Horses give functionally relevant responses to human facial expressions of emotion: a response to Schmoll (2016)

2016 Horses give functionally relevant responses to human facial expressions of emotion: a response to Schmoll Our paper demonstrates that horses show a quicker increase in heart rate when presented with photographic stimuli depicting angry versus happy human faces. We also use lateralized looking at each stimulus as a means of investigating how the image is perceived and found a strong left gaze/right hemisphere bias to angry stimuli but no significant lateralized response to happy or significant difference between responses to angry and happy. While we accept that a more extensive exploration and discussion of the results would have been useful, our original findings still stand. Schmoll [1] voices two main concerns: the inclusion of three very short latency-to-response values in the heart-rate data and our interpretation of the lateralization results. Dealing first with the heart-rate data, Schmoll raises a useful point about whether very short latency values (less than 1 s) can be attributed to an immediate response to the stimuli and we are pleased to explore this possibility. Our original analysis included a few seconds while the stimulus photo was being turned towards the subject and was only partly visible. To ensure that our latency-to-maximum heart-rate values best reflect direct responses to stimuli, we re-analysed our heart-rate data with a more conservative start point—the heartbeat immediately before the stimulus fully faced the subject. In this re-analysis, two horses (Rose: happy trial; Willsie: angry trial) whose heart rates did not increase from the test start were excluded. With this start point, heart rate still increased faster towards angry photographs compared with happy (angry: mdn ¼ 14.7 s, happy: mdn ¼ 29 s; Wilcoxon test: z ¼ 22.39, p ¼ 0.015; figure 1). Only one subject has a response latency under 5 s (Jack ¼ 1.6). While our ongoing work suggests horses' heart rates can respond within 1.6 s, even if Jack is removed the effect remains (angry: mdn ¼ 17.7 s, happy: mdn ¼ 29.10 s; z ¼ 22.17, p ¼ 0.03). Thus on the basis of heart-rate data alone, horses discriminate between angry and happy human expressions. On Schmoll's second point, we accept that our lateralization results cannot confirm a between-groups difference and for most of the paper we discuss the lateralization results independently for the two emotions. Our discussion [2] opens: " The behavioural and physiological results reported here support the hypothesis that horses are able to …