THE contemporary American artist Chuck Close has helped to disprove ideas dating back to Aristotle about how we perceive shapes, according to a study by a professor armed only with a tape-measure.|
Up close, the artist's "block portraits" appear to be an array of flat,
painted squares but as the viewer walks back from the canvas, a human face emerges.
Until now, scientists have adhered to the same-distance regime, first discussed by Aristotle in the fourth century. This states that we perceive things the same way at all sizes - so that you would recognise your friend for example, whether he was in the distance or looming large.
But Close's portraits, which can be up to 8ft tall, shows that a shape's size does affect how we see it, it is claimed in today's issue of Science. His "duality effect" - solid from afar and flat from near - was investigated last year by a study of the Chuck Close Retrospective at the Museum of Modern Art in New York, an exhibition which is now at the Hayward Gallery, London.
Using only a tape-measure, Prof Denis Pelli, a vision scientist at New York University, showed that the nose of a Close portrait can be made out at relatively short distances from the canvas (typically less than 18ft), when the individual squares are still in sharp focus for a person with normal vision.
Prof Pelli said: "It is commonly assumed that scientific research today can only be done in laboratories with complex equipment. But the essence of science is careful observation of the world around us. And in this sense, the best science and the best art share a common methodology."
He has discussed his work with Close, who pays meticulous attention to the size of the marks in his paintings. But Prof Pelli said: "He never tells people how to interpret his work. He has refrained from making comments on whether this theory is right or not."
In his study, Prof Pelli asked five observers to do a "nose test" on 33 Close paintings. The observer moved forward and backward to find the point where the nose emerges from each canvas, and these "critical distances" were recorded along with the size of the blocks making up the painting.
Prof Pelli then calculated a "critical angle" between the line of vision at the critical distance and the edges of the squares. To his surprise, it was about 0.3 degrees for all the paintings, despite their variety in face size, block size and number of blocks per face.
Prof Pelli's results also disprove two theories proposed in 1973 for why the face is obscured by the blocks when seen from near and revealed when seen from far.
The first proposes that a certain number of blocks per face are required before we can see the face. The second assumes that backing away reveals the face because our eye blurs the blocks together. But this would demand a critical angle of about 0.1 deg only.
Prof Pelli says Close deserves credit for these findings. "One might suppose that he was a naive artist, obsessed by grids, who innocently produced the coarsely gridded paintings that we use here to reveal the size dependence of shape perception. In fact, Close has devoted his career to studying just that."
Pointing out that Close increased the block sizes by 15 per cent each year, Prof Pelli said: "He was more thorough than his scientific colleagues."
28 July 1999: [Features] Maps of the human face
10 July 1999: [Features] Head and shoulders above the rest