Black Rose
An unbreakable bond
As a brain talks to itself, it can form high metacognitive processes and also errors in its epistemological understanding of causal factors. This means a structural understanding of intelligence is necessary.
On creativity, there are millions of ways to get from A to Z. What matters is how we use our resources and tools to do so. By looking into what the mechanism of intelligence is we find that control theory is important. Control theory requires a meta-process. Trial and error to get things to work. But then we can combine aspects of different experiments to come up with novel ideas and solutions. Foxes can do multilevel reasoning in tricking predators and prey.
The key aspect of intelligence tests is the manipulation of information. This requires understanding how results are achieved within a given time frame. The causal links between different patterns. In fact, the only criteria for these tests is the computation of multiple factors at once. What is not seen however is that by only taking simple samples of certain kinds of information we miss how much information can be manipulated as to its kind. There are 180 regions of the neocortex that work together seeing patterns that are not possible to sample on tests, not with accurate precision.
How does cognitive control work then? One factor must be constant, well others do the trial and error. Then planning comes in. Once all variables have been accounted for execution happens. And when mistakes are made a contingency is set in place based on a priority of needs. Systems and subsystems play a role in this. We get things to work by a top-down approach. What comes first second then third and what if we need to learn something new? All this takes place at the same time in an almost brute-force fashion. Because we are looking for structures that build on each other once thoroughly tested.
At face, we look for algorithms that can be combined with other algorithms recursively and simulated in our heads before implementation. There might not be perfect answers but it is possible to adjust to circumstances and find the best fit for that moment. There are metrics for how long something can be sustained and when failures begin to appear. The ability to calculate in the brain is the bottleneck for what can be done in the production of stable algorithms.
It is possible to map this to a brain's memory network taking into account the parallel nature of executive control. All that is needed is a well-defined exploratory environment. We want to know the capability of it as it gains complexity. To do so tasks must be made in all kinds of symetries. The most intelligent network can perform in all situations.
On creativity, there are millions of ways to get from A to Z. What matters is how we use our resources and tools to do so. By looking into what the mechanism of intelligence is we find that control theory is important. Control theory requires a meta-process. Trial and error to get things to work. But then we can combine aspects of different experiments to come up with novel ideas and solutions. Foxes can do multilevel reasoning in tricking predators and prey.
The key aspect of intelligence tests is the manipulation of information. This requires understanding how results are achieved within a given time frame. The causal links between different patterns. In fact, the only criteria for these tests is the computation of multiple factors at once. What is not seen however is that by only taking simple samples of certain kinds of information we miss how much information can be manipulated as to its kind. There are 180 regions of the neocortex that work together seeing patterns that are not possible to sample on tests, not with accurate precision.
How does cognitive control work then? One factor must be constant, well others do the trial and error. Then planning comes in. Once all variables have been accounted for execution happens. And when mistakes are made a contingency is set in place based on a priority of needs. Systems and subsystems play a role in this. We get things to work by a top-down approach. What comes first second then third and what if we need to learn something new? All this takes place at the same time in an almost brute-force fashion. Because we are looking for structures that build on each other once thoroughly tested.
At face, we look for algorithms that can be combined with other algorithms recursively and simulated in our heads before implementation. There might not be perfect answers but it is possible to adjust to circumstances and find the best fit for that moment. There are metrics for how long something can be sustained and when failures begin to appear. The ability to calculate in the brain is the bottleneck for what can be done in the production of stable algorithms.
It is possible to map this to a brain's memory network taking into account the parallel nature of executive control. All that is needed is a well-defined exploratory environment. We want to know the capability of it as it gains complexity. To do so tasks must be made in all kinds of symetries. The most intelligent network can perform in all situations.
