Skip to main content
. 2009 Nov 11;29(45):14160–14176. doi: 10.1523/JNEUROSCI.1916-09.2009

Figure 2.

Figure 2.

Effect of cue on fraction of speed pulses detected (FD). A, B, Distribution of performance improvement with cue for individual behavioral sessions (valid-cue FD − invalid-cue FD). Data are shown separately for no-switch (A) and switch (B) trials. The triangle marks the median value for each monkey. For details, see Results. C–F, Fraction of speed pulses detected is plotted as a function of the time the speed pulse occurred. Data from all behavioral sessions were combined. Data are shown separately for speed pulses occurring at the cued dot patch versus uncued dot patch. For no-switch trials (C, E), speed-pulse times are plotted with respect to dot onset. For switch trials (D, F), speed-pulse times are plotted with respect to the cue switch. Data were smoothed using a sliding window four video frames (53 ms) wide. Shaded regions are 95% confidence intervals for the mean. G, H, False-positive response rate as a function of time relative to dot onset (G) and cue switch (H) for monkey M (red) and monkey B (blue). The pattern of false-positive responses is similar for the two animals despite a fivefold difference in average false-positive rates. Below each panel is the hazard rate function for the speed pulse. The hazard rate is the probability that the speed-pulse would occur at that time given that it had not occurred previously.