In his latest book The Mind-Body Problem Explained: The Biocognitive Model for Psychiatry (Future Psychiatry Press, 2012), psychiatrist Dr. Niall McLaren discusses how the principles of information processing give a formal theory of mind that generates a model of mental disorder as a psychological phenomenon. The following excerpt from the fifth chapter of his book, “The Nexus Between Brain and Mind”, introduces readers to the data-processing features of the human brain.
~ ~ ~
“Where in my thesis is there a weakness that someone else might find—because I sure better find it before they do, because if they find it and I’m not prepared, I’m in deep trouble.”
Carl Sagan, talking of PhD candidates.
It is reputed that, sometime in the late nineteenth century, a man named Buckley announced that he would walk to China. “So?” you may ask, “happens all the time,” except that Mr. Buckley was then a native of the fair city of Sydney, and there is a lot of water between Sydney and China. Undeterred by the scoffing and derision, he set off but was never seen again. He thus achieved a kind of fame in the Antipodes, where a person who has no chance of success in some scheme is told he has “Buckley’s chance.” William James was of the view that anybody attempting to resolve the mind-body problem had Buckley’s chance: “Nature in her unfathomable designs has mixed us of clay and flame, of brain and mind, that the two things hang indubitably together and determine each other’s being, but how or why, no mortal may ever know” (Principles of Psychology, Ch. VI).
In my previous publications, I outlined essential principles by which James’ flame may arise from the clay and then act back to control it. The first four chapters of this book set the scene, so now I have to justify my claim of a theoretical resolution of the mind-body problem. We will be relying on several areas of knowledge, none of them developed by me. All the principles are in use in other fields of knowledge: there is nothing new in this thesis, just a different way of looking at old problems that leads to the resolution we need. This is a bold claim so those of you who are reading a borrowed or purloined copy can scoff now, while anybody who paid full price can only hope you get value for money.
The first area of knowledge is the vast and ever-growing field of the neurosciences, but I can’t claim to be keeping up with developments as it moves too fast. The second is the principles of information theory developed by Turing, Shannon, von Neumann and so many others. Again, I am not an information theorist and sometimes make a mess of the updates on my Linux system. Third, there is the field of philosophy, especially of mind, science and language. Once more, I’m not a professional philosopher. My undergraduate units were decades ago and hard core philosophy papers certainly tax my ageing brain. The penultimate is the sprawling field of “human studies”, including psychology in its many forms, sociology, epidemiology, general medicine, history and politics, not to overlook “near-human” studies such as ethology, primatology, and so on. We mustn’t forget basic chemistry, physics and biology, but mathematics now gives me a headache. Finally, there is my own discipline (and I use the term advisedly) of psychiatry, in which I can claim some expertise (although I am certainly not an academic). The real test of any resolution of the mind-body problem, meaning the ultimate test of any theory of mind, is that it must generate a plausible and testable theory of mental disorder.
That is, I will be relying on very broad swath of human knowledge, without being an expert in any of the preliminaries to my theory of mental disorder. Does that give me Buckley’s chance? Then read on.
In data-processing machines, there is no conceptual discontinuity at any stage between the point of entry of the data input and the point of exit of the final instructions at the interface with the peripherals. Data enters the processing system at dedicated points via sensors able to register just that particular form of energy. The sensors are in fact transducers, converting the data into the common codes of the system to send it through the system. At no point is there a change: it remains in the form of the codes until the peripheral is reached. While in the “common currency”, any and all forms of input data can interact with information already in the system (instructions, memory etc) in an emergent informational space. Critically, instructions are encoded in the same form as the data they are to process, otherwise there could be no interaction. There is nothing supernatural or anti-materialist about any of this: once data from a range of different sources has been transduced into the system’s codes, the data-processor is able to manipulate it to achieve an unexpected but rational outcome.
The data processing machine uses the laws of the physical world to create and acquit a higher-order function or virtual machine which is itself related to but separate from the physical realm. The realm of information constitutes an emergent state which exists only while a suitably-constructed switching machine is actually engaged in switching data. The emergent virtual machine can only interact with the physical world if the switching devices in its computational substrate are connected to suitable conduits, usually the same type as the switches themselves. This also applies to two separate informational spaces: they remain completely private until they are connected by suitable conduits.
Data flow can be detected at any point in the system by probes but, unless the codes used by the system are known, any signals detected remain meaningless noise. Once within the computational section of the machine, data are manipulated according to the semantic rules of the system and are therefore able to drive the system to work against the laws of thermodynamics at the expense of increasing entropy. The entire system exploits the natural laws of the material universe to achieve ends which are thermodynamically vanishingly unlikely, but never impossible.
It is common for people to claim that, since a physical machine can acquit conceptual tasks of the highest order, therefore, conceptual matters are themselves physical. This is a variant of the monist argument but it misunderstands the actual case. The switches of a data processing machine function in the physical realm, according to the laws of physics, in order to implement logical operations in an informational realm, according to the rules of its semantic system. In classic terms, the function of the physical machine constitutes the efficient cause of the computation, while the semantic instructions constitute the final cause. The two realms are incompatibly different and can only interact at the point where the computational neurons contact the conduit neurons.
All data processors operate at exceedingly fine tolerances, so that any disturbance of the physical system will quickly be evident in the system’s output. The more complex systems almost always have a high level of inbuilt redundancy such that minor physical faults can be bypassed, either by physical or by informational means. Because of the nature of the informational “space”, a complex system may well have the capacity for self-assessment and self-rectification but this is optional and certainly not necessary. A system may have no capacity to inspect its own data stores but would still function perfectly well. All of this is a natural and predictable consequence of the concept of information-processing as an ontologically separate realm from the physical world that implements it.
The human brain meets all these criteria. It is a hugely complex data processor with an afferent input, intervening computational systems, and clearly-defined efferent tracts that activate peripheral effector organs. The “common currency” of the human nervous system, neuronal impulses, is initiated at the level of the distal sensory receptors, when specific energy inputs are transduced into data flowing to the central centers for processing. In their passage from input side to output, the impulses may undergo many changes of neurotransmitters at different points in their tracts, but the significance of the impulses as carriers of information does not change at any point: in health, neurons are capable of being influenced by impulses from their neighbors, and impulses can influence neighboring neurons. The actual form of transmission between neurons is of great interest but it is not significant in terms of the brain’s capacity as a data processor. Neurons themselves could be replaced by silicon microprocessors and the new owner would not be able to tell. It is feasible that we will eventually devise self-aware computers with the capacity to experience perceptions.
~ ~ ~
Visit the author’s website http://niallmclaren.com to learn more about his work and publications.