It is rarely the case that a new topic of discussion in education research has not been studied in the past already. Of course, such research may have been marred by small sample sizes and we might have gleaned additional insights by now, but it can be useful to look back and consider on whose shoulders we stand.

Take, for example, the recent discussions around personalised learning. Important research that could contribute to these discussions is work on so-called aptitude treatment interactions (ATI) research, which took place in the 1970s and was conducted most markedly by Lee Cronbach and Richard Snow.

According to Snow, the aim of ATI research is to predict educational outcomes from combinations of “aptitudes” (a child’s ability to complete the task) and “treatments” (the teaching strategy or intervention used).

The research highlights how the effectiveness of instruction will depend on aptitude (I take this to include prior knowledge) and that it therefore is important to make a good estimate of what individuals’ aptitude in a classroom is.

If we assume everyone in a classroom is different, then one extreme action could be to personalise for each student. This is not an uncommon conclusion from research.

Benjamin Bloom is known for formulating his “2 sigma problem”, which refers to his observation that the average student who was tutored one-to-one using mastery learning techniques performed two standard deviations better than students who learned via conventional instructional methods.

Professor Anders Ericsson also says that his “deliberate practice” approach involves individualised training of a trainee by a well-qualified teacher (Ericsson and Harwell, 2019). And from the more sensitive topic of genetics, Robert Plomin also emphasises our individual uniqueness.

So, what does that mean for teaching? On the one hand we have individual differences and, on the other, we have the practical challenge that 20-30 of those individuals are in front of a teacher. It is not realistic, in my opinion, to differentiate for all the students, as this will simply lead to more workload.

But just as unrealistic would be to adopt just one teaching strategy - the ATI research tells us this will lead to less effective teaching where every child’s aptitude is different.

What normally happens is that a teacher makes a professional judgement on the approach to take (or how many different approaches), based on the children in front of him/her. The key question here, in my opinion, is where the “hinge” point for “treatments” is: where does teaching become (in)effective, and how does it relate to the amount of guidance and scaffolding?

Rather than a blunt “children learn differently” or “children learn the same way”, this question enables us to make a more nuanced approach. I think that this nuance does justice to the “voyage” we all make from not knowing much in the beginning to becoming more (relative speaking) expert from the moment we start to learn.

Christian Bokhove is associate professor in mathematics education within Southampton Education School at the University of Southampton

This article originally appeared in the 3 April 2020 issue under the headline “A supporting role that makes all the difference”