varying sequences, with distinct cues indicating the relevant task on each trial, allowing us to investigate associative processes during task set reconfiguration. In previous experiments with this procedure, remapping the cue-task assignment after initial practice yielded a performance cost that was larger in task switch trials than in task repetition trials, compared to replacing the cues with novel ones. However, this research could not distinguish task switch costs from cue switch costs. We applied a 4:2 mapping between cues and tasks to separate task switch costs from cue switch costs by analyzing only data from trials associated with a cue switch. After an initial practice phase, we exchanged two cues between tasks in a final transfer phase, resulting in two types of cues: exchanged and constant. Exchanging cues between tasks impaired performance in trials featuring an exchanged cue compared to trials featuring a constant cue. Moreover, performance impairment was not smaller in task repetition trials than in task switch trials. This was the case even when only examining task switch trials featuring a switch of cue type matched with task repetition trials that were always associated with a cue type switch. These results suggest a broader impact of disrupted associative processes, extending beyond task set reconfiguration mechanisms. While the exact mechanisms driving these effects remain elusive, the findings underscore the importance of further investigating the dynamics of altered cue-task associations by considering the type of cues across different trial types.