Big History

Check out a video I made to submit to the Bill Gates Big History Competition.

Here is a drawing that I created. It is obvious that there are certain features on this drawing that cause it to seem like it is female, but there are also features that cause it to be viewed as a male. 

Facial Recognition by Drawing Faces

Facial recognition is a powerful unconscious capability. Distinguishing between faces for those with vision, or distinguishing between voices for those who are blind allow humans to effectively communicate. How do we distinguish faces? Is distinguishing a face like decoding a written set of symbols that our brains learn to do at a young age?

I took a novel approach to understanding facial recognition. Two years ago I took a drawing class, and I used to think drawing an object was an innate talent that only certain people have. Just like anything else, it took a large amount of practice and time to figure out a viable method that produced an accurate picture.

But, in order to accurately draw an object, especially a face, takes understanding the intricacies about the face. These intricacies are easily calculate by our brains because our ability to distinguish intrafacial features as well as interfacial features. I only noticed certain facial features when attempting to re-create them in a two dimensional medium.

I noticed that proportion play a significant role in the accuracy of a face. Placing the eyes, ears, nose, mouth, eyebrows, and hair in the correct places relative to the jawbone, forehead, and chin caused the face to not appear distorted. The size of these facial features must also be in accordance to one another when attempting to recreate a face.

After relative proportion of size and placement of facial pieces, I noticed a pattern in shading. There was always light shading on the forehead, cheeks, lips, and chin. Accurately shading dark spots around the eyes, mouth, and nose also played a significant role in accurately depicting the face.

A drawing can never look exactly like a real face because there are thousands of various shade spots, proportion nuances, features that are derived from various expressions on that the human face conveys. But, with consistent shading in a given set of areas (forehead, mouth, nose, chin, and jaw), a sufficient depiction of the face can be drawn. Our brains have the capability of understanding the shade spots, and proportions of facial structures, and can then make conclusions based on them. What makes facial recognition powerful is our capability of distinguishing the nuances of different people's faces.

I noticed that every face is consistent with the way it looks in any sort of angle. Our brains understand that consistency, and remember the consistency of shading and proportion.   This is proof that our brain remembers the detailed proportion, and three dimensional intricacies among the thousands of faces we encounter on a daily basis. But, through practice of experiencing a face, we remember the formula behind each face.


It is an amazing task that our brains accomplish on a daily basis.

Collaboration on Art

As a Pre-Medicine student extra free time is a commodity. Being in California for a week allowed me to explore ideas that I have thought about, but not executed. Most of my ideas fall under the category of  self-quantified analytics, which is an individual's ability to view his or her's own progress on any platform. 

I was sitting in my Music History class a few weeks ago and it was the end of a long academic day. It was one of those days that required no thought. I went to school to hear my teachers regurgitate facts instead of inspiring the mind to create. We were having an interesting discussion on why Beethoven became depressed, and I thought to myself if five people communicate about one subject through language can they communicate in any other way? Can people communicate with each other without using figures, body language, sound, or smell?

After dosing off in class and thinking about this topic I began to brainstorm on my notepad. What if two users contributed to one blank canvas? Two minds collaborating to create a piece is a form of communication just like two people contributing to a conversation. They both are adding to a totality of one subject, or one canvas. 

A tablet, or smartphone application that allows two people to collaborate on one canvas with turns is one way this idea can be executed. One user will randomly be selected to go first and have their influence on the blank canvas. Each person will have a twenty second time limit to place something on the canvas. Twenty seconds to draw will limit the amount of material each person can place on the canvas, and also causes each user to anticipate what the other person is placing on the canvas. After going through a series of timed exchanges the piece will be finished when both users decide when it is finished.

This basic platform that allows two minds to collaborate makes culture and language obsolete as a person from Japan can exchange creatively communicate to another person across the world in Brazil. Observing how people begin drawings, and how people react to certain shapes will also give me insight on the human thought process. This can only be accomplished by storing and analyzing large quantities of data. 

Artificial Intelligence and Wernicke's and Broca's Area

Speech production in Artificial Intelligence is an issue that scientists and engineers have come across for a great period of time. Like any form of technology, we must observe nature to imitate it for the future of Artificial Intelligence. Our brain contains two areas that contribute to our fluent speech in a respective language. These areas are called Wernicke's area and Broca's areawhich are both located on the left hemisphere. Wernicke's area stores information required for speech content, arranging the words of a learned vocabulary into meaningful speech according to rules of grammar, and Broca's area is instructed by Wernicke's area to move the tongue, lips, and other speech muscles in harmony.

Manipulating a machine posed as a person to speak a language and behave in a social way must contain a system that emulates Broca's area and Wernicke's area. This system would be the one which would output the articulate sets of words for a fluent speech. Our frontal left hemisphere(holder of Wernicke's and Broca's area) contains about five billion cells which explains our capacity for a diverse set of words for speech. A compartmentalization of sets of words in forms of digital storage in an intelligent machine could act as Wernicke's area. Different categories of words such as adjectives, nouns, and verbs are to be organized in storage and located through an algorithm which would facilitate response. our brain works exactly like this but it is hard to grasp its multitude of responses that are the result of billions of intricately organized neurons. It would take years to produce a storage system and sets of responses to millions of different sensory inputs just as it takes years for a child to develop speech recognition and response.

Broca's area would be emulated b a high quality speaker system and a sensitive coordination of lips. After sensory input has been processed by the machine to produce a response, an output of words through a speaker to emulate the harmony of a tongue, throat and lips would occur. Lips along with a set of fake teeth would be installed on the machine.

A system like the human body contains many different underlying systems which work in coherence to produce what we are. Imitating Wernicke's and Broca's area plays no role on solving the problem of how an intelligent machine could learn through experience because survival is not on its mind. Humans learn to keep reproduction, survival, and pleasure afloat which is something a robot cannot do.

Cultural Relativism and the Brain

Cultural Relativism plays an important role in understanding our brain. Within a span of ten cultures, there are a number of common denominators that make these cultures similar to one another. For example, burping after a meal is polite in one culture while saying "your meal was very tasty" is polite in another. The common denominator of these two cultures is politeness which is an extension of a survival technique in the human brain. Cultural Relativism tells us as humans what our common forms of behavior are which can be physiologically explained due to various portions of the brain along with hormones.

Amongst a variety of cultures, there are various methods to signify respective messages amongst one another. Output through gestures, language, and limb movement are how all these messages are carried out exemplifying a commonality in terms of a physiological explanation. One might consider a form of saying "Hello" in Chinese much differently than saying "Hello" in Arabic but both must share common neural structure in the brain. A cultural and social aspect of a simple thing like saying "Hello" is a form of survival in the modern world. Social behavior such as greeting, thanking, or loving are all examples of human survival and each is carried out differently in differing cultures. These behaviors are tied to the limbic system which is responsible for emotion. Because messages are carried out to one another in society for a purpose, what gives each message a purpose is an emotional response that is tied to survival. It is in the limbic system where we learn to cry, laugh, love, and fight for different reasons amongst different cultures.

Cultures contain different mental and physical skills for survival. It is because of our memory that we are able to learn and repeat these skills on a day to day basis. Short term memory recites immediate sensory input while long term memory recalls sensory input from weeks or even years of previously learned knowledge. The brain distinguishes between facts and skill memorization such as the difference between shooting a basketball and memorizing a phone number. For factual memorization, sensory information is transmitted from the sensory regions of the cerebral cortex to the hippocampus and amygdala, two components of the limbic system which also function in emotions. Learning skills according one hypothesis states that it is due to changes in the structure of dendrites. Neural input causes a postsynaptic cell in the brain to take up calcium, which in turn activates enzymes that alter the cytoskeleton and change the shape of the dendrite in such a way that future transmission across that synapse is enhanced. This might occur when a young tribesman in Africa learns to weave a basket or kill prey, a skill as daunting to the mind as figuring out an algebra problem for a child in the U.S.

Learning

Learning new knowledge comes from rigorous repetition and time. Human beings learn new concepts based on knowledge that they previously know. A reorganization of previously learned knowledge intuitively is how we grasp a new concept. A step by step foundation of knowledge is built from childhood to adolescence and finally to adulthood.

When we are infants, we are like empty canvas's solely seeking items necessary for survival. It is food, water, and the nurturing presence of our Mother's that occupy our simple minds. A foundation of knowledge which helps us gain more knowledge is what we build as children. The first words we learn are usually "Mom" and "Dad" because they are the ones who give nourishment for survival. These words which are one syllable each are easy for a child to manipulate muscles in their lips and vocal chords to produce the sound. After a making the sound "Mom" or "Dad", a reaction occurs from a respective parent producing a neural arrangement that solidifies the manipulation of muscles in the vocal chord and lips. Physiologically, a cluster of neurons work together to transmit acetylcholine(ACH) to the muscles in the vocal chords and the mouth to pronounce a synchronized sound. An astounding way this could have been learned is through the mirror effect. A rigorous observation of the bay analyzing an external human saying "Mom" or "Dad" could attribute to its learning of the word.