According to Piaget, the newborn baby relies on innate reflexes and has limited knowledge. In the sensorimotor stage, the child learns through movements and sensations. Knowledge simply arises from looking, touching, hearing, sucking, grasping and putting things in the mouth. The child gradually comes to have an idea of what different objects are like. To begin with, the child’s movements are not at all purposeful because he or she has no control over them. However, when children acquire more control over their bodies, they learn that a specific movement - like bashing a toy or pulling the tail of the cat - will have specific consequences.

At around eight months old, children develop object permanence - that is, the idea that objects continue to exist even when they can no longer be seen. Piaget found that a 4-month-old will not look for a toy if it is hidden under a cloth. It is as if the rule “out of sight, out of mind” applies at this stage. However, an 8-month-old will lift the cloth, knowing that she will find the toy underneath. Between 8 and 12 months, the child will keep looking for the object in the place where he or she found it the last time. This happens even if the child watches someone hide it in a new place. According to Piaget, it is not until they are around 18 - 24 months old that toddlers possess fully developed object permanence.

One of the challenges to Piaget's theory is that the stages may not be as rigid as he believed.  Researchers like Baillergeon wanted to see if children actually developed object permanence earlier than Piaget claimed.

Baillargeon (1986) wanted to test Piaget's claim that children do not demonstrate object permanence until around 6 - 8 months. To do this, she used her impossible situation test.

The sample was made up of 40 healthy, full-term infants. 20 were 5.5 to 6-months-old and 20 were 7 to 8-months-old.  The sample was obtained by reading birth announcements in the local paper and then contacting the parents by letters and follow-up phone calls.

The infants sat in front of a small screen; to the left of this screen was a long, inclined ramp. Before the experiment began, the infants watched as the screen was raised to show them that there was nothing behind it - and then lowered again. A toy car then rolled down the ramp, passing behind the screen and exiting on the other side. The infants watched this twice.

The infants were then shown one of two conditions. 

Condition 1: The situation was the same, but when the screen was raised, a box had been placed behind the screen but not on the ramp. This was the "possible event."

Condition 2: In this condition, the box stood on top of the ramp, blocking the car's path.  This was the "impossible event."

The infants looked longer at the impossible event than the possible event, suggesting that they were surprised to see the car emerge on the other side when the box was clearly blocking the car's path. This indicates that the infants realized that the box and car continued to exist behind the screen and understood that the car could not pass through the box.

This means that the infants understood that:

• the box continued to exist, in its same location, after it was hidden by the screen
• the car continued to exist and followed a certain path
• the car should have been blocked by the box when it was sitting on the ramp.

• The study is well controlled and has high internal validity.
• The study made used a cross-sectional design.  There was no evidence of change over time.  However, this was not necessary for this experiment.
• The study was a repeated measures design; each infant was exposed to both conditions.  This controls for individual difference/participant variability.
• The study of infants is difficult because they are not able to communicate verbally with us what they are actually thinking.  Baillargeon used the length of a child's gaze to measure their level of surprise and understanding of the actual situation. There is some debate as to whether this is a valid measure of their understanding of object permanence.
• Baillargeon and her team have done several versions of this study which have yielded the same results - in some cases, showing children as young as 4.5 months showing surprise at the impossible situation test. This means that the results of the study have been shown to be reliable.