The snowman is at the centre of a circle of children and they are trying to work out whether putting a coat on him is a good idea.

One child says: “Don’t put the coat on the snowman - it will melt him.”

Another replies: “It will keep him cold and stop him melting.”

And finally, one child takes the middle option: “I don’t think the coat will make any difference.”

No, this is not an off-the-cuff primary science lesson in the snow - it’s a concept cartoon by Stuart Naylor and Brenda Keogh. In each of their cartoons, children explain a scientific phenomenon: one offers the scientifically accepted answer, the others are plausible but, at the current moment, scientifically false (a misconception).

In the snowman cartoon above, the children are presented with two alternative hypotheses (a difference will be made) and the null hypothesis (nothing will change). They are then encouraged to discuss their thoughts.

I used this with my Year 5 class, the majority of whom believed that putting a coat on the snowman would melt him. The discussions and reasonings for why they believed specific ideas were amazing. Quite a few stated that having a coat on warms you up, which led to a rich discussion about body temperature. It created the sort of positive atmosphere we want in a classroom, following a social constructivist environment (Jenkins, 2000).

It led me to wonder how I could develop scientific reasoning among my primary-age pupils even further. And I came to the conclusion that combining these cartoons with a well-used classroom questioning technique would be a very positive way forward.

Sock it and see

Typically, “what-if” questioning is used as a formative assessment strategy, gauging pupils’ theoretical understanding of phenomena. It can be seen quite clearly with a child’s understanding of the commutative law in maths, especially if rote learned.

For example, they may jump at the answer for 2 x 7, but then take a little while longer for 7 x 2, thinking it’s a completely different question. Another example is showing the calculation for “1,000 more than 2,100” as 1,000 + 2,100, then posing the “what if” question: “What if it were 1,000 less than 2,100?”

The “what if” question provides that extra step in conceptual thought and understanding.

It struck me that this technique should also be used to stimulate interest in scientific phenomena. We can consider the sound of a drum beat: we know that this is the drum vibrating and the air vibrating to the ear, producing the sound. But what if we remove the air and replace it with water? What would be different? What if we replace the air with a vacuum?

Doing this provides the children with an epistemological thought process through which they can explore how to test the question.

But it doesn’t have to be a testable hypothesis. When studying sound, I could pose the classic philosophical question: “If a tree falls in the woods with no one to hear it, does it still make a sound?” Although this activity requires quite a bit of guidance, it allows the children to view ontological perspectives of the notion of sound. The idea is that, with enough practice, the children will start to think like this automatically when a concept is placed in front of them, asking “What if this happened?” and “What if I do this instead of that?”

Rafetseder and Perner (2018) found that counterfactuals (possible alternatives to reality such as “what happens to falling objects without air?”) can be used to promote reasoning skills, as well as social cognitive abilities (such as social interaction and listening and taking on board another’s opinion - in this instance, scientific understanding).

So, could I plan science lessons to incorporate both concept cartoons and counterfactual questioning to promote scientific reasoning and curiosity? I gave it a go.

I started by consulting Keogh and Naylor’s concept cartoon set. In some ways, this planned the lesson for me as it would act as the pivotal hook at the start, generating conversation within the classroom to lead to the question: “How can we find out?”

This would lead to a discussion about what we could do to test the hypothesis in question, looking at independent, dependent and control variables to promote our scientific rigour. I could then provide groups of children with an A3 version of the concept cartoon and Post-it notes for them to place on to it detailing their ideas about the method and variables. And then they would investigate.

For the concept cartoon about the snowman (thermal insulation), for example, the children wrote their hypotheses for a similar experiment: the thermal insulation effects of a sock with ice inside.

Then it was time for the investigation. Ice was put in socks and children used stopwatches to measure the changes over time and measured the volume of water that the ice produced. They were very engaged, discussing different ideas, even asking about manipulating different variables, such as: “What if there were more layers? When I’m cold, I put more layers on ...”

It was paramount that effective and reflective questioning was taking place. “Have I measured this correctly?”, for example, and, “When should I start using the next variable?”

Next came the nailbiting (for me) process of collating the results: would they find the scientific accepted answer? (I’m in favour of celebrating when the children find something completely different. It’s still a learning moment.) How did they find this answer? What did they do differently? What could this mean if it was universally true?

I have found that this approach does indeed open up scientific questioning and curiosity. Most importantly, it opens their experiments up to evaluation, where they are able to critique their methodology and reflect on their hypothesis.

For the ice socks, the majority of the children found what was to be expected as per thermal insulation. However, some children found (when comparing two variables, a control and independent) that they had similar results. We took a step back and thought: what could have happened?

Some were quick to notice that they may not have placed the ice cubes in the sock correctly, resulting in human error that affected the results. This offered a valuable insight into working scientifically and measuring and controlling variables accurately, progressing their mindsets around working scientifically.

Despite the positives, working in this way did present challenges. For example, one very simple issue was my own subject knowledge. But this can be an opportunity, too. For example, when reading a concept cartoon about bungee jumping and air resistance, I had to re-evaluate my current scientific understanding - some of the plausible explanations made sense to me. This allowed me to take a step back and think how children may perceive this, too, and how we could investigate this.

Another difficulty was that I found the closed-mindedness of some students challenging; they would be adamant that there were no alternatives and no “what-if” questions could be considered. To address this, I brought it down to a concrete basis: “What if you were writing with a pen today instead of a pencil? Would your work look different or the same? Would you feel any different from this?” This provided some wiggle room for counterfactual reasoning.

Overall, the approach did have a positive impact. And I knew I had really got through when some children started to discuss other scientific events they had heard in the news. When the first image of the black hole was released, one child asked me: “What if the black hole were closer to our solar system? Is there a specific point it has to be for us to be in danger?”

This is a win for me. Not our demise from the universe, but allowing the children to promote their own science capital. And I now incorporate this type of questioning naturally into other curriculum lessons, ranging from English to geography to PE.

Callum Woolman is science lead and classroom teacher in an inner-city primary school

This article originally appeared in the 29 November 2019 issue under the headline “The wonder of ‘what-if’ in science lessons”