A Drop in Performance Can be a Sign of More Advanced Thinking
School of Psychology
Center for Academic Studies
Branco Weiss Professor of Research in Child Development and Education (Emeritus)
School of Education
Tel Aviv University
We all know that children get better at solving problems as they get older. Learning is always upwards and onwards. Children get better in their understanding over time. For example, children age 6 can solve all the problems they were able to solve at age four, and then some. This commonplace understanding of learning on the part of educators, parents, etc. is confirmed in our everyday observations.
But there is a surprise here. A line of research I began in the 1980’s, and which continues to this very day, shows that what we take for granted is not always the case. Studies of cognitive development indicate that, for some tasks, children have what is called U-shaped behavioral growth. What this means is that younger children solve a task correctly, older children solve the same task incorrectly and still older children solve it correctly.
Here’s an example. Let’s say we have three cups, two of which have the same amount of water at the same temperature and one of which is empty. We tell the children that the water in the two cups is cold and that they are equally cold. We then pour the water from those two cups into the third, empty cup and proceed to ask the children what the temperature in now. Children age around 4 say, correctly, that it’s the same temperature because all we did was mix same temperature water. Older children around age 6, say that the mixed water is twice as cold as the original water because there is now twice the amount of water. And children around age 8 return to the correct answer that it is the same temperature as the original water because even though there is the more water, that doesn’t mean the water is colder. It’s just more cold water at the same temperature.
Lest the reader think this is an isolated phenomenon that is found only for temperature this surprising finding has been found for tasks that tap children’s understandings of other physics concepts, such as viscosity, sweetness of water, density and pressure. And U-shaped behavioral growth has been found in other domains, as well, such as language learning, the use of metaphors and more.
So how does this happen? How is it that our commonplace understanding of always getting better has sometimes been shown not to be the case? How is it that children are getting worse in problem solving over time?
One answer to these questions is that children actually do improve their underlying thinking over time, but sometimes an advance in what gives rise to answers leads to a drop in their performance in problem-solving. For example, to return to our case of temperature, the youngest children do not pay attention to the amount of water; the older children do pay attention to the amount of water but erroneously think that more of one thing (amount of water) increases another thing (temperature); and the oldest children also pay attention to the amount of water but they don’t think that it affects the temperature.
Notice that not paying attention to the amount of water (that leads to a correct answer) is less advanced than attending to the amount of water (that leads to an incorrect answer). What that means is that in tasks such as this, as our thinking advances, there is a drop in performance.
Normally, were we to see a child solving a task correctly and then after a while she solves it incorrectly, we might get worried. But the way I showed how this drop works, we would understand that that drop in performance is a sign of cognitive advance.
What this implies is that, when teaching, we should pay attention to children’s reasoning about a problem more than if their answer to that problem is correct or not.
Strauss, S. (with R. Stavy). (Eds.). (1982). U-shaped behavioral growth. New York: Academic Press.
Sydney Straus is a member of the OLPC Learning Board.