The finding that changing-state irrelevant sound is more disruptive to working memory than steady-state sound is known as the changing-state effect. Two of the dominant accounts differ in whether they expect the changing-state effect to be task-specific: The duplex-mechanism account posits that the changing-state effect is limited to tasks involving order processing, while the attentional capture account suggests it occurs regardless of the role of order information.
As previous evidence is inconclusive and primarily based on a single task each with and without order processing, we investigated the role of order information in multiple tasks and also considered whether participants applied order-based strategies.
In Experiment 1, tones served as irrelevant sound in two tasks with and two tasks without order processing requirements. Bayesian effect size modeling with a prior favoring the duplex-mechanism account was preregistered to assess potential null effects. Surprisingly, the changing-state effect was observed only in one order-based task and not in the commonly used serial probe task.
Therefore, Experiment 2 used words as irrelevant sound and focused on the serial probe and missing item tasks. In both tasks, a changing-state effect was found. While Experiment 1 leaned slightly towards the duplex-mechanism account, Experiment 2 aligned more with the attentional capture account. Additionally, participants' strategy choices did not consistently impact the changing-state effect.
Therefore, we are currently conducting a third experiment replicating the other two tasks from Experiment 1 with the irrelevant sound used in Experiment 2.