Aesthetic Evaluation of Computer Icons: Visual Pattern Differences Between Art-Trained and Lay Raters of Icons

Adopting eye-tracking measures, we explored the influence of art experience on the aesthetic evaluation of computer icons. Participants were 27 college students with art training and 27 laypersons. Both groups rated icons of varying complexity and symmetry for “beauty” while we recorded participants’ eye movements. Results showed that art-trained participants viewed the icons with more eye fixations and had shorter scanning paths than participants in the non-art group, suggesting that art-trained participants processed the icons more deliberately. In addition, we observed an interaction effect between art experience and symmetry. For asymmetrical icons, art-trained participants’ ratings tended to be higher than those of lay persons; for symmetric icons, there was no such rater difference. The different visual patterns associated with aesthetic evaluations by these two participant groups suggest that art experience plays a pivotal role in the aesthetic appreciation of icons and has important implications for icon design strategy.

The Role of Motion in Object Recognition

Do motion cues influence object recognition when contour information is available? Three experiments examined four motion conditions for a variety of objects (no motion, random motion, atypical motion, and typical motion) when contour information was also available. A typical motion pattern was one that would normally be associated with the moving object, whereas atypical motion involved a regular motion pattern that was typical for one object in the set of 15 used in the experiments, but wasn't associated with the object in motion. In Experiments 1 and 2, the objects were made difficult to recognize, by eliminating vertices and by using small representations, respectively. In Experiment 3, large, complete contour wire frame pictures were used. In all experiments, recognition speed and accuracy were best for the typical motion condition and second best for the atypical motion. With easily-recognized objects, random motion led to faster recognition than no motion, whereas, with difficult recognition, random motion led to slower response times than no motion. The results are interpreted with a three-process model. Applications to the design of computer icons, signage, and camouflage are discussed