The Multi-Item Localization (MILO) task is a variant of visual search in which participants have to select multiple target items in a pre-defined order. While MILO has much in common with other visual foraging tasks, the use of a sequence adds the ability to precisely explore within-trial temporal context. Here I will focus on retrospective within-trial context, the influence that previously selected items have on search for the current target. A very robust finding is that reaction time patterns are almost identical for trials in which items vanish when selected and trials in which previously selected items remain visible. This suggests we can effectively ignore old targets, probably via some form of inhibitory tagging.

In the current presentation, the results of two ongoing studies will be reported. The first project asks whether there is a limit to the number of items that can be tagged. Previous MILO studies have used sequence lengths between 6 and 12 items. Do the RT functions for vanish and remain trials continue to overlap with 18, 24 or 36 items, for example? The second project asks whether inhibitory tagging relies on motor commands directed to a specific location, for example touch responses or mouse clicks. What happens to vanish/remain patterns when items are selected indirectly? This can be explored using keyboard mappings to secondary characteristics – such as randomly varying color or Landolt C orientation – that can be assigned to objects independent of their position within a sequence.