5 October 2008

stochastic nature of behaviour

As a first blog, I don't mean to come off as pretentious, but I suspect I may very well. With a grandiose title such as the one I used, I guess I'm setting myself up from the start. But bollocks to that - maybe I am pretentious. At the least, I hope that my ramblings are of at least a little interest to someone - anyone - besides myself.

So behaviour. From a psychology perspective, this is the be all, end all. From a non-psychology perspective, I suspect that few give it much more than a passing thought. As such, I hope the reader will at the least forgive my pre-occupation with behaviour and at most, perhaps share my wonder at discovering it's intricacies.

Introductions aside, let's move on to the notion of the stochastic nature of behaviour. The word 'stochastic' was unknown to me until very recently, but embodies a concept that I (and am sure the reader) have been mulling in my brain for a while now. It's to do with the concept of randomness.

My (naive) worldview is a deterministic one. In other words, given access to all information, we should be able to predict with 100% accuracy how things will occur. This can be generalised to the vastness of the universe, and degraded down to the movement of atoms. But using behaviour as our focus, the idea of stochasticity is counter to the notion of determinism. It suggests that while a portion of behaviour is indeed predictable (e.g. if I know what you ate for breakfast today, I will have a good chance of predicting what you will eat tomorrow), a portion of it is unpredictable, i.e. random. From a deterministic mindset, seems to be a 'god' variable. It's explaining something by something that is itself unexplainable. Therefore, not helpful. BUT, just because it's not helpful, that doesn't make it not true.

So it's really an empirical question: does human behaviour have an element of stochasticity to it?

To answer the question, I think we need to answer a more fundamental question: can we have access to all the information. If so, and some level of randomness still exists, I would be convinced of the stochastic nature of behaviour. Short of this, we can never be sure that what we know allow for in terms of randomness cannot itself be explained if we had access to a bit more information.

If you're still reading this, and a bit lost, maybe an example will help. Let's say I wanted to predict what you'll eat for breakfast tomorrow, and I had a bit of information about you: age, gender, food preference. These might predict what you eat tomorrow. But if I don't know anything else, then I might conclude that what you choose tomorrow will have an element of randomness. But if I got a bit more information, like what you ate today, or your attitude towards eating breakfast, or your intention to, then the randomness I assumed was there due to my lack of information would be considerably attenuated.

So can behaviour have an element of randomness? I think we are barely scratching the surface on what is knowable - so if it is stochastic, my suspicion is that this will not last.

No comments: