Learning Counterintuitive Concepts

When learning new concepts in science and maths, pupils must be able to inhibit prior contradictory knowledge and misconceptions to acquire new knowledge successfully. This skill of “interference control” varies between pupils, with variation evident from an early age. Disadvantaged pupils seem to have weaker control skills than their wealthier peers.

The Centre for Educational Neuroscience, a collaboration between Birkbeck College, Institute of Education and University College London, will develop a computer game to train pupils’ ability to control such interferences. Following its development, pupils in up to 100 primary schools will undertake 15 minutes of exercises 3 times a week, at the beginning of maths or science lessons. In the game, a child-friendly character will try to solve problems with help from the player, providing prompts and suggestions. The aim is to train the pupil to inhibit their initial response, and instead give a more delayed and reflective answer. Exercises will relate to specific maths and science content. For example, exercises will help pupils to realise that mice and elephants have the same-sized cells, or that the world is round despite seeming flat. Teachers (or TAs) will receive a half-day training workshop to understand the context and background, but the hypothesis is that interference control improves best with practice, not through a change in pedagogy. 

Why are we funding it?

This project has been funded as part of joint initiative with the Wellcome Trust to explore how insights from neuroscience can be used to improve education. You can read more about this here.

Evidence from neuroscience research supports the hypothesis that inhibition control is necessary to develop the reasoning skills required in maths and science. But studies of interventions designed to improve such “executive function” skills have shown improvements on outcomes like working memory, but have often failed to show an impact on broader attainment measures. Emerging neuroscience research suggests that the inhibition needs to happen in the networks that are specific to the skills being developed, thus the need for exercises to be related to specific subject knowledge.

As a result, this project will focus on developing exercises that are more closely related to the curriculum areas of science and maths. We will therefore be able to test whether practising these skills leads to improvements in attainment in subject tests.

How are we evaluating it?

It will be independently evaluated by the NFER, led by Simon Rutt. The main trial will be an efficacy randomised controlled trial. Efficacy trials aim to test whether an intervention can work under ideal conditions, as in this case where the intervention’s original developer is closely involved in delivering the programme to schools.

The project will start with a development phase (January 2015 – July 2016). The team will work with 8 primary schools to develop and test the content and delivery of the computer programme.

From January 2016, the team will begin to recruit schools for a randomised controlled trial, which will start in September 2016. The exact design will be finalised in the development phase, but up to 100 primary schools will be recruited, and then most will be randomly allocated to either have Year 3 or Year 5 as their intervention year, with the other year group acting as a control. Some of the control pupils will receive just business as usual, but some will receive computer-assisted support on social/emotional skills, which the computer platform was originally designed to improve. This means that we can measure the impact of the “inhibition control” elements of the game, rather than the additional engagement and motivation caused by playing a computer game.

All pupils will be post-tested in summer 2017 on a maths and science measure, as well as more specific measures of their executive function skills. In addition, fMRI scanning will be carried out on a subset of children pre and post-intervention to understand the brain-level impacts.

When will the evaluation report be due?

The evaluation report will be published in Spring 2018.