In my previous post (here) I tried to explain how bar-model supports learning using dual-coding. In this post, I want to use Cognitive Load Theory to explain that bar-models reduce cognitive load. (I should point out that as of now, I have no research evidence to show that using bar-model leads to improved long-term learning and improved problem solving – but I’m working on it).
This diagram represents the three elements of cognitive load (I’m referring to the book Efficiency in Learning, Clark, Nguyen and Sweller – 2006).
Intrinsic load is the difficulty imposed by the task. In a multi-step science question, the intrinsic load is high – the learner has a lot to hold in her head at the same time. The teacher can reduce intrinsic load, for example by breaking a complex task into smaller steps.
In this post, I argue that algebra, the gateway to understanding the universe, increases intrinsic load when a learner hasn’t mastered it. If a learner has to concentrate on interpreting the algebra, the intrinsic load is increased. She will struggle to learn because the overall cognitive load is too high. I believe this is a core barrier to learner success in science.
Using a bar-model reduces the intrinsic load because it replaces un-mastered algebra with a visual representation.
For example, you can reduce the intrinsic cognitive load of a latent heat/specific heat capacity question by representing the overall problem as a bar-model (thank you @ruthyie ).
Instead of using working memory to interpret the algebra:
ΔQ = mcΔT + LΔm + mcΔT + LΔm + mcΔT
the learner is provided with a visual representation of the algebra, with a reduced intrinsic cognitive load.
After Easter I will be asking for volunteers to carry out a bar-model experiment with their classes to test for long term retention. If you would like to take part, please contact me.