Ethogram of the Ridden Horse Part 2: The Facial Expression Ethogram

“If a measure is to be useful as a welfare indicator it must be valid (meaningful in relation to animal welfare), reliable (produce consistent results when used by different observers/scorers) and feasible (practical for contemporaneous assessment in the field)” (Waran and Randle, 2017, p.75).

The recently developed facial expression ethogram of ridden horses can aid in the early detection of lameness and low-grade musculoskeletal problems across the board, from the backyard pony to the elite performance horse (Dyson, 2017a; Dyson, 2017b; Mullard et al., 2017). One of the goals of the project is to bring people’s awareness to the possibility that changes in facial expressions may very well be a manifestation of pain. If owners are aware of this, the hope is that they will seek advice from a veterinarian before the underlying problem turns into an advanced issue that is more difficult to resolve (Dyson, 2017a). The observational methods can be readily taught and made available to all equestrians – owners, riders, trainers, veterinarians – who may not know how to recognize signs of pain in the facial expressions of their horses (Dyson, 2017; Dyson et al., 2018).

The low/no technology nature of these observational methods removes barriers associated with the collection and analysis of physiological measures, enabling adoption by the broader equestrian community. The early detection of pain in a horse provides an opportunity for early diagnosis and treatment, and can prevent a problem from escalating to a point where the horse is experiencing reduced welfare or an unsafe situation arises for the rider or handler (Mullard et al., 2017; Waran & Randle, 2017).

Motivations to develop the ethogram

It has been well documented that horses exhibit changes in their facial expressions when experiencing pain or discomfort associated with colic (Dyson, 2017a). Changes in behaviour and facial expressions have also been observed when horses are ridden, which is likely a manifestation of pain related to low-grade lameness or a musculoskeletal problem. However, owners and riders are generally not very good at detecting low-grade lameness and are slow to notice physical problems that are compromising their horse’s performance and have a tendency to label horses as ‘naughty’. Alternatively, trainers place blame on the riders. Either way, the possibility that the horse may be experiencing physical discomfort is completely overlooked (Dyson, 2017a, Mullard et al., 2017). “In a survey of 506 sports horses in normal work and presumed to be sound, 47% were overtly lame either in hand and/or ridden or had other pain-related gait abnormalities (Dyson et al., 2018, p.47).

Developing the Ethogram

In the first part of their endeavor to create an ethogram of the ridden horse, Sue Dyson (2017b) and her colleague studied numerous photographs of both lame and sound horses to develop the markers of facial expressions for the ethogram. In order to show it would be useful, they set out to demonstrate that a number of different individuals could apply the ethogram in a consistent way. They assembled a team of 14 people made up of vets, horse owners, instructors, and gave them a training manual with descriptions of every single marker in the facial expressions ethogram of the ridden horse accompanied by a lecture to introduce the concept and procedure. Participants were given a series of photos of 30 horse’s heads and asked to apply the ethogram to each of the individual photographs.

Upon observing some anomalies in the interpretation of the ethogram, it was revisited and changes were made before proceeding with a test on a larger group of different horses. Results were statistically analyzed to determine consistency in the application of the newly developed ethogram, indicating a high degree of consistency in the interpretation of the photographs, proving that it can be applied accurately by a range of people from different professional and nonprofessional equestrian backgrounds (Dyson, 2017b).

Applying the Ethogram

The second part of the study was designed to see if by using photographs of the heads of the horses, both lame and sound, whether or not the two groups could be differentiated based on facial features alone. A pain score was developed for each marker of the ethogram. For example, when looking at ear position if both ears are forward, a score of  0 would be given (normal), whereas if both ears are back, that would be a score of 3 (most likely an indicator of pain). Scoring was done for each facial feature to develop a pain score for each anatomical area which could then be tallied to produce a total pain score for each head. Statistical analysis was applied and results indicated that there was a highly significant difference in the pain score in lame versus sound horses. To support the findings, a small portion of the lame horses were administered nerve blocks. These horses were photographed both before and after the nerve blocks had been administered to observe whether or not the pain scores changed and to determine whether it was in fact pain from lameness that was causing the change in facial expression (Dyson, 2017b).

The next part of the study, which is currently in development, involves assessments of whole horses by video. Dyson and her colleagues are looking at all aspects of behaviour (facial, body, and gait) in both sound and lame horses, and timing actions, such as how long the ears are in a specific position, tail movement, head movement, and reluctance to move forward in response to a  rider’s cues. The significant markers are being identified and modified for accuracy. When complete, the ethogram will serve as a valuable tool to evaluate pain scores and differences in pain scores between lame and sound horses (Dyson, 2017b; Dyson et al., 2018).

A complete ethogram of the ridden horse vs. other methods of behavioural interpretation

Traditionally, evaluations of the training event, equipment used, rider actions and resulting performance and mental state of the horse have been observed to develop a record of behavioural observations in the ridden horse as an indication of welfare. Specific behaviours associated with pain, discomfort, anxiety, fear, and distress have been assessed by trainers and judges. However, there has been a high degree of subjectivity and inconsistency, as illustrated in Part 3: Common Methods of Behavioural Interpretation. A consistent approach across disciplines has the potential to significantly improve the welfare of the ridden horse.

Working toward the visual assessment of pain, fear, and stress across the spectrum of equestrian events, it is perhaps more prudent to focus on methods that could be learned and applied by the everyday equestrian; methods that are not dependent on technology and instead focus on teaching people how to observe and interpret facial expressions, posture, gait changes, and tail positions of the horse that are indicative of pain when under saddle. If the goal is to have a more robust application of the ethogram, technology such as that frequently used in the collection of physiological data may be a barrier.

Part 1: Introduction

Part 3: Common Methods of Behavioural Interpretation


Dyson, S. (2017a). Facial Expressions Research – is your horse trying to tell you he’s in pain? Apr 28, 2017. Retrieved from

Dyson, S. (2017b). Facial Expressions Study – Developing and Applying the Ethogram. Jun 1, 2017. Retrieved from

Dyson, S., Berger, J.M., Ellis, A.D., & J. Mullard. (2017). Can the presence of musculoskeletal pain be determined from the facial expressions of ridden horses (FEReq)? Journal of Veterinary Behaviour 19 (2017) 78-89.

Dyson, S. Berger, J., Ellis, A.D. and J. Mullard. (2018). Development of an ethogram for a pain scoring system in ridden horses and its application to determine the presence of musculoskeletal pain” Journal of Veterinary Behaviour: Clinical Applications and Research 23 (2018) 47-57.

Mullard, J., Berger, J.M., Ellis, A.D. & S. Dyson. (2017). Development of an ethogram to describe facial expressions in ridden horses (FEReq). Journal of Veterinary Behaviour 18 (2017) 7-12.

Waran, N. & H. Randle. (2017). What we can measure, we can manage: The importance of using robust welfare indicators in Equitation Science. in Applied Animal Behaviour Science 190 (2017) 74-81.


Karina Bech Gleerup. Retrieved from

One thought on “Ethogram of the Ridden Horse Part 2: The Facial Expression Ethogram

Comments are closed.