Purchase this article with an account.
Audrey Michal, Stacey Parrot, Steven Franconeri; Two modes for seeing relations between objects. Journal of Vision 2016;16(12):1272. doi: 10.1167/16.12.1272.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Past work has revealed two distinct types of spatial relation representations: categorical (e.g., 'X left of Y') and coordinate (e.g., 'X is 1 cm away from Y'; Kosslyn et al., 1989). Does a similar dissociation exist for relations based on features other than spatial location, such as size? By measuring eye movements, we show that people adopt distinct attentional modes when representing different types of magnitude relations in the same display (two bars). In Experiment 1, when asked to judge a directional relation ('Are the bars arranged short-tall or tall-short?'), participants systematically isolated one bar with their attention. In contrast, when asked to judge a nondirectional relation ('Are the bars the same or different heights?'), participants systematically shifted their gaze toward the right. A rightward shift would place the stimuli in the left visual field, allowing the bars to be processed more holistically by the right hemisphere. In a second experiment, we encouraged one group of participants (holistic group) to extract relations by imagining a line connecting the tops of the two bars (directional: 'Is the line sloped positively or negatively?'; nondirectional: 'Is the line flat or sloped?'). As a control, we included an individuated group of participants who used the same perceptual framing as Experiment 1. The holistic framing changed participants' eye movements such that directional relations were extracted more similarly to nondirectional relations than directional relations in the individuated group. Together, these results reveal two distinct modes for extracting magnitude relations (locally versus holistically), and that these modes map onto how categorical and coordinate spatial relations are processed, respectively. They also show that people can be induced to extract relations locally or holistically for the same display, either by manipulating the type of judgment (directional/nondirectional) or the perceptual framing (individuated/holistic).
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only