In 1984, Richard Koestner, Richard Ryan, Frank Bernieri and Kathleen Holt sat six- and seven-year-olds from a school near Rochester, New York, in front of some watercolour paints and asked each of them to paint a house they would like to live in. Before they started, some children heard this:
“Before you begin, I want to tell you some things that you will have to do. They are rules that we have about painting. You have to keep the paints clean. You can paint only on this small sheet of paper, so don’t spill any paint on the big sheet… In general, I want you to be a good boy (girl) and don’t make a mess with the paints.”
Others heard a different version:
“Before you begin, I want to tell you some things about the way painting is done here. I know that sometimes it’s really fun to just slop the paint around, but here the materials and room need to be kept nice for the other children who will use them… I know that some kids don’t like to be neat all the time, but now is a time for being neat.”
A third group was given no rules at all.
The rules in the first two scripts are the same. Small sheet, clean brush, no mess. What changes is everything around them: a reason, an acknowledgement that the child might feel differently, and the absence of “have to” and “be a good boy”. These two scripts speak to a question that sits underneath much of what I write about the difference between compliance and character. If we care about children owning their choices, does that mean fewer rules?
What the painting study found
Koestner and colleagues (1984) left each child alone for a few minutes afterwards with the chance to carry on painting or do something else, and measured how long they chose to keep painting. Children who had been given the controlling limits showed less intrinsic motivation than both other groups, and their paintings were rated as less creative and of lower quality. The children given informational limits were not significantly different from those given no limits at all. In the authors’ words, the results suggest “that limits can be set without undermining intrinsic motivation if they are informational in nature.”
It was a small experiment, with 44 children and a single activity on a single day. Painting is also something most young children already enjoy, which is not true of lining up or waiting their turn. I would not build a behaviour policy on it alone. But it was a genuine experiment, with children randomly assigned to conditions, and it points to a distinction that later research has developed a great deal.
Structure is not the same as control
For years, research on parenting tended to use “control” as a catch-all, covering everything from clear rules and supervision to guilt-tripping and domination. In a 2009 paper in Child Development Perspectives, Wendy Grolnick and Eva Pomerantz argued that this was causing confusion. They proposed keeping the word control for parenting characterised by “pressure, intrusion, and domination”, and describing parenting that mainly offers guidance as structure. On their account, the opposite of control is autonomy support. Structure is a separate dimension altogether.
That reframing matters because it means a child can have firm boundaries and a strong sense of choice at the same time. The research in classrooms backs this up. Hyungshim Jang, Johnmarshall Reeve and Edward Deci (2010) had trained observers rate teaching in 133 high school classrooms in the American Midwest, while 1,584 students reported on their own engagement. Autonomy support and structure were positively correlated rather than opposed, and both predicted the engagement observers could see. Only autonomy support predicted how engaged students said they felt. The authors describe the two as complementary rather than antagonistic.
A study of 1,036 secondary pupils by Maarten Vansteenkiste and colleagues (2012) found something similar. Pupils who saw their teachers as both autonomy-supportive and clear about expectations had the most positive pattern of motivation, self-regulated learning and behaviour. Pupils who saw neither had the most negative pattern.
Both studies are correlational and both are with teenagers. They show which classroom experiences tend to go together, not that changing one will produce the other, and I work with much younger children. Still, the picture fits the painting experiment. Clear rules did not seem to be the problem. Pressure did.
Why this matters for my doctorate
My Ed.D. focuses on children’s self-endorsed motivation, the kind of reason that sounds like “I do it because it matters to me” rather than “I’ll get in trouble” or, as I explored recently, “I feel bad if I don’t”. The study is still in progress, so nothing here is a finding of mine. What the wider literature does offer is a challenge to a false choice that often shapes conversations about behaviour in schools: either we are strict, or we are child-centred.
The structure and control distinction suggests the more useful question is how expectations are communicated. The research tends to describe structure as clear expectations beforehand, guidance during and informative feedback afterwards. Control tends to show up in the language around a rule: commands without reasons, rewards and threats as the main argument, approval made to depend on obedience. I have argued that behaviour systems on their own don’t build character. This research gives me a clearer way to say why. The problem is rarely that a system has rules. It is that the rules often arrive wrapped in pressure.
What it might sound like in school
Take a hypothetical Year 2 class getting ready for a trip to the local park. The teacher needs everyone to stay within sight of an adult. She could say, “You must stay with your group. Anyone who wanders off will miss golden time.” Or she could say, “The park is big and busy, and I need to be able to see everyone so I know you’re safe. I know it’s tempting to run off to the climbing frame, but today we stay with our group.”
The boundary is identical. The second version gives a reason, recognises how the children might feel and leaves out the threat. It is not softer. It is still a firm limit, and the teacher would still act if a child wandered off. It simply treats the rule as something the children can understand rather than something they must obey to avoid a consequence.
The same applies after things go wrong. When a rule is broken, the conversation can return to the reason behind it rather than the child’s standing with the adult, which is the thinking behind the questions I ask after a behaviour incident.
A question for leaders too
One more study complicates the picture in a useful way. Bart Soenens and colleagues (2012) looked at what lies behind psychologically controlling teaching, meaning intrusive behaviour that pressures students to act, think and feel in particular ways. Teachers who reported a pressuring school environment, and teachers whose own motivation for teaching was less autonomous, were more likely to teach in this way. Pupils’ own low motivation was not related to it.
Again, this is association rather than proof of cause. But it suggests that controlling classrooms are not only a matter of individual teachers’ habits. Pressure can travel down through a school. If leaders want teachers to give reasons, acknowledge feelings and hold boundaries without threats, it seems fair to ask whether staff experience the same from us, which links closely to what I have written about behavioural integrity in school leadership.
The two painting scripts take roughly the same amount of time to read aloud. One tells a child what to do. The other tells them why it matters and trusts them to see it. Both asked for exactly the same behaviour. The difference showed up afterwards, when the adult had left the room and nobody was asking the children to do anything at all.
References
- Grolnick, W. S., & Pomerantz, E. M. (2009). Issues and challenges in studying parental control: Toward a new conceptualization. Child Development Perspectives, 3(3), 165–170. https://doi.org/10.1111/j.1750-8606.2009.00099.x
- Jang, H., Reeve, J., & Deci, E. L. (2010). Engaging students in learning activities: It is not autonomy support or structure but autonomy support and structure. Journal of Educational Psychology, 102(3), 588–600. https://doi.org/10.1037/a0019682
- Koestner, R., Ryan, R. M., Bernieri, F., & Holt, K. (1984). Setting limits on children’s behavior: The differential effects of controlling vs. informational styles on intrinsic motivation and creativity. Journal of Personality, 52(3), 233–248. https://doi.org/10.1111/j.1467-6494.1984.tb00879.x
- Soenens, B., Sierens, E., Vansteenkiste, M., Dochy, F., & Goossens, L. (2012). Psychologically controlling teaching: Examining outcomes, antecedents, and mediators. Journal of Educational Psychology, 104(1), 108–120. https://doi.org/10.1037/a0025742
- Vansteenkiste, M., Sierens, E., Goossens, L., Soenens, B., Dochy, F., Mouratidis, A., Aelterman, N., Haerens, L., & Beyers, W. (2012). Identifying configurations of perceived teacher autonomy support and structure: Associations with self-regulated learning, motivation and problem behavior. Learning and Instruction, 22(6), 431–439. https://doi.org/10.1016/j.learninstruc.2012.04.002