A parametric analysis of procedural fidelity errors following mastery of a task: A translational study

Falakfarsa et al. (2023) asked what happens when you let small mistakes slip in after a learner has already mastered a task.

They worked with college students who had just learned a matching task. The team then ran short sessions where the experimenter followed the script perfectly, made tiny errors, or made bigger errors.

The students’ accuracy and speed were tracked to see if post-mastery slip-ups mattered.

Even after the students hit mastery, higher fidelity kept their performance high.

When the experimenter drifted—even a little—the students started to match more slowly and make more errors.

The takeaway: mastery does not protect against the damage of sloppy delivery.

Sarber et al. (1983) saw a similar fade-out effect, but with probe difficulty instead of fidelity. They warned that hard probe trials after training can fake you out with false errors; Falakfarsa shows the errors can also come from the teacher’s side.

Parry-Cruwys et al. (2022) proved that short, mastery-based online modules can quickly bring grad students to 90 % plus accuracy. Falakfarsa adds the next step: once that accuracy is reached, you have to keep running the script exactly or it will drop.

Together the three studies draw one clear line: mastery is not a finish line—it is a maintenance zone that still needs precision.

If you run fluency checks, maintenance trials, or parent training, keep your own behavior tight. A drift in prompt timing, wording, or reinforcement size can quietly undo the skill you just taught. Build a simple fidelity checklist and use it every time you revisit "mastered" targets.