Searching through sequences using the MILO task

Visual search tasks have taught us much about how we explore the world. A useful extension to the standard paradigm – where a single target must be located among a variable set size of distractors – is to consider search episodes that contain multiple targets. Here I will describe work using the MILO (multi-item localisation) task where participants must select randomly positioned targets in a pre-defined order (e.g., find the digits 1 through 8; Thornton & Horowitz, 2004;2020). Using a sequence makes it possible to explore the full temporal context of search. That is, the speed of selecting a given target in the sequence could be influenced both by where you’ve been (i.e., the previous target locations or retrospective context) and where you need to go next (i.e., the future target locations or prospective context). In this talk, I’ll focus on one specific finding from MILO studies: when searching through a sequence you have almost perfect memory for – and thus can effectively ignore – the locations of all previously selected items. I’ll argue that this implicates the use of a very precise inhibitory tagging mechanism. I’ll show how such tagging can be disrupted through increased cognitive load – either from inherent task demands or cognitive decline – and discuss new work exploring the possible connection to inhibition of return (IOR) as a foraging facilitator (Klein & MacInnes, 1999).