Semantic Feature Training for the Treatment of Anomia in Alzheimer Disease

Kieran J. Flanagan, David A. Copland, Sophia van Hees, Gerard J. Byrne, Anthony J. Angwin

This is a preliminary investigation into the effectiveness of semantic feature training for the treatment of anomia in Alzheimer disease (AD). Anomia is a common clinical characteristic of AD. It is widely held that anomia in AD is caused by the combination of cognitive deficits and progressive loss of semantic feature information. Therapy that aims to help participants relearn or retain semantic features should, therefore, help treat anomia in AD. Two men with AD and one man with progressive nonfluent aphasia received 10 treatment sessions focused on relearning the names of 20 animals and 20 fruits. Within each category, half of the items were of high and half were of low typicality. We individualized treatment items to each participant, using items that each had not named correctly at baseline. Treatment sessions consisted of naming, category sorting, and semantic feature verification tasks. Both participants with AD showed post-treatment improvements in naming, and one maintained the treatment effects at 6-week follow-up. The semantic category of the treatment items influenced post-treatment outcomes, but typicality did not. In contrast to the participants with AD, the man with progressive nonfluent aphasia had no improvement in naming ability. Our results suggest the potential viability of semantic feature training to treat anomia in AD and, therefore, the need for further research.