What Is Errorless Learning?
A plain guide for parents and teachers: what the term means, where it came from, how the prompts are supposed to be removed, when it beats letting a child make mistakes, and the one failure mode that everybody runs into.
The short answer
Errorless learning means arranging a task so the learner gets it right almost every time, by giving enough help up front that a wrong answer is very unlikely, then removing that help a bit at a time. The mistake is prevented rather than corrected.
The name oversells it slightly. Nobody claims zero errors ever; the aim is to keep them rare enough that the learner mostly experiences being right, and never spends long practising the wrong thing.
Why preventing an error can beat correcting one
The intuitive model of learning is that you try, you get it wrong, someone corrects you, and you remember the correction. That works well when two conditions hold: you can remember what you did, and you can tell the correction apart from the mistake afterwards.
When either condition fails, trial and error starts working against the learner. The wrong attempt gets practised and remembered alongside the right one, and later the learner cannot recall which was which. They just recall doing something, with a faint sense of familiarity attached to both.
There is a second, less technical reason, and for most children it is the bigger one. A child who has failed at this kind of task repeatedly does not arrive neutral. They arrive braced, and a great many stop trying long before any memory effect matters. For that child, a run of correct answers is not just efficient teaching, it is the thing that gets them to sit down at all.
Where it came from
Three stages, roughly.
- Pigeons, 1963. Herbert Terrace showed that if you fade a difference in gradually rather than presenting it all at once, a pigeon can learn a discrimination while making almost no errors, compared with hundreds of wrong responses under the standard method. That is the origin of the term.
- Rehabilitation, 1990s. Alan Baddeley and Barbara Wilson adapted it for people with amnesia, and found they learned considerably better when prevented from guessing wrong in the first place. This is where the memory argument above comes from, and it is still the strongest evidence base the method has.
- Classrooms and clinics, since. It became standard practice in special education and applied behaviour analysis, and has been reviewed specifically for teaching children with developmental disabilities, including Mueller and colleagues' 2007 review in Psychology in the Schools.
How it is actually done
Two levers, usually used together.
1. Make the task easy enough that failing is hard
You start with a version of the task the learner cannot really get wrong, and you make it harder in small steps. Teaching a child to pick out the cup:
- The cup, alone on the table. Nothing to choose between.
- The cup and something wildly unlike it. A choice, but not a hard one.
- The cup and something a bit similar.
- The cup and three things, one of which is a different cup.
At no point is the child left guessing between things they cannot yet tell apart. This is the same logic as building up a visual discrimination from obviously different shapes to nearly identical ones.
2. Prompt immediately, then fade the prompt
The other lever is helping before the learner has a chance to go wrong, and then making that help smaller. Two standard ways to fade:
| Method | What changes | Example |
|---|---|---|
| Most-to-least prompting | The prompt gets weaker each time. | Hand over hand, then a touch on the elbow, then pointing, then a glance, then nothing. |
| Time delay | The prompt stays the same but arrives later. | Prompt straight away, then wait 2 seconds before prompting, then 4, then 6, giving the learner room to answer first. |
| Stimulus fading | A feature of the material itself is dialled down. | The right answer starts larger or brighter than the others, and that advantage shrinks until every option looks the same. |
The fading is not an optional extra at the end. It is the method. Errorless teaching without fading is not errorless teaching, it is doing it for them.
The prompt dependency problem
Prompt dependency is when the learner stops answering and starts waiting. Not because they cannot do it, but because they have learned that if they pause long enough, the help arrives. It is the single most common way errorless teaching goes wrong.
It is entirely rational behaviour on the learner's part. If waiting has always produced the answer, waiting is the efficient strategy, and they have correctly worked out the rules of the situation you built.
The prevention is unglamorous: decide up front how the prompt will be faded and on what evidence, then actually fade it. Time delay is popular for exactly this reason, because the pause deliberately creates a moment where the learner can get in first. If you find yourself six weeks in and still giving the same level of help, that is not a plateau in the child, it is a plan that never had a next step.
Errorless or trial and error?
The honest answer is that it depends on the learner and the task, and that anyone who tells you one is simply better than the other is not describing the evidence.
| Errorless tends to win when… | Trial and error tends to win when… |
|---|---|
| The learner cannot reliably remember which attempt was the right one, as with significant memory impairment. | The learner can remember what they tried and compare it with the feedback. |
| Mistakes cause enough distress or refusal that the session ends. | Mistakes are tolerable and recovery is quick. |
| The wrong version is dangerous or hard to unlearn. | The wrong version is harmless and informative. |
| You are establishing a brand new skill from nothing. | You are strengthening or generalising a skill that already exists. |
| Confidence is currently the bottleneck, not ability. | The learner needs to build up tolerance for being stuck. |
There is also a real argument on the other side that is worth knowing about: for learners who are not memory impaired, effortful retrieval and even guessing wrong can produce better long-term retention than being handed the answer. Struggle is not automatically waste. So errorless is best understood as the right tool for a particular situation rather than a superior philosophy of teaching, and a child who is now succeeding easily is a child who should be moved on.
How these games use it
A softened version, and it is worth being precise about which parts are and are not the real method.
What the games do take from it: the difficulty starts low and moves in small steps, from two choices to four, with wrong answers that begin obviously different and become similar. A wrong choice is never punished, never buzzed at and never scored. It is named out loud and set aside, and the child carries on with what is left, so the round always ends with them getting it right rather than with a failure screen. And there is a help button that highlights the correct answer and names it, which is a prompt available on demand rather than one imposed.
What they do not do: fade prompts on a schedule, collect data, or run a protocol. There is no plan tracking a particular child across sessions, because a web page does not know who is holding the tablet. That part is yours, and it is the part that stops prompt dependency, so if you are using these deliberately rather than casually, the thing to watch is whether the help button is being pressed before the child has looked at the screen.
What this page is: general information written by the person who builds these games, not by a clinician or a behaviour analyst. If you are choosing a teaching approach for a specific child, that is a conversation with the professionals who work with them.
Common questions
What is errorless learning in simple terms?
Teaching arranged so the learner gets it right nearly every time, by giving enough help up front that a wrong answer is very unlikely, then removing that help gradually. Mistakes are prevented rather than corrected.
What is an example of errorless learning?
Teaching a child to point to the cup by first putting the cup alone on the table, then adding one very different object, then a similar one, then three. At no stage are they guessing between things they cannot yet tell apart. A matching game that starts with two obviously different choices and works up to four similar ones is the same idea.
Is errorless learning better than trial and error?
Not always. It has the strongest support where the learner cannot reliably remember which of their attempts was correct, such as significant memory impairment, and it is widely used with children who disengage after mistakes.
For a learner who can remember and recover from a mistake, effortful attempts and even wrong ones often produce better long-term retention than being handed the answer.
What is prompt dependency?
When the learner waits for the help instead of answering, because the help has always arrived. It is the main failure mode of errorless teaching, it is caused by never fading the prompts, and it is entirely rational on the learner's part.
Where did errorless learning come from?
From Herbert Terrace's pigeon discrimination experiments in 1963, which showed that fading a difference in gradually could produce learning with almost no errors. Alan Baddeley and Barbara Wilson adapted it for people with amnesia in the 1990s, and it is now common in special education and applied behaviour analysis.
Do the games on this site use errorless learning?
A softened version. Difficulty climbs in small steps, wrong answers are set aside rather than punished so every round ends in success, and a help button gives a prompt on demand. They do not fade prompts on a schedule or collect data, which is the part a person has to do.