Brain-Computer Interfaces for Cognitive Rehabilitation: A Narrative Review

Cognitive diseases/disorders are a major cause of disability affecting millions of people worldwide due to pathologies or conditions such as Alzheimer's disease, Parkinson's disease, vascular cognitive impairment, and attention deficit/hyperactivity disorder. The effects of these diseases/disorders can range from subjective cognitive complaints to mild cognitive impairment and even dementia that can affect multiple cognitive domains, causing disability and alterations in daily functioning and quality of life. Cognitive training has been researched as a potential therapeutic option for treating cognitive disability. In particular, cognitive training delivered using brain-computer interfaces (BCIs) represents a promising technology for cognitive rehabilitation. For this reason, this work presents a narrative review aimed at bridging the clinical aspects of cognition performance and the technical features of BCI systems. Additionally, it reports on the current advances of BCIs for cognitive rehabilitation in the pathologies/conditions that are the main contributors of cognitive impairment globally. Specifically, Alzheimer's disease, Parkinson's disease, mild cognitive impairment, cerebrovascular diseases, and attention-deficit/hyperactivity disorder were included in the review; other pathologies that can also cause cognitive impairment such as multiple sclerosis, traumatic brain injury, brain tumors, infectious diseases, and psychiatric conditions such as schizophrenia or autism spectrum disorder, or drug or toxin-related causes were outside the scope of the present review. Across the literature, studies have reported that BCI cognitive training can improve cognition performance for mild cognitive impairment, stroke, and attention-deficit/hyperactivity disorder, showcasing its feasibility as a complementary therapy. However, the amount of evidence is still scarce, with few reported randomized controlled clinical trials and all of the applications comprising non-invasive systems using electroencephalography, while invasive or other non-invasive approaches remain largely unassessed. Therefore, further research regarding the design of BCIs for targeting specific cognitive domains is necessary to allow for the future introduction of these types of systems into clinical rehabilitation environments, rather than only demonstrating their feasibility, as has been mainly done to date.