Babies get plenty of practice hearing, touching and even coordinating physical motions while still in the womb, but they don’t get to try out their new eyes until the moment they are thrust into the world. That may be, in part, why they are initially worse at making sense of visual information than they are at other aspects of terrestrial life, says Scott Johnson, PhD, a psychology professor at New York University.

“Newborns are not blind, but they don’t see very well,” says Johnson, who studies brain and visual development. “Their distance perception is not very good; their color perception is not terrific.”

Upon that slight foundation, most infants build sophisticated visual processing systems, notes Paul Quinn, PhD, a psychology professor at the University of Delaware. Their color perception sharpens, their nearsightedness fades and, perhaps most impressively, they begin to see the world in terms of distinct objects.

In fact, new research by Quinn and Ramesh Bhatt, PhD, a University of Kentucky psychology professor, shows that infants just three months old can easily group objects by lightness or darkness—a skill that allows them to, for instance, see that a black filing cabinet is a separate object from a white wall. And by six months, babies are able to group objects by shape, according to the study, published in the October issue of the Journal of Experimental Psychology: Human Perception and Performance (Vol. 32, No. 5, pages 1221–1230).

“We come into the world ready to use some kinds of information, but other kinds of information that may be equally important...appear to require specific and longer experience,” Quinn notes.

Gestalt baby steps

The ability to group objects by shape is what allows us to see a building instead of a pile of bricks. It’s how we are able to separate out two trees when their differently shaped leaves overlap. The ability is so fundamental, many believed it to be inborn. But Quinn and Bhatt’s study suggests otherwise.

To get at how infants see the world, the researchers took advantage of the fact that babies will generally look longer at novel pictures than familiar ones. The researchers showed the infants solid vertical or horizontal bars (see figure, this page) and let the infants gaze at them for six 15-second study periods. They then showed the infants objects that had been organized into columns or rows. If the infants gazed longer at the objects arrayed in a different way than the bars, it showed they had grouped the items correctly and were appreciating the novelty of the contrast.

The researchers showed 128 three- to four-month-old and six- to seven-month-old infants lines of Xs and Os, or filled and unfilled shapes. The three-to four-month-old infants were able to organize the filled and unfilled shapes, a test of their ability to group by lightness or darkness. However, they were not able to line up the Xs and Os, a test of their ability to group by shape. The six- to seven-month-old infants, however, performed well on both tests.

The results suggest that babies are either born with—or develop early on—the ability to group objects by brightness, but grouping by shape similarity requires more extensive experience.

“The beauty of the paper is that they have been able to show that not all grouping principles are created equal for infants,” Johnson notes.

“Shape organization requires a little more experience. It’s not as automatic as the luminance system,” Bhatt adds.

Streams of development

Some adults, however, lose the ability to group objects by shape due to brain injury, says Marlene Behrmann, PhD, a psychology professor at the University of Toronto. Research by Behrmann on such adults—known as visual agnostics—has shown results that dovetail with those of Quinn and Bhatt. While some people with visual agnosia can’t group objects by shape, they are able to group them by luminance, according to a 2003 paper by Behrmann and her colleague, Rutie Kimchi, PhD, a psychology professor at the University of Haifa, published in the Journal of Experimental Psychology: Human Perception and Performance (Vol. 29, No. 1, pages 19–42).

“This is all in contrast with existing ideas that grouping principles and perceptual organization are monolithic processes,” Behrmann notes.

Behrmann and Kimchi gave two visually agnostic patients a series of tests of object recognition and shape grouping. They found that, like the three-month-old infants, the adults were unable to group objects with similar shapes. What’s more, the participants could not identify line drawings of familiar objects. One thought a harmonica was a cash register. The other thought it was a keyboard.

The deficits are probably linked, Behrmann says. The ability to see that a row of rectangles makes up the air holes in a harmonica rather than the buttons of a cash register requires that you see them as a group.

“They are a little bit like infants in that they have the individual elements of the picture...but they can’t put them together and understand what it is.”

Both studies are part of a surge of interest in how humans develop the sense we rely upon the most, Johnson notes. Such research could lead to insight into why, for instance, infants that later develop autism don’t cue into faces. It’s possible that they are not seeing them as cohesive wholes, which could set such children back in learning a whole host of social cues, he notes.

“Unless people are studying babies and how the normal visual system works, we aren’t going to get at what happens when it goes awry,” he says.