Mind Uploading at the International Journal of Machine Consciousness

Whole brain emulation is the topic for today: being able to run all of the processes of a brain on some form of computing machinery other than the evolved biological structures we presently possess. Considered in the long term this is an important line of research, as radical life extension will ultimately require moving away from flesh and into some more robust form of machinery in order to better manage the risk of fatal accidents. 'Ultimately' here is a long way into the future, centuries or more, long after we have solved the basic problems of repairing our aging biology so as to attain continual youth. Some people will be satisfied with copying themselves from their biological substrate into a machine substrate and letting that machine copy continue on, but that seems to me little more than an expensive form of procreation - continuation of the self requires a slow transformation of the original, not a quick cut and paste of data to a new computing device. But this is an old and often rehashed argument between identity as pattern and identity as continuity.

Here are some past posts on whole brain emulation if you'd like to do some background reading:

Regardless of how people decide to use the ability to host a conscious individual somewhere other than a human brain, the technologies of whole brain emulation will have to be built. They are a precursor to any program of replacing the brain's present biological machinery with something better. From where I stand, brain emulation is also the most plausible path to true artificial intelligence, which at this time looks far more likely to arise from attempts to duplicate and then improve on the operation of human brains than from efforts to improve expert systems of varying sorts.

Reasonable people differ on this, of course, as even a brief survey of publications on artificial intelligence will tell you.

If you find this topic interesting, you might look at the latest issue of the International Journal of Machine Consciousness, featuring many of the usual suspects from the transhumanist community - folk who have been putting in time on AI and molecular nanotechnology research for some years. A couple of the more intriguing items:

Non-Destructive Whole-Brain Monitoring Using Nanorobots: Neural Electrical Data Rate Requirements

Neuronanorobotics, a promising future medical technology, may provide the ultimate tool for achieving comprehensive non-destructive real-time in vivo monitoring of the many information channels in the human brain. This paper focuses on the electrical information channel and employs a novel electrophysiological approach to estimate the data rate requirements, calculated to be (5.52 ± 1.13) × 10^16 bits/sec in an entire living human brain, for acquiring, transmitting, and storing single-neuron electrical information using medical nanorobots, corresponding to an estimated synaptic-processed spike rate of (4.31 ± 0.86) × 10^15 spikes/sec.

Why Uploading Will Not Work, or, the Ghosts Haunting Transhumanism

Transhumanists tend to have a commitment to materialism and naturalism but nonetheless pursue goals traditionally associated with religious ideologies, such as the quest for immortality. Some hope to achieve immortality through the application of a technology whereby the brain is scanned and the person "uploaded" to a computer. This process is typically described as "transferring" one's mind to a computer.

I argue that, while the technology may be feasible, uploading will not succeed because it in fact does not "transfer" a mind at all and will not preserve personal identity. Transhumanist hopes for such transfer ironically rely on treating the mind dualistically - and inconsistently with materialism - as the functional equivalent of a soul, as is evidenced by a carefully examination of the language used to describe and defend uploading. In this sense, transhumanist thought unwittingly contains remnants of dualistic and religious concepts.

A Framework for Approaches to Transfer of a Mind's Substrate

I outline some recent developments in the field of neural prosthesis concerning functional replacement of brain parts. Noting that functional replacement of brain parts could conceivably lead to a form of "mind-substrate transfer" (defined herein), I briefly review other proposed approaches to mind-substrate transfer then I propose a framework in which to place these approaches, classifying them along two axes: top-down versus bottom-up, and on-line versus off-line; I outline a further hypothetical approach suggested by this framework. I argue that underlying technological questions about mind-substrate transfer, there is a fundamental question which concerns our beliefs about continuity of identity.

On this last topic, present developments in neural prosthetics are well worth the time taken to investigate. Being able to replace some lesser pieces of the brain in the event of damage is on the verge of being a going concern - sometime within the next twenty years there will be a fair number of people walking around with implanted medical devices in their brains. Those devices will replace or augment the function of one or more component parts of the brain, allowing these patients to live where they would otherwise have died or suffered a lower quality of life. This is the start of the next wave of mapping the physical structure of the brain to its function, and that field of research will expand and accelerate just like all other areas of medicine, driven by the ongoing biotechnology revolution.


The 'why uploading will not work' argument in the quote above seems incoherent, with the writer fundamentally not understanding what is being proposed. Take for example:

"Transhumanist hopes for such transfer ironically rely on treating the mind dualistically - and inconsistently with materialism - as the functional equivalent of a soul, as is evidenced by a carefully examination of the language used to describe and defend uploading."

At no point do uploading researchers treat the mind in a dualistic fashion. In fact, they are painfully aware of just how much of the physical mind must be simulated to get proper functioning. And certainly no upload researcher would consider the "mind" to be some immaterial thing which cannot be measured or interacted with, as is typically claimed for souls.

Rather, organic minds are a physical information processing system, and like any other information processing system can be simulated entirely in software. Real arcade machines from 20 years ago are now regularly and completely simulated on desktop computers; there is no reason to believe that the same techniques, when applied to an organic machine such as a brain and body, cannot also simulate those. In fact, the laws of physics as we currently know them would have to be changed for simulation to be impossible.

Much like the arcade machine does not lose its "soul" when we simulate it, neither does a human brain/body. The reason why is because neither one has a soul to begin with.

(The standard rejoinder I would expect from the author of the piece would be "nevertheless, this is what it looks like, and the parallels with the religious quest for immortality are still there". My response would be, "you only see those parallels because you don't know enough to realize how different the two situations are".)

Posted by: Dentin at June 14th, 2012 11:39 PM

Only in the twentieth and twenty-first centuries would a certain kind of human mind come up with this kind of idea of "mind transplanting" or, more exactly, information transplanting.

There was a time when human beings did not equate mind=brain. Only for the past few hundred years, especially with the rise of rationalism and Western science, has mind become equated with brain. In the present decade, neuroscientists are intent on making mind even more specialized to equal electrical circuitry and interacting hormonal pathways. In other words, mind as electrochemical activity.

The more analytic the investigation, the more minute the units being studied, the worse the myopia, until we now have mind = information units of some kind supported or associated with distinct electrochemical activities in the brain. If we can only duplicate, electrically perhaps, those same kinds of waves, we might be able to transplant whole minds. Something like that.

I. Hope. Not.

Do you seriously believe that you--I mean the "you" of you--is merely the sum total of your electrochemical processes and information units? In other words, is that ALL you believe you are?

The whole process of intellectually conflating mind with brain is a developmental myopia that, as I said, emerged with the rise of modern science in the West. Essentially, it is materialism at its best (or worst).

I believe the entire problem of whether mind can be conflated to brain is based on how an individual or society views materialism versus immaterialism.

Rather than go through that long argument into religion (the soul), spirituality (spirit), philosophy (noumenon), psychology (self) and neurophysiology (brain neurochemical activities) and AI (artificial intelligence), I leave you with this question:

What is the weight of a thought?

You can't deny you think. You can argue that it is nothing more than electrochemical activity in your brain. You can argue that if you shut your brain down you will shut down thinking. But none of this proves anything more than correlation, not causation, and certainly not identity of mind with brain, thought with electrochemical activity.

Consider this. Everything material occupies space and time. Electrochemical activities occupy space and time. They are material, by definition. The brain, as a lump in your skull takes up space and exists over time. It is material.

What is thought? Does it occupy space? Is a thought of a mountain bigger than a thought about a flower? Is the thought about the Milky Way galaxy larger than a thought about your living room? Clearly,not. Thought does not exist in space! It takes up no room in your brain!

Does thought occupy time? Yes, it does. You can have a short thought or a long one.

All material things occupy three-dimensional space and the fourth dimension of time. That is, they take up room and persist for a certain duration.

Thought exists only in time. Therefore, thought is non-material. It exists in some non-material realm that we experience routinely. Therefore, thought cannot be "transplanted" from one place or space to another. The correlated electrical impulses and information units, maybe. But these are most decidedly NOT thought. They are correlates, "shadow thoughts" if you will, but NOT thought itself.

More challenging: what is this non-material realm in which thought resides? No brain ever excised and opened ever revealed a single thought! Where did thought go when the brain died? Are we absolutely certain thought died? Are we perfectly certain the realm from which thought emerged died with the brain?

I ask again: What is the weight of a thought?

Posted by: David Dressler, BA, RMT at June 17th, 2012 9:52 PM
Comment Submission

Post a comment; thoughtful, considered opinions are valued. New comments can be edited for a few minutes following submission. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.