On things that are better together

The amplifying power of autonomy and structure

Better together - caramel and salt.

Some things are just better together. Freddie and Brian, Elton and Bernie, Dr Dre and Snoop Dogg. Individually, they’re ok. Together, magic.

Self-determination theory by Rich Ryan and Edward Deci (2017) lays out the basic human needs people require to flourish—not just to survive but to do great things like learn, achieve, and set goals. It’s a fascinating field for teachers but I’m not persuaded that it’s been translated very well, except of course for people like Peps Mccrea who has done a fantastic job of distilling the research in his book.

SDT identifies three basic human needs: autonomy, competence and relatedness (Deci & Ryan, 2000). These needs are common to everyone and transcend cultural boundaries (Chen, 2015). In the classroom, these needs are typically operationalised as autonomy-supportive teaching behaviours, teacher-provided structure to support competence, and a sense of belonging fostered through safety, inclusion, and positive interactions. The theory assumes that the capacity for learning is innate but as teachers, we can either satisfy or thwart these needs.

Teachers of Year 9 might find it hard to believe that students are “innately curious, interested creatures who possess a natural love of learning” (Niemiec & Ryan, 2009, p. 133). Still, self-determination theory holds that we all have an innate tendency towards growth. Consider this: when that drive is completely absent in an individual, we tend to view it as a sign of dysfunction. An extreme example would be hikikomori, a condition where Japanese youth withdraw completely, refusing to leave their parents’ home or engage with society. On the other hand, simply attending school is a decent starting point and indicates at least some natural inclination to be part of a school and learn.

Autonomy has been the hero of self-determination theory for many years, which is problematic. We see the term rear its head in discussions about teacher work, but also in relation to students. For starters, it’s highly impractical to give students free choice about what they want to do. Not to mention the fact that their choices are unlikely to be conducive to much real learning. Even slightly constrained choice can be difficult. If we ask a room full of 14-year-old boys if they would like to study A Midsummer Night’s Dream or do something else— well, if you’re a teacher reading this, then I don’t need to state the obvious.

The other problem with lionising student autonomy is that it devalues the role of teacher knowledge and expertise, potentially disregards the carefully considered scope and sequence of the curriculum, and diminishes the role of the adult in the room, who should control the learning. Students like sitting in groups with friends; teachers know that learning occurs more efficiently in rows. Research backs me up here, going all the way back to Rosenshine (1978) and Brophy (1986), who looked at exactly this and found that if the adult controls the content, pace and environment, students learn more. Simple.

Some elements of autonomy support work well, for example, providing rationales for the learning (Reeve & Jang, 2006). Hollingsworth and Ybarra (2009) recognised this, and theirs is one of the few models which takes this into account. Rationales could be about relevance to the student’s world; connections to prior learning and how the lesson follows on; enthusiasm by the teacher about the intrinsic worth of studying Shakespeare; and for some students, “It’s in the test” will be enough to motivate.

Acknowledging rather than dismissing negative feelings or opinions about learning can also show students that you value them as a separate entity from your middle-aged nerdy self. We can acknowledge and then still teach Midsummer. They’re not mutually exclusive. We can also use invitational language in our interactions with students. I like to think of this as the illusion of choice, for example, “You might want to start with an adverbial phrase about the author’s context.”

You might be seeing my broader point emerge about autonomy—it must be within the constraints of the classroom and learning. Teachers are the best sources of decision-making about what constitutes a reasonable range of options in the classroom. Anything else becomes laissez-faire and we know that unbounded classrooms are surprisingly demotivating (Aelterman et al., 2019). Teacher-provided structure, on the other hand, leads to student competence, which is understandably need satisfying.

Compared to the sexier sounding and more free-wheeling autonomy support, the teacher-provided structure by which we enable competence has a bad rap. Many progressives would see it as purely teacher-talk—dry, didactic and antithetical to critical thinking and creativity. It’s no friend to the fun-looking, chaotic Rube Goldberg-style lesson, the kind that looks great on LinkedIn, but is the learning equivalent of teenage daycare.

Structure involves clarity of explanations; it requires teachers to be organised and consistent; it features simplification and chunking of new skills and knowledge. There’s emerging research that suggests structure reduces cognitive load (kind of obvious, but until recently, underexplored) (Evans et al., 2024), but also that this teaching style is motivating. There’s also emerging evidence that cognitive load comes at a motivational cost (Feldon et al., 2019). So, when you provide structure, but in an autonomy-supportive way, you get the best of both and the benefits surge.

Load reduction instruction (Martin & Evans, 2018) is a simplified model of explicit teaching that takes students from teacher support to independence, taking into account both Kalyuga’s expertise reversal effect (2007) and the ultimate goal for all who teach explicitly—student mastery and independence. The field is still in its early days but it seems load reduction influences motivation and achievement (Evans & Martin, 2023; Martin et al., 2021, 2023).

This combination of structure and autonomy support is very exciting, and is central to my ongoing research. You could say it’s a new direction for motivation, even though the scholars in this area are relatively few. Researchers of the past have thought of these two motivating styles, structure and autonomy support, as antagonistic. But in fact there is a curvilinear relationship between the two—too much structure can feel controlling, while too much autonomy can feel chaotic (Jang et al., 2010). We’re after the Goldilocks combination.

Since unbounded choice is antithetical to a structured environment (in my view, anyway), structure needs to precede later autonomy, or at least be present. I think of structure as a prerequisite. It helps me to organise my thinking and teaching. However, the research doesn’t confirm my thinking, but rather suggests that we can display situation-specific styles in tandem with one another, even when they seem antagonistic (Aelterman et al., 2019). A good example would be providing student autonomy-support and a minute later, giving contingent behavioural praise, which is controlling. We’ve all done it!

So what’s the ideal mix? The way we talk about learning, invite students to participate and interact, and exhibit openness to their views are enough for students to feel like the learning and the class environment are worthwhile. And the competence they feel as a result of teacher-provided structure means that later down the track, for example at the end of a unit where critical thinking and transfer are required, they can make genuine choices and be truly autonomous. Without competence, all we can offer is the illusion of choice.