In our daily life, we use our senses to acquire information about the objects that surrounds us. However, the information processing that allows the recognition and consecutive classification into categories of those objects remains unclear. In this study, we analyzed quantitatively and tested experimentally the visual properties of bottles silhouettes responsible for categorization in soft drinks. First, we took pictures of all personal bottles of soft drinks available in the local market. Then, we extracted the silhouette and analyzed its physical characteristics using a cluster analysis. The silhouette image analysis revealed the physical characteristics that separate the categories according to the real market. To evaluate if these parameters correlate with the consumer´s perceived categories, we conducted a psychophysical experiment where subjects chose the category of a randomly presented bottle silhouette image. The experimental results support the hypothesis that the width of the lid and the shape of the bottle are important visual factors for categorization. Furthermore, we present metrics for assessing the categorization which enable direct correlation with physical attributes of the bottles silhouettes. These findings permit implementation of guidelines for new bottle designs that capitalize on existing categorization rules based on consumer visual perception.

Arboleda, A. M., Arce-Lopera, C. A., Okajima, K. (2014). In Pion (Ed.), Model for the categorization of bottled soft drinks using their silhouette (vol. 37, pp. 108). Belgrade: European Conference on Visual Perception. http://pec.sagepub.com/content/43/1_suppl.toc

Check out the Conference Proceedings