Last update: August 11, 2015
This is the preliminary version of an original cognitive model to demonstrate or experiment with the very basics of an attract and avoid network "internal world model" navigation system. During human cognitive development this network is a central part of what provides children with an imagination that causes them to stretch out their arms and say "I can fly!" while making believe that they are a bird or airplane.
THEORY OF OPERATION – HOW IT WORKS
The Intelligence Design Lab-5 has behavior that is guided by an internal (neural) representation of the external world in activity patterns that recreate physical properties at each place mapped into the network. Signal flow propagates outward from an attractor (food) location and around places to avoid (bump into or periodically shocked by) along the way. Its confidence in motor actions depend on the magnitude and direction it is actually traveling matching the magnitude and direction of the signal flow at the corresponding place in its internal representation.
To establish a benchmark that assumes error free signals from parts of the brain that use dead reckoning to convert what is seen through its eyes into spatial coordinates in its external environment the program simply uses the already calculated X,Y positions that are used to place things in the virtual environment. In the real world our brain oppositely converts visual signals to the spatial X,Y locations, which a virtual environment has to instead start with. Where that were added to this model and working perfectly that's what you would get for coordinates. The best of both worlds provides ideal numbers to work from, which in turn gives this an excellent sense of where visible things are located around itself even though in this Lab its eyes cannot see anything.
To test place avoidance abilities a hidden moving shock zone slowly rotates counterclockwise, while the critter chases food in a clockwise direction heading straight towards the hazard. Though the test is demanding the confidence system of this intelligence strives for perfection, as does a human athlete. The relatively high confidence levels essentially indicate that it's having fun. In a research paper the arena and some of the navigational network is based upon some live rats preferred to chase after the treats even though they are not hungry enough to need to eat, while others preferred to remain in the shock free center zone. Even a rat has to first be willing. For the virtual critter the several If-Then statements that compare actual travel magnitude and direction to that of the internal representation makes it want nothing else but to chase the food around its arena.
Getting out of the way of an approaching shock zone requires a good temporal sense of what is expected to soon happen. This was added by alternating between maps of both current cue card angular time and the next time frame ahead. Either way the time dependent room related memory RoomAvoidBit(X, Y, Time) has to be given a time frame to recall, even where that is present time. Only difference is that more than one moment in time is recalled. It this way ahead of time knows when it's in the way of the shock zone and gets out of there pronto.
After avoiding being surrounded by the approaching zone it has to have the common sense to go around to the safer zone behind and wait for the food to be in the clear, while knowing where the food is located even when it's surrounded by places to avoid that can (where signal timing pattern is not right) block its signal activity. Where the signals from attract and avoid locations combine: the wanting to go both towards and away from the food results in it becoming nervously anxious, skittish, as are real animals with such a dilemma.
The signal timing that was found to work best closely follows Hebbian Theory where here neighboring cells that fire (or not) together, wire together a network with activity patterns that recreate the physical properties of what is in the external environment. It can also be conceptualized as a conservation of energy strategy where at each place in the network an incoming charge from neighbors is sent to uncharged neighbors in the opposite outgoing direction.
It's surprising how something this simple can organize into a network that provides navigational intuition like we have. Even where some tweaking is possible they still navigate well. This helps explain why animals (insects are also animals) seem born with an ability that is there from the start.
The origin of this behavior was believed to be a learned instinct that slowly developed over millions of years of time from blundering animals passing on slightly less blundering behavioral traits to offspring. But if as this model suggests it's actually the result of activity patterns in a network of cells that for the most part always had the ability to provide these internal representations then it's much more likely to have existed when multicellular animals first appeared. There are then no complex brain centers that had to slowly be programmed or hardwired. The origin of our complex navigational behaviors is here best explained by the activity patterns in relatively simple cellular networks.
The program was written in Visual Basic 6. Commented code is in the ".frm" files. Most controls have on screen information that appears when cursor is on top of it.
Forming Sphere with Self-Learning Particle-Bots.
5 years ago