Tonight I attended the Francis Crick Prize Lecture, held at The Royal Society, London, UK. The talk was ‘The Puppet Master: How the brain controls the body’ presented by Professor Daniel Wolpert, University of Cambridge.
The event was recorded and will be available on the Royal Society’s web site.
Here are some notes and thoughts from the lecture:
Why do we have a brain?
It gives us movement – all of our senses depend upon movement. This is a simple statement but in this information-obsessed world, we tend to forget it.
When Deep Blue defeated Garry Kasparov at Chess, it was headlined by many as the brain’s last stand. But was it? Deep Blue was able to win the game from an information processing perspective, but what if it had also been required to use a robot to move the chess pieces? If the whole point to having a brain is to give us movement, then perhaps that chess match was not as seminal a moment as many believed. Professor Wolpert then made a great quote (well, I thought so):
“We have become data rich, theory poor”.
Movement is surrounded by uncertainty, noise, that affects and influences our senses. It is very difficult to identically repeat the same movement. (The example that came to mind during the talk was that scene from the film ‘The Titanic’ when Kate Winslet’s character practices aiming an axe using the cupboard…) Our brains have to adapt to the noise that influence our senses. (This is one reason why multi-tasking is so inefficient, each task has its own noise and we have to keep switching, filtering and adapting…)
Bayesian inference is used to diagnose what information is relevant to a given query. The purpose of this research was to examine if the same formula can be applied to brains. Thomas Bayes identified two concepts: the process of inference (diagnosis) and decision theory (making the optimal decision)
The slide had a great perception example (once the web cast is archived, I’ll try and grab the screen shot) of how our brain interprets ‘cos the explanation won’t make sense without it).
When two kids fight, what will you hear them say “he pushed harder than me”, “no she did”, “no he did”… So if they are both your children, who’s telling the truth and who’s lying. As it turns out, they are both telling the truth! Because our brains automatically filter and adapt to noise that our senses pick up, the force we use typically feels less than what we receive – we genuinely don’t think we hit/push/whatever as hard as it feels to the recipient.
The talk described a test environment with two people. Each had to place a finger under a lever. To start the test, pressure was applied to the lever (pressing down onto the finger) for person A. Person A then was then asked to apply the same amount of pressure to the lever onto the finger of person B. Person B then had to repeat the process with the lever onto the finger of person A, and so the test continued. In theory, the pressure should stay the same throughout the test (i.e. a straight horizontal line on a chart). No surprises, in practice, the line was not horizontal! Each person had been given their instructions independently – to attempt to apply the same pressure to the lever as the pressure they felt on their finger. At the end of the test, each person was asked what instructions they thought the other person had been given. Both answered the same: the other person was told to double the pressure! 🙂
Different movements have different optimum goals. For example, walking effectively involves minimising the energy required (unless you are doing fitness walking). If you are eating, the goal is to be accurate – get the food in your mouth, not down your shirt.
The criteria for making the best decision is not always obvious. The example given in the talk: If you want to travel from London to New York, there are various ways. You could take the car and go the whole way round the world; you could take a boat. Most people fly. Why? The immediate assumption is that flying is the quickest way to get there, so what other reason could there be? Well, if you took a scheduled flight, it wasn’t the fastest method – a Harrier Jump Jet would get you there quicker, but might leave a bigger dent in your finances. So money is involved. The optimal travel method involves speed tempered with a ‘reasonable’ cost.
Playing tennis requires lots of different variables – not only predicting where the ball is going to land, how high and fast it is going to bounce etc. but also where you want to send it when you hit it. Our brains have to rapidly decipher all of this information and determine (and then carry out) the optimal movement.
There’s lots of talk about computers taking over human brain capacity in the near future. But how true is that belief? Computers may exceed our abilities to store and process information, but I’d like to see Deep Blue win a tennis match, climb a mountain, discover that a drug intended to help treat cardio-vascular diseases had unexpected side effects (they ended up calling it Viagra).
So if noise influences and interrupts our senses, and our brains have to adapt to it in order to make optimal decisions about movement, why don’t our senses do a better job of filtering and reducing noise? It is probably because there are times when we need noise – without it, we could never multitask. Without it, my amygdalas wouldn’t spot that spider that just started walking across the carpet 10 yards away from my computer desk. Without it, parents probably wouldn’t wake up when the baby starts crying.
There’s more information in the web cast and I’d recommend watching it – very entertaining and informative. I’ve added in some notes here, having just read Daniel Pink’s ‘A Whole New Mind‘ and being half way through Dr Richard Restak’s ‘The New Brain‘ Both books reference exercises mentioned in the talk.