I remember marveling as a child at the fact that each word used in the dictionary was also defined in that same dictionary. At the time, it seemed like one could learn English (or any other language) simply by taking each word and looking up all the words used in the definition, then all the words used in those words’ definitions, ad infinitum. Obviously, this is no way to actually learn a language; one needs a base vocabulary of at least several thousand common words to glean much from dictionary definitions. However, the dictionary does still display an interesting property of interconnectedness in the way that all its words are tied up with each other. Each word maintains a consistent “shape” (or multiple “shapes”) in all its uses, and that same “shape” is also formed by the series of words representing its definition. For example, every place the word “dog” is used, it sits between other words which it “fits” correctly if the meaning of “dog” is something like “a highly variable domestic mammal (Canis familiaris) closely related to the gray wolf” (from Merriam Webster).
One way to think about this interconnectedness is to view the dictionary as a puzzle, with each word represented by a piece, and each piece fitting neatly with all the pieces it relates to (i.e. the ones where it’s a part of the definition).
Luckily for dictionary writers, these connections are intrinsic to the language, rather than a unique property of the dictionary. Creating a consistent dictionary only requires identifying the definition of each word as it is used (though in certain instances it can be quite difficult to identify all the uses, like for the word “set” with its 25 definitions). Dictionary writers don’t need to worry as they’re writing about whether the entire dictionary will turn out to be consistent, since this consistency is enforced by usage – by the fact that each word means something.
In many ways, the brain can be seen as a similar type of interconnected system. Each of our ~100B neurons is connected to ~1,000 – 10,000 other neurons, making for an almost unimaginably large puzzle. Within this vast system we find multiple levels of “consistency”.
At the level of neurons, consistency is related to Hebb’s Rule, which states that “neurons which fire together, wire together”. The neural pathways which are frequently activated will be strengthened, and those which do not will gradually fade away. We can actually see this “consistency” develop in the brain during the early childhood years, as the large amount of connections developed during our first couple years get pruned down to only the most necessary ones.
Returning to the puzzle analogy, it’s as if the borders of each puzzle piece are overlapping with each other to start, and then get refined to neatly fit with each other according to the patterns of activity that actually arise.
We can also see a degree of consistency at the higher level of concepts. This type of consistency seems similar to that observed in the dictionary; our concepts all mean something, which results in them all being interconnected with each other in a coherent way. We can rapidly switch between congruent concepts without issue – for example, a car can effortlessly be viewed as an assemblage of car parts when at the repair shop or as a warm refuge when seeking shelter from a cold winter day. While similar to the dictionary consistency, this version runs deeper, as the concepts we hold are far richer than dictionary definitions.
Moving up one level further, we come to the self, which is often treated as the hallmark of consistency; individuals are considered to be unitary. However, it seems that at this level the brain actually displays a remarkable degree of inconsistency. The things we want can be completely different from one day to the next, and the way we see ourselves and others can shift equally as quickly. Even the things we believe about the world can vary depending on our current context or mood. The attributes which constitute us don’t fully permeate the brain, and are instead limited in scope to specific areas of varying sizes. For some attributes, these areas can be quite large (for example, a parent’s desire to be a good parent), but in few cases will they span the entire space (as the parent may discover when their child is being especially difficult).
This inconsistency is observed to the extreme in individuals with multiple personality disorder. For these people, it’s as if multiple selves occupy their single brain, each with their own mannerisms and memories. Returning once more to the puzzle analogy, it’s almost as if several different puzzles reside in their brain, each representing a particular individual, but almost fully distinct from each other.
Luckily, for most of us the inconsistency is not quite so drastic. However, our goals and drives can still shift from day to day, which can be jarring if one is holding onto the idea of a consistent self too tightly. It seems important to remember that we are not like the dictionary, with each of its words neatly fitting together – rather, we’re made up of an amalgamation of parts that arise to address various contexts. From the perspective of the brain, there’s no need for a sentiment like “I am a good parent” or a goal like “I want to raise my child well” to permeate across all contexts. Remembering this disparate nature of self can help one be more gentle with oneself when two different parts come into opposition; after all, both parts are “you”!
The first part about going to the definition of all the words included in the definition of the first word reminds me of the answer to a puzzle. 100 prisoners are numbered, and their numbers are randomly put into 100 boxes. They have to go in 1 by 1 and they can open 50 boxes. If all prisoners find their number they are all set free. The best path would be to open a random box, then go to the number insides box, and so on
You’re right, that does feel analogous in many ways! Veritasium has a great video on the topic – interestingly, the prisoner’s strategy only works if all the boxes are not interlinked with each other (i.e. not part of the same loop).