Babies are born with certain inherent survival mechanisms. One of these prevents them from crawling off of cliffs. How do you test this? Take a baby to the edge of a cliff, let it crawl around, and see if it drops off? Well, you could. A lot of people might take exception to this, though. No. What you do, is take the baby to a room with a partially glass floor, and let it crawl around. If the glass is clean enough, i.e. pretty much invisible, the baby won't crawl on it. You have to be really sneaky when dealing with babies. Another bizarre psychology experiment.

Well I must disagree with Gnomatron on this point (view all node entries for more info), I have seen this experiment done (on video) to demonstrate that babies will indeed crawl off a cliff.

The experiment is set up pretty much as gnomatron describes. Imagine a one metre by one metre by one metre wooden box (3 feet in imperial measurements), painted in a black and white checked pattern. Immediately next to this is another box with the "floor" painted in the same checked pattern, but with all other sides being made from transparent glass/perspex (capable of supporting the weight of the subject). The glass should be cleaned and lighting of the room appropriate to make the glass as invisible as possible to the subject.

The version of the experiment I saw had the subject placed on top of the checkered box, with a parent of the subject on the other side of the clear box. I.e. the subject would have to crawl across the invisible glass to reach their parent (ie, crawl off the cliff).

The parents were offering as much encouragement as possible to the subject (verbally, visually, tone of voice, offering favourite toys/trinkets etc.)

Here are the results (ages are slightly approximated due to my rusty memory).

A six month old baby will not pause at all, and will crawl straight to their parents.

An 18 month old child will hesitate, but given enough constant encouragement will crawl across the cliff to their parents.

A 3 year old child will generally not crawl across to their parents, no matter the amount of encourangement given. Most will become confused, some will cry, and some will try and find a way off the box (remember, it's 1 metre high, which is a fair distance for a small child).

My only qualm on this experiment is that the vision of extremely young children is impared until a certain age. Unfortunately, I am no expert on this, but I do wonder if this fact has any bearing on the experiment results.

Alternatively, the version of the experiment I saw may have been more an experiment regarding the level of dependancy or trust a baby has in their parents. (Put it this way, if you were standing at the edge of a cliff and God/Jesus/Allah/Buddha/David Hasselhoff told you to "WALK ACROSS TO ME", you'd probably do it.

I am hoping someone with some genuine early-childhood studies qualifications can help to clarify this.

The experiment being referred to was devised at Cornell University by Eleanor Gibson and Richard Walk (1960). The objective was to see if depth perception was an innate ability we had from birth, or something we learned as we matured.

In the test, Gibson and Walk placed infants ranging from 6-14 months on a "visual cliff" - a long glass table about 1 meter high, with an apparent drop off. Gibson and Walk then had the babies' mothers coax them to crawl out onto the glass. In the experiment, most infants refused to do so, indicating that they could perceive depth.

One explanation could be that by crawling age infants had learned how to perceive depth. However, newborn animals with very little visual experience, including young kittens, a day-old goat, and newly hatched chicks, respond similarly. (Myers, 2002) In addition, tests have shown that during the first month of life human infants turn away from objects that are on a direct collision course with them, but not from objects that will not hit them (Ball and Tronick, 1971).

Depth Perception is accomplished in the human body by a combination of binocular cues and monocular cues. It is hypothesized that biological maturation predisposes our wariness of heights, and that experience then amplifies it. As was shown in a video presentation of this experiment, the older a child was, the more likely the child would respond negatively to the experiment. In 1992, a group of researchers determined that when infants' movement is enhanced by a walker, they become even more wary of heights (Campos and others, 1992).

So, in conclusion, babies very well may crawl off cliffs. They are babies after all. Our biological predisposition will usually be enough to prevent them from doing so, but not it all cases. So just like kids on the internet, or around the kitchen, or a busy street, keep an eye on your kids if you want them to stay safe.


  • Myers, David G. (2002) Sensation and Perception Exploring Psychology Michigan: Worth Publishers (pp.164-165)
  • Gibson, E.J. and Walk, R.D. (1960, April). The "visual cliff." Scientific American, pp. 64-71
  • Ball, W. and Tronick, E. (1971) Infant Responses to impending collision: Optical and real. Science, 171, 818-820.
  • Campos, J.J., Bertenthal, B.I. and Kermoian, R. (1992). Early experience and emotional development: The emergence of wariness and heights. Psychological Science, 3, 61-64

Log in or register to write something here or to contact authors.