The benefits of prior sequence learning on a serial reaction time performance in Alzheimer's disease: Comparison of two learning methods

It is well known that patients with Alzheimer's disease (AD) are able to acquire new perceptual-motor skills (e.g., Rouleau et al., 2002). However, implicit learning methods should be favored, because they reduce the intervention of controlled processes related to working memory (Van Halteren-Van Tilborg, 2007). We compared two learning methods (implicit vs. declarative) of a perceptual-motor sequence in 12 patients with AD and 12 healthy older adults. In the implicit learning condition, subjects were simply asked to perform the sequence several times by pushing on the keyboard key corresponding to the stimulus on the screen. In the declarative condition, subjects learned the sequence by trial-and-error. The impact of the two methods was compared in a subsequent serial reaction time task, in which subjects had to respond as quickly as possible to the previously learned sequence. Results show that prior implicit learning is effective in both groups (p<.05). In contrast, in the declarative condition, while the two groups showed improving performance during the learning phase (p<0.01), only the control group benefits from this knowledge during the SRT task (p<0.01). In conclusion, our results show preserved perceptual-motor learning in AD when the method induces the intervention of non-declarative, automatic memory processes.