These two pictures represent the eye motions of two viewers as they scan a work of art with the goal of remembering it later. One of them is a trained artist, and the other is a trained psychologist. Can you tell which is which?
How about for this picture?
Art teachers have noted that when beginning students attempt to draw accurate portraits, they tend to exaggerate the size of key features: eyes and mouths are too big relative to the size of the head. Trained artists learn to ignore these temptations and draw the world as it really appears. Even world-famous artists such as Leonardo da Vinci have had to resort to tricks such as looking at their subject through a divided pane of glass in order to render proportions accurately. As you can see from the two examples above, even when looking at a picture, artists look differently. So which is which? I'll let you know at the end of the post.
Stine Vogt and Svein Magnussen showed 16 pictures including these two to trained artists and non-artists (psychologists) enrolled in Norway's top graduate programs in their respective disciplines, using eye-tracking cameras and software to monitor where they looked. The viewers were unaware of the purpose of the test -- they were told the study was about pupil size and response to pictures. In the first phase of the experiment, viewers simply looked at each picture in random order, and in the second phase, they were asked to view the pictures again, but to concentrate in order to remember them.
Vogt and Magnussen defined key areas of each picture - small regions around focal objects such as human bodies or faces. This graph shows how often artists and non-artists looked at these areas:
In both cases, non-artists spent significantly more time looking at these key areas than artists. Interestingly, even artists spent more time looking at these areas when trying to remember the pictures.
So who was better at remembering pictures? At the end of the study, participants gave unprompted verbal descriptions of as many of the pictures as they could, in any order. Overall, the artists remembered more details from the pictures, but surprisingly, non-artists actually remembered more of one type of pictures: abstract pictures with no recognizable objects. Artists were able to change their viewing strategy to remember the non-abstract pictures, but with no recognizable objects in abstract pictures, there was no change in strategy.
So why do artists look at pictures - especially non-abstract pictures - differently from non-artists? Vogt and Magnussen argue that it comes down to training: artists have learned to identify the real details of a picture, not just the ones that are immediately most salient to the perceptual system, which is naturally disposed to focusing on objects and faces. With this in mind, there's little doubt which pictures above show the artist's eye movements -- they are the ones to the right, which sweep across the whole picture, not just the human face and figure.
Vogt, S. & Magnussen, S. (2007). Expertise in pictorial perception: Eye-movement patterns and visual memory in artists and laymen. Perception, 36, 91-100.