Things are looking (farther) up: Upward gaze orientation is overestimated.

In the current work, we examine upward head orientation. Previous work has shown that downward gaze orientation is overestimated by a factor of ~1.5, while azimuth gaze orientation is overestimated by a factor of ~1.2. In Experiment 1, we compared head orientation in the upward, downward, and azimuth (leftward and rightward) directions. The factor by which observers overestimated upward head orientation was 1.56, similar to overestimates in downward head orientation in this and downward gaze orientation in previous work. Azimuth gaze orientation was overestimated by ~1.26, similar to previous work. In Experiment 2, we examined upward head orientation with a smaller range of angles and found that the factor increased slightly (1.7) but not statistically so from Experiment 1 or from previous work. We also examined the perceptual head orientation boundaries for upward and downward gaze orientation. These indicate a far more limited range than 90° up and down. In Experiment 3, participants gave head orientation estimates while blindfolded. Their estimates of upward, downward, and azimuth head orientation matched our current and others' previous work. Independent of gaze, head orientation is overestimated by the same factor as is gaze orientation. The upward head orientation overestimates predict well observers' estimates of heights of familiar suspended objects and fits with why some familiar suspended objects appear smaller than actual. The common scale expansion for upward and downward head and gaze orientation is a useful perceptual regularity that reliably predicts how we spatially map the environment and how we perceive the world.