Proto-arithmetic Task with Empty Sets in Newborn Chicks (Gallus gallus)

Numerical cognition research has documented proto-arithmetic abilities across a wide range of species. Yet, the status of zero (the representation of absence as a numerically meaningful quantity) remains poorly understood, particularly in non-human animals without prior numerical training. Here, I tested whether newborn domestic chicks (Gallus gallus) could spontaneously integrate empty sets into arithmetic operations, and whether their performance across different empty-set vs. non-empty set comparisons would show the distance effect characteristic of magnitude-based numerical processing. Newly hatched chicks were reared with imprinting objects which later acted as stimuli in the arithmetic test. On each trial, stimuli were hidden behind two opaque screens and subsequently moved; at the end of the sequence, one screen concealed either one or three objects while the other remained empty. Subjects chose spontaneously between the two screens. In both comparison conditions (0 vs. 1 and 0 vs. 3), they performed significantly above chance (respectively EMM = 0.585; p < .0001; EMM = 0.603; p < .0001), with no significant difference in accuracy between conditions (χ_1^2 = 0.0083; p = .9272) and no evidence of reliance on a last-movement strategy. These results indicate that sensitivity to absence is present from the earliest days of post-hatching life and does not require prior training. The absence of a distance effect suggests that zero may be processed through a qualitatively distinct mechanism, a categorical "something vs. nothing" distinction, rather than as a point on a graded numerical continuum. These findings extend previous evidence on innate proto-arithmetic abilities in chicks and contribute to the broader question of whether the concept of zero has deep evolutionary roots across vertebrates.