highly significantp < 0.0001
The median overlap was much larger for Stimulus Set 1 than for Stimulus Set 2, and such a difference was highly significant ( p < 0.0001, Mann–Whitney U -test test; see Figure 11A ).
The median overlap was much larger for Stimulus Set 1 than for Stimulus Set 2, and such a difference was highly significant ( p < 0.0001, Mann–Whitney U -test test; see Figure 11A ).
As for Object 4, the correlation with the ideal saliency map was either null, or failed to reach significance, in all those cases where the average map was highly scattered (i.e., see the default, size, azimuth right, and position right views shown in Figure 12B , top row).