12 Video Views·Aug 17, 2025
Since the 1960s, developmental psychologists point to the "Visual Cliff"—an experiment that plops babies on a fake precipice—as proof that infants learn to fear heights as they learn to crawl. Yet, over the past 25 years, a series of rigorous (and adorable) experiments by Karen Adolph of NYU's Infant Action Lab has shattered this myth, revealing that while babies can learn from experiences near high ledges or narrow bridges, it's not a phobia they acquire.
Produced by Luke Groskin
Starring Derek Hough, Tessa Rose Confessore, and Clarabelle Kaufman
Music by Audio Network Footage
Stills and Additional Footage provided by Karen Adolph and the NYU Infant Action Lab,
Eleanor Gibson and R.D. Walk
KTCA Twin Cities Public Television
Ira Flatow
Glacier National Park (C.C. 3.0)
Everyone has had some kind of an experience feeling something kind of like fear of heights
and it makes perfect common sense that we shouldn't jump over the edge of a drop-off
because it's dangerous and if it's dangerous then we should probably be afraid of it.
Probably. Exactly how and when this hypothetical fear develops makes for pretty gleeful research
inside Dr. Karin Adolf's lab.
We're discovering some surprising things that are flying in the face of you know how people
have studied locomotion for the last 100 years. Including what babies do and feel
when they're on a ledge. So the story about fear of heights is that there was a very very famous
classical study in developmental psychology by Gibson and Locke in 1960 that showed that human
infants and infants of a number of different species will avoid crossing over an apparent
drop-off from what they called a visual eclipse. As soon as the baby can move about he can be
observed in the same way as other animals. It's a glass table and on one side there's a checkerboard
patterned surface right under the glass. The other side of the table the patterned surface is way
down on the floor so visually it looks like a big three-foot drop-off. So when human babies first
begin crawling most of them will crawl over the apparent drop-off but after several weeks of
crawling experience then they'll begin to refuse to crawl over the drop-off. And the conventional
interpretation of this was that infants are avoiding the drop-off because they're afraid of
height. She's backing away. She's not interested. She wants no part of that side. The experiment
became the shaky foundation for the myth that crawling innately teaches us to fear heights.
Babies avoid going over the drop-off because they're afraid of heights. How do we know they're
afraid of heights? Because they avoid crawling over the drop-off. So as you may have noticed the
argument is circular. As it happened the visual cliff was a better test of an infant's depth
perception not their fear of heights. And if you look closely you may have actually noticed that
the babies don't look particularly scared. It is a perfectly safe surface of support so after one
trial human infants cleverly figure this out and then they will crawl over the drop-off. So how do
you really test if a baby is scared of heights? You can test babies at the edge of a real cliff
where there's no safety glass it's just an actual drop-off and the drop-off can be adjustable so it
can be little or really really big. And why stop there? Inside Dr. Adolph's infant action lab
babies are tested on a variety of apparatuses. You can test babies at the edge of a slope so it can
be steeper steeper steeper crossing bridges that are wide or narrow over a surface of support and
so on. If there's no safety glass you can test the same baby dozens on dozens and dozens of trials
in the same session and that means that you can determine for each individual infant when they
think that that surface is safe and when they think it's risky within a couple of degrees of
accuracy like two degrees of accuracy for a slope. Dr. Adolph has spent 25 years running
tests like these and the resulting data sets are both convincing and adorable. Infants spend most
of each trial right at the edge of the precipice. They're reaching their arm down over the drop-off
they are putting their little feet or their little hands on the bridge and taking little tiny steps
at the edge and all the while they're doing it with either neutral or positive facial expressions.
What's more surprising is how their approach changes as they learn different forms of
locomotion. So what happens is that when babies first begin crawling they'll crawl right over the
edge then over weeks of crawling their responses get more and more accurate until finally they're
at adult-like levels so that they can tell precisely whether a slope is safe for crawling
or whether a drop-off is safe for crawling. But then the next week when these babies stand up and
now as a new walker they walk right over the edge. So in fact there seems to be four learning
curves in development as babies learn to sit learn to crawl as they learn to cruise and then finally
as they learn to walk. So if they were afraid of heights you should just be afraid of heights
in the rain and on a train and in a tree you shouldn't see four separate learning curves.
So if babies aren't learning to be afraid of heights what are they learning? I think that
babies are learning over weeks of sitting crawling cruising walking is they're learning
to perceive the relations between their bodies and the environment. They're discovering what's
possible and what's just a little too far. And you can learn that without feeling any sort of fear at
all. In fact Eleanor Gibson the person who first invented the visual cliff wrote later in her life
as a goat is peering over the edge of a steep crack it knows not to walk off the edge and she
said I don't think it's feeling any emotions at all it just knows not to go and that's what we're
seeing in babies who probably know not to go. Probably. For Science Friday, I'm Luke Groskin.