Cognitive mechanisms of errorless learning

In this article, neuropsychologist Lidia García Pérez explains the cognitive mechanisms of errorless learning and considers when it would be a good option for cognitive rehabilitation.

Errorless learning is a method commonly used in the neuropsychological rehabilitation of memory, given its advantage for amnesic patients over traditional learning, which allows and takes advantage of errors. However, there is evidence that its results can vary significantly depending on various factors such as the severity of the damage, the specific learning task, or the patients’ neuropsychological profile [1].

So, is errorless learning the best therapeutic option for memory deficits? And for other deficits, such as executive function deficits?

In this post, I review the cognitive mechanisms that have been proposed to explain the advantage of errorless learning over errorful learning in memory rehabilitation, so that this update can help determine when this type of technique may benefit patients in cognitive neurorehabilitation.

What is errorless learning?

Errorless learning refers to a form of training that tries to prevent the person who is starting to learn certain information from making errors, as opposed to traditional learning, where errors are part of the learning process itself [e.g., 1].

In clinical practice, this approach is applied as a guiding principle in the rehabilitation of patients with severe memory deficits, since they are at particular risk of learning the errors they make (rather than learning from them).

Method of errorless learning

Although there are different procedures, the standard errorless learning method involves presenting the patient with a fragment of the information to be learned (for example, a degraded image of an object or the stem of a word), immediately afterward also presenting the complete correct answer, and then making sure the patient encodes it again in some way (repeating it aloud, writing it down, etc.).

This procedure means basing learning exclusively on the (repeated) processing of information, avoiding retrieval from long-term memory through free recall in order to minimize the probability of error.

However, we know that one of the well-established principles of learning and memory is that retrieval practice in itself benefits the recall of that information (and not because of the additional processing it involves [2, 3]). Therefore, to the extent that it is restrictive and avoids attempts to retrieve information, errorless learning is a passive way of learning that doesn’t take advantage of the effects of retrieval practice.

Trial-and-error methods, on the other hand, do encourage the patient to make retrieval attempts, which helps consolidate learning.

A recent systematic review of studies testing the advantage of errorless learning over errorful methods in different populations with memory problems [1] concluded that, although the evidence generally suggests that this approach can be useful in memory rehabilitation, not all groups with this type of impairment seem to benefit more from errorless learning than from errorful learning, and that errorless learning also has some limitations.

So why is errorless learning considered the method of choice for rehabilitating primary memory disorders?

Cognitive mechanisms behind the advantage of errorless learning

Errorless learning began to be developed in the rehabilitation of patients with severe anterograde amnesia, after it was observed that, despite their severe explicit memory impairments, their procedural memory remained intact [4, 5].

The first studies comparing memory performance after errorless and errorful learning [4, 5] found that, indeed, patients with amnesia learned more words when they were trained so that they didn’t make errors than when a procedure in which they had made errors was used.

Working memory and short-term memory: definition and differences

These early studies interpreted the superiority of errorless learning as being due to how implicit memory works: during implicit learning, encoding depends on the strength of the response to the stimulus, regardless of whether the response is correct or incorrect, so implicit recall doesn’t distinguish between errors and correct answers, but between strong and weak associations. Therefore, the authors of these studies concluded that eliminating errors improves amnesic patients’ learning because it avoids the influence and confusion caused by the activation of errors.

Since then, the debate about the memory mechanisms behind the advantage of errorless learning has basically revolved around two positions: the implicit memory–based explanation and the explicit memory–based explanation [e.g., 1, 6, 7].

Implicit memory–based explanation

The implicit explanation states that it is the explicit memory impairment of patients with amnesia that prevents them from remembering the errors they make, so since they can’t remember them and take them into account on later occasions, they keep making errors, because they can only base their learning on implicit memory.

Explicit memory–based explanation

The explicit explanation states that these patients benefit more from errorless than from errorful learning because they make use of residual explicit memory. Its argument is based on the observation that patients with better-preserved explicit memory performed better after errorless learning than patients with more severe explicit memory impairments [7].

Source memory–based explanation

A third explanation refers to source memory, which involves recalling the context of an event separately from recalling its content [1, 7].

According to this proposal, impaired source memory would cause a specific difficulty in distinguishing items learned through the therapist’s feedback from self-generated items, so the person confuses the information to be learned with their own errors. Thus, to the extent that this type of memory is impaired, this hypothesis predicts that errorless learning will be a better option than errorful learning because it avoids the interference of these errors with recall.

 

Other possible applications of errorless learning: attention or executive function impairments

The ability to detect and monitor errors and to modify behavior based on feedback therefore seems to be central to explanations of why errorless learning works better than trial and error in patients with amnesia [1].

On the other hand, it has been suggested that source memory is partly mediated by the frontal lobe [1, 7], particularly by the left and right prefrontal cortices, which are areas involved in attention and executive control functions, as well as in error detection and adjustment by comparing inputs or incoming stimuli with internal representations in long-term memory.

All of this has led some authors to consider the involvement of executive functions in the advantage of errorless learning. Recently, a key role has been proposed for attention and executive control processes, which would influence basic information processing, working memory, and episodic memory [1].

Specifically, the authors suggest that, through the allocation of sustained attention resources, executive control processes “guide stimulus processing, enable goal-directed access to and manipulation of internal representations, and support the maintenance of those representations. They therefore enable the comparison of an external stimulus with stored internal representations and the integration of memory for the content of an event with knowledge of its source or context (…). Consequently, the effectiveness of approaches (…) may vary in part depending on the degree to which they make appropriate demands on, and support the optimal functioning of, attention processes” [1].

Therefore, according to this proposal, insofar as attention processes and error monitoring are critical for learning, and insofar as they are compromised in neurological patients, it will be important to consider applying errorless procedures not only in patients with primary memory disorders, but also in the rehabilitation of patients with impairments in sustained attention and/or executive functions.

References

• Clare L& Jones RS (2008). Errorlesslearning in therehabilitation of memoryimpairment: A criticalreview. NeuropsycholRev, 18:1-23.
• Carrier M&Pashler H(1992). Theinfluence of retrievalonretention. Memory andCognition, 20: 633.
• Ruíz Rodríguez, M. (2004). Las caras de la memoria. Madrid, España: Pearson Educación.
• Glisky EL, Schacter, DL, Tulving E (1986). Learning and retention of computer-relatedvocabulary in memory-impairedpatients: method of vanishingcues. Journal of Clinical and Experimental Neuropsychology, 8(39): 292-312.
• Baddeley A & Wilson BA (1994). Whenimplicitlearning fails: amnesia and the problem of error elimination. Neuropsychologia, 32(1): 53-68.
• Anderson ND&CraikFI (2006).Themnemonicmechanisms of errorlesslearning. Neuropsychologia, 44(14): 2806-13.
• Anderson ND, Guild EB, Cyr AA, Roberts J, Clare L (2012).Contributions of frontal and medial temporal lobefunctioning to errorlesslearningadvantage. NeuropsycholRehabil, 22(2):169-86.

If you enjoyed this article about the cognitive mechanisms of errorless learning, you may also be interested in the following articles:

“This article has been translated. Link to the original article in Spanish:”
Mecanismos cognitivos del aprendizaje sin error