Humans may have the ability to smell fear in horses, the findings of a preliminary study suggest.
The natural body odour of animals and humans consists of a wide range of volatile compounds that carry a range of information.
Agnieszka Sabiniewicz and her fellow researchers said the body odour of mammals can convey cues to other members of their species about their emotional state.
However, little attention has been paid to the communication of emotions through odour between species, and no studies have examined whether humans are able to recognise animal emotions from body odour.
The study team, reporting in the journal Animals, set out to investigate. “Considering the intimate and long-lasting relationship between human subjects and domesticated animals, the question that arises is whether the ability to recognise emotions conveyed by chemosignals exists only within species or can it also occur between species?”
They noted that horse domestication started about 6000 years ago and, since then, they have been used for transport, herding, food, trade, welfare, competition, and recreation. The study team decided to use horses as odour donors, because of their long history of close coexistence with humans.
Body odour samples were collected from 16 two-year-old Thoroughbred horses in fear and non-fear situations. The fear situation involved a race scenario early in their training. Their ear position, head position, and body tension during this exercise confirmed their expectation.
The non-fear situation involved a relaxing walk.
The odour in each case was collected in near-odourless fleece material placed under new saddle pads.
The horse odour samples were then assessed by 73 human odour raters, comprising 49 women and 24 men.
The odor-impregnated fleece samples from each horse were cut into equal portions and put in two separate closed glass jars of 0.5 litres – one holding the fear-related sample, the other the non-fear sample.
Each odour rater was allowed to briefly sniff at both samples for a maximum of two inhalations and was then asked to assess which of the two jars contained the fear odour and which one contained the non-fear odour.
The average accuracy achieved across all raters was 65.7%, which was statistically greater than chance (that is, 50%).
Eleven out of 73 subjects (4 men and 7 women) performed significantly above chance in the odour rating task, with some achieving an accuracy of 80% or more. None of the odour raters performed significantly below chance.
“Therefore, we provide evidence that human individuals are able to correctly categorise horses’ fear vs. non-fear odours,” the researchers reported.
“We found that humans, as a group, were able to correctly assign whether horse odour samples were collected under a fear or a non-fear condition.
“They perceived the body odour of horses collected under the fear condition as more intense, compared with the non-fear condition.”
An open question remained as a result of the study, they said. Can humans simply distinguish between little versus much sweat, and between high intensity versus low intensity, or were they able to recognise horses’ fear and non-fear emotions?
“These results appear to fit the notion that the ability to recognise emotions in other species may present an advantage to both the sender and the receiver of emotional cues, particularly in the interaction between humans and domesticated animals.”
The findings, they concluded, indicate that smell might contribute to the human recognition of horse emotions. However, the results should be addressed with caution given the preliminary nature of the study.
Discussing their results, the researchers said a variety of communication channels have evolved between humans and horses since domestication, mostly involving visual, touch or vocal cues.
“But the smell of horses has also been shown to contribute to human reactions to horses,” they noted. One study found that the smell of horses in a therapeutic context led to a decrease in levels of the stress hormone cortisol and to higher self-reported well-being in human patients.
The authors acknowledged there were limitations in their study. Firstly, the horses’ reactions in the fear state were validated only through behavioural measures. “Thus, it may be insufficient to conclude whether this condition actually generated fear and, if so, to what extent.”
Future research should include other options, such as measuring cortisol levels.
Participants were not asked to openly attribute a quality to the odour samples but had to choose between limited options and closed responses — intense or not; fear or not.
Describing odours can be tiring and challenging, they said, especially when participants are presented with 32 samples. The study had been designed to make this task easier for the participants.
Thirdly, the fear condition contained a strong element of physical activity which increased sweat production, raising the question of whether humans could simply distinguish between little versus much sweat and between high intensity versus low intensity.
“While future research in this field should control for the level of physical activity in both conditions, it should be noted that experiencing fear per se leads to increased sweating.”
Future studies with carefully developed methodology should explore this issue further, they said.
The study team comprised Sabiniewicz, Michał Białek, Karolina Tarnowska, Robert Świątek and Piotr Sorokowski, all with the University of Wroclaw in Poland; and Małgorzata Dobrowolska, with the Silesian University of Technology, also in Poland. Sabiniewicz is also with the Department of Otorhinolaryngology, Smell and Taste Clinic, part of the Dresden University of Technology in Germany.
Sabiniewicz, A.; Białek, M.; Tarnowska, K.; Świątek, R.; Dobrowolska, M.; Sorokowski, P. A Preliminary Investigation of Interspecific Chemosensory Communication of Emotions: Can Humans (Homo sapiens) Recognise Fear- and Non-Fear Body Odour from Horses (Equus ferus caballus). Animals 2021, 11, 3499. https://doi.org/10.3390/ani11123499