There are only a few acts more daring than encouraging people to think for themselves, but what does it mean to ‘think’? By following the computational theory of mind, we are exploring cognition the same way we would explore any information processing system. That is, we start describing ‘thinking’ as computing and ‘beliefs’ as information. This has very interesting implications as we start exploring this metaphor more profoundly and look for the analogous counterpart of other elements of an information processing system.
In philosophy of mind, the computational theory of mind (CTM), also known as computationalism, is a family of views that hold that the human mind is an information processing system and that cognition and consciousness together are a form of computation.
When dealing with data, we are able to synthesize new information by processing it through functions, that is through relationships or expressions involving one or more variables. We can even further develop this process by nesting these functions and creating ‘meta-functions’. I would say that this is how algorithms are built, which conceptually matches how our minds work. We create these heuristics (or functions) which through experience can be developed into more complex structures that allow us to interact with the world. We are constantly trying to optimize these models so that we can maximize the utility we get from our cognition through reinforcement learning, which put into simple terms is just trial and error. I do not think that it is a coincidence that some of the state-of-the-art machine learning algorithms are based on this optimization process, such as Long Short-Term Memory (LSTM) networks.
Long Short-Term Memory (LSTM) networks are a type of recurrent neural network capable of learning order dependence in sequence prediction problems. This is a behaviour required in complex problem domains like machine translation, speech recognition, and more.
We can also think of cognition in a more technical frame, and start considering what would be the equivalent of ‘variables’ and ‘operations’ in our minds. I believe that the former are fabricated from our sensible experiences which are later named/transformed by another pre-existing function. This also has interesting implications. If our cognition is built upon previous information-processing models, how were we able to build our first mental-function? Well, there are brilliant minds, such as Noam Chomsky, who actually claim that we are born ‘wired’ with these structures which ultimately allow us to develop language. Similarly, the ‘operations’ that allow us to play with information are best described as a kind of universal syntax. In a way I would like to believe that Ancient Greek philosophers already had some awareness of this as they elaborated on the notion of ‘logos’, that is the divine reason implicit in the cosmos, ordering it and giving it form and meaning.
Chomsky based his theory on the idea that all languages contain similar structures and rules (a universal grammar), and the fact that children everywhere acquire language the same way, and without much effort, seems to indicate that we’re born wired with the basics already present in our brains.
This post has been very casually and selfishly written so that I could explore some of the connections I see between cognition and data processing. I believe there are many more similarities, and even differences, between our minds and information-processing systems which I did not elaborate on. A lot of the ideas I touched in were gathered through my knowledge in psychology and theories from Steven Pinker and Noam Chomsky. I hope to continue to develop these ideas as I explore some of the applications of computational modelling in social sciences.
emiliovel 