Psychologist logo

Stop, think, and overcome intuitions

Ella Rhodes reports from the annual Learnus lecture, by Professor Denis Mareschal.

01 December 2020

Understanding concepts which may, at first glance, seem entirely counterintuitive is a vital part of learning and development. In the annual Learnus lecture Professor Denis Mareschal (Birkbeck, University of London) shared insights from an intervention which aimed to encourage children to stop and think when faced with such a concept. 

Mareschal started by highlighting the difficulty in bridging the gap between findings in cognitive neuroscience and practice in the classroom. He said the Wellcome Trust and Education Endowment Foundation had come together to tackle this by funding intervention trials. This partnership funded Mareschal’s UnLocke intervention, along with support from Birkbeck, University of London, the UCL Institute of Education and Learnus, which aimed to improve maths and science reasoning in primary school children using cognitive neuroscience evidence.   

Mareschal pointed to some of the concepts which we learn in science and maths which can seem entirely counterintuitive to children: for example, while we know the Earth orbits the Sun it seems as if the Sun orbits the Earth when we look at the sky. Even as adults we may be occasionally fooled by these common misconceptions we still hold on to, ‘In order to utilise, our formal scientific knowledge or formal scientific training we have to overcome these ideas that pop into our minds all the time.’ 

Inhibitory control, Mareschal pointed out, may be key to inhibiting incorrect ideas or inappropriate strategies when trying to reach a correct answer. He said there was some evidence that inhibitory control had a role in academic achievement – particularly in primary school maths – and had been linked to performance in standardised maths tests, procedural maths skill and conceptual maths knowledge. 

Mareschal said there was another type of interference which may prevent children from accepting counterintuitive concepts. ‘In physics if you're trying to teach… a seven-year-old, that the earth is round – this seems unbelievable because on a daily basis they get perceptual reinforcement that the earth is flat… and yet maybe 10 minutes in a whole week someone is telling you that it's round. Somehow they have to be able to extract that as being the most important piece of information… Similar things exist in mathematics, there are plenty of misconceptions in maths where the perceptual evidence appears to contradict what the formal conclusions are.’ Mareschal gave the example of shapes which have larger perimeters being seen to have larger areas than shapes with smaller perimeters, where this may not be the case. 

‘The question is how do we overcome these intuitions, these perceptual biases? Again, it may have something to do with inhibitory control – being able to inhibit the answer that's being driven by the perceptual information in order to select the answer that is given by analysis.’ 

The UnLocke intervention started as an educational challenge, he said, asking how children can be encouraged to take on counterintuitive concepts in maths and science and let go of intuitive explanations of the world. He and his colleagues began looking at the science in the area to find a causal mechanism that drives this behaviour to target in an intervention. 

One study, published in 2014, found that when presented with a counterintuitive problem experts showed more activation in areas of the brain associated with inhibitory control compared with novices. ‘The conclusion I get from this paper is that improving your ability to inhibit those misconceptions would improve performance on academic tests. So this was the neuroscience evidence that we took to construct a classroom-based intervention that would hopefully help children, improve their academic performance in math and science tests.’ 

The UnLocke intervention was developed over 18 months with an aim to encourage primary school children to use their existing inhibitory control skills in the context of maths and science. After several pilot trials and consultation with teachers the team developed a computer game in the style of a game show. ‘Stop and Think’ was ‘hosted’ by the character Andy, who was joined by three game show contestants. Andy would pose questions, using information from the national curriculum, and guide the contestants and children through different ways of thinking encouraging them to stop and think for a few seconds before answering. 

‘We're trying to get the children to stop, in order to be able to stop what we want them to do is to take a short breather, wait four seconds before they answer, and this allows the alternative answers, the less frequent ones, the less associated ones, to bubble up and come to the forefront.’ The intervention was run using a randomised control trial design three times per week for 10 weeks during the first 12 minutes of a maths or science lesson to children in years three and five.  

Those classes in a control group either carried on with business as usual or played a similar game called See + which included questions about the PHSE curriculum. Once the 10 weeks had ended half of the children who took part were evaluated using standardised tests of maths and the other half were tested on science. All of the children were also tested on their inhibitory control skills using a ‘chimeric Stroop test’, in which children were shown an animal made of one animal’s head and another animal’s body and asked which animal’s body they could see. 

Projects funded by the Education Endowment Foundation are also evaluated by independent bodies – in this case the National Federation of Educational Research (NFER), which carried out the random allocation of classes to either intervention or control groups, delivered the assessments and also gathered qualitative evidence from teachers about their experience using the game in lessons. The NFER and Education Endowment Foundation decided in its analysis to compare the Stop and Think group to both control groups combined. 

The trial recruited more than 6,600 children across England from 87 different primary schools. The results showed that, when year five and year three groups were combined, the children who took part in Stop and Think had an average of a month’s extra progress in maths (not a statistically significant difference), and two months better performance in science (which was significant).  

When the children’s performance was broken down by year group it showed that these improvements were mainly driven by year five pupils, with little improvement in year three pupils’ performance in maths or science. To assess whether inhibitory control was important in the improvements seen, children in the Stop and Think group were compared with the active control condition, See +. It emerged that those in the experimental condition had significantly better performance in science and maths after the intervention. 

‘The children were better than those in the active control, suggesting that it was indeed inhibitory control that was driving this improvement in performance… that this is really driven by the year five pupils, and the year three pupils aren't really showing much of an effect.’

While the qualitative data collected by the NFER through interviews with teachers uncovered many positive comments, a majority did not think it should be rolled out in its current form. Some said it would be hard to fit the game into a school day and others highlighted software issues or said some children found it too simple and boring. 

Mareschal and his team are working on a larger trial as well as trying to resolve software issues. ‘The cycle doesn't end there, the cycle of development and discourse and conversation with practitioners… we continue to improve our intervention and our understanding of the mechanisms that underlie this intervention.’