Superhuman reflexes

Though perhaps best exampled by Neo in the 1999 film The Matrix, superhuman reflexes to some extent are an attribute of so many superheroes that it’s basically a default superpower. But just how realistic would it be for us to expect genetic enhancement to enable people to have such quick reflexes? That’s what I’m going to attempt to answer in this post.

First we need to define some terms. Biologically, a reflex is a simple and automatic response to a stimulus. If you touch something painfully hot you will automatically withdraw your hand, so it’s a reflex. You can still learn or unlearn a reflex, but the important thing is that a reflex is done without thinking: your brain isn’t really involved. If, out of the corner of your eye, you see something falling off a table and quickly grab it, that’s not a reflex. You don’t automatically grab any falling object, and if that object was a knife you’d leap away from the table rather than try to grab it, which is a sure sign your brain is processing the situation rather than it being automatic.

As you’re probably aware, there is a delay between the stimulus (e.g. touching something painful) and the response (e.g. pulling your hand away) of a reflex. The main factor determining this delay is how fast the signal can travel along the nerve cells (neurons) from the hand, to the spinal cord, and back down to the muscles (there’s a slight delay as the signals cross the synapse from one neuron to another, but its contribution is minor). In humans, the neuronal conduction velocity varies between different nerves, but let’s just look at the ones sending the signal from the hand to the spinal cord (which travel at about 50 meters per second) and those that send signals from the spinal cord down to the muscles (which travel at about 100 meters per second). So travelling a meter along your arm and then another meter back means the fastest possible reflex would occur after 30ms (it’s a bit slower than this in reality, but let’s go with this).

So the question now turns to how we can get neurons to conduct signals faster. There are three tactics that animals use to do this. The first is making the neurons thicker in diameter, which is why the neurons that control our muscles are the some of the thickest in our body at up to 20 micrometers. But some animals, like squid, have neurons that are are 500 micrometers, but our neurons are still 4 times faster. This is because animals like humans have a second strategy, which is to insulate the neurons with a substance called myelin. This allows the neurons to be much smaller while still being really fast, which is essential for packing as many brain cells as possible into a small skull. But the record for the fastest neuronal conductional velocity in the animal kingdom belongs to a penaeid shrimp, which not only has fairly big neurons (at 120 microns) and myelin insulation but also has a third mechanism where in between the neuron and the insulation is a gap filled with super conductive salty fluid that further speeds up how fast the signals can travel (for a discussion of these mechanisms, see Castelfranco & Hartline 2016). Their neurons can send signal at 200 meters per second, twice the speed of our fasted human neurons.

Given that we need to fit our nerves through the holes our vertebrae (the neural foramina), we can’t really rely on increasing the size of our neurons. And we’re already fairly well insulated. Maybe we could use some shortcuts like the penaeid shrimp do, but without the increase in size it’s unlikely we’d get to the velocities of 200 meters per second the shrimp achieve. Any genetic enhancement to human reflexes is realistically going to be much more modest.

But let’s say we do manage to double the speed of our neurons, what would that be like? Well, a simple doubling of conduction speed would more or less halve our reaction time. Our reflexes would be faster, but so too would the speed at which we could think and perceive things. Our sense of time wouldn’t change, but given the limitations of our current neurons the closest we can come to seeing what the world would look like with double speed neurons is watching a video at half speed (which you can easily do on YouTube). This would give a massive advantage in martial arts and many sports like sprinting, fencing, tennis, football and baseball.

But would this magnitude of reflex enhancement be enough to dodge a bullet? Almost certainly not, and not least because we have only speed up the nerves while leaving the muscles as slow as ever. The simplest test of visual reaction time of the sort you’d need to dodge a bullet is the ruler drop test, where you try to catch a measuring stick as quickly as you can after noticing it has been dropped. This therefore captures both the ability to visually detect movement and perform a simple movement (a pinch grip) in response. The average reaction time, on this test, for athletes is about 200ms. Even if your reaction time has been enhanced to be half that of a normal man, a Glock 17 has a muzzle velocity of 375m/s so in the 100ms it takes you to react by moving your muscles the bullet would have traveled 37.5 meters. The gunman would have to be quite a distance away to allow you to duck behind cover as soon as you see the gunshot. And the bullet would still be traveling too fast for you to see, because we only enhanced reflexes not your visual perception speed (which would involve making photoreceptors in the retina work more rapidly), so you would still be attempting to dodge a bullet that you cannot actually see.

So genetic enhancement of reflexes has many practical limitations in comparison to the fanciful portrayals we see in fiction, but would still be immensely advantageous in competitive sport or hand-to-hand combat where every millisecond matters.

In order to get extremely rapid reflexes, we’d need to do away with the limits of biological systems altogether and transmit signals through electronic circuits, which could a million times faster. When it comes to quick reaction times, robots or perhaps cyborgs will have a massive advantage. At the moment robots already have faster “reflexes” than humans, but the human brain still outperforms artificial intelligence if the stimulus requires complicated visual or spatial processing or if performing a novel movement requiring coordination. But maybe one day, robots won’t only be beating us at chess but also fencing, tennis, baseball and martial arts. And by then, maybe we’ll be able to implant that technology into our own body, and truly gain reflexes faster than anything the biological world can offer.

Man controls new prosthetic leg using leg muscles

Though reported as ‘man controls prosthetic leg using thought’, the man is controlling the prosthetic leg using his thigh muscles, which he is controlling using the nerves that used to go to his lower leg and foot (but were surgically re-routed to the thigh). The prosthesis detects the electrical activity in the muscles, rather than the nerves themselves.

So it goes like this:

brain → spinal cord → motor nerve → thigh muscle → EMG in prosthetic limb → movement

Not even interfacing with the motor nerves directly, let alone any kind of a brain-computer interface. This is NOT a mind-controlled prosthetic, any more than any other prosthetic.

The first transhumans will be gene-modded

Biotechnological interventions are, in the short term, the most likely method to obvious human enhancement.

I’ve said it, and that’s also the conclusion of Kyle Munkittrick’s post “From Gears to Genes: A Sea Change in Transhumanism“:

Transhumanism is the idea of guiding and improving human evolution with intention through the use of technologies and culture. If those technologies are not robotic and cybernetic but, instead, genetic and organic, then so be it. And that seems to be the way things are going.

I totally agree. But for different reasons.

Yes, it’s totally true that genetics is advancing faster than cybernetics, nanotechnology or artificial intelligence. Unlike Kyle, I wouldn’t say those fields are without progress, but they are moving more slowly and, to top that off, are farther away from being able to produce an enhancement.

On the other hand, mice have already had their memories, strength, endurance and lifespan enhanced by genetic or pharmaceutical means. In fact, so have humans  – by evolution. The reason humans are smarter and longer lived in comparison to mice (and in comparison to the common ancestor between humans and mice) is entirely genetic. It seems stupid to come up with a different paradigm when you already have a proven technique for enhancing intelligence and lifespan.

I’ll concede that biology has its limitations (No gene can make you bulletproof), and for the fancy enhancements of posthuman sci-fi biotechnology won’t be enough. And this is probably why some let their fantasies cloud their judgment and continue to believe they will be able to upload themselves into a virtual reality powered by cold-fusion powered quantum computers implanted in their brain within a decade or two.

I can say with confidence that the first clearly enhanced humans, with really obvious improvements in intelligence/lifespan/athleticism will be a product of genetic enhancement alone.

Hattip to IEET: http://ieet.org/index.php/IEET/more/munkittrick20100611/

Boosting brainpower

The practical and ethical issues with intelligence enhancement are receiving more attention, with a recent article in New Scientist titled “Will designer brains divide humanity“.

For the most part, the article is quite basic, but I have an issue with one part in particular:

The next stage of brainpower enhancement could be technological – through genetic engineering or brain prostheses. Because the gene variants pivotal to intellectual brilliance have yet to be discovered, boosting brainpower by altering genes may still be some way off, or even impossible. Prostheses are much closer, especially as the technology for wiring brains into computers is already being tested.

This is none other than cybernetic favoritism! I mean sure, genes effecting intelligence aren’t obvious, but it’s also not obvious how and where to interface a brain chip to increase intelligence. And though neural prostheses are being tested, no neural prosthesis has increased any aspect of intelligence in any brain, whereas there have been 33 genetic alterations that increase the learning and memory of mice (not to mention that all the differences in intelligence between animals are genetic in origin). Considering the annoyance of having surgery for neural implants compared to the ease of a simple injection for genetic modification, I would personally put my money on the genetic enhancement of intelligence. Nonetheless, both avenues should be pursued, and might eventually complement one another.

Onto the ethical issues discussed in the article, most are fairly basic. Starting with human dignity, referring to comments made by Dietrich Birnbacher, a philosopher at the University of Düsseldorf in Germany:

One potential problem arises from altering what we consider to be “normal”: the dangers are similar to the social pressure to conform to idealised forms of beauty, physique or sporting ability that we see today. People without enhancement could come to see themselves as failures, have lower self-esteem or even be discriminated against by those whose brains have been enhanced, Birnbacher says.

These concerns are all quite valid, but aren’t necessarily impossible barriers. If enhancement technology was supported by the government, then no people wanting such technology would be left without it. And the discrimination I will deal with in a minute, after looking at the next section:

The perception that some people are giving themselves an unfair advantage over everyone else by “enhancing” their brains would be socially divisive, says John Dupré at the University of Exeter, UK. “Anyone can read to their kids or play them music, but put a piece of software in their heads, and that’s seen as unfair,” he says. As Dupré sees it, the possibility of two completely different human species eventually developing is “a legitimate worry”.

I do actually worry about enhancement being socially divisive, but I am not sure this would occur only by discrimination of the enhanced towards the un-enhanced. As I have argued previously, it’s entirely possible that the enhanced will be viewed as unnatural disgraces to humanity, and the pure, natural humans would discriminate against them because of it.

The rest of the article deals with issues such as brain plasticity, evolution and epigenetics. These are not particularly relevant to any ethical concerns and neither will they significantly enhance the intelligence of the average reader of this blog, so I’m not going to address them here.

Nanotube-based neurotechnology

A recent study in Nature Nanotechnology shows augmentation of neurons by growing them on a base of carbon nanotubes. The author’s hypothesis is that this is due to the carbon nanotubes conducting electrical impulses from the axon back to the dendrites, thus acting as a shortcut for the normal process of back propagation of action potentials.

Back propagation (propagation in the direction opposite to the arrow in the above diagram) occurs when the action potential flows not only along the axon, but also back to the dendrites. Back propagation of the action potential is required for synaptic potentiation, as it allows for extra excitation of the dendrites at those occasions when the axon has fired, thus leading to better coincidence detection via Hebbian learning (Magee and Johnston, 1997). Neurons strengthen connections based on the close timing between presynaptic and postsynaptic excitation (‘fire together, wire together’), so it is important for signals of postsynaptic excitation to propagate back to the dendrites where the synapses are, so that the synapse may be strengthened or weakened accordingly.

In a sense, then, the augmentation of neurons in this study is similar to that seen in NR2B transgenic mice (aka ‘Doogie mice’), which overexpress the gene for a subunit of NMDA channel – a channel involved in long-term potentiation (LTP) of synaptic connections. The facilitation of more back-propagated action potentials would also result in enhanced learning and memory, if carbon nanotubes were seeded into key parts of the brain. This would, however, cause similar problems to the NR2B Tg mice. There is more to learning than simple recall, as forgetting is important too. We have a limited number of neurons to play with, and therefore a blanket increase in memorisation is an inferior solution compared to a highly-regulated memory augmentation, where things we need to remember are remembered and things we don’t wish to recall are forgotten.

But, this study was basically a random arrangement of nanotubes and neurons, which created some interactions which proved to be functional. It will be very interesting to see future neurotechnology based on carbon nanotubes, as these ‘wires’ are small enough to connect not just to a single neuron but between individual parts of that neuron. Yet still many of the issues of brain-computer interfaces will exist even with the enhanced biocompatability afforded by the carbon nanotubes – one needs to be able to interpret signals used by neurons and compute them, and this computation needs to be done with a small, preferably implantable computer. Still, I have a feeling nanotechnology will be essential for any cyborg (which is the main reason I had nanotechnology as my second field of study on my Bachelor of Science, after my first love – neuroscience).

Reference: Cellot et al (2008) Carbon nanotubes might improve neuronal performance by favouring electrical shortcuts, Nature Nanotechnology, AOP 21 December 2008, doi: 10.1038/nnano.2008.374

Cyborgs vs fyborgs, modifications vs medications

Throughout the twenty-first century, we fyborgs will find ourselves deeply integrated into systems of machines, but we will remain biological. And as long as this is true, the primary changes to our own form and character will arise not from implants but from direct manipulation of our genetics, our metabolism, and our biochemistry.

The above paragraph comes from Gregory Stock’s 2003 book Redesigning Humans, in the chapter (chapter 2) entitled ‘Our commitment to our flesh’, and argues that enhancement of our bodies will occur by biological means like genetic modification, because any cybernetic technology will be abandoned in favour of the fyborg alternative.

Coined by the late transhumanist Alexander Chislenko, the term “fyborg” is a portmanteau of ‘functional’ and ‘cyborg’. It refers to the utilisation of technological tools external to the body, which is supposedly a more popular notion than having surgery to implant the technology. So, while a cyborg would use a mathematical processing chip implanted into his brain, a fyborg would use a calculator or notebook computer to perform any difficult calculations. A cyborg may have an artificial eye overlaying an interface onto the world, but a fyborg may achieve the same thing by wearing high-tech glasses.

Gregory Stock would do well to realise that a similar divide as exists between cyborgs and fyborgs can be seen in the field of genetic technology, with the competition between modifications and medications. Currently, a lot of biotechnological interventions are reduced to a form where they can be taken as a regular injection or an oral pill, supposedly because people don’t want to mess around with permanent changes to their body or genome. So, while one could get genetically modified to have super-fast reflexes, another may take a drug with similar effects. A lot of the same benefits of fyborgisation, and the disadvantages of surgery, will be likely present in biological interventions as well.

Gregory Stock outlines a perceived problem with cyborg technology quite well in the following excerpt:

Enticing as a direct brain linkage sounds in the abstract, virtually every scheme for one has this flaw. Healthy individuals are not going to allow some cyber-surgeon to hack into their brains to bring them enrichments that are largely obtainable in other ways.

This first appeals to a visceral reaction to surgery, which compares quite unfavourably to simply buying a high-tech gadget. Yet I could see the same reaction occurring when a parent is presented the choice between allowing his healthy young daughter to have her chromosomes hacked by some genetic engineer just to bring a benefit that could be given by a pill every morning and night. All the same, this distinction is not purely based on emotions, for there are good sound reasons underlying this gut reaction.

One of the strongest real benefits of fyborg technology over cyborg technology is that it is non-invasive, which primarily makes it far less risky than the alternative. Stock writes, rightly, that we “are not cavalier even about proven technologies”, and therefore a drastic procedure like surgery will risk too many problems. The same could be said, however, of genetic technologies in comparison to pharmacological enhancements. That a drug has only brief effects has the benefit of making any side-effects similarly short-lived. Therefore, by Stock’s own logic, the genetic interventions will be cast aside in favour of pharmacological equivalents.

Stock also explains that a fyborg device can be more easily repaired and updated than the cyborg alternative, allowing for the enhancements to keep up with the changing pace of technology. Yet this too applies to biological interventions, because it is indeed far easier to switch to the newer and more advanced medication than it is to have one’s genes edited to a more effective or safer version. And likewise for any regrets one has about an enhancement, which would be far easier to rectify if such enhancement were non-invasive fyborg ones or temporary pharmacological ones.

Is that to say that in the future we will just carry around move advanced gizmos and consume better versions of coffee and Viagra? Not at all, and Stock does explain perhaps the main reason why anyone would choose a permanent body modification – when it either is the best way to produce that enhancement (or the only way).

Implantable technology would benefit strongly from the close interaction with the body. As Stock points out, you can’t have a wearable pacemaker, because it needs to be implanted. Likewise, an interface directly with the brain or body would be far faster and more effective. A calculator takes time to use, but if the brain could instantly connect with a mathematical processor, this would be far faster. To put on an exoskeleton to lift a heavy object takes time, but super-strong prosthetic limbs would always be on call as part of your body. Not surprisingly, this applies to genetic modification as well. Not all biological modifications are amenable to being taken in a pill or potion, due to the need for the contents of a pill to avoid being broken down by the digestive tract and travel through the stomach wall to the correct site in the body without being diluted too much. Further, though most temporary modifications would be injectable, so too would a well-made permanent gene mod (and a once-off injection is better than regular injections, I’d say).

In addition, the fact that fyborgs and pharmaceuticals are not permanent additions to your body is not only its greatest benefit, but also a source of great weakness: any non-implanted technology can be more easily lost or fall into the wrong hands. In a world where enhancements may be expensive, highly-coveted and possibly illegal (or abhorred by some), this is not a trivial issue. Surely many would sleep more soundly knowing their enhancements are hidden inside their body, not lying around to be stolen by thieves or confiscated by police. The same could be said of a bottle of pills, which share these same problems and in addition have only limited uses. (Although, the theft of something from inside another’s body, or the forced removal of genetic modifications, may be far more dangerous and gruesome).

That said, I do agree somewhat with Stock’s conclusion that genetic enhancement will be more influential than cybernetic technologies, in that I think it will upon us far sooner and, in the near future, be more effective. And, to my surprise, Stock nails the reason quite well too:

Our flesh is a dense three-dimensional matrix of biological cells, ill-suited for a permanent, working union with broad arrays of sensitive electronic probes.

The problem of interfacing with the moving target that is the self-organising network of neurons we call the brain is a mammoth one, and the simple brain-computer interfaces that exist suffer from incalculable technological and methodological barriers before they will reach the level of our brain, let alone supersede it. Genetic interventions, on the other hand, really have only a single large hurdle – targeting of genetic elements to the correct location on the genome (to avoid cancer-causing mutations).

Electrogenic humans

Electrogenesis refers to the production of electrical activity in living tissue. In a sense, we humans are already electrogenic; each of our brain cells (neurons) produces about 70 millivolts of electric potential and our muscle cells about 95 millivolts. They do this by using chemical energy (in the form of ATP) to power electrogenic pumps, most commonly the Na⁺/K⁺ ATPase (sodium-potassium transporter, but usually abbreviated to NAKA)- this enzyme is so common in the massive human brain and muscles that it is responsible for using up to 40% of our resting energy consumption (and the man who discovered it, Jens Skou, was awarded the 1997 Nobel Prize in Chemistry. This pumps out 3 Na⁺ atoms in exchange for 2 K⁺ atoms. This is a good deal, because three positively charged ions (cations) are exchanged for two, which causes the electric potential to drop below zero. With enough of these going, it will drop give the cell a slight charge.

But some marine creatures really excel in this aspect, with specialised cells called electrocytes dedicated entirely to the production of an electric potential, which they use primarily to communicate and sense prey but also to stun other animals (prey or predators). These include the electric rays (Torpediniformes) which can produce a potential of around 200 volts, the electric catfish (Malapteruridae) which are capable of producing 350 volts and, perhaps one the best known (erroneously as the electric eel), the electric knifefish (Electrophorus electricus), capable of a shock of up to 600 volts (Mermelstein et al, 2000). The electric knifefish/eel has been proposed as next on the list of animals to have their genome sequenced (Albert et al, 2008), which I think will speed this sort of research along quite nicely.

These animals have a specialised organ, called (of all things) the electric organ, which is made up of thousands of these electrocytes (sometimes also called electroplaques or electroplaxes) organised in series (with each series stacked in parallel to sum currents), with each cell producing a potential about 150mV (actually, there are two potentials, one 65mV and the other 85mV. For details, read Jian Xu’s scientific paper, which I link to below). That doesn’t seem like much, but stack ten thousand of these in one organ and, if all those cells discharge at one, it will produce a 1500V electric organ discharge (EOD).

Just last month, researchers Jian Xu from Yale University and David Lavan from the National Institute of Standards and Technology (Maryland, USA) published a paper in Nature Nanotechnology outlining a theoretical upgrade to the electrocyte, able to produce 28% more power and to use chemical energy to do it with 38% greater efficiency. It was theoretical, however, but they will probably try it (or somebody else will) for real some time soon. And probably the electric fishes themselves will be slowly evolving towards this outcome themselves.

In the meantime, we can consider how awesome it would be to generate an electric current ourselves. Most in the Western world are experiencing an obesity epidemic, so we have plenty of chemical energy to spare for producing an electric potential. The most likely and practical option will be to have a small patch of electrogenic cells surrounding any electronic implant, like the prosthetic arm and cybernetic implants we will all have by that stage. They may also prove useful in biological pacemakers, if the heart was surrounded with electrogenic cells to provide impulses (though, it would probably prove easier to just repair the heart).

More interestingly (at least in my opinion), we could genetically engineer (or implant) our very own electric organ. If the electric organ was just below the skin of our chest and arms, but very well insulated except for at ends of our fingers, we’d literally have the full current and voltage of the electric organ at our fingertips. We’d also need to wire up the brain control mechanisms of this organ, specifically some brainstem nucleus to act as a pacemaker to ensure all those electrocytes fire together, so that we could control when the electric shocks occur.

Now, before we get all excited about throwing lightning bolts, let me first remind everyone of how physics works back in the real world. As I said, bioelectrogenesis has evolved only in aquatic organisms, because water is a good conductor. Air, on the other hand, is a damn good insulator until a phenomena known as dielectric breakdown occurs (this is when the air ionises to become conductive). The dielectric strength of air is 3,000,000 volts per metre (3MV/m), but greater on hot or humid days. This means that to throw an arc discharge (i.e. a zap of lightning) across on metre of air, you require 3 million volts. You can throw a much smaller arc with much less, as anyone who as zapped themselves with static electricity on a doorknob will know (noting Paschen’s law, which shows that even a small voltage of 500V will be sufficient to cross a gap of half a centimetre). If I hold my thumb and forefinger just a centimetre apart, I will still require 30,000 volts to throw an arc between them. To produce that with a current of 1 amp (which probably won’t be necessary, but let’s assume that for the sake of ease of calculation) would require about 150,000 cells (assuming each produces 200mV), which is far more than the electric eel has (and, mind you, the electric eel dedicates about 80% of its 25kg body to its electric organs, and still only manages about 4000 cells per series).  So in other words, you’re not going to be able to throw lightning (which makes sense, because we don’t even have a practical ‘lightning gun‘ yet, let alone a biological version).

But I don’t really care. I’d settle for being able to make a lightbulb glow or give somebody an annoying shock if they irritated me (essentially a biological electroschock weapon). So, where’s my bioelectrogenesis?

[Hat tip to fayyaad at Utter Insanity for inspiring me to try to explain how human electrogeneration is possible and bringing my attention to this recent news]

Rat brained robots and reporters

As news stories have been reporting for the last 36 hours or so, researchers at the University of Reading in the UK (home of the Project Cyborg, led by Prof. Kevin Warwick) have made a robot controlled by rat neurons.

But many reporters, being fundamentally rat-brained themselves, have made the mistake of saying silly things like these:

To build the cyborg, Warwick, his colleague Ben Whalley and their team dissolved a fetal rat’s brain. They then put the free-floating brain cells, or neurons, into an electrode-ringed Petri dish, where the cells quickly reassociated with each other and began randomly firing electrical signals. Electrical pulses sent through the ring of electrodes calmed down the neurons, and after a while the cells began to “learn” patterns among the external pulses. [from FOXNews]

Now the reason there are quote marks around the word “learn” is because it didn’t really happen.

It is true that a network of 300,000 murine neurons (i.e. rat brain cells) are controlling that robot through 80 electrodes. It’s true that the cells, which were dissociated in solution, spontaneously formed that network (which was, in my opinion, the most exciting part). But it is not a robot controlled by a rat’s brain, for these are only brain cells – admittedly forming a primitive brain, but not a brain anything like that of a rat.

And it is most certainly not true that these neurons learnt how to control the robot – the researchers set up the robot and electrodes so that whenever the sensors detected a wall, it sent a 1 volt pulse to the neurons in the dish. And the neurons, being fully functional, were elicited into giving off an action potential. Just like neurons do when hit with an electrical stimulus of 1 volt (which is massive for such small cells, by the way). And this action potential, after it had been transmitted through a few neurons, was recorded and used (by a computer) to make the robot turn away from the wall. It is not the neurons in the dish that learnt to control the robot, but the computer that is reading the activity of those neurons (and the programming of that computer by human brains).

From a neuroscience perspective, the neurons have not learnt much. They have probably strengthened their connections to the electrodes, but it is a far cry from learning to control the robot. Those 300,000 neurons aren’t really doing much that the 302 neurons of the nematode worm C. elegans couldn’t do – in fact, they are probably doing far, far less.

Aside from the fact that the neurons formed a network spontaneously, I am unimpressed.

UPDATE: Steve Potter, a neuroscientist from Georgia Institute of Technology has commented on the blog Neurophilosophy, saying:

I see nothing new here beyond what we and others have been doing for the past 5 years. Believe what you read in peer reviewed papers.

So that explains why I, being a neuroscience student, wasn’t at all impressed.

We can’t rebuild him. We don’t have the technology.

Actually, we have some technology, but it’s not very good at the moment. That is the conclusion of a recent technology review in New Scientist, which asks the important question: Do we have the technology to build a bionic human?

First on the list, replacing bones with metal alloys. Duncan Graham-Rowe, the author of this New Scientist tech review, hits upon a very important point:

But artificial bones are not perfect. One idea that may see them match natural bone’s strength and lightness is to build implants by zapping titanium powder with a laser. That can makes pores of different sizes in different areas of the finished product, controlling strength and stiffness in the same way as real bone

The part in bold is important, because an artificial bone has to be much the same as the original, despite fictional cyborgs usually getting some enhanced strength or resilience from their metallic skeleton. This is because a strong metal (titanium alloys have a yield strength of about 800MPa) in contact with weaker bone (cortical bone has a yield strength of around 100-200MPa) can cause a lot of stress on the bone at the contact point, causing wearing and severe pain. Only by replacing the entire skeleton could a strong alloy be used in a cyborg.

The review then moves onto tissue engineering – rebuilding humans with the same materials they were made with in the first place: flesh and blood. That is a promising and fast growing field, because it prevents a lot of foreign body immune reactions, but it doesn’t really get to the heart of the bionic human question (pun not intended).

This is quickly remedied as the discussion turns to neural interfaces, specifically cochlear and retinal implants, as well as the a hippocampal prosthesis. All of which suffer from the same problems. First, they require a lot of understanding about how the neural circuitry works. Second, fine stimulations require very small electrodes, and interference starts to become an issue at those scales. After all, most neurons are only a dozen microns across. (but highly variable, like any biological tissue).

Artificial limbs are up next, which involve all of the above issues, as they must not only join to the skeleton to transmit their load effectively to the body, but also be controlled by the nervous system. Needless to say, while solutions look promising, there is still a long way to go, for much the same reasons as mentioned above. Advanced arm bionics usually are controlled by the nerves from the missing limb, albeit rerouted to control chest muscles which in turn activate movement sensors which drive the limb. And they usually require a shoulder strap to transfer their load onto the body, though in the near future they will likely be connected directly to bone. But hey, we can’t regrow limbs (yet), so this is a good stop-gap solution.

Finally, the review ends with a discussion of power requirements. Powering the prosthetic from the body is good in theory, but biological systems are far, far less power hungry than artificial systems (the human brain uses 20-25W of power, while a Playstation 3 or high-powered PC can use up to 200W, and IBM’s Roadrunner supercomputer that recently set a processing record trying to simulate the mammalian visual cortex uses over 2,000,000W). So it is likely that trying to run any advanced neural interface or bionic limb will use up far more than the human body could generate, and even small systems could drain the area around the implant of all available energy. It will be batteries or fuel cells for a while yet.

Actually, the review ends by talking about how bionic systems could be vulnerable to electrical interference, but the human body isn’t exactly invulnerable, so I don’t think this a fair comparison. It is merely trading one weakness for another – a problem for those with pacemakers, who end up with the weaknesses of both worlds – but not for somebody who was shot in their bionic leg and doesn’t bleed to death from the femoral artery.

AI is NOT part of transhumanism…

..well, at least not proper transhumanism as alluded to by Huxley and refined thereafter (AI may be transhumanistic under that perverted transhumanism espoused by Max More).

Transhumanism is about extending human capacities. However, artificial intelligence research is about making beings with capacities approaching, and eventually surpassing, what we know consider human capacities. They are parallel routes to superintelligences and even to a technological singularity, but they are not the same thing. How can AI be ‘more than human’ if it is something different entirely? Is an apple ‘more than an orange’? One may taste better, and one may be juicer, but an apple is not an ‘enhanced orange’ nor is an orange an ‘trans-apple’.

An AI with human-like capacities would not be human, because it is not an organism with the designation Homo sapiens (it may, however, qualify as a person worthy of the same rights we give to adult humans). A robot with some capacities greater than those possessed by humans would not be a transhuman, because it never was human nor descended from humans, so is not an extension of anything human (except in the abstract sense, under which my scientific calculator is an extension of myself). An upload of a human mind would be transhuman (as would a human using a brain-computer interface), however, because he or she would still have been a human, but extended into an artificial substrate.

The same thing goes for posthumanism. A posthuman is a human extended to such a degree, with so many capacities vastly beyond human limits, that the ‘human’ designation isn’t doing it justice. And an AI superintelligence would not be a posthuman, because it would not be a radical extension of humanity (at best it is a post-computer, because it is such a radical extension of my desktop PC).

To be fair, the same arguments that are used for transhumanism also make good ones for extending the capacities of any other sentient being, whether they are humanoid robots or non-human animals like dolphins and whales. But that wouldn’t be transhumanism, it would be ‘transandroidism’ or ‘transcetaceanism’. Or perhaps more generally ‘transsentience’ or something.

Julian Huxley, when he first used the term, described transhumanism as “man remaining man, but trans­cending himself, by realizing new possibilities of and for his human nature.” Transhumanism is exclusively about making better humans. Unless humans can merge with computers, artificial intelligence don’t enter into it.