I know it’s a controversial question. But theoretically, could a computer develop awareness? I ask myself this all the time. I was reading a scientific review and well… there’s a more profound question to answer first.
Artificial intelligence technology has advanced in many ways. However, robots are a long way from becoming humans.
Do you remember that robot Sophia, developed by Hanson Robotics Limited?
She can easily fool you with her facial expressions, her ability to track and recognize faces, and to sustain a conversation with you. But although the dialogue might be very astonishing — or shocking, she’s just a machine with complex algorithms running many processes to sustain the conversation.
It’s like a chatbox with your mobile carrier but more advanced. It’s not consciousness at all.
Besides, human consciousness is not only about recognizing patterns.
The scientists outlined three levels of consciousness in a review published in Science Magazine. You must understand how awareness arises in the human brain if you envision to build a conscious robot.
According to the review, the first level of consciousness is C1. It’s all about the ability to make decisions after drawing a vast repertoire of thoughts, considering multiple possibilities. The scientists suggest that this level of consciousness has a relation with the behaviors.
C1 is seen in human babies as well as in animals.
They gave an example of thirsty elephants and how they know to locate and move straight to water well even if they are 50 km away!
This decision-making requires a sophisticated neural circuit. The elephant needs memory information, to select the best choice from a set of available options, to maintain this decision over time and to coordinate a variety of operations, including to navigate to the well.
In humans and other primates, the pre-frontal cortex serves as the center for information processing where many of the actions described in consciousness C1 occur. It might be possible to derive the computational principles for encoding if we thoroughly analyze those responsible neurons circuits.
The second level of consciousness is C2. It’s the ability to monitor your own thoughts calculations, that is to be self-conscious.
C2 level awareness results in feelings of certainty or error that is, the mechanism helps us to perceive the mistakes and correcting them. Moreover, self-awareness helps us to discover what we know and what we don’t know, which leads to curiosity.
The wanting to know more.
“We cannot be conscious of what we are not conscious of”
I loved this sentence. Related to the C0 level of consciousness, it refers to the unconscious operations occurring in the human brain, such as recognition of the face and expressions. Most calculations performed by the brain occur at this level. You seem to automatically recognize the face and speech of your family without consciously thinking about it, right?
Have you ever heard about Convolutional Neural Networks?
It’s an astonishing deep learning algorithm you can use that learns how to recognize patterns mainly for image processing but also autocorrelated data. It’s truly incredible. Most of the artificial intelligence machines nowadays operate at this C0 level. These systems can perform many of the calculations of the C0 level, including facial recognition from our smartphones.
C0, C1, and C2 levels are based on a hypothesis that sees consciousness as a result of computations physically done by the brain.
According to the review,
Although centuries of philosophical dualism have led us to consider consciousness as unreducible to physical interactions, the empirical evidence is compatible with the possibility that consciousness arises from nothing more than specific computations.
Human consciousness could emerge from a set of specific calculations in the brain. It might be the day we could minutely explain the processes that lead to conscious in humans.
After that, to code this in robots may not be so difficult.