Goats who stare at video screens – assessing behavioural responses of goats towards images of familiar and unfamiliar con- and heterospecifics

Abstract

Many cognitive paradigms rely on active decision-making, creating participation biases (i.e., drop-outs due to lack of learning on how to indicate a choice) while once-learned contingencies might bias the outcomes of subsequent similar tests. We here present a looking time approach to study goat perception and cognition, without the need to extensively train animals and no reliance on learned contingencies. In our looking time paradigm, we assessed the attention of 10 female dwarf goats (Capra hircus) towards images of familiar and unfamiliar con- and heterospecifics using an experimental apparatus containing two video screens. Subjects were confronted with images on either the left or the right screen of the apparatus while the other screen remained white. The presented images were single faces of either familiar or unfamiliar goats or humans, presented for 10 seconds each. Each subject received 8 trials for each stimulus type, resulting in a total of 32 trials. Spontaneous behavioural reactions to the presented stimuli, including the looking behaviour (direction of first look, looking duration) and the time spent with the ears in different positions (forward, backward, horizontal, others), were analysed using linear mixed-effects models. We found that goats looked longer at the video screen presenting a stimulus compared to the screen that remained white. Goats looked longer at images depicting other goats compared to humans, while their looking behaviour did not significantly differ when being confronted with familiar vs. unfamiliar individuals. We did not find statistical support for an association between the ear positions and the presented stimuli. Our findings indicate that goats are capable of discriminating between two-dimensional con- and heterospecific faces, but also raise questions on their ability to categorise other individuals regarding their familiarity using 2D face images alone. Our subjects might either lack this ability or might be unable to recognise the provided 2D images as representations of real-life subjects. Alternatively, subjects might have shown an equal amount of motivation to pay close attention to both familiar and unfamiliar faces masking potential effects. The looking time paradigm developed in this study appears to be a promising approach to investigate a variety of other research questions linked to how domestic ungulate species perceive their physical and social environment.