WASHINGTON,
D.C. — Something extraordinary happens between 18
and 24 months of age. In that brief span, toddlers take giant strides in their
ability to visually inspect, recognize and manipulate various items. This
achievement puts toddlers on par with adults given comparable tasks.
This developmental leap has been tracked for the first time
by researchers at Indiana University in Bloomington
who have created a special head-mounted camera that toddlers wear as they play
with toys.
The device allowed the team to measure expansion of object
recognition among toddlers. This development feeds into rapid advances made during
the same time period in the ability to learn objects’ names and to engage in pretend
play, a basic form of symbolic thinking, psychologist Linda Smith said while
describing the work July 24 in Washington,
D.C., during the annual meeting
of the Cognitive Science Society.
“Something very important happens in visual object
recognition during this six-month period that then affects many other
developing cognitive systems,” Smith asserted.
These latest findings support the view that general mental
capacities, such as perceiving and exploring objects, give rise to specific
skills, such as recognizing objects’ names, remarked psychologist Jeffrey Elman
of the University of California, San
Diego.
An influential opposing view holds that higher-order cognition,
such as language and symbolic thought, stands apart from visual perception and
other basic mental functions.
In order to track what children look at as they explore
objects, Smith and her colleagues constructed a comfortable, head-mounted
apparatus for toddlers. Each child wears a headband with a tiny, adjustable
camera placed between and just above the eyes. A cord connected to the headband
transmits video signals to a computer for a thorough visual analysis.
One head-cam study, led by Indiana’s Alfredo Pereira, found that, from
15 to 30 months of age, children begin to prefer viewing objects from the same
angles that adults do. This visual orientation, known as the planar view,
focuses on an object’s top, bottom or side.
Making natural head movements while looking at real-world
objects, along with viewing objects from a planar perspective, yields copious
information about those objects, Smith said.
Visual orientations that show more than one side of an
object, such as a three-quarter view, prove especially useful in assessing
static items, such as those in photographs, she noted.
The researchers gave eight objects, one at a time, to each
of 30 toddlers to play with as they sat on a chair near a parent. Half of the
objects were familiar, such as an airplane and a cup. The rest were items that
preserved the basic structure of airplanes, cups and other familiar items but
lacked identifying details.
Head-cam data indicated that, from 15 to 30 months of age,
children increasingly favored planar views of both familiar and novel objects. This
trend especially characterized toddlers who had vocabularies of at least 100 nouns.
Planar views may offer a prime look at buttons to be pushed,
handles to be grabbed and other clues to objects’ functions, Smith suggested. That
knowledge then fuels learning of objects’ names, in her view.
A second study, led by IU Bloomington's Sandra Street, examined the ability of 42
toddlers to insert three differently shaped objects into appropriately shaped openings.
Participants were split between 18-month- and 24-month-olds. This ability has
often been used to test for developmental delay in toddlers, but little is
known about how the ability develops over time.
The younger kids “failed miserably,” Smith said. Older
children immediately succeeded on this task, performing at an adult level.
Only the 24-month-olds looked at objects with a consistently
planar view.
Intriguingly, pretend play — say, using a laundry basket as
if it’s a car — also emerges between 18 and 24 months of age. Upon learning to
recognize objects having abstract shapes, toddlers may be prepared for symbolic
forms of play, Smith hypothesized.
Found in: Behavior and Humans
Please login or register to participate.