A cognitive approach to Korsakoff's syndrome.

Ten Korsakoff and ten alcoholic subjects were compared on their ability to take advantage of meaningful and linguistic factors in five verbal memory experiments. Korsakoff subjects were deficient in the use of visual imagery, in remembering random letters and in learning relatively easy paired-associates; there was a suggestion that they did not benefit from variation in word class. The effect of word frequency, the ability to remember very easy paired-associates, and possibly the effect of pronounceability were preserved. These findings were interpreted as supporting a model which emphasised the disruption of some higher cognitive structure whilst leaving intact simpler memory functions.