Juvenile pigs use simple geometric 2D shapes but not portrait photographs of conspecifics as visual discriminative stimuli

Several animals living in social groups, such as monkeys, cows and sheep, have been shown to use facial discrimination for social recognition. Whether pigs can discriminate between faces of conspecifics purely based on visual stimuli provided by 2D portrait photographs, has not yet been investigated. Therefore, in this study piglets with a large birth weight range were trained in a visual discrimination task. Piglets were derived from different litters; from each litter same sex siblings with a low (LBW) and normal birth weight (NBW) were selected. With this setup it could be clarified whether pigs are able to discriminate between 2D photographs of conspecifics, and if LBW animals have more difficulty doing so than NBW siblings. Because pigs learn visual discrimination tasks slowly, we started with a simple discrimination task involving one of two geometric black-and-white stimuli, followed by a simultaneous discrimination task in which one of the black-and-white stimuli served as Splus, the other as Sminus, followed by a reversal. Pigs needed on average 337 trials in the simple and 98 trials in the simultaneous discrimination task to reach criterion. Only 1/3 of all pigs reached criterion on a reversal (average of “learners”: 276 trials to criterion). None of the pigs learned to discriminate between 2D photos of heads of conspecifics, even after 289 trials, when training was discontinued. Birth weight did not affect learning. We conclude that pigs need input from more modalities than vision alone to enable discrimination between conspecifics.