Categorizing helps us to cope with the vast variety of objects in our environment. Although categorization represents a core cognitive function, stimulus features that drive behavior and underlying strategies for categorizing objects often remain elusive.

Categorizing helps us to cope with the vast variety of objects in our environment. Although categorization represents a core cognitive function, stimulus features that drive behavior and underlying strategies for categorizing objects often remain elusive. To elucidate these issues, we performed behavioral experiments with pigeons - classic model animals to investigate perceptual categorization. We generated two categories of artificial stimuli called digital embryos and analyzed the pigeons pecking behavior using machine learning. Our results show that pecking is indicative of the upcoming choice and thus related to features of interest. However, individual animals use different stimulus aspects to base their decision on. By using defined artificial stimuli in addition with a detailed analysis of the pecking behavior, our study paves the way to pinpoint stimulus features as well as individual strategies to solve the task.

Pusch, R., Packheiser, J., Koenen, C. Iovine, F. & Güntürkün, O. (2022) Digital embryos: a novel technical approach to investigate perceptual categorization in pigeons (Columba livia) using machine learning. Anim Cogn. https://doi.org/10.1007/s10071-021-01594-1