An algorithm for consciousness

[An open letter to Sam Harris]

Dear Dr. Harris.

Please allow me to introduce myself – I’m a keen follower (and fan) of your views on religion, the evolution of ethical tendencies in human beings and the need for secular moral standards in modern society, from Colombo, Sri Lanka.

I’d like to make a comment, pertaining to your opinion about the nature of “Free Will”. As a reference point, lets take this talk as your general approach to the subject: http://www.youtube.com/watch?v=dodTNPp12rg&feature=related

I’ll begin my argument by introducing my paradigm of nature, which is an informational paradigm (or algorithmic paradigm). I have observed, Dr. Harris, that some fundamental principles in nature that are strongly supported by empirical evidence are non-atomic. Lets look at a couple of examples, beginning with Darwinian evolution. I would define evolution as an increase in the environmental durability of a replicating molecular structure, due to its differential destruction by environmental forces, which in turn happens because the offspring structures are not identical to the parent structures due to random molecular events in the replication process. (My apologies to Charles Darwin and Carl Segan, I’ve borrowed definitions from both and amalgamated them)

What sort of concept is “evolution”? Is it a concept like General Relativity? I.e. if we visualize space-time as 3-D surface and massive bodies as curvatures of this surface in a 4th mathematical dimension, we can then explain the observed movement of bodies to a greater degree of accuracy then if we thought of them as Newton did (mutual attraction of bodies observing an inverse square “law” of attraction). I propose that evolution is not a concept like the aforesaid. Instead, its an informational “state change” principle, or to put it more crudely, an algorithm that is apparent in nature for physical bodies (replicators) that satisfy certain ground-conditions (the right molecular makeup). There is no physical “Law” in evolution akin to (say) the inverse square law: instead you need all the components of an algorithm like informational state change (the mutated genes of the offspring) and a clock (the replication itself). It’s a higher-level non-atomic truth about nature.

Now we come to how this bears upon free will. Take the concept of “self preservation” in living organisms. A living organism will actively attempt to preserve its physical makeup, from our conscious observational viewpoint. Do we want to deny this fact that is easily supported by empirical evidence? Do we want to say, for example, that in a deterministic world of increasing entropy, the probability of a “smart” molecule running about dodging danger and grabbing molecular components (food) to maintain its makeup is so low that its just an “illusion”? No we obviously don’t, we accept it as a higher-level fundamental truth brought about life facilitated by the lower-level algorithmic truth about nature – the evolution of replicators. Why don’t we apply the same thinking to consciousness? Qualia (which to me seem like an early adaptation to our sensory nervous system) and even advanced decision-making might be a fundamental reality of the universe, generated by evolution, just like the “higher level” reality of self-preservation. The building blocks would be an algorithmic ratchet mechanism, and the laws observed would be informational laws. We just have yet to discover them. We don’t have to be servile to the old “atomic” laws of physics that usually require the measurement of a single parameter, and a causal explanation for that parameter which is a “physical” explanation.

I fear people would misunderstand that current science definitively tells us a non-deterministic process simply cannot arise from a Universe that has deterministic explanations for mundane sensory observations (e.g. the apple falling), simply because of one (albeit rather interesting) bit of evidence provided by Benjamin Libet (more about this later). We’d actually shoot-down evolution itself if we did that. Consider the problem of the evolution of taste reaction [1][2]. I.e. the conjecture that bitter substances are poisonous and that substances that generate a subjectively pleasant mix of taste reactions are of nutritional value. Lets just assume for purposes of argument that:

A) There exists a very strong empirical correlation between taste qualia and nutritional value of the substances tasted within a random sampling of a large population of humans, and

B) There exists a strong agreement amongst a sufficiently large sample population of humans about the pleasantness or unpleasantness of a varying degree of sample substances tasted.

Can we reconcile such a finding with the notion that the decision-making is decoupled with the qualia, and that the qualia is somehow “experienced” after we decided to spit out the nasty-tasting substance from our mouth? If that is the case, what is our explanation of the evolution of those taste qualia? Or are we saying that taste experiences didn’t evolve, and instead the autonomous taste reactions did – and the experience itself is somehow an “epiphenomenon” that coincides with the autonomous reaction post-fact, due to the “vast complexity” of the autonomous reaction? (I can just about hear your objections aloud, that my conjecture about the outcome of taste research might prove to be wrong – please bear with me a moment longer).

The crux of my point is that popular science might be taking a U-turn for the surreal once consciousness is described as an epiphenomenon created by unconscious deterministic computations of sufficiently high complexity, that cause the survival machine (our brains) to have a sort of “diluted subjective experience” of the said computation, post hoc. The conscious experience actually doesn’t add significant analytical value (aka “free will”) toward the problem being computed, we are told. Reactive actions to physical stimuli take place in our neural networks according to deterministic physics and chemistry, and the conscious experience just happens due to the immense complexity involved in the computation of the reactive response.

I think not. I think our brains have the evolutionary adaption of consciousness, which is essentially an advanced algorithm, and that our brains also possess many autonomously functioning survival adaptations, and we tend to observe a mix of them in action in our laboratory tests. This could be the root cause of Libet’s experimental results – an unconscious decision superseding our (slower) conscious input to a problem at hand, and hence signals being sent to the motor centers before we can express our thoughts on the subject (amongst other possibilities, e.g. refer to Dennett’s interpretation). It does not render conscious input frivolous or absent of Free Will in any sense. Consciousness is a value addition to the survival machine, and mimics many other physical adaptations by being imperfect.

In conclusion, I feel that in our keenness to “take god out of the brain”, we have unwittingly introduced a ghostly replacement – hard determinism, yet another iron age paradigm (e.g. hethu-pala waadaya in Buddhism). Lets do better, and focus on how both elementary qualia (like pain or sexual pleasure) and more complex qualia (manipulating sensory input in abstract ways and making choices) can be emulated by a learning information system, that has the flexibility to enhance its hardware based on what it learns. After all, the evolving brain is just this – a physical computer whose information states ultimately cause changes in hardware, through natural selection. Certainly far more flexible than the computers we have today, which are virtual machines that, even if programmed sufficiently well with an advanced suite of learning algorithms, suffer from an inability to tweak their hardware based on their internal binary states. The Cloud offers some hope of “physical computing” – but of a very limited kind.

Ruwan Rajapakse, Colombo, Sri Lanka.

Advertisements

4 Responses to An algorithm for consciousness

  1. insomniac says:

    Good job! Once we start to look at the Universe as a computation instead of an explosion, a lot of old anomalies disappear. Are you familiar with the work of Brian Whitworth? Here are three of his papers you might like.

    # Chapter 1.The emergence of the physical world from information processing, http://brianwhitworth.com/BW-VRT1.pdf Also at http://arxiv.org/abs/1011.3436
    # Chapter 2. Simulating space and time http://brianwhitworth.com/BW-VRT2.pdf Also at http://arxiv.org/abs/1011.5499
    # Chapter 3. The light of existence. http://brianwhitworth.com/BW-VRT3.pdf Also at http://arxiv.org/abs/1011.5705

    more at http://brianwhitworth.com/

    Then there is my book/blog, LifeOS: exploring the system that executes DNA.
    Looking forward to more.

  2. James says:

    After everything you say, there is still no reason to invoke a free will. What you say about conciousness being an advanced algorithm.. this could all be true, but still pertain in a world where we don’t have free will. Please correct me if I have misunderstood something critical.

    • Well James, let me ask you this – can you come up with a coherent argument as to how conscious experiences clearly seem to have evolved to augment the physical needs of replicators, if they (i.e. conscious choices like spitting out bitter food) are “experienced after the fact”? There seems to be a vacuum in explanation there. What I’m also saying is, we won’t be so hung-up about trying to square consciousness with atomic concepts in classical physics, which tend to be deterministic in nature, if we viewed consciousness as an information system. I haven’t a theory of consciousness (far from it), but I have a paradigm from which I view the problem, and it leaves room for conscious choice (and not just the illusion of it). At the very least, we don’t have enough evidence just from Libert’s experiments to dismiss the role of conscious choice. Sam’s intentions are good, but I don’t think we are in a position to leverage Libet’s experiments to dismiss free will. It’s only my opinion, I may be missing something.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: