Purchase this article with an account.
Brian Levinthal, Sumeeth Jonathan, Jason Scimeca, Steven Franconeri; Competition limits spatial selection. Journal of Vision 2011;11(11):232. doi: 10.1167/11.11.232.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Comparing, tracking, or monitoring multiple objects may require that we select multiple spatial locations simultaneously. Previous studies show that this ability is possible for at least 2 and possibly up to 8 locations. What is the limited resource that determines these upper limits? We suggest that this independent resource is competition for space itself. Because selected locations are known to have suppressive surrounds, then if these surrounds have a minimum size, as selected locations become too close together in space they will compete and cause interference. The hemifield and quadrant boundaries may buffer this effect by blocking this competition, potentially explaining past results where these regions appear to have ‘independent selection resources’. This competition account makes a unique prediction that a more general resource account does not - that even within a single visual quadrant, decreasing the distance between two selection regions should impair performance. We asked participants to monitor a stream of letters (180 ms/letter) and to determine whether letters in two locations were the same or different. To ensure that participants selected only the relevant locations, letters were flanked by distractor letters. When these locations spaced far apart within a quadrant, performance was higher (70%) than when they were placed closer together (60%). Placing the letter streams across the hemifield boundary eliminated this proximity effect, such that performance was equally high at both distances (76%). Our results are consistent with a mechanism of selection that is limited by competition for selection within a spatiotopic map. We will also briefly discuss new work showing that these same limits on selection within static displays can also explain limits on selection in moving displays (multiple object tracking tasks).
This PDF is available to Subscribers Only