Skip to main content
Fig. 4 | BMC Neuroscience

Fig. 4

From: A dimension reduction technique applied to regression on high dimension, low sample size neurophysiological data sets

Fig. 4

Direction obtained for the RoLDSIS procedure for a representative participant (the same as in Fig. 2. The results of the RoLDISS for the physical (\(\Phi\)) and psychophysical (\(\Psi\)) observations are shown in the top and bottom panels, respectively. In each panel, the time-domain representation of the optimal direction vector, obtained by applying the inverse DWT on the RoLDSIS result is shown by the black line, which is atop of the scalogram (time/frequency representation) of this direction vector. The amplitudes of the DWT coefficients are represented in a color scale, negative values in blue and positive values in red. The more saturated the color in a cell, the higher is the magnitude of the DWT coefficient associated with that cell. Frequency bands of the DWT are shown in increasing order from bottom to top (V8: 0–9.76 Hz, W8: 9.76–19.5 Hz, W7: 19.5–39.1 Hz, W6: 39.1–78.1 Hz, W5: 78.1–156 Hz)

Back to article page