
Good teaching brings the science of learning and relational pedagogies together
I had the opportunity to speak to a group of secondary school DPs and APs at the ASDAPA mini conference in Auckland at the end of last week. They’d given me the title; “Finding the Balance, Science of Learning and Relational Pedagogy”. This is a question that can sometimes be framed as a choice (should teaching focus on the science of learning, or on relationships?) but in my presentation I chose to make the argument that we need both. [See my slides at the end of this post.]
I started with some simple definitions – the science of learning (SoL) helps us understand how people acquire knowledge, use memory, build on what they already know, and develop through practice and feedback. It asks…
- How does the brain learn and remember?
- How do what we already know, our attention and memory affect learning?
- How do teaching, practice, feedback and recall support learning?
- How can teachers make learning more effective?
Relational pedagogy (RL) helps us understand how trust, identity, belonging and a learner’s experiences shape their participation in learning. It asks…
- How do trust, belonging and relationships affect a learner’s engagement?
- How do a learner’s identity and experiences shape how they learn?
- How can we help learners feel safe to ask questions, take risks and make mistakes?
- How can teachers give learners more voice while keeping clear expectations?
These are different perspectives on teaching, but they are not competing approaches. Good teaching takes both seriously.
Evidence needs professional judgement
The first point I made was that the science(s) of learning is/are not a teaching method or a ready-made recipe. They draw on research from fields including cognitive psychology, educational psychology and neuroscience. That research can help teachers understand learning processes, but applying it in a classroom takes professional judgement.
The history of reading research offers a useful example. In 2000 the report from The National Reading Panel found benefits from explicit instruction in phonemic awareness and phonics. Those findings are important and were soon embraced by jurisdictions across the US and in included in legislation pertaining to how teachers should teach. But the panel also cautioned that phonics should not become a complete reading programme. Teaching the code is important; so is giving learners a broader, purposeful experience of reading.
In his 2003 research paper Tim Shannon summarises myths about the national reading panel report highlighting the problem with using the NRP document as an unyielding statutory rulebook, resulting in laws being passed that mandated a rigid “phonics-first-and-only” approach, overlooking the panel’s original, balanced recommendation to weave code instruction seamlessly into a holistic, comprehension-focused literacy environment.
The recent release of a 7-year audit of the NPD report by Denny Taylor highlights the problems that have emerged when the findings of the original report have been used to support just one particular perspective here, with Denny concluding that “…the NRP P/A and phonics meta-analyses have been used to gaslight the public.”
The point I am making here is that we need to understand how evidence can inform teaching without dictating every decision. We need to attend to what research tells us, what it does not tell us, and what we can see in front of us as learners respond.
The same care applies to relationships. Strong relationships do not replace subject knowledge, purposeful practice or effective instruction. Belonging on its own cannot resolve a mathematical misconception. But a learner’s willingness to reveal uncertainty can help a teacher discover what is going wrong and decide how best to respond.
What a wrong answer can tell us
To give this a practical perspective I used a simple classroom scenario. Imagine a student repeatedly making the same error when attempting a long division problem. The teacher has explained the process, modelled examples and tried again. Still, the student makes the same mistake.
It might be easy for both teacher and student to conclude that the student simply “can’t do it.” But the real issue may be that the learner is simply applying a rule that made sense in a different context, but not in the one being introduced now. Simply repeating the procedure won’t help if that underlying idea remains unseen.
In this situation understanding prior knowledge and mental models helps the teacher recognise why an error might persist. A trusting relationship makes it easier for the learner to explain their thinking without fear of embarrassment. Together, these perspectives point toward a more useful response with the teacher asking the student to show how they approached the problem, identify where their reasoning diverges, and choose an explanation or example that addresses that specific idea.
The relationship helps make the learner’s thinking visible. Pedagogical knowledge that draws on the sciences of learning helps the teacher decide what to do with it.
Making room for challenge and voice
In another example I used I looked at what might happen when a student challenges a teacher’s explanation of something in class. A challenge might be poorly supported, or it might open up a worthwhile question. The teacher’s task is to distinguish between the two and guide the discussion with clear expectations.
Learning often involves comparing ideas, evaluating evidence and revising what we think. But discussion does not automatically lead to understanding. Learners need enough background knowledge, thoughtful guidance and a clear sense of how to reason well.
Relational pedagogy makes space for disagreement without turning it into a personal threat. The teacher might ask what evidence supports the student’s alternative, invite the class to consider it, and help them weigh the explanations. Sharing power with learners does not mean abandoning the teacher’s responsibility or expertise. It means using that expertise to help learners develop their own judgement.
Learning is cognitive and social

Neither the brain nor a relationship operates in isolation from the world around it. What learners bring with them, the experiences they have, and the support they receive all shape what becomes possible.
This doesn’t mean that students must always feel comfortable before they can learn. Learning can involve uncertainty, effort and discomfort. The question is whether learners have enough support to persist, ask for help and recover from mistakes when the work is demanding.
That is one reason to move beyond a simple opposition between teacher-directed and learner-centred teaching. A teacher can explain a concept explicitly and then invite students to explore its implications. They can set clear learning intentions while offering choices about how to approach a problem. They can provide structured practice and ask learners to explain, question and apply what they understand.
The point is not to choose who leads every moment. It is to design learning that builds understanding, capability and increasing independence.
From connection to capability
In the presentation, I described this as a continuous cycle:

- Know the learner: attend to their identity, experience, prior knowledge, strengths and challenges.
- Design the learning: decide what learners need to know, what should be taught explicitly, and how they will practise and apply it.
- Respond to the learner: listen, observe and assess; use what you learn to adjust the support or challenge.
- Develop agency and capability: help learners build the knowledge, confidence and judgement to take increasing responsibility for their learning.
Each part informs the next. Knowing the learner shapes the design of the learning. The learning gives us opportunities to see how the learner is thinking. Our responses help them develop capability, which in turn gives us a deeper understanding of what they can do and what support they need.
Relationships are not simply a means to academic achievement, either. Care, challenge, support, shared power and expanded possibilities can help learners experience both belonging and growth. Care does not mean we lower expectations. Similarly, high expectations need not imply control or emotional distance.
It’s all in the mix!
The science(s) of learning reminds us that learning is not magic. There are things we can understand about how knowledge develops, how memory works, and how teaching and practice can support learning.
Relational pedagogy reminds us that a learner is not simply a cognitive system to be optimised. Each learner is a person with a history, an identity, relationships, aspirations and a developing sense of what they are capable of becoming.

Think of this using a cooking metaphor. No single ingredient makes the meal, and simply using more of something isn’t necessarily better either. The recipe isn’t fixed. Our thinking needs to shift from “Which ingredient should we be using more of?” to “What is the right mix of ingredients to create the conditions for great learning?”
We don’t build relationships instead of teaching well. We build relationships as part of teaching well. And we teach well not only to help learners know more, but to help them become more capable and independent in how they use what they know.
That is why both SoL and RL need to be present in our teaching practices and in the way we design curriculum. The useful question is not which one matters more. It is: “how can the quality of our teaching and the quality of our relationships work together to help every learner thrive?”
Access the slides I used…


One reply on “Finding the balance”
Ata mārie Derek. Tau kē – Awesome. A great set of slides which demonstrate the balance in pedagogy for our tamariki. No one size fits all. Whatever the research demonstrates has to be moderated by the informed classroom teacher to met the needs of their ākonga in their context. Thank you for the concise presentation visuals.