A Simple Introduction To Mental Models
People have ideas. That sounds really simple, and it probably is, but sometimes it’s slept on. People do have ideas. Ideas about how the world works, ideas about how other people work, ideas about how things work. Those ideas are called mental models.
What are Mental Models?
Mental models are simple approximations of how the world works (”approximations” here is key because it captures a central premise of mental models: they cannot, and do not explain everything). They summarize the world around them, the various parts of that world and the interplay between them, and their own role in it.
Mental models are like architectural blueprints: they represent an external, greater reality. They serve to simplify the complex into the simple and understandable. The key test of a mental model is its utility: the extent and frequency of its success at doing the task it’s designed to do.
What Do All Mental Models Have In Common?
- Incomplete foundations. Some facts are obscure, like Norway’s annual defence budget. Other facts are unquantifiable, like the importance of positional discipline as a defender facing a quick transition in football. Even more, facts are incomplete: the size of the universe. The majority of facts, however, are simply unknowable. Mental models incorporate this uncertainty into their framing.
- Flexible. Like a chameleon, mental models are flexible, adapting and blending in different circumstances, to fit different needs. Mental models are also flexible enough to cognitively hold multiple opinions, sometimes contrasting, at the same time. The wide spectrum of use cases a mental model can apply itself to means that the variability can spill over into the negative sometimes: a mental model can be more minus than plus. Some of the best mental models can be applied to almost any situation but can be catastrophic in others.
- Sieves. Since mental models are just that: models, the extent of their functionality is dependent on just how much information can be filtered through that model. For example, if someone thinks that all vaccines are a plot by Bill Gates to reduce the human population and subjugate what’s left, every single thing they read or hear about relating to vaccines will pass through that “Bill Gates microchip” filter. Information that may contradict that will get filtered out, and only information that supports that hypothesis passes through.
What Are Some Examples of Mental Models?
- My personal favourite is Hanlon’s Razor. In one sentence: “Do not ascribe to malice what can be more easily explained by stupidity”. Useful for thinking about Nigerian inefficiencies.
- Emergent properties. Essentially, the whole is greater than the sum of its parts. Occurs when an entity has characteristics its individual parts do not have in isolation. Think consciousness in biological systems.
- Regression to the mean. I first encountered this in a Yale Introductory Psychology lecture on YouTube. In sum, extreme events or statistical outliers in any context (job performance, grades, goal-scoring)are more likely to be followed by events that are closer to the average. A real-world example is when a striker has a hot streak at the start of the season, overperforming their expected goals (or xG), but at the end of the season, their goal total is much closer to their total xG than you would have thought based on the early season form.
The Most Important Mental Model
“All models are wrong, but some are useful”. Mental models are, in short, approximations of the world around us. Models can’t capture everything, and can’t be flexible enough to accommodate everything. Models cannot describe the complexity of reality completely, and, as such, will fail at points. Some, though, are better than others, and the best models approximate reality to a sufficiently useful degree.
Importantly, a single model cannot explain everything. Information undoubtedly will get filtered out, even crucial information. That’s why a larger set of models is required. More models are better than one, and increasing returns to scale (another mental model) means that the efficiencies of single models are multiplied,