This article discusses experiments in synthetic psychology. The author has been dealing for years with certain structures within animal brains that seemed to be interpretable as pieces of computing machinery because of their simplicity and regularity. The article talks about machines with very simple internal structures, too simple in fact to be interesting from the point of view of mechanical or electrical engineers. Interest arises, when we look at these machines or vehicles as if they were animals in a natural environment. To describe the behavior, psychological language will be used.
The first example is Vehicle I, which is equipped with one sensor and one motor (a simple connection). The vehicle will move, wherever it is, in the direction in which it happens to be pointing. It will slow down in cold regions and speed up where it is warm. If you saw this vehicle swimming in a pond, you would think that it is restless and does not like warm water. It is not smart enough to turn back to a nice cold spot it overshot. People would categorize this vehicle as alive, since you have never seen a particle of dead matter move around quite like that.
The second example, Vehicle 2, you may think of it as a descendant of Vehicle I. The more the sensors are excited, the faster the motor runs. By watching these vehicles, one would say that they both dislike their source. One vehicle is a coward because it becomes restless in the vicinity of the source, tends to avoid it and escapes until it safely reaches a place where the influence of the source is scarcely felt. The second vehicle would be considered aggressive, because it gets excited in the presence of the source, but turns toward it and hits them with high velocity.
The third example emulates love in that both vehicles will slow down in the presence of strong stimulus and race where the stimulus is weak. They will therefore spend more time in the vicinity of the source than away from it. They will actually come to rest in the immediate vicinity of the source. Another mutation is the explorer which constantly searches for a stronger source. There can also be an example of knowledge since the vehicle will react to a light bulb be running into it. It seems that the vehicle ascertains a threat and responds accordingly.
The next example discusses instincts of the vehicles. A very life like pattern would be: no activation up to a threshold value of the stimulus and increasing activation beyond the threshold, starting with a certain fixed minimum. Whatever the origin, thresholds in some behavior patterns make a lot of difference in the eye of the observer. These creatures, the observer would say, ponder over their decisions.
Now we can evaluate evolution within this framework of vehicles. If the lucky accidents (not falling off the table) live on forever, they will also have a multitude of descendants, for they will stay on the table all the time while the less lucky ones come and go. Therefore, they have a much greater chance of being picked up by the copyists as models for the next generation. This is the model of Darwinian evolution. The wiring that produces this behavior may be so complicated and involved that we will never be able to isolate a simple scheme.
Finally, we evaluate if a vehicle could contain some type of neural network which could store information to be retrieved later with the proper context. Orderly representation of space in a vehicle is more than just convenience of construction. It provides for easy tests of reality. If these can be taken as images of objects in the world outside, the velocity of the movement of images will stay between certain reasonable bounds, dictated by the physical laws governing the movement of the objects. Continuity of movement, no matter at what velocity, is a primary criterion for the physical reality of an object. This too could be fairly easily detected by a network with two dimensional connectivity.
Reference:
- Braitenberg, Valentino. “Vehicles: Experiments in Synthetic Psychology.” <http://mitpress.mit.edu/books/vehicles>
