Fig. 7From: A simulation study on the effects of neuronal ensemble properties on decoding algorithms for intracortical brain–machine interfacesJoint space of PWTN and UWTN for each decoder. The top row contains AD topological plots, which represent the relationship between UWTN and PWTN. The character K, O, and P corresponds to the KF, OLE, and PVA, respectively, and denotes the other decoder whose performance is statistically different from the given decoder. For example, P on the KF plot indicates that the AD of the PVA is statistically different from that of the KF (p < 0.01). The bottom row indicates an AD difference map between an upper triangular space and a lower triangular space divided by a black-dotted diagonal boundary. Asterisks indicate that two conditions symmetric with respect to the diagonal boundary yielded statistically different ADs (Watson’s U2 test, p < 0.01)Back to article page