Errorless learning of a video call in Alzheimer Disease: verbal motor error dissociation as a potential marker of procedural learning (a pilot study)

Background: Dementia is rising worldwide, and in Alzheimer Disease (AD) the loss of communication adds to the burden. In AD the explicit, error correcting memory fails early, whereas procedural and implicit memory is relatively preserved, which is the basis of errorless learning. Objective: To test whether people with AD can learn an everyday communicative task, placing a video call, through errorless learning, and to characterise the learning curve and the pattern of verbal and motor errors. Methods: Twenty patients with AD learned a three step video call on a touchscreen through errorless learning, over two sessions of 15 minutes with three trials each, separated by a 15 minute interval. Execution time was recorded for each trial. Verbal errors, a correct gesture with an incorrect verbalisation, and motor errors, a wrong, approximate, or absent gesture, were scored for each step. Means are reported with 95% confidence intervals. Results: Mean execution time fell from 41.7 to 23.3 seconds and levelled off over the last two trials of the second session. Verbal and motor errors followed significantly different courses across steps: verbal errors were present from the first trial and fell close to zero by the last step of the second session, whereas motor errors increased. Conclusion: People with AD acquired the procedure, and the dissociation between verbal and motor errors is consistent with a dissociation between procedural performance and explicit verbal report during acquisition. These preliminary findings support the use of preserved procedural memory in rehabilitation.