Singing mice

Get a group of mice, tailor their genomes for increased rate of mutation, selectively breed for novel traits and let go. It's evolution writ fast in an artificial environment, and it's what Japanese biologists at the University of Osaka did in a nice bit of open-ended research.

Some of the mice ended up with longer limbs, odd tails, but they've had at least one surprising result. One sub-population of the mice can now sing - video here.

Aside from the core appeal of the idea of a mouse that sings (or chirps, as the case may be), these mice might help further our knowledge of how structured sounds (eg. birdsong), language and even the capacity for speech can evolve, especially in terms of the genes involved.

The researchers are quoted as claiming that the discovery may lead to discoveries on how human speech developed, but this seems just a tad too far-fetched for me. Mice and humans may be just too distant, phylogenetically speaking, for there to be any guarantee of a connection between the genes responsible for human speech and these mice singing.

The mice were genetically modified so that they were prone to DNA replication errors, improving the probability of mutations arising, some of which would then lead to interesting, novel traits. When the researchers identified novel traits, they would selectively breed the mice to carry the traits forward. It is a matter of semantics whether this deserves to be called accelerated evolution, since some would argue that natural selection must be involved for the term 'evolution' to be appropriate, and there is only artificial selection at work here.

In this case however, I'm happy with using the term accelerated evolution to quickly convey what was done, even though many writers have misused the idea of accelerated evolution to refer to something that happens to individuals (i.e. "he's evolving"), rather than populations over generations.

It is also intriguing to think of what applications this technique might have in creating novel traits in future. Who knows what kinds of traits could be imbued in future engineered pets or lifestock animals?

The lead scientist on the project: "I know it's a long shot and people would say it's 'too absurd'... but I'm doing this with hopes of making a Mickey Mouse some day."

Scales

It might not seem like it, but I am not an individual. I am a conglomeration of tiny single-minded agents, each acting according to its own interests (programmed in by generations of differential gene survival). None of these agents know I exist. They are incapable of even conceiving that something like me might exist. And yet, the aggregate of their interactions with each other equals me.

The exact biological identity of these agents is not really important. It depends on exactly what I consider the core of my being and what is accessory. It might be reasonable to treat every cell in my body as a single tiny agent, contributing in its own way to generating me. In any case, the facts are that I am not an individual, I am a group.

As I already said, not one of the neurons in my brain knows that I exist. Not one of the cells in my blood knows that it is helping to keep me alive. Not one of my muscle cells knows it is helping me move. They just selfishly follow their programmed instructions.

At this point, it is worth clarifying what I mean by 'selfish'. Each and every one of these agents is selfish in the sense that it follows its own instructions, and only its own instructions. They do not go out of their way to assist other agents, except if their programmed instructions demand so. A white blood cell may assist other cells by engulfing pathogens, but it has no concept of the implications of these actions. All it 'wants' to do it eat pathogens.

So if these agents can generate my mind as a side effect of their selfish actions and interactions, though they have no concept of doing so, is it reasonable to posit that the same can work on different scales? Treating individual humans as agents, say.

I'm intrigued by the idea that society itself might be conscious. After all, it certainly acts intelligently. The society as a whole is orders of magnitude more intelligent than any individual human within it. In fact, due to the added benefits of individuals working together, the intelligence of the society is greater than the sum of the intelligences of the humans that comprise it.

But can our pedestrian concept of consciousness even apply to something like a society? Can a society be self-aware, for instance. If a system makes predictions of a future state of the environment, with itself in that environment, we might infer that it is self-aware. Society certainly does this, so might it be self-aware?

With these kinds of examples, we are running headlong into the Chinese telephone brain - a well-known thought experiment in the philosophy of the mind.

Suppose that every citizen of China is issued with a telephone and a book of rules. With these tools, each citizen will reproduce the activity of a neuron in the human brain, with telephone calls taking the role of synaptic signals. Obviously, this would be much slower than the human brain, but would the system as a whole be able to think?

Many people have an instinctive bias against even considering the idea that systems outside our own everyday scale might be conscious. We are groups of simpler agents, and there are greater, more sophisticated groups above us. If the groups above us are more intelligent than we are (as their behaviours testify), then why should we deny them the possibility of consciousness a priori?

Perhaps, by drilling down into what we really mean by consciousness and trying to generalise our definitions onto exotic systems, we might discover something surprising about what our own minds really are.

The idea that every individual neuron in our heads might be, in a sense, conscious is ludicrous to us. Perhaps a larger societal mind might find the idea of each human comprising it being conscious to be similarly ludicrous.

Cyanotheces 51142

When Craig Venter made synthetic life earlier this year, he gave possible future applications of synthetic biology. High on the list of near-term applications, he mentioned designing microbes that use sunlight to produce hydrogen. Well, nature has beaten him to it. Sort of.

These handsome things belong to a species of cyanobacteria (blue-green algae) called Cyanothece 51142. By day, they make glucose with carbon dioxide and sunlight, like most plants. But by night, an enzyme uses the glucose plus nitrogen in the air to produce ammonia, which gives off hydrogen as a by-product.

What we have on our hands is a bacterium that makes hydrogen using energy from sunlight. This is significant because hydrogen is likely to serve a major supporting, if not leading role in our future energy economy.

The idea goes like this. Energy is very difficult to store in any great magnitude for a prolonged period of time. Renewable energy sources like solar power and wind power are intermittent, so a long-term means of storing huge amounts of energy is necessary. Hydrogen gas can be created in endothermic (energy-consuming) reactions, and can later be combined with oxygen in a fuel cell to give energy and water.

There are no natural sources of hydrogen gas, so a big problem with this concept is the present lack of an efficient means to make hydrogen. The amount of energy you put in to create the hydrogen vastly outweighs the amount you get back at the end. If we could bring the values closer to parity, it would go along way to making a hydrogen economy feasible.

By genetically tweaking organisms like Cyanotheces 51142, we might gain an efficient way of using the energy of sunlight to produce hydrogen.

Of course, there are other problems with a hydrogen economy. There are concerns over hydrogen's energy density, our ability to store it safely, and our ability to make efficient fuel cells to recombine the hydrogen at the end of the process.

But nonetheless, the discovery of this little critter can only be good news.