Andrew McLean looks at a new way of understanding the way we relate to our horses…
I’m interested in the area of Attachment Theory, as it applies to the horse/human bond in the hope it may shed light on the elusive (objectively speaking) qualities of trust, rapport and bonding. Applying Attachment Theory to horse training is new ground, but I think it applies well, as the horse is a social animal too. It has been recently extended to describe not only the quality of mother–infant relationships and other caregiver-infant relationships, but also to relationships between human adults and more recently human-pet relationships. There’s a lot of evidence that dogs and humans attach to each other in a very similar way that humans attach to each other and I think that it’s likely that any social animal that’s in our domain, would share at least some of these characteristics.
The horse is a good candidate for that because of two factors – one the one hand he is the most fearful of domestic animals. The horse has the biggest amygdala (the part of the brain that modulates fear) of all domestic animals yet on the other hand his heart rate is significantly lowered when groomed by other horses at the base of the neck. It has also now been shown that the same lowering occurs when humans stroke horses along the neck at the base of the withers (many of you already do that). Stroking, oxytocin and heart rate lowering are all connected. So it seems that stroking at the base of the withers is an antidote for stress.
For horses there are important health benefits in being able to allogroom (stroke/scratch each other). Group-housed horses for example show significantly lower incidences of stereotypies (stress related issues such as cribbing, weaving etc.). As an antidote for fear and insecurity, it seems this is a really important site of attachment and bonding. These are areas that we haven’t been able to explain scientifically, that enhance or inhibit learning: trust, bonding, rapport, could be explained to some extent by these areas.
Attachment is the big one I’m looking at with horses, because Attachment Theory has provided psychologists with valuable assessment and treatment tools. I’m particularly interested also because the history of horse training is as much about not touching horses: we lead, we drive, we lunge, we round pen, all of those things are very much at a distance. Even when we ride a horse we have layers between us, it’s not a tactile attachment which has been shown to be important in close relationships. Touch is soothing. The need for touch is a well that empties fast and thus can never be filled, unlike other primary needs.
Touch is really important and when we do we really touch horses apart from grooming? People commonly tend to pat, and patting at worst tends to make horses more reactive, or vigilant and more alert, and at best is a neutral signal that doesn’t enhance learning. Yet we can’t stop patting horses, dogs and elephants it seems. It’s such a waste of time and emotion. In the end animals become habituated to it, but if we didn’t pat for good behaviour but just stroked at the base of the wither and said in a distinct voice, ‘good boy’ as a secondary reinforcement before we stroked, we would be more successful. Especially so if we ensure we never say ‘good boy’ without stroking directly after.
This was the sole positive reinforcer Manuela and Jo Formosa used to train Jo’s Worldwide PB to variations of light whip-tap cues for gait, tempo and stride length changes and to transform these to seat cues.
One French study purported to show that wither stroking as a positive reinforcer is a poor reward, however for some obscure reason the researcher only gave three rubs on the withers – conversely one of my Master’s students at Nottingham-Trent, Emily Hancock, has shown that some horses’ heart rates don’t fall for two minutes in the early stages, probably for the same reasons as isolated children sometimes resent touch. Social animals need it but don’t always seem to want it if they are detached. For me however, the importance of touch is far more than just reinforcement. It’s about security for an animal that may live in a world with no touch, yet craves it as a social animal. I think the genius of Kell Jeffery’s revolutionary breaking-in method that has since been embraced by myself and others such as Parelli, lies in the bareback bonding effect where the breaker spends a lot of time lying on the horse’s back sideways, longways, stroking and soothing all the time. Bucking is an absolute rarity in these early bareback sessions.
We should use touch, especially wither stroking more, and we should teach this in pony club, and all Federations as rewards. Because when you think about it, even with light aids, what’s rewarding about training from the horse’s point of view? Social contact via touch is natural and I have seen many horses that don’t initially like wither scratching with other horses (most likely because we rear them in isolation so much), yet when housed with other horses, they may learn to enjoy touch from herd mates. Those horses that I have seen that don’t initially seem to like human touch can learn to do so if you persist, and I think for their health it’s important you should. Their heart rates soon lower by the standard 10 beats per minute if aroused.
Other reasons that intrigued me to start thinking about Attachment Theory with other social animals such as horses and elephants was that I read a book about Harry Harlow’s famous experiments with Rhesus monkeys that showed what other researchers of children already knew – that if infants were deprived of touch from a caregiver they became lifeless after a few months and never developed physically or in language and after longer periods, they couldn’t be retrieved and many would die. The second intriguing thing was the new studies on synchronicity between humans and horses. Many studies have shown that when the partnership is good between rider and horse their heart rates tend to be synchronous. Linda Keeling in Sweden did some fascinating experiments (that we are about to replicate) showing that when a person leading a horse or riding a horse expects something bad to happen, both their heart rates increase synchronously. Maybe it’s just as simple as riders stiffen and horses notice it. But it is interesting and it does suggest we have selectively bred into domestic horses other interesting traits as we have with dogs that we don’t quite understand as yet.
To my mind, getting our training right by using the correct Operant Conditioning tools at the time, and having regard for the horse’s arousal, affective states and attachment needs, sums up perfect training. It’s the best we can do. It’s been a great journey for me and I’m happy to be always learning. I think we need to be honest about training too. We should recognise that where the aids that we use in horse training originate from pressure-release, then even the lightest aid is still a shrunken version of Negative Reinforcement. If we do it badly it’s crippling for the horse and leads to problems.
We should not abandon Negative Reinforcement for Positive. To do so suggests a poor understanding of cognitive training processes and worse still denies the fact that horses are dangerous animals if you don’t have control. Control has been facilitated for centuries by Negative Reinforcement and has allowed horses to be controlled in the terror of battle. When done well it can produce calm confident horses that you can rely on for safety – you won’t fear death and catastrophe to put your children on well trained horses in suddenly challenging circumstances. So the important thing for me is not to deny Negative Reinforcement, but to embrace it and be always mindful of doing it to the best of our ability, to shrink pressures to the lightest of aids and to reward with touch. We should also utilise Positive Reinforcement such as clicker training to facilitate training not only of hard to train movements but also for no good reason other than to teach your horse useless tricks that enhance your bond, like pushing a ball around or picking up your hat.
Andrew and Da Lyla
To really make good ground in our relationships with horses we should replace the shallow and simplistic notions of respect, dominance and submission (I’d like to wipe that one out of the dressage test) and just use Learning Theory correctly. The more you know about it, the more it explains the things that entrap your mind to thinking of submission and respect. Even when a horse invades your space he’s just learning by Negative Reinforcement to displace you. When he displaces you in the saddle, he’s doing the same. When he doesn’t load in the float, it’s a failure of the Negative Reinforcement of leading pressures. When he’s head shy, aggressive, lazy, go-ey, strong or heavy: same thing.
I know some people hate science, because their brains just don’t seem to see the world in a systematic way. But there is a saying: ‘what we can’t measure we can’t manage’, and the comforts we enjoy post hunter-gatherer are largely a result of categorising and measuring. It helps us understand the world so long as were not blinkered to assume that that’s all that’s important and we are open to our own arrogance. I’m always amused though, that the people who distrust the veracity of a scientific approach seem happy enough to jump on a plane. But when it comes to animals sometimes what seems soft and kind (ie allowing random acts of acceleration or line or displacing you on the ground or under saddle) are not healthy because they make horses insecure, and Learning Theory explains this well. But even in the absence of any deep understanding of learning processes, we need to remember that our job is to simply teach the horse what we want him to learn as effectively and thoroughly as we can and make him feel as secure as possible.