Figure 4 presents mean sensitivity (i.e., average number of noise dots for ∼79% accuracy) as a function of group (novice, expert), object domain (bird, human), and object orientation (upright, inverted). The sensitivity data were analyzed in a 2 × 2 × 2 mixed-design analysis of variance (ANOVA) with group as a between-subjects factor and object domain and orientation as within-subjects factors. The main effect of group was not significant,
F(1, 38) = 2.64,
p = 0.112. Furthermore, group did not interact with either object domain,
F(1, 38) = 0.006,
p = 0.941, or object orientation,
F(1, 38) = 1.84,
p = 0.183. Crucially, the three-way interaction between group, object domain, and object orientation was not significant,
F(1, 38) = 2.17,
p = 0.149. Thus, expertise did not modulate the effects of the other factors on motion sensitivity. The same pattern was found when removing four experts with an expertise score lower than the highest-scoring novice and their age-matched controls (see
Supplementary Analysis).
There were main effects and interactions for object domain and object orientation. The significant main effect of object domain, F(1, 38) = 59.10, p < 0.001, generalized eta2 = 0.22, showed that sensitivity was higher to humans (M = 21.12 dots, SE = 3.61 dots) than birds (M = 10.64 dots, SE = 1.89 dots). The main effect of object orientation was significant, F(1, 38) = 49.30, p < 0.001, generalized eta2 = 0.16. Sensitivity was higher for upright (M = 20.08 dots, SE = 3.65 dots) than inverted (M = 11.68 dots, SE = 2.12 dots) PL stimuli. These main effects were qualified by a significant interaction between object domain and object orientation, F(1, 38) = 38.82, p < 0.001, generalized eta2 = 0.12. Sensitivity was higher to upright humans (M = 28.97 dots, SE = 3.58 dots) than inverted humans (M = 13.28 dots, SE = 2.37 dots, t(39) = 6.92, p < 0.001). However, sensitivity did not significantly differ between upright birds (M = 11.19 dots, SE = 2.01 dots) and inverted birds (M = 10.08 dots, SE = 1.79 dots, t(39) = 1.42, p = 0.163).