The neonatal period in humans and in most mammals is characterized by intense mother-young interactions favoring pair bonding and the adaptation of neonates to their new environment. However, in many post-delivery procedures, human babies commonly experience combined maternal separation and intense handling for about one hour post-birth. Currently, the effects of such disturbances on later attachment and on the development of newborns are still debated: clearly, further investigations are required. As animals present good models for controlled experimentation, we chose domestic horses to investigate this issue. Horses, like humans, are characterized by single births, long lactating periods and selective mother-infant bonds. Routine postnatal procedures for foals, as for human babies, also involve intense handling and maternal separation. In the present study, we monitored the behavior of foals from early stages of development to “adolescence”, in a normal ecological context (social groups with adults and peers). Experimental foals, separated from their mothers and handled for only 1 hour post-birth, were compared to control foals, left undisturbed after birth. Our results revealed short- and long-term effects of this unique neonatal experience on attachment and subsequent social competences. Thus, experimental foals presented patterns of insecure attachment to their mothers (strong dependence on their mothers, little play) and impaired social competences (social withdrawal, aggressiveness) at all ages. We discuss these results in terms of mother-young interactions, timing of interactions and relationships between bonding and subsequent social competences. Our results indicate that this ungulate species could become an interesting animal model. To our knowledge, this is the first clear demonstration that intervention just after birth affects bonding and subsequent social competences (at least until “adolescence”). It opens new research directions for studies on both humans and other animals.

Stereotypic behaviours, i.e. repetitive behaviours induced by frustration, repeated attempts to cope and/or brain dysfunction, are intriguing as they occur in a variety of domestic and captive species without any clear adaptive function. Among the different hypotheses, the coping hypothesis predicts that stereotypic behaviours provide a way for animals in unfavourable environmental conditions to adjust. As such, they are expected to have a lower physiological stress level (glucocorticoids) than non-stereotypic animals. Attempts to link stereotypic behaviours with glucocorticoids however have yielded contradictory results. Here we investigated correlates of oral and motor stereotypic behaviours and glucocorticoid levels in two large samples of domestic horses (N Study1 = 55, N Study2 = 58), kept in sub-optimal conditions (e.g. confinement, social isolation), and already known to experience poor welfare states. Each horse was observed in its box using focal sampling (study 1) and instantaneous scan sampling (study 2). Plasma samples (collected in study 1) but also non-invasive faecal samples (collected in both studies) were retrieved in order to assess cortisol levels.