Quantifying the face inversion effect in nonhuman primates: a phylogenetic meta-analysis

Face recognition is important for primate social cognition, enabling rapid discrimination between faces and objects. In humans, face recognition is characterized by certain cognitive specializations such as face-specific sensitivity to upright faces. The face inversion effect reflects the disproportionate inversion cost (i.e., poorer recognition of inverted compared to upright stimuli) for faces compared to non-face objects. Numerous studies have evaluated the face inversion effect in various nonhuman primates, but the findings are highly variable and mixed, especially in monkeys. To address this inconsistency, the current study employed a multilevel, phylogenetic meta-analysis on 52 effect sizes across 16 studies to quantify the magnitude of the face inversion effect in nonhuman primates. Overall, the difference in inversion costs for faces compared to nonface stimuli was small but not statistically significant (b = 0.31, se = 0.16, 95% CI [- 0.01, 0.62], p = 0.06). Additionally, this did not systematically vary between the family of species (chimpanzee or monkey) or type of face stimuli (conspecific or heterospecific). Ultimately, it cannot be concluded that the face inversion effect is a reliable phenomenon in either chimpanzees or monkeys, suggesting that nonhuman primates may not use similar face-specific processing strategies as humans.