Abstract
‘Subitizing’ refers to the rapid and accurate ability to enumerate small sets of objects (1-4) with a significantly smaller cost in reaction time for each additional object than in larger object sets. The processes underlying this phenomenon are still under debate. One popular theory suggests that a fixed number of object-tracking ‘fingers of instantiation’ or FINSTs allow for pre-attentive selection of objects, facilitating the speedy enumeration of object sets up to the available number of FINSTs (Pylyshyn 1989). This theory suggests that subitizing would not occur over features, since FINSTs mark proto-objects without carrying knowledge of their properties. Evidence shows that subitizing does not occur while enumerating features such as color and orientation (Watson, Maylor, & Bruce 2005). However, features such as color and orientation do not intrinsically contain information about spatial boundary, a quality that has been shown to be important in number judgments (Franconeri, Bemis, & Alvarez 2009). Here we investigate the presence of subitizing during the enumeration of object features that hold unique points in space, yet are not spatially segmented from each other. In a series of behavioral experiments, participants were asked to enumerate control displays containing 1-8 spatially distinct shapes, as well as test displays of 1-8 spatially connected protrusions. We looked for evidence of subitizing in both conditions based on statistically greater slopes for larger numerosities than those in the subitizing range. We observed subitizing for both test and control conditions, indicating that subitizing can occur across features such as protrusions from a single object.
Meeting abstract presented at VSS 2013