Abstract
Animate and inanimate objects differ in their intermediate visual features. For instance, animate objects tend to be more curvilinear compared to inanimate objects (Levin et al. 2001; Perrinet and Bednar, 2015; Long et al. 2016). Recently, it has been demonstrated that these quantitative differences in the intermediate visual features of animate and inanimate objects are sufficient for categorization: human participants can classify synthesized images of animate and inanimate objects that differ largely in the amount of these visual features significantly above chance (Long et al. 2016). A remaining question, however, is whether this observed categorization is a consequence of simple, top-down cognitive strategies (e.g. rectangular shapes are less likely to be animals) or a consequence of bottom-up processing of their intermediate visual features, per se, in the absence of top-down cognitive strategies. To discriminate between these two alternatives, we repeated the classification portion of Long et al. (2016) but matched the synthesized stimuli, on average, in the amount of image-based and perceived curvilinear and rectilinear information. Additionally, in our synthesized stimuli, global shape information was not preserved and the images appeared as texture patterns. These changes prevented participants from using top-down cognitive strategies to perform the task. We found that participants (n=20) still classified these synthesized stimuli significantly above chance, even though they were unaware of their classification performance. For both object categories, participants used the curvilinear, but not the rectilinear image-based information present in the stimuli for classification. The perceived measures of curvilinearity and rectilinearity did not predict classification performance. Surprisingly, the stimuli most consistently classified as animate corresponded to the most dangerous animals in our sample of images. We conclude that bottom-up processing of the curvilinear features present in the visual input conveys information associated with the valance/arousal of the stimuli and is sufficient for animate/inanimate object categorization.
Meeting abstract presented at VSS 2018