What Self-determination Theory Actually Says About Why People Do Things
In most L&D contexts I’ve worked in, the implicit model of motivation is binary: people are either intrinsically motivated, doing something because it matters to them, or they’re extrinsically motivated, doing it because they have to. The learning designer’s job, on this reading, is to push people from the second category into the first. “Make fun and engaging” is set as the design paradigm for motivation. And usually leads to the wrong learning design decisions.
Research on Motivation and Self-determination (Ryan, Deci and colleagues) showed that external rewards can undermine motivation for tasks people already found interesting. That was an important finding: the type of motivation matters. But because we tend to binary thinking, we transformed that finding into a rule: extrinsic bad, intrinsic good. And that rule leads learning designers to chase a version of intrinsic motivation that most people in most professional contexts simply aren’t starting from and won’t even reach.
The Continuum
Self-determination Theory (SDT) describes motivation as a spectrum of six types, ordered by how much a person has genuinely taken a value or behavior on as their own. Where someone sits at any given moment depends as much on the conditions around them as on their individual disposition. To make this concrete: imagine a cohort working through a mandatory company-wide AI literacy program. The deadline is set, participation is mandatory, and the cohort spans everyone from skeptics to enthusiasts.
😐 Amotivation is the state where someone isn’t really acting at all, just going through the motions. It comes from not seeing the point, not feeling capable, or not expecting the activity to lead anywhere. One person in our cohort has already decided AI tools are a passing trend or that it’s just too much for them. And they’ll click through the modules.
👉 Side note: design can produce amotivation actively. A 60-minute module with no skip option for content someone already knows signals that their existing knowledge doesn’t matter and that nobody designed this with them in mind.
🙄 External regulation is behavior driven entirely by external requirements like compliance. Several people in our AI cohort are here. They’re not opposed to AI literacy; but they have thirty other things on their plate. Research is consistent that behavior driven by external controls tends not to persist once those controls are removed. Most people arrive at mandatory training from external regulation, they’re there because they have to be. The question is whether the design gives them any reason to move in the motivation continuum.
😕 Introjected regulation is motivation that has moved inside the person but in the form of pressure rather than genuine ownership such as guilt, anxiety, fear of falling behind. The driver is internal, but it doesn’t feel like yours. One person in our cohort feels vaguely anxious about not knowing enough about AI. They show up, complete activities, maybe even ask questions. From the outside this looks like engagement. But research associates this pattern with higher anxiety and poorer coping when things go wrong and it doesn’t produce the kind of ownership that transfers to real work.
🤔 Identified regulation is where the person has consciously decided the behavior is worth doing for their own reasons. They may not find the activity interesting, but they’ve connected it to something they actually care about. The project manager in our cohort has recently struggled to evaluate AI-generated outputs in client proposals. She’s not fascinated by language models, but she’s made the connection. This is where design starts to have real leverage, the conditions that support identification are things a well-designed experience can actually provide.
🙂 Integrated regulation is where a value has been fully absorbed into a person’s sense of self. Integrated regulation is still technically extrinsic: the behavior isn’t driven by enjoyment of the activity itself, but by its alignment with who the person is. One person in our cohort is a learning designer who has been thinking about what it means to be a responsible practitioner in an era of AI-generated content. Engaging with this training is continuous with commitments she already holds.
😊 Intrinsic motivation is behavior driven by genuine interest in the activity itself. Even autonomy-supportive conditions won’t generate it where that interest isn’t present. A few people in our cohort are here. They find language models fascinating and have been experimenting with them outside work. The training is, for them, a low-bar version of something they’re already doing. This is why designing for fun is a weaker goal than it sounds. Enjoyment, where it appears, tends to follow from need satisfaction rather than being engineered directly.
What Drives Internalization
SDT states three basic universal psychological needs: competence, autonomy, and relatedness. When these are satisfied, people move toward more internalized motivation. When they’re blocked, people move toward the controlled end or disengage. The person producing amotivation isn’t disengaged by temperament. Something in the design is failing to support at least one of these needs.
Competence
Can the person feel effective in relation to this content? A module pitched at beginners for someone who already works with AI tools daily answers this question badly.
Autonomy
Does the person have any meaningful say in how they engage? This requires that choice, where it exists, is real rather than cosmetic, and that rationale is framed in terms of what the learner values rather than what the organization needs from them.
Relatedness
Does the design take the person's actual situation seriously? For learning design it means that people have to feel that the experience was made with someone like them in mind. When the design talks past the real context, people notice. It doesn't feel like it's for them, so they don't engage as if it is.
So what actually moves someone like the project manager from external compliance toward identification? Research points to three specific conditions: a meaningful rationale connected to what the person already values (not what the organization needs from them), acknowledgment that the activity might not be interesting rather than overselling it (counterintuitive as it sounds, it’s what makes a rationale credible), and real rather than cosmetic choice wherever possible.
The Design Question
The continuum is not a stage model. People don’t move through it in sequence, and context can shift them in either direction. A well-designed experience can move someone from external compliance toward identification. A controlling one can move a curious person toward disengagement. Internalization is something a person does. Design can make it more or less likely, but can’t do it for them. The design question shifts from “how do we make this engaging?” to “what conditions would support internalization from where people in this context are actually starting?”
Understanding where people sit on the continuum is the first step toward a more people-centered learning design. I’ve created a one-page design audit for stakeholder conversations that helps map a cohort, check a design against the three basic needs, and apply the three conditions in the way we communicate and give space before building a project.
Key Research: Self-determination theory (Ryan & Deci, 1987 and 2000), Self-determination theory and work motivation (Gagné & Deci, 2005), Internalization (Deci et al. 1994; Ryan & Connell, 1989).
