Fig. 2

Diagram of the set of five cue selection experiments used to evaluate different cue types. There were 100 participants in total (n = 20 per cue selection experiment); each pair of bars represents one experiment, each with a different group of participants. The y-axis on each bar graph indicates the difference in reaction time (in seconds) between the invalid and valid cue conditions, with positive values indicating slower reaction times in the invalid cue condition. Note that in the main experiment, the RT difference between valid and invalid spatiotemporal cues was approximately 120 ms. Error bars indicate ± 1 SEM. a The first set of three experiments measured the cue validity effect for the static versus flashing zebra cues, the expanding versus contracting rings, and the red bounding box versus flashing red dot. b The three cue types with the larger RT difference within each paired comparison (the static zebra, the expanding ring, and the red bounding box) were used in two additional paired comparisons: static zebra versus expanding ring and expanding ring versus red bounding box. In all 5 experiments, all pairwise differences between cue types conditions were not statistically significant, but the cue with the largest invalid-to-valid RT difference within the second set of experiments—the expanding ring—was chosen for the main experiment