About Cognitive Evaluation Theory

Most of us assume that if we want someone to do more of something, we should reward them for doing it. Pay a salesperson more per sale and they will usually sell more. Give a bonus for good attendance and people tend to show up more reliably. This is usually true, and it’s the basic logic behind most incentive schemes in a workplace.

But there’s a specific finding in organizational behavior research that complicates this picture. Sometimes, adding an external reward to a task people already enjoy doing for its own sake can reduce their motivation to do it. That’s the core idea behind cognitive evaluation theory (CET), developed by psychologists Edward Deci and Richard Ryan.

What Is Cognitive Evaluation Theory?

Cognitive evaluation theory is a sub-theory within the broader self-determination theory (SDT), which Deci and Ryan developed to explain what drives human motivation. CET looks specifically at what happens to a person’s intrinsic motivation (the enjoyment that comes from doing a task for its own sake, rather than for some outside payoff) once an external reward gets introduced.

We need to separate two kinds of motivation before we go any further. Intrinsic motivation is doing something because we find it interesting or satisfying in itself, like a hobby. Extrinsic motivation is doing something because of an outcome separate from the task itself, like money, a grade, or avoiding a telling-off from a manager. Most jobs involve a mix of both. A software engineer might genuinely enjoy a hard coding problem while also caring about their salary and performance review.

CET’s argument is that external rewards, deadlines, close supervision, and even competition can shift how we interpret our own behavior. If we start to feel we’re doing a task for the reward rather than because we chose to, our sense of control over our own actions drops, and intrinsic motivation can drop with it, sometimes even after the reward disappears.

Where Did This Idea Come From?

Edward Deci ran one of the founding experiments in 1971. College students solved a puzzle called the Soma cube across a few sessions. In the middle session, one group was paid a dollar for each puzzle solved, while a second group wasn’t paid at all. Then, during a free-choice break where the experimenter left the room, Deci watched how much time each group spent on the puzzle once nothing was requiring them to touch it.

The unpaid group kept playing with the puzzle during the break, much as before. The paid group spent noticeably less free time on it once the payment stopped. The task had gone from something they did because they found it interesting to something they did to earn a dollar, and once the dollar disappeared, a chunk of their motivation went with it.

Why Would a Reward Make Motivation Go Down?

CET explains this through two psychological needs: the need to feel autonomous (that we’re choosing our own actions) and the need to feel competent (that we’re good at what we do). Whether a reward helps or hurts intrinsic motivation depends on which of these needs it touches, and how.

Deci and Ryan draw a distinction between two ways a reward can be experienced. A reward can feel controlling, like it’s being used to manipulate or pressure us into behaving a certain way. Think of a manager who says, “Finish this report by Friday and there’s a bonus in it for you.” That reward pulls the reason for doing the work outside of us. We start doing it for the bonus, not for the report.

A reward can also feel informational, meaning it mainly tells us something about our own competence, without much pressure attached. A coach who says, “That was a well-executed play, you read the defense really well,” is confirming skill rather than dangling a payoff. Feedback like this can strengthen intrinsic motivation, because it satisfies the need to feel competent without threatening our sense of control over our own choices.

The same reward, framed two different ways, can have opposite effects. A bonus tied tightly to output and delivered with controlling language is more likely to undermine intrinsic motivation. Recognition, some choice in how the work gets done, and feedback that confirms mastery are more likely to support it.

A Real Example: The Day-Care Late-Pickup Fine

One of the clearest real-world illustrations of this effect comes from a study of ten day-care centers in Israel, run by economists Uri Gneezy and Aldo Rustichini. Some parents were regularly a few minutes late picking up their children, leaving staff waiting around. The centers introduced a small fine for late pickups, expecting lateness to fall.

It rose instead. Once being late had a price attached, some parents seem to have reinterpreted their obligation. Before the fine, showing up on time was a social matter, tied to respect for the staff’s time. After the fine, it became a transaction: pay a small amount, arrive whenever. When the researchers later removed the fine, lateness stayed high. The sense of obligation did not simply return once the price disappeared.

This is a field result rather than a controlled lab experiment like Deci’s, but it shows a similar mechanism at work. Attaching an external price to behavior that was previously guided by an internal sense of obligation changed how people thought about the behavior itself.

Does This Mean Managers Should Never Use Rewards?

No, and we need to be careful about over-applying the theory here. CET does not say rewards are always bad. Plenty of tasks at work have very little built-in intrinsic appeal to begin with. Nobody finds data entry or repetitive compliance checks fascinating on its own. For tasks like these, there isn’t much intrinsic motivation to undermine, so extrinsic rewards do exactly the job we need them to do.

The undermining effect mainly shows up when a task already has genuine intrinsic appeal, and the reward is experienced as controlling rather than informational. It also matters whether the reward was expected in advance. An unexpected bonus after strong work tends not to damage intrinsic motivation the way a reward promised up front, tied explicitly to hitting a target, tends to.

How a reward is delivered matters as much as whether it exists at all. Two managers can offer the exact same bonus. One frames it as, “I’m watching your numbers closely and you’d better hit this target.” The other frames it as, “You’ve been doing strong work, and I want to recognize that.” Same amount of money, a very different message about who’s in control.

What Does This Mean for a Manager Making Real Decisions?

As a manager, this can put you in a genuinely awkward spot. Say you run a design team that mostly loves the creative side of the work, but clients expect delivery on time. If you introduce a strict deadline bonus, tied closely to output and checked week by week, you risk turning a task people found satisfying into one they now do mainly for the payment. Miss paying it out once, or remove it later, and motivation may end up lower than where it started.

One option is to lean on informational framing wherever we can: recognition, some autonomy over how the work gets done, and feedback tied to skill and quality rather than raw output numbers. Atlassian, for example, runs dedicated days where engineers pick their own project to work on, without a manager assigning the task or attaching a payment to a specific outcome.

We also have to think about team culture over time. If pay-for-performance schemes creep into every corner of a job, including tasks once done out of professional pride or care for colleagues, we can gradually turn a workplace built on shared commitment into one where nothing gets done unless it’s separately priced. That’s a slow, often invisible shift, and much harder to reverse than it was to create. It’s worth asking honestly, before introducing a reward system, whether a task already carried enough intrinsic pull that the system might damage motivation more than the output gain is worth.

Key Points to Take Away

Cognitive evaluation theory, developed by Deci and Ryan, explains how external rewards can increase or decrease a person’s intrinsic motivation for a task, depending on how the reward is experienced.

Rewards experienced as controlling (used to pressure or direct behavior) tend to reduce intrinsic motivation. Rewards experienced as informational (confirming competence, offered without pressure) tend to support it.
The effect mainly matters for tasks that already carry genuine intrinsic interest. Routine or unpleasant tasks with little natural appeal aren’t really at risk of this kind of undermining.
Real-world cases, like the day-care late-pickup fine study, show that attaching a price to behavior can permanently change how people think about that behavior, even after the price is removed.

Managers deciding whether to introduce performance pay, bonuses, or closer monitoring should consider not just whether the reward is fair, but how it’s likely to be interpreted, and what it might do to motivation that already existed before the reward showed up.