Advocates of innate perception have based their conclusions on previous research, which typically measured perceptual abilities of four-month-olds and older infants. However, Scott Johnson, a professor of psychology and neural science who conducted the study, compared these abilities in both two- and four-month-olds, finding distinctions in the perceptual skills of the two groups.
"These results are only a part of the larger literature on perception, but this study does provide a very important piece of the puzzle," said Johnson. "It is now clear that theories of innate knowledge do not hold up under scrutiny. Instead, the developing visual system seems to build object representations from smaller, visible components, such as the visible portions of a partly occluded object. Isolating how and why this occurs should be the focal point of subsequent scholarship."
The question of how humans develop knowledge of the perceptual appearance of objects--such as the realization that a view of an object that is partly obscured does not match its true extent--is an ongoing puzzle in cognitive science. Previously, scientists believed that infants learned this concept through manual exploration in order to gauge the actual physical size of the objects they see. But subsequent research had indicated that infants seem to know about object occlusion even before they can reach, leading to the theory that object knowledge is something that infants are simply born with. These latest findings provide a new view, implying that perceptual development in infants emerges from a combination of experience and brain development.
"How and when we develop knowledge of the mechanics of how the world works are fundamental questions in psychology research," Johnson explained. "Previous studies showing perception of occlusion in four-month-olds concluded that these characteristics must be innate because of the young age of the research subjects. However, these abilities hadn't been explored in younger babies."
Johnson's study explored how infants process two parts of a visual environment--specifically, when they see two elements moving, can they determine if the elements are part of a singular object. Theories of innate object knowledge have long held that if infants can detect "correlated" motion of two or more elements in a visual scene, then the elements are automatically thought to be part of the same object. Johnson tested whether this is actually true by measuring both motion perception and unity perception in the same display, consisting of a partly hidden rod whose visible parts protruded from behind a box. He reported that even the two-month-olds reliably detected the rod parts' motion, but nevertheless could not join the visibly moving parts into a coherent object except under limited circumstances.
Johnson measured the infants' ability to make these distinctions by showing them a series of computer-generated displays. Those who recognized each display as different from the previous ones--and responded by viewing it for sustained amounts of time--were those who comprehended the characteristics of the moving rod. Johnson did not find differences associated with the gender of the infants in his research.
"Another implication of these findings is that infants do not necessarily benefit in any meaningful way from stimulating toys or exercises," Johnson added. "Babies learn these concepts quickly through visual observation rather than enrichment, direct instruction, or manual object manipulation. Parents need not feel that they can improve their infants' IQ or brainpower through special procedures or materials. Social interaction, however, is a different story--infants benefit greatly from one-on-one time with other people."
He added that his and other studies may serve as a precursor to research that can shed light on the factors contributing to developmental delays in cognitive skills.
"As we broaden our understanding of neurological development in infants, we can advance our comprehension of many developmental difficulties, such as autism, and devise early interventions," Johnson noted. "This is currently an area of intense research focus."
For more information or for a copy of the paper, please contact Johnson directly at 212-998-3794 or sj36@nyu.edu.
The study was funded by grants from the National Science Foundation and the National Institute of Child Health and Human Development.
Journal
Psychological Science