Table 1 Table describing the six cues used in the cue selection experiments and the reasoning behind their inclusion in the set of candidate cues
From: Effects of temporal and spatiotemporal cues on detection of dynamic road hazards
Cue type | Reasoning | Description |
|---|---|---|
Red bounding box | Likely highly salient as compared to the larger road environment (e.g., green plants or grey/black roadways); (Bauer et al., 1996) | Red rectangular outline shown for 33 ms (1 frame), corresponding to the annotated location, drawn with a 5 px stroke width |
Static zebra-striped bounding box | Maximizes luminance contrast of the cue versus background for arbitrary background luminance; increases cue salience (Engmann et al., 2009) | Black and white frame shown for 167 ms (5 frames), corresponding to the annotated location, with adjacent concentric alternating black and white outlines (white-black-white-black at 5 px each, for a total stroke width of 20 px) |
Flashing zebra-striped bounding box | Maximizes luminance contrast of the cue versus background for arbitrary background luminance, may capture attention better than a static version (Stolte & Ansorge, 2021) | As above, but the black and white elements reversed on every frame (white-to-black and black-to-white), with the cue shown for 167 ms (5 frames) |
Expanding ring (spatiotemporal cue used in main experiment) | Looming stimuli are known to capture attention and may be more salient superimposed on a complex, dynamic scene (Franconeri & Simons, 2003) | Red circular outline shown for 167 ms (5 frames). The diameter of the ring on the first frame was equal to the average of half the annotated height and width of the hazard (or distractor) and increased by 40% on each consecutive video frame |
Contracting ring | Some evidence suggest that motion onsets alone may capture attention and may be sufficient for a cue (Abrams & Christ, 2005) | Red circular outline shown for 167 ms (5 frames). The diameter of the ring on the last of the five frames was equal to the sum of the annotated height and width of the hazard (or distractor), and the diameter on each preceding frame was 20% larger |
Flashing red dot | Used on grounds of being perhaps the easiest to implement, but likely the least salient or prone to capture attention of any cue in the set | Red filled dot on the center of the annotated hazard location (22 pixels in diameter), appeared and disappeared on alternating frames for 7 frames (4 on-frames) |