Cognitask: BCI system based on P300 for cognitive therapies

Brain-computer interfaces can be used within a therapy rehabilitation of cognitive functions such as attention and working memory. This paper presents the design, implementation and preliminary evaluation of Cognitask, a BCI for cognitive rehabilitation. It consists of a modified version of the P300-based speller BCI that allows the presentation of different cognitive tasks. During a task, the patient must complete a certain pattern by ordering a set of images presented disorderly in a visual stimulation matrix. The selection of each image is recognized using the P300 elicited when the patient looks at the matrix where images are flashing and selects one by attending to it. Cognitask consists of four modules. The first acquires and conditions the electroencephalography signal from six electrodes positioned at the fronto-central and parieto-occipital regions. The second processes the signals and translates them into a control signal. The third consists of a user-interface that presents the cognitive task to the patient. The last module consists of a graphical interface for the professional that allows configuration of the session parameters. Two evaluations of Cognitask were performed. The first evaluated the operational functioning of the system using input signals generated by a software. The second evaluated the performance of the system in three healthy volunteers. The results showed that Cognitask had a correct functioning and a maximum average success rate of 92%. These results suggest that Cognitask can be transferred to the clinical setting for evaluation in adults with cognitive deficits.