An article by Lucy Rees for the Equine Behaviour Forum;
A kind-eyed lady on a charming little Spanish grey: not the sorts to invent themselves problems. But he would not turn right. When she asked him, rather crudely pulling her right rein and jamming her left foot in his belly, he slewed his head left and followed it. They circled as if round a roundabout, took the last exit, and ended up where they wanted to go, but whichever way she asked him to turn, he only went left.
I looked in his mouth. He was in a classy dressage school so a simple tooth problem seemed unlikely – but then, so did her wooden aids. He was 7, his teeth perfect: out of luck. .
In dealing with “problem” horses, diagnosis is all-important. The commonest cause for bad behaviour is pain, so we eliminate that first. Another is physical inability; in this case, maybe, neck vertebrae dislocated when wrestling with men or posts during halter-breaking. But he turned his head equally easily both ways when offered carrots by his belly… Over to me.
When I rode him, he did turn right, but oddly. In response to the inside leg he flexed nicely to the right before suddenly jerking his head left and continuing right, leading with his shoulder. For a carrot held by my right foot, he turned his head, then a look of terror overcame him. He snatched the carrot and whipped his head left, quite differently to when I’d offered it from the ground. I took the saddle off, in case it was jabbing him, examined his faultless back, and jumped on bareback. Same deal. With me mounted and Eva offering the carrot from the ground, his problem was obvious. He couldn´t bear to see the rider with his right eye.
Loose in a round pen, he wouldn´t go right either; nor would he let me approach from the right, but swivelled to get me in his left eye. Bingo.
The problem, although common in much-handled colts, is rare in a 7 year old. We are obsessed about handling, leading and mounting only from the left, so the horse habituates to seeing you only with his left eye. Like many trainers, I´d thought that what horses learn with one eye they don´t transfer to the other, but a paper by Elaine Hanggi on interocular transfer of discrimination learning proves me wrong. When a horse with one eye blindfolded learns a visual discrimination task (for instance, that a bucket marked X always has rewards in it, but a bucket marked O never does), he still performs correctly even when the blindfold changes sides. But discrimination learning is not habituation, which possibly uses different pathways in the brain.
Horses are extremely acute about where in the visual field a stimulus (or person) can be considered safe or dangerous. When a horse has had a bad experience preceded by a visual cue in a certain part of the visual field (for instance, when being branded), he may happily permit you to be anywhere else but at that point. This can give rise to surprise panics, particularly if you persist in thinking that horses are controlled by ropes, not position and body language.
Similarly, if your approach to his left side always has a favourable outcome but that to the right is unpractised and unknown, the horse will turn to keep you in his left eye. The problem is so common, and I have overcome it so many times by treating it as a non-transference of visual habituation, that it seems likely that this is a different phenomenon to Hanggi´s discrimination, although I will freely admit that things ain´t always what they seem. We need some more experiments, as always.
Always leading a horse from the left, especially with pressure on the rope, creates physical problems as well as visual ones. Watch such a horse being led away: often, he is not moving straight, but with his hips skewed to the right. Every time he is stopped he turns left slightly. He may even be doing a shoulder-in, that best of suppling exercises, to the left. If he spends all his youth doing this, it´s not a wonder he lunges and turns to the left with ease, but not to the right. Significantly, horses that have grown up wild do not show this bias.
When you start riding, you are usually instructed to mount from the left, so that you don´t do yourself a nasty in the goolies with your sword, an indispensable piece of equipment in the twenty-first century. Since you insist that the horse stand still, he braces himself one-sidedly. This strengthens not only one side of his body but yours too. My friend Esteban Labari, a horseman and osteopath, has found that all professional riders have twisted hips, the right leg shorter than the left, and lop-sided back muscles, from repeating this nonsensical asymmetrical exercise many times daily.
Unconsciously we can create even more asymmetries. Many riders sit lop-sided or ride with one stirrup shorter than the other: usually it´s the left that´s longer, since the leather stretches gradually and imperceptibly with all that mounting. They do not notice, though the horse does. It would seem that we, being two-legged, would have more sense of balance than a horse, but we haven’t. The cerebellum, the wrinkly blob at the back of the brain that controls movements in equilibrium, is bigger and better developed in a horse´s brain than in ours.
Furthermore most of us are right-handed, and stronger on the right, so we ride with more pressure on the right rein than the left, especially at first. Although Eva didn´t sit lop-sided, she clutched her reins very firmly. As Spain has no teaching qualification, the few qualified teachers there are often have B.H.S. qualifications, and that criminal idea about contact being a kilo bag of sugar in each hand (sugar: why not razor blades, or dynamite?) What Eva perceived as a kilo in her right hand was considerably more than the kilo in her left. Since Querido didn´t want to turn his head right, he´d learned to stiffen all his right side against this uneven pressure, adding to his difficulties. And since she rode rigidly upright, looking fixedly between his ears at all times, Querido, lacking any clue as to where they were going, simply stiffened himself more as she applied more pressure. The riding instructor’s solution was to wrench him forcefully to the right, making him hate right turns even more, but Eva, being weaker and kinder, could not do it.
Once we´d sorted out his visual problem she found, to their mutual delight, that she could ride him bareback without reins and turn as easily to the right as to the left, using only her body. Curiously, and significantly, with the reins in her hands she couldn´t: she pulled, and he went left. We are creatures of great habit – we, not horses. Well, paciencia. At least she knows it´s possible.
But is all asymmetry in horses our fault? Pluvinel, one of the great 18th century masters, thought not. When his pupil Louis XIII asked him why most horses turn left easier than right, he replied “because they lie curved to the left in the mother´s womb”. For 11 months, without moving? He had not spent the hours we did in the days before scans, watching mares we hoped were pregnant to see the asymmetrical bulges in their bellies. Feet appear one side one day, the other the next. They move, those foals. It´s one of those pseudoscientific pieces of gobbledegook that horse people solemnly repeat to each other, delaying any advance in knowledge or technique.
Horses might, though, have hand preferences, as we do. About 85% of us are right-handed. As motor instructions from one side of the brain cross over to the other before being expressed, this indicates that most of us tend to use the left side of the brain, which specialises in logic, language and mathematics, more than the right, which deals more with spatial organisation, emotional response and insight. A disproportionate number of architects are left-handed, though few lawyers are. As horses are notably better at right-brain responses than left, we might expect them to be basically left-handed.
Paul McGreevy has recently studied this, in a wonderfully simple way. He watched youngsters at pasture and noted which foot was forward when they grazed. He watched them resting, to see which hip they rested. He also offered them a smell, to see which nostril they used first.
Nearly 50% of them spent more time with the left foot forward; about 40% were ambidextrous, and the remaining 10% were right-footed. (If you look at the feet of long-legged colts at pasture, you will often see that the left hoof is concave, long in the toe and short in the heel, where the right is stubby-toed, convex and long-heeled, due to this natural left-footedness. Sometimes, of course, it´s the other way round, but seldom.) As the youngsters grew up, being led from box to pasture daily, the left-side bias grew stronger, mostly due to the ambidextrous ones getting more left-footed.
In resting, they used both sides equally, which is not surprising unless they¨re lame.
In smelling, they tended to use the right nostril. Since the olfactory nerves don´t cross over, this is also what we might expect. Surprisingly, though, there was no correlation between an individual being left- or right-footed and which nostril he used.
This bias, though, does not necessarily mean that a horse is naturally suppler on the right, though it might mean that on flat ground in a straight line he might take the left lead more easily at the gallop. A ridden horse is likely to have rather a lot more reasons for such behaviour.
I´m not taking sides about this, of course, but until you´ve stopped dragging your horse around curved to the left, seen an osteopath, done months of exercises to readjust your musculature, learned to mount from both sides, checked your stirrup leathers and saddle flocking, and ridden with spring balances in both reins, don´t put all the blame on the horse for going one way more easily than the other.
The simplicity of McGreevy’s method gave me an idea. The first asymmetrical movement that foals practise is suckling. I watched my 4-month-old foal: he chose the left side of his mother 25 times, the right 23. He’s an ambisuckler. But are they all?
It seems to me that we could investigate this as a group, simply noting which side a foal suckles, and pooling the results.
True scientific investigation has its difficulties. One is that you have to have clear-cut questions. Ours is simple: do foals have preferences about which side of their mother they suckle?
Secondly, you aim to have numbers. Fortunately there isn’t any doubt about what we are counting, and there aren’t any half-measures, although we might add a category for rear-end suckling. This eliminates a problem we often find, that you can’t compare studies because they have not used the same criteria. For instance, in dominance hierarchy studies, some count only aggressive moves like ears back or head thrusts, others add other measures like tail lashing, some count moving away as submission, others score different moves as more or less dominant or submissive, and so on. It makes comparison impossible.
Third, you have to work out whether your results are due to chance, or whether they mean something more: statistical significance. If you toss a coin 6 times, in theory it should come down 3 heads, 3 tails. If you got 5 heads, 1 tail you’d think it was luck. If you threw it 600 times and got 500 heads, you’d think something was up. At what point should you start thinking you’ve got a loaded coin?
There are statistical tests which you apply to your numbers, which show you the possibility of your results being due to pure chance. P (probability)= 0.5 means that in half of all such cases (for instance, throwing a coin 100 times) chance alone would account for your results. In other words, it’s likely that nothing more than chance is at work. P=0.05 means that in only 5 cases out of a hundred your results could be due to chance: pretty unlikely. Anything below this level of probability is accepted as being indicative that something more than chance is at work. At p=0.01, it’s fairly certain that chance alone could not have given you those results.
There are a lot of statistical tests, designed to handle different sorts of measurements. Do foals have preferred sides to suckle? requires a different type of test to Does that change with age? One of the first things you do before starting an experiment is to find out whether there is a statistical test you can apply to your intended measurements, and what is the minimum number of repetitions required for that test. Always, the greater the number of observations you have, the more sure you can be that your results are not just due to chance.
For his foot-forward test, McGreevy used a minimum of 50 observations on each horse. This is easy for anyone who has a foal at home. However, if you made one-off observations on 100 foals, and 99 of them were suckling on the left, that would also be an unlikely result to come up with by chance. Whether we can use such one-off observations, and how; whether we can use strings of five, ten or more observations, to what extent we can lump them together, are beyond my feeble grasp of statistics. Do we have a statistician among our readers?
Another problem that frequently arises is observer bias. We have different eyes. McGreevy did all the observing himself. He knew how far forward a foot had to be before he counted it as being in front of the other, and being alone he could stick to his criteria. Someone else might have drawn the line somewhere else. Since one of the basic tenets of scientific investigation is that it should be repeatable, we have to make our lines clear. If someone else can’t repeat our results he will, after trying, usually contact us to see what we are doing differently.
As a first-year student I worked for my professor on an experiment that involved putting lizards, some of which had brain operations, in a Y-maze and seeing which way they ran. Sometimes the prof. did it, sometimes his assistant, while I did it on weekends and holidays. We kept the results hidden from each other, with a coding system so that we didn´t know which lizards had been operated. After three months we compared notes.
Almost all the prof’s lizards had a strong left-hand bias. Almost all the assistant’s had a strong right-hand bias. Most of mine didn’t care which way they ran, although some had a slight bias to one side or the other, and often they didn’t run at all.
The trouble was, they were the same lizards.
The prof. swore, the assistant protested, and I trembled like a good little student. Had we all placed the maze with the same orientation? Were the lights on or off? Were the mystery drillers at work on the walls? We got out the maze and the lizards.
The prof reached into the carrying box, caught a lizard, gripping it firmly in his right hand with his thumb pressed against its jaw, and released it in the maze. It ran left. After it had rested the assistant caught it again, with the same constrictive grip except that he was left-handed, and released it. It ran right. I scooped it up, letting it sit on my palm without gripping it at all, and released it. It just sat there. We all swore, and went to the pub to console ourselves for the loss of three months’ work. Stupid, isn’t it? It taught me much more than the experiment did.
In theory, the more observers we have, the more that observer bias should cancel out. In fact, the only bias that I can think of is that someone may be so convinced of the fact that they have a side- preferring foal that they ignore its suckling on the other side. Don’t do it. Every observation counts. What we do in everyday life is mislead ourselves by making up our minds and ignoring what doesn’t fit our theory. In science you can´t: it’s impartial.
Our criteria must derive from thinking about situations in which something other than free choice is affecting the foal: that the mother is tied up, or standing by a wall, or there is a person nearby. So we should all make our observations from, say, more than 20 metres away when the mare is free in a field, more than, say, 3 metres from a wall or fence.
Finally, there might possibly be factors that affect the foal’s choice. Maybe colts develop preferences more than fillies; maybe preferences increase or decrease with age; maybe Arabs have different ideas to other breeds. By recording the approximate age of the foal, its sex, and its breed or type, we might get enough numbers to do correlation tests.
In thinking this over, I am strongly reminded of Moyra Williams, the founder of this group and the author of several books about horse psychology. A human psychologist by training, she had a deep appreciation of scientific investigation and frequently had her students watching and counting to discover what horses were really doing, even in the middle of the night. She would have been pleased and proud to see her group undertake such a study.