Jeonju City Blog photo of horses at the Jeonju Riding Club.

Linking Equine Temperament and Neurochemistry

Korean researchers explore the role of neurochemistry on the temperament of riding horses.

 

What if a simple blood test could predict the docility and friendliness of a horse? Would such a test become standard in the pre-purchase exam?

Journal Reference:

Kim, J., Park, Y., Kim, E. J., Jung, H., & Yoon, M. (2021). Relationship between oxytocin and serotonin and the fearfulness, dominance, and trainability of horses. Journal of animal science and technology, 63(2), 453.

Kim, J., & Yoon, M. (2022). The effect of serotonin and oxytocin on equine docility and friendliness to humans. Journal of Veterinary Behavior, 50, 18-22.

Song, Y., Kim, J., Park, Y., & Yoon, M. (2023). Association between the plasma concentration of melatonin and behavioral temperament in horses. Journal of Animal Science and Technology, 65(5), 1094.

Choi, Y., Wickens, C. L., Park, Y., & Yoon, M. Evaluating Hormonal Levels in Plasma and Saliva as Temperament Indicators in Horses. Available at SSRN 4758263.

Lee, G., & Yoon, M. (2021). Association of plasma concentrations of oxytocin, vasopressin, and serotonin with docility and friendliness of horses. Domestic animal endocrinology, 74, 106482.

Summary:

Researchers at Kyungpook National University Horse Science Lab have published a number of papers examining the neurochemistry and temperaments of the riding horses in their horse program. While many of these studies also look at variation between breeds, genders, and age, the Horse Science Lab keeps searching for correspondence between behavioral characteristics and neurochemistry. The lab is headed by Minjung Yoon, a PhD graduate of UC Davis, who is the supervising coauthor of each of these studies.

His lab gets a lot of practice annually looking at the blood and saliva of the animals in the KNP Horse Industry Center. The horses in that stable might feel like pincushions.

The underlying intent is to explain the variation in temperaments of these patient animals and the real challenge to the scientists isn’t drawing blood, but capturing temperamental assessments that reflect the animals permanent general demeanor. Over time, the methods have evolved from simple ranking, to scoring high/medium/low, and currently to a full scoring system that combines more than a dozen individual assessments for a number of traits. As these personality assessments have been refined, the lab is now spending some effort toward validating each behavioral characteristic used to describe the animals. This is a daunting task, but the horse world can expect to benefit from the developments this lab is making through this long-range effort.

In study number 1, Kim et al (2021) found that plasma oxytocin (OXT) concentration was negatively correlated with fearfulness and dominance, but positively correlated with trainability in horses, while plasma serotonin (5-HT) concentration was negatively correlated with dominance and positively correlated with trainability in horses.

In number 2, Kim & Yoon (2022) concludes that the plasma concentrations of serotonin (5-HT) and oxytocin (OXT) can act as determining factors for equine docility and friendliness towards humans, with 5-HT and OXT concentrations differing by breed and increasing with age, and being positively correlated with each other and with equine temperament.

Study number 3 reported null results. Song et al (2023) found that melatonin concentrations differed by horse breed, but were not associated with horse temperament traits like docility, affinity, dominance, or trainability, suggesting melatonin is unlikely to be a useful biomarker for predicting horse temperament.

In study number 4, Choi et al (2024) highlights the potential use of salivary cortisol and oxytocin as non-invasive markers for assessing specific temperament traits in horses, which could improve horse management and welfare practices. Cortisol levels (in plasma and saliva) were negatively correlated with concentration, trainability, cooperativeness, and gentleness, but positively correlated with excitability. Oxytocin levels (in plasma and saliva) were positively correlated with friendliness.

A final paper, here number 5, was not freely available and only its abstract could be compared with the other studies. Lee and Yoon (2021) did not find friendliness or docility related to vasopressin or serotonin concentration, but noted that oxytocin and docility were correlated in Thoroughbred horses.

The following table summarizes the behavioral characteristics measured against the correlated neurochemicals. Numbers refer to the reporting study. Vasopressin and melatonin are not included on the chart as they showed no correlations in these studies.

TRAIT Observable Behavior Direction of correlation and study effect was found in.	Oxytocin	Serotonin	Cortisol
DOCILITY	+5
Calmness (2)       Remains calm in new or startling situation (4)	+2	+2
Ease of Handling (2)       Overall gentleness and ease of handling (4)	+2		-4
Behavioral response in an unfamiliar situation (3)
Time it takes to catch (3)
Tendency to become easily excited (4)			+4
FRIENDLINESS
Willingness to follow (2)	+2	+2
How the horse communicates with humans  (2)	+2	+2
Distance maintained from humans  (3)
Time it takes to get familiar with strangers  (3)
Friendly and approachable, interacts in friendly ways (4)	+4
FEARFULNESS
Keeping distance from unfamiliar persons (1)	-1	+1
Rapid and excessive reaction to new stimuli (1)	-1	+1
Ability to tolerate various stimuli without distress (4)
Tendency to become fearful easily (4)
DOMINANCE
Trying to gain upper hand over unfamiliar people (1)	-1	-1
Trying to gain upper hand in fight for food (1, 3)	-1	-1
Displays competitive behavior with other horses (4)
TRAINABILITY
Willing to participate in training (1,3)       Willingness to work and follow instructions (4)	+1
Time to achieve training goal (1,3)	+1
Maintains concentration with being distracted by the environment (4)		-4	-4
Learns and adapts to new training easily and quickly (4)		+4	-4
Shows resistance or reluctance to obey commands (4)

1. Kim et al, 2021 scored high, medium, low
2. Kim and Yoon 2022 ranked by 3 professors, average rank.
3. Song et al 2023 scored by 3 professional trainers 0 to 5 points, with low (0-1 points), medium (2-3 points), and high (4-5 points)
4. Choi et al 2024 assessed by 10 students and 4 trainers 1 to 5 scale
5. Lee and Yoon 2021 assessed by 3 trainers, details unknown

Presumably, these studies have used many of the same horses. It would be very informative to know how much variation there is within each individual over time. Without understanding the normal fluctuations of an individual horse in response to life, using a neurochemistry work up to assess an individual’s temperament makes little sense. In some sense, these studies show that, when a trained professional can simply handle the animal, they can predict its neurochemical state.

What this study does help with is understanding that the animals are reflecting their current neurochemical status. When science gives trainers the tools to monitor neurochemical changes in real-time, we might finally be able to operationalize the human-equine bond and optimize the emotional state of the animals that we trust our lives to. Until then, we look forward to following the continuing work of Dr. Yoon and his amazing lab.

Keywords: oxytocin, equine personality, equine temperament, serotonin

---

Trust Science Scorecard:

• The studies started out with a proper set of hypotheses.

In some sense, these studies are pioneer efforts into a new paradigm. The temperamental variables are not well validated. The weakness of the hypotheses reflect this necessary pioneering stage.

• The method to measure OXY seems legitimate.

The methods appear to be valid.

• The sample of animals is adequate for the question.

This is the strength of these studies. They all have large samples.

• The statistics don’t involve a large number of variables but just those relevant to the hypothesis.

Study 2 focused on cross tabulating age, sex, breed, and temperament. The sample size did not seem adequate to keep all of the categories balanced for analysis. The other studies inspired more statistical confidence.