Location memory in visual search : the multi-item localization (MILO) task : a tool for investigation

Abstract

In the current study, the Multi-Item Localization Task (MILO; Thornton & Horowitz, 2004) was used as a tool to explore location memory in visual search. Previous studies provide evidence that whilst visually searching for a specific target, participants are able to ignore locations that have already been visited, making their ongoing search more efficient (Thornton & Horowitz, 2004). This is referred to as location tagging and the current study sought to explore if this ability could be disrupted. The overall goal was to increase knowledge regarding the resources needed for location tagging to take place. Specifically, we manipulated concurrent spatial working memory workload while performing the MILO task to examine if location tagging could be disrupted. In Experiment 1, participants (N = 12) had to select eight numerical targets that appeared on an iPad screen by touching each item in sequence. There were three conditions. On Vanish trials, items disappeared when they were touched, removing them from the search array. On Remains trials, items did not vanish, leaving them available to interfere with subsequent search. Previous studies found identical response time patterns for these two conditions, the signature of location tagging. Here, we also introduced Chess trials where memory workload was heavily increased as participants were asked to simultaneously remember a chess layout whilst completing the Remains MILO task. Results for the Vanish & Remains conditions replicated previous studies, indicating the presence of location tagging. However, the workload added in the Chess trials was not enough to disrupt location tagging. In Experiment 2 participants (N = 12) completed the same three conditions, but had to select six targets made up of chess pieces in the game’s precedence order. This was a novel sequence, that would also be expected to interfere with memory during the Chess trials. Again, there was no evidence that location tagging was disrupted. However, when comparing the results of Experiments 1 and 2, it appeared that overall search-behaviour was affected by using the chess items for the MILO task as patterns relating to future planning were noticeably different. In terms of our main goals, however, we can conclude that participants were still able to remember where they have been before when searching for the current target, even when the memory workload was increased, suggesting that location tagging may be an automatic process that either does not make demands on central resources or at least uses resources that are different from those needed during explicit location memory.