Since elderly people suffering from dementia want to go on living independently for as long as possible, they need to be able to maintain familiar and learn new practical skills. Although explicit or declarative learning methods are mostly used to train new skills, it is hypothesized that implicit or procedural techniques may be more effective in this population. The present review discusses 23 experimental studies on implicit motor-skill learning in patients with Alzheimer’s disease (AD). All studies found intact implicit motor-learning capacities. Subsequently, it is elaborated how these intact learning abilities can be exploited in the patients’ rehabilitation with respect to the variables ‘practice’ and ‘feedback.’ Recommendations for future research are provided, and it is concluded that if training programs are adjusted to specific needs and abilities, older people with AD are well able to (re)learn practical motor skills, which may enhance their autonomy.

Older patients with mild dementia are able to acquire new skills that are relevant for daily life, showing a similar rate of learning regardless of whether implicit or explicit learning techniques are used.

Experimental studies show intact implicit motor learning in patients with Alzheimer's disease (AD) but the results for patients with Parkinson's disease (PD) are inconclusive. This study tests implicit sequence learning in AD and PD patients, and healthy controls, using the classical Serial Reaction Time Task (SRTT), and a somewhat similar Pattern Learning Task (PLT), which involves stylus movements in different directions, and which allows detailed movement analysis. As expected, the time measures showed less implicit motor learning in the PD patients relative to the other groups in both tasks, but their error percentages increased when the sequence changed from a fixed to a random order, which is indicative of implicit learning. The AD patients showed a reversed pattern of results. Arguably, errors and time measures may reflect the involvement of separate processes, e.g., spatial and motor components, which could be differently affected in AD and PD.

Patients with amnesia have deficits in declarative memory but intact memory for motor and perceptual skills, which suggests that explicit memory and implicit memory are distinct. However, the evidence that implicit motor learning is intact in amnesic patients is contradictory. This study investigated implicit sequence learning in amnesic patients with Korsakoff’s syndrome (N = 20) and matched controls (N = 14), using the classical Serial Reaction Time Task and a newly developed Pattern Learning Task in which the planning and execution of the responses are more spatially demanding. Results showed that implicit motor learning occurred in both groups of participants; however, on the Pattern Learning Task, the percentage of errors did not increase in the Korsakoff group in the random test phase, which is indicative of less implicit learning. Thus, our findings show that the performance of patients with Korsakoff’s syndrome is compromised on an implicit learning task with a strong spatial response component.