Abstract
Humans often use visual objects as landmarks to navigate through their environment. Several qualities of an object can make it useful as a landmark; for example, (i) being positioned at a navigationally relevant decision point, or (ii) being large and stable. Is there a common representational code for these landmark properties? To investigate this question, we presented subjects with a movie depicting a route through a virtual museum, with objects placed at intersections (navigational decision points) or at simple turns (non-decision points) along the route. They were asked to pay particular attention to half of these objects, indicated by the category vehicle or animal. Following learning, subjects performed a magnitude estimation task in which they judged the size of the objects on a scale of 1-9. Although the actual size of the decision and non-decision point objects was matched, we found that objects previously encountered at decision points were remembered as being larger than objects previously encountered at non-decision points. In contrast, the attentional manipulation had no effect on remembered size. This finding indicates that there is a common representational code for the properties that make an object a landmark, such that if an object attains salience as a landmark by having one of these properties, is it more likely to be remembered as having another as well. Based on previous neuroimaging work showing that the parahippocampal place area (PPA) responds strongly to both objects at navigational decision points (Janzen & Van Turrenout, 2004) and to large objects independent of their explicit navigational utility (Konkle & Oliva, 2012; Mullally & Maguire, 2011), we suggest that the PPA is the neural locus of this common landmark code (Troiani et al., in press).
Meeting abstract presented at VSS 2013