As reported in Cordis, the robot will be designed and built in a new state of the art robotics laboratory at the University of Essex, scheduled for completion in 2004. Meanwhile, a team at the University of Bristol’s psychology department, led by Professor Tom Troscianko, will develop those parts of the robot’s ‘brain’ that deal with vision.
Professor Troscianko is an expert in the neuropsychology of primate vision. Much of what is known about consciousness comes from the study of visual experience and visual imagination, and the team will attempt to make the artificial systems match the systems of humans and apes as closely as possible.
They will then place the robot in complex environments where it must imagine itself trying out various actions before choosing the best one. Powerful computers will analyze and display what is going on in the robot’s brain, and the team will use the data to look for signs of consciousness.
As reported in The Guardian, such signs of consciousness must meet Aleksander’s five axioms of consciousness: Axiom 1: a sense of place
We feel that we are at the centre of an “out there” world, and we have the ability to place ourselves in the world around us; Axiom 2: imagination – We can “see” things that we have experienced in the past, and we can also conjure up things we have never seen. Reading a novel can conjure up mental images of different worlds, for example; Axiom 3: directed attention – Our thoughts are not just passive reflections of what is happening in the world – we are able to focus our attention, and we are conscious only of that to which we attend; Axiom 4: planning – We have the ability to carry out “what if?” exercises. Scenarios of future events and actions can be mapped out in our minds even if we are just sitting still; Axiom 5: decision/emotion – Emotions guide us into recognising what is good for us and what is bad for us, and into acting accordingly.
This effort was one of just 13 selected from a total of 700 applicants for the Adventure Fund. Good luck, guys!
I fully expect that someday we will succeed in creating a conscious machine, and that it will likely emerge on its own, starting out as a relatively simple neural network that evolves to something hugely complex as a result of experiences over time. But I also expect that we’ll try a lot of dead-ends before we hit upon the right strategy, and that it could wind up costing hundreds of billions of dollars. 700,000 euro can get used up pretty quickly. That said, their plan of focusing on the visual aspects of consciousness seems quite interesting. I can’t wait to see what results they come up with. And as they say, their results will be valuable regardless of the outcome.
Those five axioms seems more the thinking of an engineer than of a cognitive scientist.
If you start by defining consciousness in terms of what an advanced robot might conceivably be expected to do, then if your effort to produce such a robot is successful you can say you have a conscious robot. This is merely a matter of semantics. It has little or nothing to do with whether the robot is actually “conscious” in the sense that living beings are conscious. And in fact, since it’s admitted that we don’t understand much about our own consciousness, we have no way to measure whether a robot possesses the equivalent or not.
We can, however, see that this set of “axioms” isn’t even close to being definitive. A baby can’t plan, and probably can’t conjure up mental images of things it has never seen, either. Animals like cats and dogs can make short-term “plans” based on past experience but can’t imagine what they have not experienced. Yet it can’t be doubted that babies, cats and dogs are conscious. Thus consciousness precedes the development of these abilities; it does not follow as the result of them. Furthermore, while emotions sometimes guide us to do what is good for us and avoid what is bad for us, very frequently they do the opposite; emotion is more than a guide to action.