Method, device and database for reconstructing intended activities from neural signals

A signal processing means (SVE) for reconstruction of intended activities, which can be coupled to a database (DB) and an interface (I), wherein the interface (I) is adapted to receive signals (S) from a man-machine interface (MMI), wherein the signals represent a neural activity of the brain of a human being, and wherein the signal processing means is adapted to

• generate a first representation (R1) from neural signals (S) using a predetermined method,
• determine, for second representations (Ri) from a plurality of predetermined second representations (RR), which respectively are representative for a predetermined activity, wherein the predetermined second representations (RR) are generated from predetermined activities, respectively a degree of correspondence between the respective second representation (Ri) and the first representation (R1) according to a predetermined correspondence criterion, and
• depending on the degree of correspondence, select a second representation (R2) of the neural signals (S) from the plurality of second representations (RR) as the second representation (R2) representative of the intended activity, and to reconstruct the activity represented by the selected representation (R2) as the intended activity.