I want to understand behaviors of human through mathematical analysis. I dream of a unified theory of human behaviors. It may sound ridiculous to some people. Especially considering the “Interpretivist” movements in the discipline of behavioral science, perhaps, my research is not the right direction to move on. Despite, I am interested in this topic due to the huge number of potential applications I can see. It is my way of staying foolish.
I’m an engineer working on Human-Computer Interaction where people are fascinated by both human and technology. Therefore, I want to explain human’s actions using mathematics. It is a difficult task because of the huge number of potential variables affecting human behaviors. But the task becomes easier when we try to explain the huge amount of data that people disseminate. Humans have been generating a large quantity of data every moment from the beginning of its existence (hieroglyphics, texts, audio, videos). Now with the advent of the internet, smartphones, and wearable devices it is even easier to record the data that humans generate. Analysis of such data provides a strong clue on how humans respond to various stimulants. Doing this analysis in an unconstrained setting is difficult, so I am interested in a constrained setting which also allows enough freedom for exposing enough humanly traits. One such setting is the formal human-human interaction scenarios like public speaking or job interviews. We can do experiments to answer a lot of questions: How humans respond? What makes them happy and why? We can also analyze what methods successful speakers use for better communication and why? Answering these questions would give us clues to answer more fundamental questions: why humans behave the way they do? I envision that one day these questions would enable us to see the defining characteristics of human’s responses; leading us to a formal (i.e. mathematical) and unified theory. Unlike traditional HCI practitioners, I believe it is easier to come up with a unified mathematical theory of human behavior than to come up with the same for nature.
This train of thought has many practical implications (i.e. the idea wouldn’t have to starve to death for lack of money). Explaining human behavior would enable us to measure human qualities. For example, what might be a measure of good public speaking behavior, or a good husband’s behavior, or a good teacher’s behavior? Maybe one day we’ll be able to measure those just by analyzing data. If we can measure human behaviors we might be able to predict it too. Wouldn’t it be nice to automatically predict the impact of a presidential speech before actually delivering the speech? Or, to practice and prepare for a job interview just with the help of a computer? We might be able to catch potential serial killers or greatest criminals from their behavioral data.
The recent developments in machine learning, especially Deep Learning, are enabling computers to do almost anything that humans can do. If it becomes possible to emulate human minds through machine learning, then why wouldn’t it also be possible to learn human behaviors using a machine (both as an external or internal agent of society; using the exact same technique that interpretivists utilize now)? I believe that deep learning can bring major break-through in behavioral science, clinical psychology, and medical science in general. My research can be considered that kind of research in its infancy.
Mathematical solutions might not be the most user-friendly but it is reproducible, verifiable, and formal (thus implementable in a computer). It does not depend on the clairvoyance of the human soul. The rigor of logic appears much more beautiful to me than the power play and authority of the “experts”.
At this point, a question typically arises is — Why artificial intelligence? Crowd-sourcing is currently being used to bypass many problems in AI. Why I’m not interested in solutions like that? Crowd-sourced software appears to be giving an answer without actually knowing the question. It is not praiseworthy because it looks like cheating when people solve artificial intelligence using artificial-artificial intelligence. This doesn’t mean I don’t like crowd-sourcing. In fact, crowd-sourcing is one of the tools that I regularly use. However, I want to use the crowd intelligence to improve artificial intelligence, not to replace it. For example, crowd-sourcing could be used to collect massive ground truth data. But “using people to replace Google” doesn’t seem to be a correct approach to me. I like the concept of the “people teaching machines for serving humans better”, than “converting people into humanized-machines”.