http://www.nytimes.com/2011/07/28/science/28life.html?_r=1

Consciousness, I would think that is the thing that defines life.
Or could life be a dead, lifeless, mechanical process sort of thing?
I just thought that one line in the article was a bit odd.

It remind me of Jim's post:
http://blog.theuniversesolved.com/2011/05/02/is-lida-the-software-bot-really-conscious/

Now you may find my following comments either overly complicated or a little over simplified (if not downright confusing, depending on your frame of reference) but I define sentience as simply being able to pause -- but pause for a purpose. In other words, our brains can only categorize. We do not compute facts like a computer. We categorize and then pause.

I think Consciousness defines sentient life, but other simpler things may define "life" in general. For example, an amoeba is alive, but is an amoeba sentient?

Think about it. I define sentience as simply being able to believe (i.e. to actually make a choice). Computers can randomize and give you a random result, but have no ability to actually choose. "Pick a number between 1 and 10" for example. Sure a computer will roll its dice and give a number, but did it acually "pick" a number? That is why computers (at least so far) have not been sentient.

Amoebas and higher animals like dogs, horses, and cats can make some miniscule choices (but my cat likes to drink water out of the fish bowl even though he constantly gets scolded for doing so). That is because ameobas and other animals only have a very limited ability for choice. Mostly they are like robots -- programmed to act on instinct. We humans, however, are the true great experiment of the programmers (the Akasha ... the Matrix ... God' VR world .. programmed reality).

But Human consciousness is this higher ability to think about things and then form a belief -- which to me means to make a choice -- which to me in turn simply means "pause" for a moment.

We think that we as humans compute facts (with our brains). We do not. Computers compute facts. Our Brains compute choices. "Validation" and "verification" in "Bayesian" type models are both crippled and perturbed by the endless heuristics, weights, probabilities, probabilistic quantifiers, or levels of importance. The great human experiment by the programmers was to give us ("players" in this VR game of life) the ability for "demon" arrest (i.e. in a mathematical sense to overcome the combinatorial implosions that probabilities cause). Whereby, we as humans who are the programmer's triumph of working around the combinatorial explosions (implosions) of knowledge representations within the finite computational resources of a "physical" world (local universe) Bayesian probabilities are replaced by simulated reality (non-local universe) where the belief representation (rather than knowledge representation) terminates the probabilities and cannot be crippled by combinatorial explosion.

So to paraphrase Descartes only slighly ... I believe, therfore, I am.

I guess I have a slightly different perspective. AI researchers often talk about consciousness as an emergent property of the complexity of the brain. I am very interested in things that are emergent - it is a concept that applies to many systems in many different fields, including software design. For example, rather than designing software up front, a design can emerge from developing software using the practice of test-driven development (TDD).

I talked about emergent properties a bit on the Reality Transformed panel last week. We could categorize emergence into two different types: continuous and discontinuous. Continuous emergence would be a property that results from some process no matter how small or large the input to the process is. If consciousness is a continuously emergent property of neural complexity, then one would have to say that any animal has some level of consciousness. Not only that, any computational device, no matter how small (a flip flop) would have some level of consciousness. Personally, I find that argument a little hard to accept to say the least.

However, one could make the argument that in "living" things only, consciousness exists at a level relative to the level of the complexity of the neurological system. Such an argument makes some sense to me. As an example, if consciousness is a manifestation of a creator experiencing physical reality through living organisms (as many philosophers and mystics have proposed), it would stand to reason that the level of consciousness that can be experienced is somewhat proportional to neurological complexity.

On the other hand, there could be such a thing as discontinuous emergence of a property or process. As an example, consider nuclear fusion in a star. You can pile more and more matter into the system and let gravity generate higher and higher heat as the matter collapses inward. But no nuclear fusion happens until the magic density and quantity of matter occurs. Then, boom, the star blinks on. Could consciousness be similar in nature; that is, it doesn't emerge from neurological complexity until a particular threshold is passed. Maybe that threshold is just below that of humans? Or some other arbitrary point in the animal kingdom?

But I doubt it. Because there is no evidence of that. How might we determine empirically whether something is conscious or not? By apparent intelligence, or ability to learn? Maybe not because synthetic neural nets can learn. By apparent ability to make decisions? Again, logic gates are simple and don't even require silicon. How about by jdlaw's method, by observing a pause before a decision? I kind of like that, but of course it isn't foolproof. Fact is, we have no better way of determining consciousness by observing empirical data that we do of knowing how somebody else feels. Still, wouldn't it seem that you would be able to notice a distinct difference between the behavior of something that is conscious and something that isn't? But we don't. Each animal of incremental complexity displays incremental changes in behavior patterns that are entirely consistent with incremental consciousness. Amoeba's have been shown to learn. Birds can work out puzzles that they haven't seen before and can think symbolically. Apes have vocabularies of hundreds of words. Some have posited that the "encephalization quotient" or ratio of brain weight to animal weight might define intelligence. If so, dolphins are very close to humans!

Finally, there is tremendous evidence that consciousness survives death (e.g. see Pim von Lommel's research). Then there is John von Neumann's estimate that we experience 280 trillion bits of data in our lifetimes. But the brain only contains 100 billion neurons. And yet, under hypnosis, we can be made to recall nearly any particular experience in full detail, not the 1 in 3000 that these numbers would imply.

So, the only thing that makes sense to me is that consciousness is not emergent, but it is greatly assisted by the brain during our lives. The seat of it appears to not be fixed in the brain, which is entirely consistent with programmed reality. And, on the evidence, other animals have it as well. Or they are so well modeled as to be indistinguishable from what human's think of as consciousness.

I am totally going to blog on this!

thanks for that, spearshaker, very interesting. another book to put on my reading list.

I'm with you, jdlaw. So much of this is semantics. One person's "life" is another person's "programmed autonomy." I think that life is actually very difficult to define. Personally, I think there is more to life than programmed autonomy, but I can't put my finger on it. Sentience to me, is synonymous with soul and free will.