Wikipedia states that “a mental model is an explanation of someone’s thought process about how something works in the real world.” More accurately it is an representation of someone’s understanding of a certain piece of the world.
This representation is composed of symbols arranged in a relationship to each other. Let’s take for example a locked door. A simple model of this real world situation might include three symbols: the door, the lock, and the key. If one does not have the key then the lock keeps the door closed and it is impassable. If one does have the key then the door is no longer an obstacle and becomes passable. The way these symbols relate to each other forms the model of the locked door. And to a degree this model matches reality and thus is true to that same degree.
However reality is more nuanced than our simple model suggests. One could more perceptively understand the lock to be a physical system. And should one understand that physical system – that is to have an accurate mental model of how the lock operates – one has the ability to manipulate it in such a way as to open the door even in the absence of the key. Or perhaps one might realize that the door is composed of particle board and will cave in to a forcibly applied shoulder.
In our first, simpler model we take the lock and door to be fundamental entities. However on closer examination find that the lock is actually a physical system and the door is impassable only as long as it remains unbroken. Putting the sturdiness of the door aside, our expanded view of the situation has more than three symbols. The door and key are still only two, but the lock has revealed itself to be a constellation of symbols: pins, springs, cams, and more. We could this much further, describing the pins in terms of the metals they are composed of, and those in terms of their chemical and physical structure, and those in terms of their atomic compositions and bonds, and those in terms of quantum mechanics, and on and on. Looking at the world like this in everyday life is of course just silly and no one does so. Likewise I don’t need to know how an internal combustion engine works to know how to drive a car. Knowing that when I turn my key the engine starts and then powers my vehicle is an abstraction, but is all that matters to me functionally.
Like me and my car, none of us understands how everything in the world truly works. We instead understand the world through our symbolic representation of it. We even model each other. We might know exactly how a friend would respond to a situation. And then sometimes they surprise us – our model was off.
The basis of our models are our premises – our fundamental assumptions. Our premises are then things we take for granted in our mental models, like how I take for granted that my car will move when I press the gas pedal.
In some models, most notably mathematics and to a lesser degree science, we explicitly state our premises – our assumptions. In math we call these axioms and in science we “define our terms”. In everyday life our assumptions are much more likely to have been taught to us as children, and these we often carry throughout life without ever stopping to question if they are true or not.
Models are a abstract representation of reality, and therefore a unavoidably partial representation. Just like the written transcript of MLK Jr.’s famous “I Have a Dream” speech cannot capture his voice the way an audio recording could, and neither could hearing the recording compare to actually being there for his march on Washington, a representation will always miss out on some aspect of reality, and to that same degree will be inaccurate.
To make sense of the sheer infinity of reality, we use models, which essentially look at reality and say: this part of reality is important, and the rest doesn’t matter. This is of course quite useful, and many models we use are quite accurate, but all models fall short of perfectly capturing that which they model. The key to building a good model is deducing which aspects of reality matter, and it is important to note that this decision happens outside the model itself. Choosing what parts of reality to include in one’s model is where assumption enters the process. The person creating the model to a degree arbitrarily determines what part of reality does and does not matter.
The arbitrariness can be tempered by comparison of the model to reality. If the model fails to predict reality at all, one can safely conclude that the inherent assumptions of the model are untrue. Counter-intuitively, models are the most dangerous to truth when they are most accurate – accurate models can lull one into a false sense of security, a false belief that the model completely represents reality as it is. When this happens the assumptions that lie at the base of the model are no longer noticed. A very huge and tragic example of this is the assumption that underlay the formula that spurred the housing market bubble and crash of 2008.
Using models is unavoidable, nor should they be avoided. However we should strive to be aware of the assumptions that underlie our models, as our starting premises will inevitably color the understanding our model produces. When we understand our models and their assumptions, they are a powerful tool for us. When we accept a model blindly without questioning its underlying assumptions, the model controls us.
I write this post to provide a base for my further ideas. I will be as diligent as possible to declare what assumptions I make in my thinking, this post itself being part of that diligence. All ideas I put forth on this blog are of course models, and thus are not perfect representations of how things are and should be examined critically. This post, along with that on critical thinking, are my boilerplate entreaties that you examine my ideas on their own merit and not just take my word for them. That being said, I believe what I have to say has value and truth, or I would not waste my time writing it nor your time reading it.