An unexpected acoustic indicator of positive emotions in horses

Assessing positive emotions is crucial for identifying how animals perceive the conditions offered. However, indicators of positive emotions are still scarce and many proposed behavioural markers have proven ambiguous. Studies have established a link between acoustic signals and emitter’s internal state, but few related to positive emotions and still fewer considered non-vocal sounds (e.g. purring in felids). One of them, the snort, is shared by several perrisodactyls and has been associated mainly with positive contexts. We hypothesized that this could be also the case in horses. These vibrating sounds produced by nostrils during expiration are often confused with the blows, another sound produced by an exhalation of the air through the nostrils, but more intense and of a shorter duration. Blows have been shown to be emitted in an alarm context and have been generally associated with an alert/vigilance posture while snorts have up to now been considered as having a hygienic function. However, pilot observations have revealed that snorts were produced more in some individuals than in others, despite identical air conditions. Here, we hypothesized that snorts in healthy animals might reflect the animal’s psychological state, following a physiological change related to mild positive excitations. We observed 48 horses belonging to four different populations: two in facilities offering restricted life conditions (e.g. single stall), two living in facilities with more naturalistic conditions (e.g. stable groups in pastures). The immediate context of production (e.g. stall/pasture) and the horse’s activity and state (ear positions) were observed using the focal behavioural sampling method. We additionally performed an evaluation of the welfare state, using common indicators (i.e. stereotypies, aggressiveness toward human, ears backwards while feeding, prolonged orientation toward a wall in stall). A total chronic stress score (TCSS) was calculated per horse from the data obtained for each of these indicators, reflecting how much the welfare state was altered. The results showed that 1) snort production was significantly associated with positive situations (e.g. while feeding: in stall (hay): 67.3% of the snorts; in pasture (grass): 69.6% of snorts), and horses expressed more ears in forward or sidewards positions (reflecting more positive internal states) while snorting compared to a basal situation (Wilcoxon test, Z=3.47, p<0.001), 2) the riding school horses produced twice as many snorts when in pasture compared to the stall condition (Wilcoxon test, Z=2.84, p=0.004), 3) the naturalistic population emitted significantly more snorts than riding school ones in comparable contexts (Kruskal Wallis test, chi-squared=12.851, df=2, p=0.001), 4) the frequency of snorts was especially negatively correlated with the TCSS (reflecting compromised welfare): the lower the TCSS, the higher the snort rate (Spearman correlation, r=-0.39, p=0.005). These four results converge to indicate that snorts could reflect a positive emotional state. This study leads us to believe that this acoustic signal could constitute a potential marker of quiet positive emotions in horses, at least outside the working situation (not tested yet). Snort could be a useful tool for improving management practices, based on the horses’ subjective assessment of the situations.