The results of this experiment appear in [this thesis.][2] This project's preregistration may be viewed [here.][1] The results of the first experiment suggested that both enhancement of the attended object and suppression of the ignored object operate independently to affect noticing rates during sustained inattentional blindness. In the previous experiment, the stimuli we used were designed to vary with respect to only one set of objects, and to remain constantly similar to the other. However, the variance still operated along the a feature dimension that could be used to separate the two object sets, namely luminance. In this control experiment, the unexpected object varies along a dimension orthogonal to that which differentiates the attended and ignored sets. We replace the previous white squares with a set of gray rectangles, which are longer than they are wide, and pit them against black and white checkerboards. The unexpected object can be either a vertical gray rectangle, perfectly matching one set, or a horizontal gray rectangle. If the filters are sufficiently sensitive, then the dual-filters hypothesis predicts the following data (see also the uploaded figures in OSF storage): * When attending to the gray rectangles and ignoring the checkerboards, enhancement of the attended set predicts that the vertical unexpected bar will be noticed more often than the horizontal unexpected bar, as the former is more similar to the attended set. Both bar orientations are equally similar to the checkerboard, so the effect of the ignored set will be held constant. * When attending to the checkerboards and ignoring the rectangles, suppression of the ignored set predicts that the vertical bar will be noticed less often than the horizontal bar, since the former is more similar to the ignored set and so should be suppressed. Again, similarity to the attended checkerboards is constant between the two unexpected objects. If the filters are insufficiently sensitive to the attended object, we may not observe a difference between the two unexpected orientations. In that case, we should see similarly high noticing for the unexpected rectangles when attending rectangles, and similarly low noticing when ignoring them. [1]: https://osf.io/wc2aa/?view_only=2a65d093c2f24affab62e9bf8dde766b [2]: https://www.ideals.illinois.edu/items/105273