Errorless learning of computer-generated words in a patient with semantic dementia

Abstract

This study explores the effectiveness and feasibility of an errorless learning approach administered via a computer-based treatment for anomia to CS, an individual with semantic dementia. Using a multiple baseline across behaviours design, we explored treatment specific effects, maintenance and generalisation of gains derived from the MossTalk Words^® therapy programme. CS was treated on three lists of words, each containing items for which CS retained some semantic knowledge and some for which he did not. CS was tested immediately after therapy, and one and three months later. Improved naming was maintained on all lists at all testing intervals. In addition, among those words for which CS retained some semantic knowledge, he maintained the ability to name all practised words, but only half of the not practised words. This study underscored the feasibility of computer-based treatments for anomia in progressive disorders, demonstrated the effectiveness of an errorless approach in semantic dementia in re-training lost words, and provided justification for training words that patients still have in their daily vocabulary. The results are discussed in relation to other treatment studies in progressive aphasia and in the context of factors necessary for therapeutic success in semantic dementia.

Keywords:

• Semantic dementia
• Errorless learning
• Computer-based approach
• Anomia
• Language therapy

Acknowledgments

We are indebted to CS for his motivation and tenacity. We also thank Ruth Fink for making MossTalk Words^® available to us. This study was supported by a Canadian Institutes in Health Research (CIHR #159058) Institute of Aging Post-Doctoral Fellowship to R. Jokel, and an NSERC grant to N. Anderson. The authors acknowledge the support of the Toronto Rehabilitation Institute, which receives funding under the Provincial Rehabilitation Research Program from the Ministry of Health and Long Term Care in Ontario, Canada. Jennifer Cupit provided valuable assistance on this project.

Notes