Progressing from programmatic to discovery research: a case example with the overjustification effect.

The team set up a simple lab task. People could press a button to earn points.

First they paid points for every press. Then they stopped paying.

They watched what happened to the pressing rate after the pay stopped.

Pressing went up when the points ended. That is the opposite of the over-justification idea.

The authors say the points had acted like a fine, not a prize. The task itself was already fun.

Sailor (1971) saw the same jump years earlier. When food was skipped, pigeons pecked faster. Both studies show that losing a reinforcer can lift response rate.

Ward et al. (2017) flipped this idea into a treatment. They gave kids a quick break from work and escape behavior dropped. Their brief wait-out is a mirror image: losing a liked activity can punish, just as gaining it can reinforce.

Griffith et al. (2012) add why timeout works. They showed rats pressed a lever to earn a break from hard work. The break was reinforcing because it cut effort, not because it removed shock. Together these papers say removal effects depend on what the client values at that moment.

Before you stick a sticker chart on a client, test if the task is already preferred. If the child likes sorting colors more than getting candy, your candy may act like a tiny fine. Run a quick paired-choice probe first. When the reward is less liked than the job, skip the reward or pick a better one. This one step keeps your program from accidentally punishing the very behavior you want to grow.