CONSCIOUSNESS

ŞEYDA DEMİROK HUMAN CONSCIOUSNESS AND THE BRAIN (2015)
25





These concepts are called as 'hard problems of consciousness'
http://plato.stanford.edu/entries/consciousness-17th/
'I think therefore I am'
X-rays, lasers…
Black body radiation, the photoelectric effect…
Umezawa, Vitiello, Freeman…
For detailed description and definition check the referances [7] page
For detailed description and definition chech the references [7] ch. 1.5: computation: top-down and bottom-up procedures
[7] P.28-33
[7] P. 33
The details of these discussions and arguments will not be mentioned here because in the book, there is a whole chapter used for these relations.
[7] P.76
[7] P. 215
Feedforward neural network, Radial basis function (RBF) network, Kohonen self-organizing network, Learning Vector Quantization…
BOGAZICI UNIVERSITY
CONSCIOUSNESS
HUMAN CONSCIOUSNESS AND THE BRAIN

ŞEYDA DEMİROK
15.06.2015


From the early times of humanity, human consciousness has always been a mystery. Nowadays, we have lots of ideas and theories over how it works. Neural networks, machine learning algorithms and increase in the understanding of the nervous system and psychology are all results of these theories and improvements of science. However, some subjects, like emotions and experience, are still unknown. The aim of this paper is connect all these areas by studying on their individual aspects over consciousness. Each scientific view has its own focus and consciousness can only be understood completely by combining all these views. Instead of the details of these views, their conceptual meanings will be the main focus of this paper.



HUMAN CONSCIOUSNESS AND THE BRAIN
Şeyda Demirok
Bosphorus University
ABSTRACT-From the early times of humanity, human consciousness has always been a mystery. Nowadays, we have lots of ideas and theories over how it works. Neural networks, machine learning algorithms and increase in the understanding of the nervous system and psychology are all results of these theories and improvements of science. However, some subjects, like emotions and experience, are still unknown.
If we look at history, we can claim that profound arguments over consciousness have started with Descartes. His famous saying 'cogito ergo sum' can be accepted as one of the biggest building blocks of the ideas over consciousness. This fact should be seen as a chain that connects science and philosophy. Since the starting point of consciousness is the result of the philosophical questions, keys to answers of the unknown parts may also be found in philosophy.
Up until now, consciousness has been a subject of philosophy, and psychology. Then, biologists started to work on this subject because it is obvious that consciousness is related with the brain and the brain is directly a subject of biology. However, recently some physicist also started to work on consciousness and how it works. This is because all matters are subject to laws of physics. Also, history has proven countless times that using physical knowledge for medical purposes have led to great discoveries and amazing improvements in biology and medicine.
The aim of this paper is connect all these areas by studying on their individual aspects over consciousness. Each scientific view has its own focus and consciousness can only be understood completely by combining all these views. Instead of the details of these views, their conceptual meanings will be the main focus of this paper.

TABLE OF CONTENTS
Introduction ……………………………………………………………………………….3
Shadows of the Mind………………………………………………………………………4
Neural Networks and Learning Paradigms………………………………………………...9
Neurons and the Brain……………………………………………………………………13
Hard Problem of Consciousness………………………………………………………….20
Discussions……………………………………………………………………………….23
Acknowledgments………………………………………………………………………..24
References………………………………………………………………………………..25

INTRODUCTION
This article will mostly focus on the physical aspects of human consciousness and the brain. The aim of this article is to understand some important arguments over consciousness and current scientific aspects of the brain. Then, suggesting some new ideas and discussing the ideas given will take a part. Also, discussions over some modern experiments and theories which led to great discoveries will be one of the subjects.
Quantum mechanics enabled us to understand lots of mysteries in science. Therefore, some people thought it can also be the key to understand consciousness. Some scientists also believe this but there are also scientists who do not agree. Roger Penrose is one of the scientists who do not believe that quantum theory can answer consciousness. In his famous book, Shadows of the Mind, he emphasizes on this idea and explains it in details. Unfortunately, in Shadows of the Mind, Penrose does not make any clear definition of consciousness, awareness, intelligence and mind. This condition sometimes led to confusion for me. However, this should not be counted as Penrose's mistake because these words are usually used on behalf of each other in many other psychological papers.
Through this paper, there will be a path to understand consciousness. Firstly, starting from Shadows of the Mind, quantum mechanical solution to the consciousness will be discussed. Then, we will move back to see the whole picture about consciousness that we have today and that is called as Neural Networks and/or Artificial Intelligence. By, working on this subject, not only present scientific improvements about how the brain works will be discussed but also the brain's connection with the consciousness will be re-considered. To understand this connection, focusing more detailed on neurons and the brain is necessary. Therefore, there is part only gives information about how neurons and the brain works. This investigation about consciousness led me to the hard-problem of consciousness which is still a partially unknown subject! This subject may be a key to our problem to understand consciousness as a whole. After studying on this subject, there will be a review and discussion of all ideas given in the paper.
Rather than scientific details of the concepts mentioned in this paper, their conceptual meanings will be the focus point. Therefore, for more detailed information about the concepts, one may want to check the references of each subject.
SHADOWS OF THE MIND
Roger Penrose is one of the most important physicists when one talks about consciousness and its relation to physical laws. He has lots of work over consciousness and Shadows of the Mind is one of his famous books that cover consciousness. This part focuses mostly on this book. To be more clear and neat, analyzing his ideas within his book part by part is necessary.
PART 1:
Penrose starts in his book by questioning 'the ultimate scope of science' and relation between our mental existence and physical laws that govern our universe. Then, he emphasizes that consciousness is a part of our universe so they are not separable from each other. This is also our standpoint about consciousness. Later, he separates the ideas about consciousness to four different groups. Although these four may not cover all different ideas, this separation is important to be clearer. [7]
All thinking is computation; in particular, feelings of conscious awareness are evoked merely by the carrying out of appropriate computations.
Awareness is a feature of the brain's physical action; and whereas any physical action can be simulated computationally, computational simulation cannot by itself evoke awareness.
Appropriate physical action of the brain evokes awareness, but this physical action cannot even be properly simulated computationally.
Awareness cannot be explained by physical, computational, or any other scientific terms. P.12 [7]
'A' and 'D' are too extreme so they should not be considered as a scientific view without any valid explanation. But, still, those who believe in 'A' may check Hard Problem of Consciousness part and those who believe in D may read the Neural Networks part. Hopefully, these two parts will show that A and D are out of scientific perspective. For me, there should be one more step between B and C. It is true that 'awareness is a feature of brain's physical action' but this does not mean that all physical actions of brain neither have to be computational. Same way, accepting 'appropriate physical action of the brain evokes awareness' may lead to computational or non-computational simulations. What about Penrose? What does he think as the best viewpoint?
"The viewpoint C is the one which I believe myself to be closest to the truth. It is more of an operational viewpoint than B since it asserts that there are external manifestations of conscious objects (say, brains) that differ from the external manifestations of a computer: the outward effects of consciousness cannot be properly simulated computationally". P.15 [7]
For now, the difference between B and middle of B and C may seem alike but in concept there is a big difference between them. We shall return to this difference later.
Penrose describes what computational means as 'a computation is the action of a Turing machine.'[7] Turing test and Turing machine are necessary requirements for an algorithm to be simulated. Then, Penrose separates the computational algorithms into two parts as top-down and bottom-up (combined) systems. Top-down systems are well-defined and clearly understood fixed computational procedures to solve some problems at hand. The main difference of bottom-up systems is having memory of their previous experiences. These kinds of algorithms form the basis of artificial intelligence. However, not all algorithms have to be computational so other non-computational cases should also be considered. This is important to decide on whether consciousness is computational or non-computational. Penrose's viewpoint 'C' requires a non-computational definition for consciousness. To support his viewpoint, Penrose gives examples of non-computational, well-defined mathematical problems: Hilbert's tenth problem and tiling problems. To construct a physical universe whose action is beyond computational simulation, Penrose mentions these problems in details in his book. Then he concludes:
'There are completely deterministic universe models, with clear-cut rules of evolution, that are impossible to simulate computationally.'[7]
This result may seem irrelevant for the understanding of consciousness. However, Penrose refers to some sensations currently being called as hard problems of consciousness like 'sweetness' and 'happiness' and asserts that these sensations cannot have a computational simulation. By using John Searle's 'Chinese room' argument and limitations of artificial intelligence at his time, he defends his idea. Then, emphasizes on Gödel's theorem and by discussing it in details with its relation to Turing machine to prove his above conclusion. During this discussion, Penrose comes to another conclusion from Gödel's and Turing's ideas:
'Human mathematicians are not using a knowably sound algorithm in order to ascertain mathematical truth.'[7]
He defends this conclusion with a long question & answer part in his book at chapter 2.
PART 2:
This second conclusion is made due to Gödel's and Turing's arguments and theories. In other words, this is the conclusion of Gödel and Turing. To find out what Gödel and Turing actually think, Penrose searches for their viewpoints about consciousness. Although both of them based their thoughts on same conclusion, they end up claiming opposite views about consciousness. Penrose interprets Gödel's view as:
'Gödel appears to have taken it as evident that the physical brain must itself behave computationally, but that the mind is something beyond the brain, so that the mind's action is not constrained to behave according to the computational laws that he believed must control the physical brain's behavior. 'P.128 [7]
Therefore, Gödel's viewpoint is similar to 'D'. In that explanation, separating the mind and the brain is remarkable. Is Gödel only person who believes in that? No, even before Gödel, there were philosophers who believed mind and the brain is two separate things. Why do people believe and think that they are separate? This idea will be discussed in the discussion part. Then, Penrose interprets Turing's view as:
'Turing made a great point of the fact that human mathematicians are very capable of making mistakes; he argued that for a computer to be able to be genuinely intelligent, it, also, would have to be allowed to make mistakes… Thus, he appears to have regarded the inaccuracy of human mathematical as essential, allowing the mind's (supposed) inaccurate algorithmic action to completely sound algorithmic procedure.' P.129 [7]
So, Turing's viewpoint is similar to 'A'. Although both claim same argument, the second conclusion, they end up defend two opposite views! Moreover, Penrose also agrees with the second conclusion but he has the viewpoint of 'C'. He explains himself at the conclusion part of chapter three by saying;
'The argument given in this chapter would seem to provide a clear-cut case, demonstrating that human mathematical understanding cannot be reduced to (knowable) computational mechanisms, where such mechanisms can include any combination of top-down, bottom-up, or random procedures.' P. 201 [7]
By saying so, Penrose stress on his view. Then, he also claims that all current physical theories must be computational in nature. This means if consciousness is not computational, it cannot be a part of current physical theories. He finally reaches a conclusion about consciousness:
'Conscious understanding, in particular, must involve some kind of non-algorithmic physical action.' P. 214 [7]
Until the end of part one, Penrose defends his view logically. However, perhaps to be more reliable, he investigates the mind in classical physics and later in quantum physics in the second part of his book.
PART 3:
In the second part of his book, Penrose searches for a new physics to understand the mind since he believes current physical theories cannot be a solution. Although there are different ideas over connection between the brain (matter) and the mind, Penrose emphasizes on the existence of the mind in our physical body. Since it is within the framework of physical laws, it should be defined and expressed by the physical laws. He expresses his prediction about the mind by saying;
'Whatever it is that controls or describes the mind must indeed be an integral part of the same grand scheme which governs, also, all the material attributes of our universe.' P. 213 [7]
Although what he says seems to gather matter and the mind to one category, Penrose also accepts the possibility of their difference.
He opposes to those who claims mind cannot be a separate and different substance from matter by pointing the Einstein's famous equation (E=mc2) and entropy (dE=TdS-pdV). Einstein's equation connects the energy, which is an abstract quantity required for doing work, and mass of a particle, which is a measure of matter. If this equation is correct, currently there is no doubt about its correctness, accepting the connection between the mind and the matter and accepting their separation is not conceptually or logically wrong. Likewise, entropy is a feature of the notion of information and it is also related to other material quantities. This understanding proves that even if mind and consciousness are separate substances from physical concepts like Gödel claimed, there is no reason to forbid the discussion of their relations with physical concepts.
Penrose agrees mind is a part of our physical bodies so he accepts it can be defined by some physical laws. However, he claims current physical laws cannot be the one. To show there should be another theory, he firstly breaks the trust for current physical laws: classical mechanics and quantum physics. He firstly shows classical mechanics is not appropriate to define a non-algorithmic system because it is perfectly computable and algorithmic.
'In classical physics, one can, at any one particular time, specify all the data needed for defining a physical system, and the future evolution of that system is not only completely determined by this data but it can also be computed from it, by the effective methods of Turing computation.' P. 215 [7]
After explaining this theory's relation to computational systems, Penrose accepts the importance of classical physics for current scientific improvements. 'Classical physics is all we need for the purposes of Artificial Intelligence.'
Then, he moves to quantum physics. Quantum physics includes more uncertainties about data. Also, equations of quantum theory (basically Schrödinger equation) provide more freedom for random nature. Therefore, it is, usually, thought to be the answer for the consciousness. However, Penrose claims that pure randomness cannot supply the needed non-algorithmic action. He refers to the earlier chapter of his book for the explanation. This chapter analyzes the behaviors of chaotic and random systems. Then, it explains the situation as:
'Since we are supposing our robot to be digitally controlled, and that, correspondingly, its environment can also be provided as some kind of digital input there will be a finite number of such possible alternatives. This number might be very large indeed, but it would still be a computational matter to describe all of them together. … the computation itself would not be "unknowable"; that is to say, one could see how to build a (theoretical) computer –or Turing machine- that could carry out the simulation, even though it would be out of the question actually to carry it out' P. 170 [7]
Penrose opposes to the idea of quantum mechanical explanations of consciousness by figuring its random nature. He stresses on a needed new physics to answer our all questions! Although Penrose opposes to these explanations, he does not take the quantum physics out of the picture. Just like classical mechanics, quantum mechanics is also a building block to understanding of human consciousness and the brain. Classical physics brought us the artificial intelligence and quantum physical explanations take us even more forward. Penrose uses quantum mechanics to describe the behaviors of microtubules within the neurons.











NEURAL NETWORKS AND LEARNING PARADIGMS
Neural Networks (NN) are one of the popular topics among the present research areas. There is huge demand for improvements in this area because this is the core of artificial intelligence (AI). AI is a key subject to understanding of consciousness and the brain. If human like robots can be built then this shows that the viewpoint of 'A' mentioned in the previous part should be correct. However, our curiosity should make us question how far neural networks came and how far NN will improve AI. Let's try to find the answer.
There are various kinds of methods for neural networks. People who studied on artificial intelligence and neural networks would agree that it is impossible to go over all of them here. Rather, analyzing the main concept and studying on a few examples is a more convenient way for the aim of this paper. Mathematical details of the methods are not important because our interest is not how neural networks mathematically work but the conceptual understanding of NN.
As far as I know, the very beginning of the methods of NN is based on the same idea. The idea is about how the system should progress. For any method, there should be some inputs and according to their value or differences they go under a certain function to hidden level. This part can be considered as the brain of the system. It interprets the values and these values also go under another function. The results of the last function are called as outputs. Hidden level does not have to consist of one layer instead it usually consists of many complicated layers. So, the schematic representation can be shown as below.

Figure 1: Neural Networks (simple demonstration)
To speak more scientifically, let us call input random variables as xi. For each xi value, there exists a corresponding weight notated as wi. The function, I (xi), uses these weight values to reach the hidden level.
xi*wi=yi
Hidden level also uses the same procedure and multiplies its input yi with its corresponding weight value denoted as Wi.
yi*Wi=zi
These results, zi's, are outputs of the system. According to these values, the system responds to the environment.
The above explanation may seem too easy to understand. However, so called weights forms the fundamental basis of the neural networks. There are various kinds of mathematical methods which are used as weight functions. Principal components, estimation theory and mean-squared error can be considered as the most common concepts used in this area. With these techniques, lots of technological devices are produced. However, what attract more attention are not these improvements in technology but the possibility of learning. Learning process is similar to the one mentioned above. But it also needs a feedback system to work.
'An accepted norm for decision rules or strategies used to classify patterns is that they do so in a way that minimizes the "expected risk." Such strategies are called "Bayes strategies" and can be applied to problems containing any number of categories.' [9]
Different learning methods can be built and used for different robotic system. However, one thing can never change. Every learning paradigm loses some information during process. Conceptually, it is impossible to find a method that sets these losses to zero. This missing part is usually called as 'forgetting' term. This makes sense since human also forget and confuse things they already learned. So, with better forgetting term, robots may behave more like human being and probably better than animals.
Erkki Oja is a Professor of Computer Science and Engineering at Aalto University. He has been working on neural networks and has lots of paper over this subject. In his paper (Principal Components, Minor Components, and Linear Neural Networks) he proposes an alternative system for neural networks to add learning process to the system. The above, demonstrated system is useful for present technology but not good enough for learning systems because it is lack of feedback. He demonstrates this proposed system as below figure;

Figure 2: The unsymmetrical SGA network. The p parallel linear neurons have outputs yj and weights wij for inputs xi. The time index k is not shown explicitly. [6]
This demonstration is more advanced than the previous one because every step is connected to each other just like neurons. Since this system is more like nervous system, it is more promising for learning paradigms. Prof. Oja starts with the very basic argument for the method of neural networks, similar to the one mentioned above, and improves it step by step. By using principal components method, matrixes and their corresponding differential equations, he proposes a learning algorithm with "forgetting" term and another term representing the other neurons effect over the output.
'The corresponding network implementation is given by the learning algorithm
wjk-1=γk[-yj(k)x(k)
+ yjk2+1-wjk-1Twjk-1wjk-1
+αi>jyikyjkwi(k-1)].
The first term in the brackets is the anti-Hebbian term and the second term is the "forgetting" term, proportional to wj (k-1) itself. Now the coefficients in the forgetting term are more complicated than in the original SGA algorithm, but everything is still local within one neuron although not within one connection weight. The third term, similar to the SGA, gives the influence of the other neurons. Assuming this modified learning rule for each neuron, the network of Figure 1 can be used to implement the algorithm (38) with obvious changes in the interneuron signals' [6]
This algorithm mentioned above is typical learning rules for the adaptive PCA (principal components algorithm) or minor component extraction problem. They are suitable for neural network implementations. However, there are many other algorithms for neural network implementations.













NEURONS AND THE BRAIN [4]
This part focuses on how neurons and brain work to understand how thinking process takes place in the brain. Neurologist Dr. Bülent Madi also works on this topic and wrote a book called "Öğrenme Beyinde Nasıl Oluşur?". Information given in this part is based on this book. To understand the thinking process, the steps below will be followed;
Kinds of Neurons and How They Work
Parts of Brain and Their Role
Neural Networks

KINDS OF NEURONS AND HOW THEY WORK
The central nervous system (CNS) is composed of two cells: neurons and glia. Ninety percentage of human brain is consists of glia cells. Some of their tasks are as follows,
They surround neurons and hold them in place and help neurons to gather information.
They supply nutrients and oxygen to neurons and protect the brain from infection.
They insulate one neuron from another.
They destroy dead neurons.
Neurons are fundamental information processing systems in our body. It has DNA in its nucleus. DNA is the fundamental source of information about genetic. There are also lots of mitochondria to produce energy; and some enzymes, proteins and neurotransmitters in its nucleus. Neurons exist in the brain, cerebellum, brain stem and spinal cord. Their aim is to gather information from inside or near outside of organism. A neuron consists of nucleus in its cell body, dendrites needed to gather information from environment, axon needed to transfer the information.

Figure 3: Neuron
There are different kinds of neurons according to their task.
Sensory neurons; takes the information coming from environment and send signals to the CNS according to the information.
Interneurons; are located in the CNS and its task is to allow information transfer inside of the CNS.
Motor neurons; takes signals coming from the CNS to the organs like the muscle or gland.
Information coming from a neuron can be transferred to another neuron, gland or muscle through synapses. Synapses are where presynaptic terminals of a neuron and dendrites of another neuron come together. This place is not physically connected but rather it is an opening. Synapses are where electrical signals coming through axon turn into chemical signals and then, again, turn into electrical signals through various processes. This is called neurotransmission.
"Neurotransmission (or synaptic transmission) is communication between neurons as accomplished by the movement of chemicals or electrical signals across a synapse. For any interneuron, its function is to receive INPUT "information" from other neurons through synapses, to process that information, then to send "information" as OUTPUT to other neurons through synapses. Consequently, an interneuron cannot fulfill its function if it is not connected to other neurons in a network. A network of neurons (or neural network) is merely a group of neurons through which information flows from one neuron to another."[10]
Synapses are important for learning process and to form a behavior. Increase in the number of synapses makes learning easier. Also, any construction taking place in synapses effects information transfer. This problem can cause different types of disorders in touching, smelling, seeing, hearing and tasting. Moreover, it may be a cause of amnesia or behavioral disturbances. Synapses can be modified due to their task; increase or decrease in number of synapse, completely disappear, change in shape. This is called synaptic plasticity. A synapse is consists of six part;
Presynaptic neuron
Postsynaptic neuron
Presynaptic membrane
Postsynaptic membrane
Synaptic opening
Neurotransmitters

Figure 4: synapse
Presynaptic neuron is the one where the signal ends, and postsynaptic neuron is the one where the new signal occurs. Neurotransmitters are chemicals used to transfer the information in the synaptic opening. When a signal reaches to a neuron, it makes presynaptic end secrete neurotransmitters having been produced and stored. Some of neurotransmitters are decomposed in the synaptic opening and some of them stimulate postsynaptic neurons. After this stimulation, an electrical signal occurs in the postsynaptic neurons. There are various kinds of neurotransmitters and all of them have different tasks. When they decrease or increase, different kinds of disorders happen. These disorders may be related with breathing, memory, behaviors, happiness, and sleeping.
The brain continuously improves and evolves by gathering information from outside since the baby is inside its mother's womb. Sensory receptors work for this purpose. They may locate every part of our body. They take all the information about heat, pressure, pain, scent of the environment vs. as a signal to the brain. The brain interprets the signals and responds accordingly. This process may be illustrated as;
SENSE PERCEPTION RECOGNITION IDEA IN MIND
Through consolidation and enhancement, neurotransmitters can learn by familiarization how to secrete, where stimulate to. Therefore, in an improved brain, thousands of connection can be made in a few second.

PARTS OF BRAIN AND THEIR ROLE
Neural systems consist of three kinds of systems.
parasympathetic nervous system (autonomic nervous system)
peripheral nervous system (PNS)
central nervous system (CNS)
These three systems have their own unique functions for the body. Peripheral and Central Nervous System are more important for our topic.
PNS is responsible for the signal transportation from CNS to body, face, arms and legs and also from sensory receptors to CNS.
CNS consists of spinal cord, brainstem, cerebellum, cerebrum (the brain) and a liquid called cerebrospinal fluid (CSF) whose function is to protect them.

Figure 5: Central Nervous System
Cerebrospinal Fluid (CSF); is produced from koroid plecsus. It is found inside and around the brain, and the subarachnoid space, between the arachnid mater and the pia mater. It also occupies around cerebellum, brainstem and spinal cord. Finally, it is absorbed by arachnid villus and sent to vein. In that circle, it is renewed 4-5 times per day. Increase or decrease in the amount of CSF causes disorders like mental disability, autism.
Spinal Cord; is responsible for taking the signals coming from arms, legs and other organs to the brain and signals coming from the brain to the body. It has its own reflects circulation. Some memory traces are also found here!
Brainstem; is found between the brain and spinal cord. It is where the nerves for breathing, hearing, tasting, eye movements, heart beats, head movement vs. come out.
Cerebellum; is responsible for tightening and loosing of muscles in less than a second. Therefore, walking easily with balance, writing correctly, dancing is related with cerebellum. There is a connection between cerebellum and frontal lobe, temporal lobe, parietal lobe. Although cerebellum is smaller than the brain, the neuron density is higher!
The Brain; consists of two hemisphere. Also, there are four lobes in the brain. Both hemispheres have their own focuses and they are connected with each other by commissural fibers. If they cannot communicate with each other, there can be problems related with the person's academic life.
Brain lobes
The brain is where information coming from the environment is interpreted and is answered. The brain can be divided into four parts called lobes according to their functions. Every lobe is responsible for another task. However, this is not absolute. There are some cases where one lobe takes another's task. Their functions and responsibilities are mentioned below.

Figure 6: The Brain
Occipital lobe; is at the back of the brain. It is divided into several functional visual areas. It is where images interpreted so seeing is possible.
Parietal lobe; is basically responsible for interpreting the sensory signals coming from various parts of the body, for knowledge of number, perception of direction and in the manipulation of objects.
Temporal lobe; is responsible for hearing, interpreting of the information heard, understanding, recognizing the face. Wernicke area places here. Wernicke area is an important part of the brain. It enables one speak meaningfully and think normally. Temporal lobe is also responsible for seeing, tasting, smelling and emotions.
Frontal lobe; may be said as the most important part of the brain. It takes up the 1/3 of the brain. It is responsible for controlling the one's own behaviors, planning the ahead of time, speaking loudly, social responses, and emotion management. It can be divided into three parts according to its main functions.
Motor cortex; takes up the 1/3 of the frontal lobe. It is responsible for muscle movements' plans and coordination.
Broca's area; is responsible for speaking. If it is damaged, a person cannot talk properly.
Prefrontal cortex; is responsible for behaviors. It decides on the behavior by organizing conscious or subconscious information stored in memory. Planning, choosing, solving a problem, guessing, interpreting, improving a project is under its responsibility. Also, being conscious and aware is said to be its responsibility!
NEURAL NETWORKS
In the brain, there is no other part specialized on a function except speaking, touching and seeing. Therefore, a function takes place by correlating parts of the brain. These systems are called as neural networks. There are many kinds of neural networks and some of them are;
Network of recognition of the faces
Language network
Attention and awareness network
Emotional network
Network of visual and spatial attention
Motion network
However, mentioning each of them would be hard so only one of them is mentioned in this paper.
Emotional Network: Emotional facial expressions show motor cortex has an important place in emotional network because marionette lines, hand, arm and leg postures show emotional condition. Emotional network is also related with limbic system, frontal lobe and occipital lobe. It is also related with parietal lobe because an object in a room may affect emotions. These examples show emotional network functions by using different parts of the brain.
Another example for emotional network is anxiety. When a person is anxiety, his mind becomes full of sad and uneasy memories. When there is a threatening case, Amigdala and other systems related with anxiety makes frontal lobe focus on the situation. The brain interprets the incoming signals and the past experience about the situation with hippocampus. If there is a stressful case, hippocampus becomes active. Anxiety starts when a person cannot decide on how to react towards the case, when he cannot interpret the situation and when he thinks he is in a dangerous position.
If Amigdala is alerted for a long time, it cannot turn into its normal condition for a long time. This situation makes hippocampus and prefrontal lobe inactive. If this happens prefrontal lobe cannot control Amigdala and panic attack and phobias takes place. Therefore, under stress, cortisol level within the blood increases. The specialized reseptors of hippocampus reacts towards adequate level of cortisol and this condition makes hypothalamus produce less cortisol. This way, Amigdala is taken under control.














HARD PROBLEM OF CONSCIOUSNESS
The brain is where all the thinking process takes place, it is where we solve our math problems and it is where we fell in love. At least this is what present science tells us. But, not every process has same kind of behavior. For example, memorizing a poem, solving a problem or categorizing objects are all called easy problem of consciousness. These are called as easy problems because we can understand how they work. They work as a function so they can be simulated. This understanding also opened new scientific areas like artificial intelligence, neural networks and machine learning. As mentioned in Neural Networks part, these problems are a part of consciousness but not enough to describe consciousness. The reason is hard problems are more different than easy ones and how they process cannot be understood as much as easy ones.
David J. Chalmers is a professor of philosophy in New York University and is the first person dividing the problems about consciousness as easy and hard problems. He mentions these problems in his paper named Facing Up to the Problem of Consciousness. Then describes the hard problems as;
'The really hard problem of consciousness is the problem of experience. When we think and perceive, there is a whir of information-processing, but there is also a subjective aspect. As Nagel (1974) has put it, there is something it is like to be a conscious organism. This subjective aspect is experience. When we see, for example, we experience visual sensations: the felt quality of redness, the experience of dark and light, the quality of depth in a visual field... What unites all of these states is that there is something it is like to be in them. All of them are states of experience.'[2]
Since hard problem is about our experiences, there is, yet, no function to define them at least for now. This statement causes the question what is a function or what kind of function we are talking about. The answer is mechanism of the process. In Neurons and The Brain part, the neurobiological explanations for the processes of easy problems are written. By simulating these explanations, a computational function can be easily derived. Why cannot a function for hard problems be found? This is because experience is purely subjective. When a hundred people look at a picture, they will all define their experience in a different way. Some or most of the people may feel similar emotions but the way they define their emotions will be unique for each person. If there is a function, for a certain input, there should be a result with some uncertainty. However, in the example of picture, there is no functional system, yet! Classical mechanical approach does not allow high uncertainties and, according to Chalmers (like Penrose), quantum mechanics also cannot be a solution to this problem because it cannot describe experiences.
In the same paper, Chalmers mentions current neurobiological explanations and claims that none of them explains the hard problem, experience. Shortly, the reason, why all the explanations are not qualified by Chalmers, is merely because all of them are based on physical theories. He defends his idea by saying;
'These new discoveries may help us make significant progress in understanding brain function, but for any neural process we isolate, the same question will always arise. It is difficult to imagine what a proponent of new neurophysiology expects to happen, over and above the explanation of further cognitive functions. It is not as if we will suddenly discover a phenomenal glow inside a neuron!' [2]
Here, 'the same question' refers to 'why should this process give rise to experience'. Why? This question is very important and also challenging. So far, neural networks and nervous system were taken in consideration but both did not explain how experience works. This question may take the understanding of consciousness one step further. But, before starting to ask 'why should this process give rise to experience', the process should be found. It is somewhere in our body for sure but can we directly relate it with neurons or with the brain? There have been countless experiments about neurons and the brain but, as far as I know, there have not been found any adequate explanation for the experience. Of course, this does not mean we are searching the wrong place but changing research methods and focuses may be a good idea. John R. Searle is an important philosopher in the University of California, Berkeley and a professor of philosophy of mind and language. He mentions on this topic in one of his papers and says;
'Consciousness is entirely caused by neurobiological processes and is realized in brain structures. The essential trait of consciousness that we need to explain is unified qualitative subjectivity. Consciousness thus differs from other biological phenomena in that it has a subjective or first-person ontology, but this subjective ontology does not prevent us from having an epistemically objective science of consciousness.' [8]
Later in the same paper, Searle emphasizes on how one should approach to the consciousness. Then, he describes two methods to analyze the consciousness and calls them as the building block model and the unified field theory. He describes the building block model as;
'The idea is that any conscious field is made of its various parts: the visual experience of red, the taste of coffee, the feeling of the wind coming in through the window. It seems that if we could figure out what makes even one building block conscious, we would have the key to the whole structure. If we could, for example, crack visual consciousness, that would give us the key to all the other modalities.' [8]
This idea seems reasonable in a lot of ways. For example, classic physics has been known for a long time when scientists had started to work on atomic scale particles. However, they observed very accurate formulas of classic physics did not work on this scale. After quantum physical laws had proven, not only atomic scale was understood but also its relation to the classic physics was understood. Likewise, if any part of consciousness can be understood, it may be understood as a whole. Finally, its relation with nervous system may be clear to us. However, understanding the one part of consciousness does not seem easier than the understanding the whole structure. After all, same questions like 'where does it happen' and 'how does it happen' still remains. Conceptually this idea may be correct but does accepting this view make consciousness any clearer than now? I could not find any reason to say 'yes'. Searle also criticizes this view by saying;
'The building block theory: The conscious field is made up of small components that combine to form the field. To find the causal NCC for any component is to find an element that is causally necessary and sufficient for that conscious experience. Hence to find even one is, in an important sense, to crack the problem of consciousness.' [8]
Here, NCC means 'the neurobiological events that are correlated with consciousness.' Searle defends his criticism by showing weak points of the scientific researches supporting this method. He mentions three researches, 'blind sight', 'binocular rivalry and gestalt switching', and 'the neural correlates of vision'. Then, he explains how they support the method and shows that they are not adequate. The details are not necessary so they will not be mentioned here. Finally, he mentions to the unified field method.
'The research program that is implicit in the hypothesis of unified field consciousness is that at some point we need to investigate the general condition of the conscious brain as opposed to the condition of the unconscious brain. We will not explain the general phenomenon of unified qualitative subjectivity by looking for specific local NCCs. ... What we are trying to find is which features of a system that is made up of a hundred billion discreet elements, neurons, connected by synapses can produce a conscious field of the sort that I have described. There is a perfectly ordinary sense in which consciousness is unified and holistic, but the brain is not in that way unified and holistic.' [8]
Here, Searle also makes a differentiation between consciousness and the brain like Gödel! This separation of mind (or consciousness) and the brain become clearer with Searle's explanation about unified field theory. This theory differ consciousness from the brain by being unified. Consciousness is unified in a sense that it does not work part by part. On the other hand, as we mentioned in Neurons and the Brain part, the brain works with neurons and specialized parts. It is possible for some part of the brain to be damaged but other parts to function normally. This is not the case for consciousness!
DISCUSSION
Through my research, there were times that I got confused. The reason was mostly about definitions of consciousness, awareness, intelligence or mind and links between them. Scientists usually took them in one box or leave them all out of the box. This means they consider them all together as if they are not separable. For me, this was hard to accept! There should be a connection between them; this is obvious. However, this connection seems to me as the same connection between steps of a stair or rings of a chain!
awareness yields consciousness yields mind(intelligence)
The importance of this connection lies under the understanding of the relation between matter and consciousness. Usually, biologists and the writers of the papers I read only concern living things. However, this may not be so true. For example, if awareness is defined as 'being aware of the environmental effects and/or environment it is in', can not one claim that 'biologically non-living' things also have awareness. This is so because they are aware of the environmental effect. They respond to these effects by moving their valance electrons from one point to another due to electrical or thermal conduction. They respond to time by ageing due to thermodynamic. Also, consciousness is not the limit of intelligence or the mind, rather it is step towards the mind. Animals also have consciousness they understand what is going on and acts accordingly. However, I do not believe they have intelligence or the mind since their living standards, eating habits, social rules do not change in time. A lion in 21th century lives exactly the same way with a lion lived in 1st century. However, the way human lives change very rapidly. This condition can only be explained by our intelligence.
On the other hand, can someone prove that hard problems of consciousness are directly related to the brain or they take place in the brain? Let us make a brainstorm about it. If someone works as a labor in a construction of a building from early times of morning till night, he should be tired and have pain all over his body when he returns home. This shows and proves that he uses his body for his work. Or, think about a student who studied on complex mathamatic problems or studied a foreign language for a whole day. Since these situations spend brain's energy, at the end of the day, he should have a headache. These examples make perfectly sense and the results are the proofs of the places used.
If so, when a person feels happy or sad, does not he feel these feelings in his heart? When we miss our home, lover or a friend, do not we feel pain in our hearts? When we feel so excited and happy, do not we feel as if our hearts race? These situations are thought as the results of chemicals produced in the brain. Well, that may be correct. However, let us be rationale. If, in other examples, people have pain or strong feelings in the places they used, why do we try to explain these feelings with chemical actions? Actually, a labor feels pain in his body because, when he uses his muscles, he spends the energy stored in his muscles. Then, his brain sends signals to his muscles to feel pain and leave the work to protect the body. Likewise, when a student studies for a long time, he has a headache to protect the brain. Maybe, the reason of the chemical reactions due to these emotions is about to protect the heart in some way. If we separate these hard problems from consciousness in that way, we can be able to understand the consciousness even deeper.

ACKNOWLEDGEMENTS
All the information given in 'Neurons and Brain' part is based on 'Öğrenme Beyinde Nasıl Oluşur?' written by neurologist Dr. Bülent Madi. Any mistake due to translation of definitions and/or descriptions belongs to me.
I would like to thank especially to my advisor Prof. ARIK for his continuous help and encouragement. Our discussions not only improved my view of consciousness but also gave me a wider view over the world.
I also would like to thank my dear friends from psychology department for their advices and for our helpful discussions. Their existence always decreases my tension and helps me study.
Without my parents' motivation, I could not possibly write this paper. I thank them for always being next to me for help and for their understandings.








REFERENCES
Bierman, D. J., Whitmarsh, S. (2006) Consciousness and Quantum Physics: Empirical Research on the Subjective Reduction of the Statevector, The Emerging Physics of Consciousness, The Frontiers Collection , pp 27-48
Chalmers, David J. (1995). Facing Up to the Problem of Consciousness. Journal of Consciousness Studies 2 (3), pp. 200-219
Horst, S. (2001) EVOLUTIONARY EXPLANATION AND CONSCIOUSNESS. Journal of Psychology and Theology, Vol. 30, No. 1, 41-50
Madi, Bülent, (2014). Öğrenme Beyinde Nasıl Oluşur? Ankara: Efil Yayınevi
McCarthy, J. (1995) Review of Shadows of the Mind by Roger Penrose, Electronic Journal Psyche http://www-formal.stanford.edu/jmc/reviews/penrose2.html
Oja, Erkki. (1992) Principal Components, Minor Components, and Linear Neural Networks. Neural Networks, Vol. 5, pp 927-935, ,pergamon Press Ltd.,USA
Penrose, Roger. (1995) Shadows of the Mind, Oxford University Press, Oxford.
Searle R. John. (unpublished) Consciousness, on his personal website under articles http://socrates.berkeley.edu/~jsearle/
Specht, Donald F. (1990) Probabilistic Neural Networks, Neural Networks, Vol. 3, pp 109-118, pergamon Press plc, USA
Stufflebeam, R. (2008) Neurons, Synapses, Action Potentials, and Neurotransmission, Consortium on Cognitive Science Instruction http://www.mind.ilstu.edu/curriculum/neurons_intro/neurons_intro.php?modGUI=232&compGUI=1826&itemGUI=3155