Effects of immediate self-correction, delayed self-correction, and no correction on the acquisition and maintenance of multiplication facts by a fourth-grade student with learning disabilities.

One fourth-grade student with learning disabilities practiced multiplication facts at school.

The teacher tried three ways to handle wrong answers: immediate self-correction, delayed self-correction, or no correction.

They used an alternating-treatments design, switching the rule each day to see which helped most.

Immediate self-correction won. The child answered more facts correctly and kept the skill weeks later.

Delayed correction and no correction scored lower and faded faster.

Xenitidis et al. (2010) later asked a new question: once you pick immediate correction, how should you pace the practice? They cut one-minute drills into three 20-second sprints and kids with ADHD learned 30% more facts. The two studies stack nicely: first fix errors fast, then chop the timer.

Stocker et al. (2019) looked at 20 years of math-fluency studies and found that most set a speed aim, like 60-80 correct digits per minute. The 1998 paper did not use such aims; adding them later could lift outcomes even higher.

Neef et al. (1986) seems to clash. They used overcorrection for spelling and also got big, lasting gains. The gap is only in the subject: both show that quick, active error treatment beats gentle or no feedback.

If a student stalls on math facts, hand them the answer right after each error and have them write it once. This tiny move beats waiting or skipping the mistake. Pair it today with short timed bursts and clear fluency aims to squeeze even more growth out of every minute.