When numbers bias chicks’ choices

A peculiar characteristic of numbers, which we usually ignore, is the spatial organization that numbers have in our mind. Most people who read and write from left to right, orientate numbers in the same direction. Thus small numbers are on the left and large ones on the right space. A few years ago, we showed that also three-day old chicks and newborns share with us the same orientation of numbers on space. A main question remained unanswered: Is the association between numbers and space driven by numerical magnitude or can it be simply explained by individual bias?

In this new paper, we show that the association between numbers and space in day-old domestic chicks is driven by numerical magnitude and cannot be explained by individual bias.

 Scientific background

In the early 1800s, Francis Galton reported that most humans represent numbers on a mental number line, along which smaller numbers are located on the left and larger ones on the right side.

Traditionally, such a spatial-numerical association has been considered a by-product of culture, mainly linked to reading/writing direction. Evidence from young infants, newborns and animal species has challenged the primary role of culture in determining the left-to-right direction of spatial-numerical association.

The evidence of spontaneous spatial-numerical association in young domestic chicks has attracted much attention, which unleashed a firestorm of controversy about the origin of such association. Many scientists sustain that early spatial numerical association in chicks is due to animals’ individual biases, and remains unrelated to the human mental number line. Others, though, have appreciated the possibility to frame evidence from preverbal infants and non-human animal species, in a single model that accounts for a common origin.

The research question

We investigated the effect of numerical magnitude in determining the spatial numerical association, also controlling for individual biases.

The association of space and numbers previously reported for human newborns and three-day old domestic chicks, according to some scientists may depend on individual biases rather than on processing actual numerical magnitudes. The effect of individual spatial biases in determining this effect has been studied here for the very first time.

The experiment and the results

We trained 3-day-old chicks to respond to a numerical target, 5. At test, chicks faced two identical numerousness, either 2vs.2, 8vs.8 or 5vs.5, one on their left and one on their right side. We computed the percentage of Left-sided Choice, LC. Chicks showed a left bias in the 2vs.2, a right bias in the 8vs.8 and no bias in the control test 5vs.5, irrespective of testing order.

Results cannot be explained by individual orienting biases or by preference for novelty. Our results show, instead, that it is the difference in the numerical magnitude between the values experienced at training and those experienced at test, which determines the direction of the bias. Moreover, we found a linear trend with three points of reference (LC2vs.2 > LC5vs.5 > LC8vs.8); which poses further restrictions in favor of a spatial-numerical mapping guided by the mental number line.

The most important findings

We provide compelling evidence that the spatial numerical association in three-day-old domestic chicks is driven by numerical magnitude. Particularly, for the first time our results allow to directly exclude any role of an individual’s side preferences on the spatial numerical association effect.

It is number magnitude that causes the side bias: after training on 5, smaller numbers, 2, induce a left bias and larger numbers, 8, a right bias. Moreover, our data highlight a linear trend so that the Left-Bias is: 2vs.2 > 5vs.5 > 8vs.8, posing further restrictions in favor of a left-to-right oriented spatial-numerical mapping.


Our findings indicate that a disposition to associate numbers onto a left-to-right-oriented mental number line exists in newborn and inexperienced animals, independently of individual side biases. This prompts further research on, respectively, the role of experience, or culture, in shaping this predisposition and the identification of the underlying neural substrates.

Scientific publication

Rugani, R, Vallortigara, G, Priftis, K, and Regolin L (2020). Numerical magnitude, rather than individual bias, explains spatial numerical association in newborn chicks. ELife 9, e54662
DOI: https://doi.org/10.7554/eLife.54662