Combining errorless learning and vanishing cues

Errorless learning and vanishing cues

This blog post is a part of Research Tuesday, which encourages bloggers to write about recent research once a month. The goal is to read more research as well as share research with our readers to improve the EBP in the therapy community.

As many of you may know, I am quite the fan of errorless learning and vanishing cues. In my experience, I have found combining errorless learning and vanishing cues can lead to better patient outcomes. So when I came across the following article investigating the success of combining the two versus isolated methods, I wanted to take the opportunity to see what the research says compared to my experience.

Haslam, C., Moss, Z., & Hodder, K. (2010). Are two methods better than one? Evaluating the effectiveness of combining errorless learning with vanishing cues. Journal of Clinical and Experimental Neuropsychology. 32(9), 973-985.

Two experiments

Experiment 1
Participants included were 60 healthy adults with the average age of 22 years. They were recruited from the university setting. Participants were asked to complete a dual task, requiring divided attention and utilization of implicit memory.

The primary task was learning a set of 16 greeble-surname associations. The secondary task was monitoring taps on the back of their hands. They were to press a foot pedal when they felt 3 consecutive taps on the same hand.

Three learning conditions were provided. First, errorless learning only. Second, vanishing cues only. And third, a combination of errorless learning and vanishing cues. Under the combined approach only the first trial was errorless, due to the nature of how they facilitated a vanishing cues method. (Their approach to vanishing cues combined with errorless learning is different than mine. See my explanation below.) Baseline measures were obtained for all participants using a trial and error learning method.

Experiment 1 results
Results for the standard task and dual task were very similar. Errorless learning was better than vanishing cues. Errorless learning was no different than errorless learning plus vanishing cues (using the Bonferroni correction for multiple comparisons). The authors concluded healthy adults do not benefit from a combined approach.

Experiment 2
Participants included 22 adults with Alzheimer’s disease and an average age of 73.5 years, who were recruited via a memory clinic. Their average MMSE score was 24 (SD=4), which is consistent with mild to moderate cognitive impairment. All participants were living at home. Sixteen of the participants were taking medication for cognitive complaints (e.g. donepezil and galantamine). A full assessment to determine cognitive profile was completed for each participant. Participants were asked to learn face-first name associations.

Four learning conditions were provided to each participant: trail and error, errorless learning only, vanishing cues only, and errorless learning combined with vanishing cues. Procedures were modeled after experiment 1 procedures; however, no dual tasks were provided.

Experiment 2 results
The authors found that errorless learning and vanishing cues in isolation provided greater results than trial and error. Combining the two methods was better than just vanishing cues; however, a combined approach was not better than errorless learning alone.

What does this mean for practice?

For me, not much. Errorless learning is still an excellent method for helping people with memory impairment. I was surprised by the authors’ approach to a combined errorless learning and vanishing cues method. The article explained their approach (in experiment 1) as, “Participants were presented with the complete greeble–surname association on the first trial, and letters were gradually removed on subsequent trials. In practice, on 70% of trials this entailed the removal of one letter at Trial 2, two letters at Trial 3, and three letters at Trial 4. However, if the correct name was not produced after 10 s on any trial, the next letters of the name were presented until the correct name was produced, and letters were removed on subsequent trials, as for the vanishing cues group.”

This is different from how I approach a combined method. I have no research-based evidence (only personal experience) to support my approach, but as I initially read research about combining the two, this is the way I interpreted the information to combine errorless learning and vanishing cues.

I think about errorless learning as targeting accuracy and fading cues as a way to improve independence. I provide as many cues a patient needs in order to consistently provide the targeted response. I usually do 4-5 trials before working on fading my cues. If the person is successful with the faded cues (and I fade very gradually considering tactile, visual, and verbal cues), then I do another 3-4 at that same level before fading again. At any point if they are unsuccessful with that level of cues, then I go backup to the prior level (where they were successful) and do another 4-5 trials before attempting to fade again.

Rather than only the first trial being errorless in the case of these experiments, I should have 3-4 trials that are correct responses. Then as I fade the cues, an opportunity for error does present, but I am quick to return to that errorless level of cues. Again, I do not have any research-based evidence to support this specific approach. This is what my brain came up with as I learned about errorless learning and vanishing cues and considered combining them.

Sometimes I add in spaced retrieval therapy. Rather than decreasing cues, I’ll increase time between trials. I don’t know that there is any specific cookbook type of approach that will work for all clinician-patient pairs, but this approach has been successful for me. Try combining several approaches together and see what it does for your patients’ success.

Want to learn more about errorless learning and vanishing cues?

What is errorless learning? on Gray Matter Therapy
Errorless learning for memory impairment on ADVANCE Outlook: OT
Vanishing cues for memory impairment on Gray Matter Therapy
Tips for using spaced retrieval therapy on ADVANCE Speech and Hearing Perspectives

Rachel Wynn
Follow me

Rachel Wynn

Speech-Language Pathologist at Gray Matter Therapy
Rachel is a speech-language pathologist and creator of Gray Matter Therapy. She started making noise as a patient-centered care advocate in 2013. She believes great care happens when patients are informed and engaged.
Rachel Wynn
Follow me