Repetition Costs in Task Switching Are not Equal to Cue Switching Costs: Evidence From a Cue-Independent Context

People tend to be more efficient when repeating the same action twice in a row, but only if their goal does not change. In task-switching experiments, response-repetition benefits are observed in task repetitions, but response-repetition costs in task switches (the RR effect). The RR effect may be explained in terms of task-response binding so that, for example, when the task repeats it retrieves the binding, and this causes a performance cost when the response switches. Some studies set out to investigate whether a task-irrelevant feature may become bound with the task and the response. The irrelevant feature (the context henceforth) was a dimension of the cue (cue modality, cue colour or cue language, depending on the experiment) which was irrelevant to the task, and that could switch or repeat in each trial like the task and the response. The most common finding was larger RR benefits in task repetitions when the context repeated than when it switched. This pattern is consistent with the context being bound with the task and the response and retrieving them when it repeats (and not when it switches). However, some studies using two cues per task found larger RR benefits when the cue repeated, and this was attributed to cue encoding priming. Thus, in this experiment, we then want to examine context binding in task switching by removing cue encoding benefits. To this aim, we will remove the cues and we will manipulate the context as a feature of the target. If we will replicate the larger RR benefits with context repetitions, we provide evidence for context binding effects which cannot be attributed to cue encoding priming. If we will not replicate them, we provide evidence that cue encoding benefits played an important role in the previous studies.