Self-Determination Theory: Needs, Motivation and Examples

Go from storyboard to an interactive course in minutes

Start Free

Picture a regional council where, every year, staff open the same online refresher on gifts, benefits and conflicts of interest. Completion sits near 100%. People click through the policy pages, the depot crews say it was written for office staff, and experienced officers mutter that they did it last year. Meanwhile the integrity team keeps receiving late declarations, or none, for exactly the grey-area cases the course is meant to cover.

I'm an instructional designer at Coursebox, where we make instructional design software that turns a brief into a course outline, so I'll mention our product where I tested it. Self-determination theory explains why a course like that refresher can be finished by everyone and still change little, and what to do about it. I cover the theory from Deci and Ryan's own papers, its six mini-theories, the motivation continuum, the debate about rewards, one council refresher worked through each idea, real workplace cases, the criticisms and two free templates.

What is self-determination theory?

Self-determination theory is Edward Deci and Richard Ryan's theory of human motivation, which holds that everyone has three basic psychological needs, for autonomy, competence and relatedness, and that settings which support those needs foster more autonomous and longer-lasting motivation, at work and in learning.

View the full-size diagram

Ryan and Deci define it in their 2017 book, Self-Determination Theory: Basic Psychological Needs in Motivation, Development, and Wellness, as "an empirically based, organismic theory of human behavior and personality development", one that "differentiates types of motivation along a continuum from controlled to autonomous" (p. 3). Two ideas carry most of the weight. The first is that the quality of motivation varies as well as its amount: acting because you value something is different from acting because someone is watching. The second is that the social setting, such as a manager, a teacher, a course or a pay scheme, can support or thwart the three needs, and that this shapes which kind of motivation people end up with.

The theory covers far more than training. Its research spans schools, health care, sport, work, relationships and well-being, and its six mini-theories each explain one part of that ground. For a training designer, the useful core is the three needs, the continuum from controlled to autonomous motivation, and what the research says about rewards and feedback.

In their 2000 paper in American Psychologist, Ryan and Deci describe the three needs as innate: when they are satisfied, people show more self-motivation and better mental health, and when they are thwarted, motivation and well-being drop (p. 68). The needs are not ranked. "Individuals cannot thrive without satisfying all of them, any more than people can thrive with water but not food" (p. 75).

Here are the three needs at a glance, with the question each one puts to a course.

Case study: Field service managers in a US office-machine company

This was an SDT-based training for managers, and the founding workplace test of the theory. When the study ran, the service division of a major US office-machine company had profitability and morale problems, with job cuts and a wage freeze. Its technicians spent almost all their time on the road and saw their managers briefly once a week. The study was funded by Xerox.

• What they did: An outside consultant spent 13 days in each branch, including a two-day off-site session and follow-ups, and watched each manager run a team meeting and gave feedback. The training had three themes: give staff more room to take initiative, give feedback that informs rather than controls, and acknowledge staff's perspective and feelings.
• Evidence and result: Deci, Connell and Ryan reported in 1989 that managers' orientation shifted towards supporting autonomy (8 trained managers compared with 13 not yet trained). In the short term, only 1 of 16 employee measures moved. Later, in the company's own attitude survey, satisfaction rose from 3.31 to 3.48 in trained branches while it fell from 3.65 to 3.44 in untrained ones (188 then 235 respondents in trained branches, 151 then 177 in untrained ones).
• Limitation: Branches were not randomly assigned, the comparison branches started at a different level, everything was self-reported and no productivity data was collected. The authors write that they "did not have an adequate test of the impact of the intervention on subordinates".
• Lesson: Managers' style can change after coached practice, but staff notice slowly when contact is thin. While pay and jobs were under threat, the managers' orientation was linked to nothing their teams reported. Do not expect a motivation course to fix a pay or job-security problem.

Who developed self-determination theory, and what to read

Edward Deci and Richard Ryan developed the theory over five decades, mostly at the University of Rochester. It began with Deci's 1971 experiments on money and puzzle-solving. Students who were paid a dollar for each puzzle they solved in one session later spent less of their free time on the puzzles, compared with an unpaid group, while in a separate experiment verbal praise had the opposite effect. Deci was cautious about his own result, writing that "there were only 24 subjects in the experiment, and the significance was only at the .10 level, so additional evidence on the validity of the finding is needed" (p. 110). That evidence came later, in a 1999 meta-analysis of 128 experiments by Deci, Koestner and Ryan.

Deci and Ryan set out cognitive evaluation theory around 1980, and their 1985 book, Intrinsic Motivation and Self-Determination in Human Behavior, framed the wider theory. Ryan and Deci's two papers in 2000 are among its most cited statements, and the 2017 book is the full account. Edward Deci died in February 2026, aged 83. He had joined Rochester in 1970 and retired in 2017, and in 2015 the two founded the Center for Self-Determination Theory. Richard Ryan is a research professor at the Institute for Positive Psychology and Education at Australian Catholic University in Sydney, and a professor at the University of Rochester.

If you want to read the originators rather than summaries, start with one of these:

• Ryan and Deci (2000), American Psychologist. Read this first. It is short, free on the centre's site, and covers the needs, the continuum and internalisation in eleven pages.
• Ryan and Deci (2020), Contemporary Educational Psychology. Read this overview for the current view of intrinsic and extrinsic motivation and its classroom implications.
• Ryan and Deci (2017), the Guilford book. Read it when you need depth on one area. At 756 pages it is a reference, with chapters on schools, work and video games.

The six mini-theories

Self-determination theory is a set of six mini-theories, each built to explain a group of findings. The Center for Self-Determination Theory lists them in this order:

1. Cognitive evaluation theory (CET) concerns intrinsic motivation: how rewards, feedback, deadlines and other events affect people's interest in an activity. Events that feel controlling tend to reduce it, and events that inform people about their competence tend to support it.
2. Organismic integration theory (OIT) concerns extrinsic motivation and how far people take a rule or value on as their own. It describes a continuum from external pressure to full integration, and it highlights support for autonomy and relatedness as the conditions for internalisation.
3. Causality orientations theory (COT) describes individual differences in how people tend to orient towards their surroundings: towards autonomy, towards control, or towards feeling that nothing they do makes a difference.
4. Basic psychological needs theory (BPNT) sets out the three needs and their links with well-being. The centre states that "all three needs are essential and that if any is thwarted there will be distinct functional costs."
5. Goal contents theory (GCT) compares intrinsic and extrinsic goals and their effects on motivation and well-being.
6. Relationships motivation theory (RMT) concerns close relationships, and argues that some close relationships are essential to people's adjustment and well-being.

The count has changed over time. A 2010 historical overview described five mini-theories, and relationships motivation theory was named in a 2014 chapter. Training design draws mostly on the first, second and fourth.

The worked example: a council's annual refresher

I'll use one hypothetical example for the rest of the guide. A regional council with about 900 staff runs a customer service centre, libraries, planning and building approvals, a works depot and waste services. Each year every staff member, including team leaders, completes the code of conduct refresher, and the module people resent most is gifts, benefits and conflicts of interest. It is mandatory, recorded for audit, and delivered as about 25 minutes of self-paced online learning plus a 15-minute team discussion run by each team leader.

The problem is not completion. It is that staff treat the module as a box to tick, and the integrity team still sees late or missing declarations for grey-area cases: a builder's thank-you hamper sent to a building surveyor, a library volunteer related to a contractor bidding for work, a supplier offering a depot crew lunch, a planning officer whose neighbour lodges an application. The rule is not hard to learn. The hard part is recognising the grey case in your own work and choosing to declare it.

So that learners can make those decisions, the course gives them the facts it relies on. The council's policy requires staff to declare any gift or benefit they are offered, even if they decline it, and any private interest that could affect their work or look as if it could. Declarations go to the staff member's team leader on the declaration form within 5 working days. Learners get a one-page policy summary, the form and the facts of each scenario. Nobody is assumed to judge a grey case well because of their role or years of service.

The intended outcome is that staff recognise grey-area gifts and conflicts in their own work and declare them on time, and that team leaders talk about declarations openly. The example suits self-determination theory because a mandatory, audited course is the hard case for it. Staff cannot opt out, and few will ever find the topic interesting. The theory still says something precise about what the course can do.

The three basic psychological needs, worked through the refresher

Each need has a short definition, then the same three labels: the originators' words, the council, and the course.

Autonomy

Autonomy is the experience of acting with volition: doing something because you endorse it, even when someone else asked for it. Ryan and Deci (2017, p. 14) write that "behaviors are autonomously motivated to the extent that the person experiences volition". Autonomy does not require choice of whether to do the task at all, which is why it applies to mandatory training.

• In Ryan and Deci's words: "Autonomy support includes affordances of choice and encouragement of self-regulation" (2017, p. 12).
• At the council: Staff cannot skip the refresher, but they can be told why it exists in terms they care about: residents' trust in council decisions, and protection for staff against later accusations. They can also be told that an annual refresher can feel repetitive.
• In the course: Open with that reason, write in invitational language ("you can", "here is why") rather than threats, and offer real choices inside the mandatory frame: which work area's scenarios to work through, and which grey case to bring to the team discussion.

Competence

Competence is the experience of being effective and able to grow. In Ryan and Deci's 2020 overview, it "is best satisfied within well-structured environments that afford optimal challenges, positive feedback, and opportunities for growth". Structure supports competence. Leaving people to find their own way through a policy does not support autonomy, and it is more likely to frustrate competence.

• In Ryan and Deci's words: "competence supports include provisions of structure and positive informational feedback" (2017, p. 12).
• At the council: Experienced staff handle the clear cases well and are bored by them. New staff need the basics first. Almost everyone struggles with the grey cases.
• In the course: Start with a short pre-check on grey-area cases, and tell people who handle them well that they can go straight to the harder scenarios. At each decision, explain the reasoning against the policy and show the consequence of not declaring, rather than marking an answer "Correct!" or adding a score.

Relatedness

Relatedness is the experience of belonging and of being cared about by the people around you at work. Ryan and Deci (2000, p. 73) call it "centrally important for internalization": people take on the values of groups they feel connected to.

• In Ryan and Deci's words: "relatedness supports include the caring involvement of others" (2017, p. 12).
• At the council: The online module is completed alone. Declarations feel like something you do to stay out of trouble, not something your colleagues and team leader do as a normal part of the job.
• In the course: The 15-minute team discussion carries this need. The team leader brings one anonymised grey case from the team's own area, shares a declaration they have made themselves, and asks colleagues how they would handle it.

Need support does not have to come only from managers. One trial in banking and property teams tested whether colleagues could provide it.

Case study: Peer support in northern European banking teams

This was an SDT-based training for colleagues rather than managers. Most need-support programmes go through managers, so researchers tested whether team members in three banking and property organisations in northern Europe could support each other's needs instead.

• What they did: Trained HR staff ran two half-day workshops, and teams then held three peer feedback sessions on their own over about seven weeks. A card exercise on everyday work behaviours built perspective taking, communication and collaboration.
• Evidence and result: Jungert and colleagues reported in 2018 that self-reported autonomous motivation rose in the trained teams, analysing 146 people in 26 teams (112 in 19 trained teams, 34 in 7 comparison teams). The authors describe the effects as small to medium, and the rise in motivation ran through higher need satisfaction. Controlled motivation did not change.
• Limitation: Teams were not randomly assigned, there were few of them, there was no follow-up, and, in the authors' words, "no objective outcomes were measured beyond motivation".
• Lesson: A short peer format run by HR raised self-reported autonomous motivation within seven weeks. Whether people's work changed is unknown, so treat a peer element as a support for motivation, not as proof of different behaviour.

Intrinsic vs extrinsic motivation: the continuum

Intrinsic motivation means doing something because it is interesting or enjoyable in itself. Extrinsic motivation means doing it because it leads to a separable outcome, such as pay, approval, avoiding trouble or protecting something you value. Those are Ryan and Deci's classic definitions, from their 2000 paper in Contemporary Educational Psychology (p. 55).

The theory's main step beyond that split is that extrinsic motivation varies in how autonomous it is. Organismic integration theory describes a continuum from amotivation through four kinds of extrinsic motivation to intrinsic motivation. The first two extrinsic kinds feel controlled, and the last two feel autonomous. In their 2020 overview, Ryan and Deci say identified and integrated motivation mean "people view the activities as worthwhile, even if not enjoyable".

That sentence sets the target for the council refresher. Nobody needs to enjoy declaring a thank-you hamper. The council needs staff who declare because they value residents' trust, which is identified regulation, rather than staff who declare to avoid trouble or out of guilt.

Ryan and Deci (2017, p. 15) also say how each type tends to be produced. Controlling reward contingencies produce external regulation. Ego involvement and self-esteem that depends on performance produce introjection. Identification is promoted "by providing convincing rationales for acting". Integration is helped by "a combination of acknowledging feelings, providing a rationale, and highlighting choice instead of control". That is close to a design brief for the first page of a mandatory course.

The evidence behind it comes from a 1994 experiment by Deci, Eghrari, Patrick and Leone, which tested three conditions for an uninteresting activity: a meaningful rationale, acknowledging people's feelings, and conveying choice. Ryan and Deci (2000, p. 74) summarise the finding: the rationale, with support for autonomy and relatedness, "promoted its internalization and integration", while controlling contexts yielded less internalisation, and what did occur "tended to be only introjected".

Do rewards undermine intrinsic motivation?

Rewards are the most argued-over part of self-determination theory. Cognitive evaluation theory predicts that rewards which feel controlling reduce intrinsic motivation, and rewards or feedback which inform people about their competence can support it. The largest test by the originators is Deci, Koestner and Ryan's 1999 meta-analysis of 128 experiments.

The same review found that positive feedback increased free-choice persistence (d = 0.33) and self-reported interest (d = 0.31). Unexpected tangible rewards had no effect (d = 0.01), and neither did rewards not tied to the task. The undermining effect also applies only to interesting tasks, because "with boring tasks there is little or no intrinsic motivation to undermine" (p. 633).

There is another side. In a 1994 meta-analysis of 96 studies, Cameron and Pierce concluded that "overall, reward does not decrease intrinsic motivation", that verbal praise increases it, and that the only negative effect, which they called minimal, appears when expected tangible rewards are given simply for doing a task. Their 2001 follow-up with Banko found that rewards for low-interest tasks enhanced free-choice motivation, and that harm appeared when rewards were tangible, expected and loosely tied to performance.

The two sides agree on more than the argument suggests. Praise and informational feedback help. Unexpected rewards do no harm. The risk sits with expected, tangible rewards for simply doing or completing an interesting task. They disagree on how large and general that harm is, and on rewards tied to performance quality.

A third meta-analysis changes the question. Cerasoli, Nicklin and Ford pooled 40 years of research from school, work and sport, covering more than 200,000 people, and reported that intrinsic motivation better predicted the quality of performance, while incentives better predicted its quantity. They conclude that incentives and intrinsic motivation "are not necessarily antagonistic and are best considered simultaneously."

Gamification fits the same pattern. In an experiment by Sailer and colleagues, badges, leaderboards and performance graphs raised competence need satisfaction, while avatars, stories and teammates raised relatedness. "Perceived decision freedom, however, could not be affected as intended." Game elements are not automatically controlling, but they did not support autonomy.

For the council, the rule is simple. No points, badges, leaderboards or competitions between departments. Feedback on each scenario explains the reasoning against the policy. Any thanks or recognition comes unexpectedly from the team leader, and it acknowledges good judgement rather than completion. Ryan and Deci (2020) make the same point about school grades: grades used as "motivators" will typically be experienced as controlling.

How to use self-determination theory to design training

Deci and Ryan did not publish a course design method, so this workflow is my adaptation of their research for a training designer, using the council refresher.

1. Name the behaviour and the motivation you want. Write the target as a behaviour (declare grey-area gifts within 5 working days) and the motivation behind it (staff declare because they value residents' trust).
2. Write the reason in the learner's terms, first. Put it on the opening page, before any policy. Say that the refresher can feel repetitive, and say plainly, once, that it is required.
3. Take the pressure out of the language. Explain the reasons and remove shaming or exaggerated threats. Keep accurate requirements, deadlines, completion reporting and any consequences staff need to know, stated neutrally.
4. Offer meaningful choices that fit the work. Choice of work area and choice of which case to discuss are real decisions. A menu of ten optional extras is not.
5. Match the challenge. Use a short pre-check, let people who handle the clear cases skip ahead, and save the hardest grey cases for last.
6. Make feedback explain. Every scenario decision gets the reasoning against the policy and the consequence of not declaring, with no score.
7. Add a relatedness element. A team discussion led by someone learners respect, who shares their own declaration, makes declaring a normal thing colleagues do.
8. Brief the team leaders and their managers. The discussion depends on how the team leader runs it, and that depends on what their own managers expect.
9. Measure behaviour, not completion. Check whether declarations become more timely in each area, the behaviour level of Kirkpatrick's evaluation model, and add a short pulse on the three needs.

For steps 5 and 6, a scenario-based learning design does most of the work, and explanatory feedback on practice is formative assessment rather than a test to pass.

The refresher course plan

Here is the refresher as a plan, with each part tied to the need it supports.

The choice finding comes from Patall, Cooper and Robinson's 2008 review in Psychological Bulletin. A choice also has to be meaningful, and people need enough information to make it, which is why each work-area path names the kinds of cases it covers before the learner picks.

Train the team leaders, not only the learners

The course can supply a reason and a choice, but relatedness and much of autonomy support come from people. In Gagné and Deci's account of self-determination theory at work, managerial autonomy support means "acknowledging their subordinates' perspectives, providing relevant information in a non-controlling way, offering choice, and encouraging self-initiation rather than pressuring subordinates to behave in specified ways" (p. 345).

That style can be taught. In education, Reeve and Cheon's 2021 review of 51 autonomy-supportive teaching interventions, 38 of them randomised trials, concluded that teachers can learn to be more autonomy supportive. The behaviours they describe include several that translate well to a team leader: take the learner's perspective, give explanatory reasons, acknowledge negative feelings, use invitational language and show patience. That evidence comes from schools, so treat it as a guide to what to practise rather than proof for workplaces.

At work, the links are clear but correlational. Slemp and colleagues' 2018 meta-analysis of 72 studies and 32,870 people found that leaders' autonomy support went with more autonomous work motivation and was unrelated to controlled motivation.

Two cautions sit beside that figure. The same review found a stronger link with self-rated performance than with performance rated by others, and the studies were mostly cross-sectional, so they show what goes together rather than what training causes. The intervention studies give more direct evidence, with different designs and more modest downstream findings.

The strongest design among them is a cluster randomised trial in publicly funded physiotherapy clinics in Dublin, Ireland. Physiotherapists in the trained clinics received eight hours of communication training based on self-determination theory. In the trial's study of staff behaviour, independent raters scored recorded sessions, and the trained physiotherapists supported patients' needs far more (d = 2.27, from one recorded session each for 24 physiotherapists). But the effect on patients' self-reported adherence to home exercise was weak (d = 0.28), it had faded by 24 weeks, and there was no effect on pain, according to Lonsdale and colleagues' 2017 report on 255 patients. Even after training, raters judged support well below ideal. The lesson for the council is to plan follow-up for team leaders, such as a second discussion and feedback on how the first went, rather than a single briefing.

Case study: Leadership training in a Swedish municipality

This was an SDT-based training for new first-line managers in a midsized municipality in northern Sweden, across childcare, education, elderly care, culture and leisure. The aim was managers who support employees' autonomy, competence and relatedness.

• What they did: As part of the municipality's mandatory leadership training, managers attended two two-day sessions a month apart and a half-day session about three months later, with cases, role play in groups of three, a book and feedback from their own employees.
• Evidence and result: Tafvelin, von Thiele Schwarz and Stenling reported in 2019 "no statistically significant improvement in managers' display of needs support or employee need satisfaction", comparing 21 trained managers with 17 from the next intake (37 analysed after one left) and surveying 442, 386 and 332 employees across three rounds. In focus groups, managers valued the role play and cases, but said senior management did not prioritise or follow up the training, and workload spiked during the training.
• Limitation: Groups were not randomly assigned, response rates differed between them, and all measures were self-reported.
• Lesson: The study detected no statistically significant improvement after four and a half days of training, and managers said senior management did not prioritise or follow it up while workload rose. For the council, brief directors as well as team leaders, and check the timing before launch.

Measuring motivation after the course

Completion is not evidence of motivation. For the council, track declaration timeliness alongside the number of situations that required a declaration, where that can be established. Counts alone can rise because more gifts were offered or because staff reported more accurately. A before-and-after pattern can guide follow-up, but it does not establish motivation or isolate the course's effect. A short pulse on the three needs can sit beside it, as a prompt for conversation rather than a score that certifies the course.

If you want a published scale, three are common. The Work-related Basic Need Satisfaction scale has 18 items, but it was validated in Dutch samples and its authors note that the English version had not been tested. The Basic Psychological Need Satisfaction and Frustration Scale has 24 items and was tested in four countries. The Multidimensional Work Motivation Scale has 19 items, tested with 3,435 workers in seven languages and nine countries. The centre's questionnaires are free for academic use, and commercial use needs its permission. Whatever you use, report the three needs separately: Van den Broeck and colleagues' 2016 review concluded that "it is not appropriate to average the three needs together or to use an overall need satisfaction score."

What course briefs ask for

To see whether people brief courses with the three needs in mind, I looked at course requests typed into Coursebox in the 90 days to 7 October 2026: one per person, from English-language browsers, with no uploaded file. Of 266 prompts, 104 were course requests. Two separate AI classification passes sorted every request against written definitions, and every request either pass flagged was then re-read by hand before a final call. Where readers differed the higher count is used, so treat the figures as classification estimates. Agreement and re-reading flagged requests cannot establish an upper bound or find every missed request.

Mandatory training looked no different. Among the 9 to 11 compliance and induction requests, such as safety inductions, patient safety and anti-harassment training, two framed the course with a must-pass test or a signed acknowledgement, none asked for a learner choice, and one was classified as asking the course to explain why the content concerned learners. Several asked for realistic workplace scenarios, which could support competence if the challenge and feedback fit. Scenario requests alone do not show whether a course supports any of the three needs.

This is not a criticism of the briefs. Most describe content, which is what a brief is for, and the sample covers requests, not finished courses, from a mix of teachers, coaches, course sellers and workplace trainers. It shows what these particular briefs said, not how the needs were supported in finished courses. If you want a course to give learners a reason, a choice and a colleague to talk to, you have to write those into the brief yourself.

Self-determination theory vs ARCS, social cognitive theory and Maslow

Self-determination theory is often taught next to other motivation models. They answer different questions. The table is my synthesis, not a ranking; I found no study comparing them head to head.

The ARCS model is the closest sibling, and the two are easy to blur. ARCS confidence is the learner's expectation of success; self-determination theory's competence is a basic need to feel effective and grow. Social cognitive theory is the better guide when the main problem is teaching the skill itself, through modelling and self-efficacy. Maslow differs in structure: Gagné and Deci note that Maslow's needs are "organized in a hierarchy", whereas self-determination theory treats all three needs as necessary at the same time. For the council refresher, I would use self-determination theory to decide the reason, the choice, the feedback and the team discussion, and ARCS to check the finished course for sections that lose attention.

Criticisms and limitations of self-determination theory

Self-determination theory has a large research base, and it has drawn criticism on several fronts.

• The reward debate is not settled. Behaviourist researchers such as Cameron and Pierce read the same experiments as showing limited harm, and the two camps still disagree on how large and general the undermining effect is.
• Culture and choice. Iyengar and Lepper found that personal choice generally enhanced motivation more for American participants with independent selves than for Asian participants with interdependent selves, and that Asian American children were most motivated when trusted authority figures or peers chose for them. Ryan and colleagues reply that autonomy is not the same as choice or independence. In a study of 559 people in South Korea, Russia, Turkey and the US, autonomy related to well-being across cultures, and the authors argue that "the opposite of autonomy is not dependence but rather heteronomy". Ryan and Deci (2020) accept that how needs show up varies between cultures.
• Need satisfaction predicts controlled motivation poorly. Van den Broeck and colleagues' 2016 review of 99 workplace studies found that the three needs "did not explain any meaningful variance in external or introjected motivation", This limits using need satisfaction to explain every motivation type; it does not by itself refute their proposed role in well-being and autonomous motivation.
• The continuum is hard to measure. A 2017 meta-analysis by Howard, Gagné and Bureau of 486 samples largely supported a continuum-like structure, but could not separate integrated regulation from the other forms, and the Multidimensional Work Motivation Scale leaves it out.
• Most evidence is self-report and correlational. Self-reports of interest and behavioural measures agree only modestly, and the workplace meta-analyses rely mostly on cross-sectional surveys.
• Workplace trials are few, small and mixed. Of the four I reviewed for this guide, three changed the people trained more than the people they managed or served. The Swedish trial found no statistically significant improvement on its measured outcomes. I found no randomised trial of manager need-support training in a private company with open full text, and no case that measured business outcomes such as sales, productivity or turnover. Nothing here shows that this kind of training raises performance or return on investment.

Using self-determination theory with an AI course builder

On 8 October 2026 I briefed Coursebox to build the council refresher. Coursebox is our product, so weigh my account with that in mind. The brief gave the policy facts, asked for a respectful tone that explains reasons and acknowledges the repetition, and ruled out points, badges, leaderboards, scores, prizes and a final graded test. It asked for an opening on why the refresher exists, a grey-area check with the option to skip ahead, a choice of three work areas with two scenarios each, a team discussion guide, and a wrap-up as the last section.

Most of that came back as briefed. The outline had no points, badges, scores or graded test, and no quizzes at all. The course description read as an invitation, aimed at council staff and team leaders "who want a practical refresher". The first section opened with the reasons for the refresher. The second offered three grey-area calls, a page explaining the reasoning, and a decision point before continuing. The third held the work-area choice and a team discussion of an anonymised local case, and a section on making your declaration this week came last.

One thing was missing. The choice step offered three work areas, but only the front counter and libraries path had scenarios. A choice that leads to one path is not much of a choice, and depot crews would have had only counter and library cases to work through.

One follow-up message added the two missing paths to the outline. I asked for the two missing paths, then reloaded the outline and checked it. It now had a planning and building path, with a builder's hamper scenario and a private interest scenario, and a depot and waste path, with a supplier lunch scenario and a contractor interest scenario, both after the choice. It came to 4 sections and 9 pages, with no quizzes and the wrap-up still last; the outline's own time estimate, about 40 minutes, is longer than the 25 minutes I briefed, so that needs trimming in the generated course. I checked the outline only: whether the skip advice, the branching and the explanatory feedback render as briefed needs checking in the generated course, and later declaration patterns would show behaviour, not motivation or causation on their own. A separate motivation measure and an appropriate evaluation design would be needed for those claims. With all three paths in place, every work area now gets its own grey cases to decide.

Putting self-determination theory to work

Self-determination theory explains why a course can reach 100% completion and change little: completion can come from pressure, and the theory predicts that behaviour lasts when people value what they do. For mandatory training, aim for staff who see the rule as worthwhile, not staff who enjoy the course. Give a reason in their terms, a few real choices, a challenge that fits, feedback that explains, and a colleague or team leader who takes part. For a refresher like this one, I would omit completion prizes and shaming threats, and do not expect a course to fix problems with pay, workload or senior managers.

Start with the first page of your next mandatory course: rewrite it as the reason a learner would accept, say that the course may feel repetitive, and replace shaming warnings with a neutral statement of the requirement and any relevant consequences.