I own the most beautiful horse in the world. Well, maybe I’m a little biased.
We all have our own biases – things that skew our attention away from seeing the bigger picture or considering other options. Sometimes our biases are just towards our likes. Most of you reading this article will be biased towards horses, for example.
I like mares. I like chestnuts. I like the chiseled features of a quality Thoroughbred. You could say that I just seem to have a natural bias towards them, but maybe it’s because of Chelsea – a chestnut Thoroughbred mare who I have owned for 23 years.
A mental shortcut with important consequences!
Biases are mental shortcuts that help us navigate our lives efficiently. They might limit our choices, but some psychological studies of happiness have found that the more narrow our options, the happier we are with our choices.
So, why should you be aware of biases beyond knowing your likes and dislikes? Well, recognising our biases can save us from making unfounded assumptions or contributing to unhelpful myths.
Scientists have described multiple types of biases. They need to be aware of the biases that might skew the findings of their research away from reality.
If biases are allowed to impact their research design or interpretation of data, then the findings of experimental or laboratory research might not reflect the objective world as much as we had hoped. The implications can be profound.
Some common bias examples
I don’t want to give my horse medication that was only tested on horses who were naturally resistant to the side effects. I don’t want to apply training techniques that were approved by scientists who only paid attention to the horses who developed quickly or who only included intelligent horses in the sample to start with. I don’t want my vet to recommend euthanasia because no other horse they have treated for a particular condition or injury has survived.
In the first two examples, the scientific experiments were affected by something called selection bias. Horses were selected to support the desired or expected outcome. At least, horses were not selected to challenge the desired or expected outcome.
The third example illustrates a bias called the availability heuristic. This is where we are biased by the examples most available in our minds.
When the vet in the hypothetical example weighs up the likelihood of a procedure’s success, they may only recall the cases that come to mind the easiest or were most emotionally charged – such as failed procedures. Of course, they may also over-inflate the chance of success because they find it easy to recall some extraordinary cases.
The vet’s recommendation could be complicated by another bias called recall bias, particularly where the most recent procedure was unsuccessful and resulted in euthanasia. Because of recall bias, we are more likely to overestimate the likelihood of an event happening when an example occurred recently – e.g. a rider fatality, a bushfire, a tricky ride on a sensitive chestnut Thoroughbred mare… Similarly, we are likely to underestimate the likelihood of something occurring when we find it hard to recall a time when it occurred.
Recall bias can explain why some people who have not experienced a bushfire, a horse with tetanus or a bad fall have not made a bushfire action plan, vaccinated their horse or routinely wear a helmet when riding.
In these instances, someone’s inability to recall the worst case leads them to a biased risk calculation, and usually a greater acceptance of risk. Like the availability heuristic, recall bias may also explain a spike in people’s bushfire, disease and personal injury prevention behaviours immediately after a horse dies due to bushfire or tetanus or a rider suffers a debilitating head injury or dies from a fall without a helmet.
These same behavioural failures (failure to make a bushfire action plan, vaccinate for tetanus or wear a helmet) could also be the result of a confirmation bias.
A confirmation bias can be seen in people when the use outlying cases to support their views, even if they only occurred once. These might be references to the time that someone had a bushfire action plan but still lost their horse, vaccinated their horse for tetanus but then the horse had a deadly reaction to the vaccine, or knows someone who died following a fall when they were wearing a helmet.
In these instances, confirmation bias causes people to over-emphasise information consistent with their beliefs and behaviours – and trivialise information that challenges those beliefs and behaviours.
Confirmation bias may be particularly influential when you have invested significant time, money and personal identity in something like a system of horse training or a ‘home-bred’ horse. It leads you to cling on to your investment instead of allowing you to consider new information that challenges (ie disconfirms) your attachment with an open mind.
The IKEA effect
In fact, the attachment that people show to home-bred horses can be explained by the IKEA effect. One of the more recently coined biases (dating to 2011), this cognitive bias refers to the high value people give to objects they have created or helped to produce themselves, often giving them some pride in their achievement. The IKEA effect may explain why some horse people value horse they have bred or trained themselves higher than if the same horse had been bred or trained elsewhere.
It may also impact our decision to persist with a horse that we have bred or trained ourselves even when we know that the partnership is not ideal, enjoyable or safe.
Let’s return to my beautiful Thoroughbred, chestnut, off the track mare. Did I mention that she was quite a handful in her youth? It is easy to think of chestnut Thoroughbred mares who are sensitive and difficult to handle, and to forget those who did not fit this profile, or discount them as anomalies or exceptions (“she was different because she was trained by so and so”).
However, if we sat down and wrote a list of all the chestnut Thoroughbred mares with whom we have interacted (assuming we have overcome our recall bias!), and put them into columns of difficult or not, we may find that there are similar numbers in both columns.
We may also find that even if the chestnut Thoroughbred mares we have known in our lifetime have been difficult, there may only be two or three that we have known – certainly not enough to generalise our experiences to all chestnut Thoroughbred mares.
Conviction, in-group and information biases
Did I mention that Chelsea has always been a windsucker? Some people think that horses learn to windsuck by copying others. If these people are recognised leaders who speak confidently and unwaveringly, we may be more likely to believe this theory due to a conviction bias. That is, if they are so convinced by their argument and willing to defend it, we feel that it must be true.
If the majority of people in our social group, riding club, online forum or agistment centre share the same view, we may be further impacted by an in-group bias which takes its strength from a human desire to fit in and feel part of the ‘in-group’.
If we are challenged by someone else who tells us that horses are not social learners and therefore cannot copy windsucking, we may turn to google to find articles that support our argument that windsucking is a copied behaviour.
However, when we do this selective searching, we introduce an information bias to our view. The problem is that whilst we find supporting information, we may not seek or acknowledge conflicting information. This makes it more likely we will find evidence of proof than evidence of error.
It may be the case, for example, that there are multiple horses on the same property who windsuck. If the horses are all related or the same breed (e.g. Thoroughbreds), another plausible theory might be a genetic element. Accepting the ‘copying’ or ‘genetic’ theories may bias us against considering that there may be different contributing factors. For example, the common approach to feeding, housing, weaning, etc., or the way in which Thoroughbreds are managed (early weaning, social segregation, stabling and feeding regimes that are low in roughage and high in concentrates) has been shown to increase the likelihood of windsucking behaviours. In other words, it’s our husbandry that is causing windsucking.
Moreover, if our online search terms were ‘genetic basis for windsucking’ or ‘horse breeds that windsuck’, the opinion we form will most likely be subject to selection and confirmation biases. In short; we find people who believe that windsucking is copied or genetic when they are the only terms we search for.
This is why experimental scientists go to great lengths to ensure that their research participants have been randomly selected, thereby minimising the effect of a selection bias.
Biases have consequences
I might hear someone talking about how sensitive chestnut horses are. A confirmation bias could lead me to remember the time I thought my mare had broken her leg but the vet discovered a hoof abscess.
Confirmation bias might lead me to interpret this event as evidence of just how over-reactive and sensitive chestnut Thoroughbreds are.
Seeing as she is a mare, I might also be influenced by social constructions of human females and also describe her as a ‘drama queen’. This would be a kind of bias called a fundamental attribution error where instead of thinking about Chelsea’s reaction in the context of what else may be occurring, such as a very painful hoof abscess or an embedded nail through her sole, I judge what I see as some innate quality within her.
The pressures of these biases on my decision-making might prevent me from instead thinking ‘wow, imagine how painful a hoof abscess must be, especially when 25% of your whole weight – roughly 125 kilos – is being borne on a structure that is impressively intricate, critical to your survival and the function of your entire circulatory system’. I might also forget that whilst Chelsea was a handful in her youth, my horse skills were also youthful and she was the first Thoroughbred I had owned.
My biases may also prevent me from recalling the occasions when this same mare has stood patiently without anaesthetic whilst I have cleaned and bandaged her exposed bone and injury without assistance.
Being aware is your challenge
Of course, biases are far from straightforward. They interact with our beliefs, cultural attitudes, social expectations and personalities.
A belief that female riders are more sensitive (a fundamental attribution error) might prevent us from being able to recall male riders who have been successful in dressage (selection bias) – a sport we often equate with communication, sensitivity and other feminised attributes (confirmation bias).
On the other hand, a belief that male riders are braver than women could directly result from only ever seeing male riders participate in rodeo (availability heuristic). These biases all prevent us not only from seeing the bigger picture (‘rough’ female riders, ‘sensitive’ male riders, female rodeo riders) but more significantly – from seeking evidence that challenges our beliefs or acknowledging that exceptions may exist in other times, places and populations.
If you take the time to question, challenge, reflect, think, ask, and research, you may find counter-cases to any and all of your beliefs.
You may be quite right in your biased assumptions, but if you entertain the idea that you may be wrong, you create the opportunity to find an exception.
Maybe, the more exceptions you find, the more balanced and accurate your view will be. You may even find that when you ride a chestnut Thoroughbred mare without expectation, she will behave in pleasantly surprising ways.
There is a lot to gain from developing an awareness of your biases and how they influence your thoughts and behaviours. In particular, most of us (that includes you) will be subject to a superiority bias. You will think that you are less biased, more objective and more open-minded than other people that you know or with whom you are familiar. You may be a better rider, a more ethical horse owner, better at towing floats or more informed than other horse people who you know.
When you first read this article, you may have been nodding to yourself as other people came to mind when you were reading about the common biases. Now read the article again and see where your own views and behaviours may have been subject to bias.
There are over 50 biases. You can find out more about them on the internet, but be careful. You are more likely to forget the information that you find online. That’s a phenomenon explained by another recently coined bias; the google effect…
Books we recommend:
This original and insightful book explores how horses can be considered as social actors within shared interspecies networks. It examines what we know about how horses understand us and how we perceive them, as well as the implications of actively recognising other animals as actors within shared social lives. This book explores how interspecies relationships work, using a variety of examples to demonstrate how horses and people build social lives. Considering horses as social actors presents new possibilities for improving the quality of animal lives, the human condition and human-horse relations.
This edited volume demonstrates the broader socio-cultural context for individual human-horse relations and equestrian practices by documenting the international value of equines; socially, culturally, as subjects of academic study and as drivers of public policy. It broadens our understanding of the importance of horses to humans by providing case studies from an unprecedented diversity of cultures. The volume is grounded in the contention that the changing status of equines reveals – and moves us to reflect on – important material and symbolic societal transformations ushered in by (post)modernity which affect local and global contexts alike. Through a detailed consideration of the social relations and cultural dimensions of equestrian practices across several continents, this volume provides readers with an understanding of the ways in which interactions with horses provide global connectivity with localized identities, and vice versa. It further discusses new frontiers in the research on and practice of equestrianism, framed against global megatrends and local micro-trends.
This article, titled “You’re biased!” was published in Horses and People September-October 2019 magazine.