Contemplating the Welfare and Training of Horses

How horses find their way home? By Suzanne Rogers.

Have you ever wondered how squirrels remember where they hid their food? Or how animals that migrate know the route to take each year? Or how birds living in large groups remember where their specific nest is among thousands? These are just some of the questions that spurred research into how the brain records information about the environment and the animal’s position in that environment. This work can help us to understand how feral horses find their way around their range and even how our domestic horses can find their way home if they escape or they end up without a rider.

Map and compass

Animal navigation is a complex process that we are only beginning to understand: first the animal determines its current location in relation to what he/she knows about and then chooses the appropriate direction of travel towards its destination. This is called the ‘map’ and ‘compass’ system (reviewed by Frost and Mouritsen, 2006). [N.B. for brevity I will use ‘she’ when referring to a horse]

Horses that live in large areas naturally ‘home range’- they move to different parts of their range as the seasons affect the available resources. Homing can be defined as navigation within near- to medium- distance unfamiliar space (reviewed in Frost and Mouritsen, 2006). Herds use the same routes every year and the alpha mare, who will have made the journey many times before, leads the movement of the herd within its home-range. So, how does the alpha mare know where to go?

Making maps

Since the 1970s, researchers have been studying the cellular mechanisms by which the processes of cognitive mapping, spatial learning and spatial memory occur. This just means studying at the level of cells how animals map their environment, learn about it and remember it.

Let’s consider the first time a horse moves to a new part of the herd’s home-range. When an animal is placed in a new environment, the brain processes sensory information about its surroundings – for horses, a great proportion of this information will be visual. Within minutes of experiencing the new ‘place’, about half of the million specialized brain cells called ‘place cells’ (first described by O’Keefe et al., 1971) in the part of the brain called the hippocampus are fired. The ‘view’ is split up into overlapping fields (‘place fields’) and each field is assigned a place cell. As the horse’s head moves into a place field, the cell assigned to that field fires and some cells stop firing, providing orientation in the field (reviewed in Rotenburg et al., 1996).

The place cells recording these overlapping fields are not actually next to each other – which is efficient because each place cell can be used in more than one spatial map. The hippocampus records the memory for several weeks and only if the information is to be retained does it move to the part of the brain where information is stored, the cerebral cortex (reviewed in Rotenburg et al., 1996). This mechanism has been proved for ‘near space’ navigation but is also relevant to longer distances because during a journey there will still be shorter-term destinations (e.g. resources along the way such as rivers to drink from).

Learning the route

Until 1996 scientists were not sure whether animals learn about their surroundings as a map or react automatically to various stimuli they come across. Two research groups helped to solve the mystery and significantly added to our understanding of cognitive mapping (Tsien et al. and Rotenburg et al. 1996). Both groups, using different methods, disrupted the way that connections between cells change in strength (called synaptic plasticity, which occurs during learning) and found that indeed, the animals were no longer able to navigate around their environments – spatial learning was affected. The involvement of synaptic plasticity shows the link between learning and spatial orientation so we now can hypothesize that animals, including horses, learn about their surroundings as a map rather than just reacting automatically to things in their environment.

Since this early work into place cells, two other types of brain cell have been discovered – head direction cells (HDCs) and grid cells. HDCs are those that fire when an animal orientates its head in a certain direction (Knierim et al., 1995). There are different HDCs for different head orientations and they are influenced by landmarks, and information concerning how the head moves.

Later, grid cells were discovered (Hafting et al. 2005). These are found in the part of the brain called the entorhinal cortex and fire strongly when an animal is in specific locations in an environment. Grid cells have multiple firing fields, which tesselate the environment in a hexagonal pattern. The grid cells are anchored to external landmarks, suggesting that grid cells may be part of a self-motion-based map of the environment. It has been suggested that grid cells add a ‘place code’ in the entorhinal cortex, providing associations between place and events (for example to associate a place with good resources) – needed for the formation of memories (Sargolini, 2006).

Finding the way

Exactly how animals navigate long distances is still being researched but there are some interesting concepts that are probably relevant to equine homing behaviour:

  1. Minimization of stimuli: We only pay attention to meaningful cues. In the same way that when asking for directions we don’t need to know the detail of who lives in all the houses you pass but just need to know where to turn right – or if you are looking for someone in a crowd who is meeting you at the airport you won’t notice the detail about the other people there – horses do not take note of all the cues in the environment, just the important ones (perhaps a big tree that is good for rubbing on or a sheltered site).
  2. Features theory: In the same way as above, we, and horses, can remember part of a route without having to remember everything about it.
  3. Context: In the same way that we might not recognize the postman out of his uniform if we see him in the supermarket there is a context-specific element to mapping.
  4. Landmarks: It is likely that horses use landmarks in recognizing their home range. The famous ethologist Tinbergen proved that animals use landmarks to find their nests. For example, if an insect’s nest is surrounded by a set formation of pine cones, the insect leaves home, you move the pine cones to a new location, the insect will return to the pine cones and not to the nest. Landmarks that horses might use could be trees, boulders, houses etc or could be patterns of smells.
  5. Prototype theory (Hernstein, 1976): if a route is followed many times then the ‘ideal’ will be remembered. For example, if you do the same journey to work every day you don’t remember the first time in particular, you only remember specific times if they were very good or bad. In the same way, if a foal takes a similar route many times she will only remember some of them.
  6. Central pattern generators (Randall et al. 1997): are sets of brain cells that activate cells resulting in movement according to pre-set patterns. When a horse sees a visual stimulus associated with its home range, central pattern generators could cause automatic walking behaviour towards that visual stimulus. Animals do form a ‘map’ as described above but there might be an element of automatic behaviour within that map.
  7. Time: Horses need to know when to move from one area at a particular time to get to the new area at the right time for plentiful resources. It is suggested that a ‘time tag’ could be added to the memory formation (e.g. when the trees are green, the journey should be started).


When a horse is moving within her home range or being ridden out on a hack, she is learning about the environment – including about information about landmarks and features – such as areas that are sheltered or particularly good grazing areas. If the horse is wild, feral or kept almost naturally in a herd, as a foal she will follow the group to different resources at different times of the year. She will remember (via prototype theory above) the ideal situation with respect to, for example, crossing a river.

Physiologically, as the foal enters new environment, place cells will be fired within the place field, and when she orientates her head in a certain position, the corresponding head direction cells will fire, influenced by landmarks and information about the position of her head during orientation. When she is in some specific locations within her home range, grid cells will fire, to add the place coding to the memory associated with the event (good areas for food, for example).

Although further research is needed into how place cells, head direction cells and grid cells work together in determining orientation in space and in navigation, we are on the way to an improved understanding of the complex mechanism of homing behaviour.

By Suzanne Rogers

© Learning About Animals 2009

(Enjoyed this article? Then please donate a little to The Equine Independent to keep us writing. You can donate via paypal to – Many thanksm Emma Lethbridge -editior)


Frost, B.J. and Mouritsen, H. (2006) The neural mechanisms of long distance animal navigation. Current Opinion in Neurobiology 16, 481-488

Hafting et al. (2005) Microstructure of a spatial map in the entorhinal cortex. Nature 436, 801-806

Herrnstein, R. J., Loveland, D. H., & Cable, C. (1976). Natural concepts in pigeons Journal of Experimental Psychology: Animal Behavior Processes, 2, 285-302

Knierim, J. (1995) Place cells, head direction cells, and learning landmark stability. J. Neuroscience 15, 1645-1659

Liang, K.C. et al. (1994) Involvement of hippocampal NMDA and AMPA receptors in acquisition, formation and retrieval of spatial memory in the Morris water maze. Chin. J. Physiol. 37, 201-212

O’Keefe, J. and Dostrovsky, J. (1971) The hippocampus as a spatial map. Preliminary evidence from unit activity in the freely-moving rat. Brain Research 34, pp. 171–175

Rotenburg et al. (1996) Mice Expressing Activated CaMKII Lack Low Frequency LTP and Do Not Form Stable Place Cells in the CA1 Region of the Hippocampus. Cell Vol. 87, Pp. 1351-1361

Sargolini F, et al. (2006) Conjunctive representation of position, direction, and velocity in entorhinal cortex. Science 312, 758-62

Tinbergen, N. (1951) The study of instinct. Clarendon Press, Oxford.