Equine flavour preferences.

Ever wondered why horses have such big, long noses? How important are smell and taste when it comes to preferences and food choices? And, do you think horses use their senses of smell and taste in the same ways we do? 

In this two-part series, nutritionist Mariette van den Berg investigates the influence smell and taste have in a horse’s diet selection.

Last month, Mariette reviewed the science behind how these senses work, and examined the anatomy of the nasal and oral cavities in horses. This month, she looks at how animals learn to like or dislike certain foods and how this knowledge can help you when managing your horse’s diet.

Last month, we discussed the function of smell and taste in humans and other animals, and how this may affect dietary preferences. We learned that animals may express preferences for foods based on a change in liking of the flavour of the food (e.g. odour, taste and texture) or based on the anticipated consequences from eating that food (post-ingestive feedback). In this section, we continue by elaborating further about how animals acquire food preferences and aversions and in particular, equine flavour preferences. 

Preferences and aversions 

Much of the research concerning conditioned (‘learned’) food preferences has involved rats and ruminants such as, goats, sheep and cattle. We know, for example, that lambs acquire preferences for flavours paired with the nutrients in that food. When lambs were offered solutions of glucose flavoured with grape (group 1) or orange (group 2) on some days, and solutions of saccharin (an artificial sweetener) flavoured with grape (group 2) and orange (group 1) on alternate days, they subsequently preferred (96% to 4%) the flavours that were paired with the glucose (1).

Rats will also eat more of a food or flavour when these are paired with calories, even when they don’t like the flavour. For example, rats acquired strong preferences (over 95%) for bitter and sour tastes when its was paired with a positive nutritional consequences (2,3).

The research shows that food intake and preferences will also depend on the nutritional status of the animal, so that when animals have nutritional deficiencies, a positive post-ingestive feedback from foods (the feedback from the gut to the brain that allows animals to sense the nutritional effect of the food ingested) will be stronger and influence the preference for that food. In nutrition these are called the hedonic responses – ‘hedonic’ relating to the pleasant or unpleasant sensations an animal can experience.

Unpleasant experiences, such as nausea or colic, can result from the ingestion of excess toxins and nutrients, which stimulates the emetic system and causes a decrease in preferences (‘aversions’).

The emetic system (responsible for causing nausea and vomiting) is very sensitive to low doses of most toxins and involves complex interactions between several brain areas. Studies have shown that ruminants can learn to dislike foods when they experience imbalances in essential amino acids and energy (i.e. the food is too high or too low in these nutrients). They also have demonstrated that rats, ruminants and even horses will learn to avoid a food when this is followed directly by a nausea-causing toxin called Lithium Chloride (LiCl) (4).

These findings highlight that flavour-feedback associations are dynamic processes – preferences and aversions continually change – which reflects the fact that most animals are adapted to eating a varied diet (including ourselves!). This is most obvious in herbivores that are exposed to complex plant environments where nutrients and toxins change from day to day and from season to season, depending on the growing conditions.

Equine flavour preferences 

When it comes to equine research, only a few studies have been able to show that horses alter their food preferences based on flavour and post-ingestive feedback mechanisms.

In 2005, researchers at Southampton University’s Equine Behaviour Centre in the United Kingdom conducted a study to determine horse flavour preferences (6). The researchers started with 15 different flavours, but eliminated echinacea, coriander and nutmeg when some of the eight horses in the study refused to eat food flavoured with those spices.

Another four flavours were abolished later because, while the horses did eat them, they were not eaten as rapidly as the remaining eight flavours. The flavours that received this unenthusiastic reaction from the horses were ginger, garlic, turmeric and, most surprisingly, the beloved apple!

In the second part of the study, the remaining eight flavours were offered in pairs to determine which flavours horses preferred over the others. This was repeated until each horse had been offered every possible pairing of flavours. The final results, in order of preference were: fenugreek, banana, cherry, rosemary, cumin, carrot, peppermint and oregano.

In the third phase of the study, researchers added fenugreek and banana to unflavoured mineral pellets, which, on their own, horses found unpalatable. It took the horses an average of 195 seconds to eat the pellets, and, in some cases, they did not finish them.

The banana and fenugreek flavoured pellets were consumed in an average of 52 and 66 seconds, respectively. These results suggest that flavours can be used in horse feeds to increase intake of foods or help with the change over to other food types. However, if feedback mechanisms work in a similar way in horses, long-term exposure to preferred flavours may actually reduce intake after a while – due to the sensory satiety of that flavour! As the old proverb says, variety is the spice of life!

Recognition of foods based on nutrient profile 

But what about flavour paired with nutrients? Can horses discriminate foods based on their nutritional content in the same way ruminants and rats can?

A study by Cairns et al. (7) showed that horses are able to select a higher energy concentrate over a lower energy one, regardless of the preferred flavour (in this case, mint or garlic), and they can form associations between a food and its nutritional composition. This was done by presenting a higher energy feed paired with one flavour, and a lower energy of another flavour and then, later in the study, swapping the flavours and energy levels. The results showed that, although initially the horses continued to prefer the flavour that was associated with the increased energy intake, once they realised it was not paying-off in energy, they switched to the higher energy diet with the other flavour.

Whereas cattle and sheep get a very direct response from the rumen, the fact that horses are hindgut fermenters and have very long gut transition times, means they may have some limitations when it comes to post-ingestive feedback from nutrients. They do, however, respond very well to toxins as they are directly taken up in the foregut system.

Another, more recent study, investigated the effect of long-term exposure of a single nutritional forage that was rich in either protein, lipids (fats) or hydrolysable carbohydrates (starches that can be digested in the small intestine). They wanted to know if in a future multiple choice session the horses would recognise and select the forages higher in nutrients (8).

The results indicated that the dietary experience (i.e., having been previously exposed to those macro-nutrients) did alter the horses’ foraging behaviour, showing a greater preference for the forages rich in protein or hydrolysable carbohydrates, whereas lipids remained constant throughout the testing period.

The authors suggest that the horses responded to the macro-nutrients in the diet and the dietary experience (single exposure to each macro-nutrient) facilitated feedback mechanisms, and hence affected their dietary preferences. This means that horses are, indeed, able to recognise foods based on their nutritional content.

There are, however, lots of other questions that remain unanswered. For example, would horses respond in the same way to minerals? And, how important is smell and taste in the actual selection process of foods, especially when dealing with a variety of plants in pasture environments?

Some researchers have suggested that because horses (being single stomached, hindgut fermenters) have a slower post-ingestive feedback, they have a higher risk of ingesting too many toxic plants before feeling the poisonous effect and, therefore, they must have adapted to rely much more on the smell (and taste) of foods, as it allows them to make the decision to eat a plant or not before it is too late (the so called pre-ingestive feedback!).

Our questions 

Let’s go back to our questions. Do you think horses use their senses of smell and taste in the same way we do? And how important is smell and taste for the selection of their diet?

We can answer with certainty that, to a large degree, the senses of smell and taste in horses work in a similar way as seen in other mammals (i.e. via olfactory and gustatory receptors that provide messages to the brain). However, the actual smell and taste sensations and abilities vary amongst species. By simply looking at the animal’s design, horses seem to beat humans when it comes to smell!

What about the second question? How important are smell and taste? It seems that this remains largely unclear – yes, horses may exhibit preferences for flavours and nutrients, but these are already established with familiar concentrates and forages. What about novel tastes and smells?

At the University of New England, we are currently conducting new research that will, hopefully, shed some more light on how smell and taste work when horses are dealing with a variety of choices and plants. The research is in its early stages, but if you would like to receive more information on this subject and research, you are welcome to contact me directly by email.

Practical applications 

Can we apply this information about equine flavour preferences to our feeding management?

Adding flavours to horse feeds can be helpful when you are dealing with foods that may be very unpalatable. Traditionally, molasses has been used extensively, but we now know that adding more sugar content to a diet is not always a wise choice. Some horse owners, for example, are dealing with horses that are sugar-sensitive and need to be fed high-fibre, low-calorie foods, such as soybean hulls or beet pulp.

A solution could be to try flavouring the feed with non-nutritive (sugar-free) flavours. One example could be lucerne juice, which you can make in your kitchen by blending and sieving some fresh lucerne plants. Success, however, will depend on the individual’s preferences, and some trial and error will be required! Non-nutritive flavours will also allow you to provide variety without changing the actual diet (this may be especially important when trying to add variety to cereal-based diets without having to change the actual feed).

These flavours can also be used when a new food is introduced or when horses are moved/transported to unfamiliar places – competition or new facilities.

Learning more about post-ingestive feedback mechanisms in horses will be very useful when dealing with pasture planning and rotation. Horses can be very selective grazers, over-grazing certain preferred plants and under-grazing others that may be good to eat, but not preferred. It may, therefore, be useful to introduce them to the idea of sampling the less palatable or unfamiliar species – for example, by spraying a familiar and well-liked flavour. This would allow the animals to try these plants and may will lead to better utilisation of the pasture and a more even pattern of grazing which, in turn, will encourage biodiversity.

It goes without saying that we should never force animals to eat a particular plant species – for example, by providing only that choice – as this may cause excessive ingestion of one particular toxin. The idea of encouraging them to try new plants is only meant to facilitate learning about other food options, so they can gain experience in mixing and matching nutrients, as well as detoxifying from their system toxins from different plant species.

The role of toxins in plants is generally misunderstood and many people think that many plants are poisonous and can’t be eaten by horses at all. We are learning more and more that relationships between plants and herbivores is much richer and more complex so, in a future series, we will explain more about toxins in plants (grasses and browse) and about how this may affect horses’ foraging behaviour.


  1. Burritt, E.A., Provenza, F.D., 1992. Lambs form preferences for nonnutritive flavors paired with glucose. Journal of animal science 70, 1133-1136. 
  2. Mehile, R., 1991. Hedonic-shift consitioning with calories, in: Bolles, R.C. (Ed.), The Hedonics of Taste, Lawerence Erlbaum Associates, Hillsdale, N.J.
  3. Sclafani, A., 1991. Conditioned food preferences. Bull. Psychon. Soc. 29, 256-260.
  4. Provenza, F.D., 1995. Postingestive feedback as an elementary determinant of food preference and intake in ruminants. Journal of Range Management 48, 2-17.
  5. Pfister, J.A., Stegelmeier, B.L., Cheney, C.D., Ralphs, M.H., Gardner, D.R., 2002. Conditioning taste aversions to locoweed (Oxytropis sericea) in horses. Journal of animal science 80, 79-83.
  6. Goodwin, D., Davidson, H.P.B., Harris, P., 2005. Selection and acceptance of flavours in concentrate diets for stabled horses. Applied Animal Behaviour Science 95, 223-232.
  7. Cairns, M.C., Cooper, J.J., Davidson, H.P.B., Mills, D.S., 2002. Association in horses of orosensory characteristics of foods with their post-ingestive consequences. Animal Science 75, 257-265.
  8. Redgate, S.E., Cooper, J.J., Hall, S., Eady, P., Harris, P.A., 2014. Dietary experience modifies horses’ feeding behavior and selection patterns of three macronutrient rich diets. Journal of Animal Science 92, 1524-1530.

View this article “Making Sense of Taste and Smell: Part 2″ about equine flavour preferences by clicking here [wpdm_package id=52961 template=”link-template-button.php”]