Cognitive Daily

When human infants are born, the physical structure of their brains has not fully developed: the human brain continues to grow for more than two years after birth. It’s very clear that newborn infants don’t have the same cognitive abilities as babies even 6 months older. For example, they can’t move their heads follow along a movement with their eyes. They appear to have very little control over their limbs. Three-month-olds have difficulty reaching and grasping objects. For humans, walking or crawling isn’t a remote possibility for more than six months. Contrast this to wildebeasts, for example, which can run within 15 minutes of birth, or whales, which can swim and follow their mothers virtually as soon as they are born.

So what is it that these animals have that humans don’t have? Research with primates has shown that the same region of the brain is activated when they watch another animal performing an action as when they perform the action themselves. So perhaps what babies lack is a recognition of the relationship between their own actions and the results of those actions: they lack goals.

If you’ve ever watched a very small infant flailing about, this certainly seems to be the case: few of their actions appear to be directed toward any specific end. So how exactly do they learn to do anything? A simple experiment by Jessica Sommerville, Amanda Woodward, and Amy Needham focused on one aspect of goal-oriented behavior: reaching and grasping.

Three month-olds aren’t very good at grabbing things. If their hand happens to clench around an object, such as one of Greta’s earrings, they can give it a good grab and pull, but typically it’s very difficult for babies this age to intentionally pick up and grab an object. Sommers’ team got around that limitation by making little Velcro mittens for them. They let 15 three-month-olds play with a little cloth ball and teddy bear for a while without the mittens and observed the typical flailing behavior for babies this age. Then they put the mittens on and watched what happened next. Since the mittens stuck to the toys, grasping wasn’t necessary, and babies spent significantly more time coordinating looking and reaching for the toys when they had their mittens compared to without the mittens. A second group of 15 babies skipped this part of the experiment.

For the second part of the experiment, Babies watched as a gloved adult “played” with bigger versions of the same toys in a systematic way. The adult reached for and grasped one of the toys repeatedly until the baby became bored watching them. Next, the position of the two toys was reversed, and six test trials were completed: in three trials, the adult reached for the same toy, which was now in a different position. In the other trials, the adult reached for the other toy, repeating the identical motion as the baby had seen initially, but reaching for a different toy. So one group got to play with the toys first and then watched the adult, but a second group of babies just watched the adult. A trained observer, who didn’t know which group each baby was in, monitored how long the babies looked at each test trial to see which motion appeared to be more interesting to the babies. Here are the results:


When babies hadn’t played with the toys first, they looked for the same amount of time at both test trials, making no distinction between watching a new motion or watching the same motion to reach for a new toy. But when babies had played with the toys, they watched the adult reaching for a new toy for a significantly longer period of time.

Sommerville’s team argues this suggests that the experience playing with the toys had informed the babies’ perception of the adult’s grasping motion. It appears that these babies were surprised to see the adult reaching for a different toy, even though it appeared in the same place as the old toy. The results suggest that these babies understand that the adult is reaching with the goal of grasping a particular toy.

Armed with this new knowledge, babies can begin to learn how to deliberately grasp objects: the limiting factor appears to be grasping ability, not the ability to reach or develop the goal to get an object. Perhaps more interesting, babies’ understanding of the goals of others appears to be directly linked to their own experience in forming similar goals. Once you’ve learned how to attain a goal yourself, you also understand that others may have the same goals.

Sommerville, J.A., Woodward, A.L., & Needham, A. (2005). Action experience alters 3-month-old infants’ perception of others’ actions. Cognition, 96, B1-B11.


  1. #1 Sameer Singh
    March 2, 2006

    I think this “lacking of goals” is a direct result of evolution, one of the reasons why humans survived and rule (sort of). The whole debate about how much is “programmed” into the baby’s brain versus how much it learns is exactly that. The more the emphasis on learning from experience, the more flexible the baby is going to be.

    So if a baby whale is put in a different scenario at birth, it will still try and swim the way it has been drilled into the brain, but the human baby will learn to adapt to the new environment. This has a downfall of course, the human baby takes longer to develop and save itself.

    The human babies, therefore, might have reached the balance between flexibiliy of adapting to new environments, and speed of development.

  2. #2 Gordon Worley
    March 3, 2006

    As I understand it, babies are thought to lack many capabilities because they must be born before their development proceeds any further due to the position of the birth canal. If we didn’t walk upright with our hips rotated we would likely give birth to toddlers rather than infants.

    Of course, since human babies go from infant to toddler outside the mother rather than insider her, humans may be capable of some very interesting development during this period that wouldn’t be possible for other animals.

New comments have been disabled.