Search for multiple targets is constrained by both retrospective (i.e., where you've been) and prospective (i.e., where you're planning to go) components of performance. Previous studies using the Multi-Item Localisation (MILO) task have demonstrated that participants accurately remember and discount locations they have already visited and that they plan future actions up to 2 or 3 items ahead (Thornton & Horowitz, 2004 Perception & Psychophysics 66 38–50). A prominent feature of the MILO serial-reaction time (SRT) function is a highly elevated, that is slowed, response, to T1 compared to T2 and all the other items. This “prospective gap” is typically between 600 ms and 1000 ms. Here we present three experiments that use the MILO iPad app to explore this “prospective gap”. In Experiment 1, we “shuffled” the position of future targets each time a response was made. This blocks planning and thus slows all responses to the level of first target, effectively eliminating the gap. In Experiment 2, participants responded to eight identical targets, removing the need to plan a specific sequence of actions. In this situation, absolute response time is greatly reduced and the T1–T2 gap shrinks to around 350 ms. In Experiment 3, participants repeated their search through the same array 10 times. Under these circumstances, the gap systematically reduced from 1300 ms on trial 1 to 300 ms on trial 10. Together, these results suggest that the previously observed prospective gap is a combination of set-up time for registering a new visual layout, response preparation, and sequence planning.
