Mind · Reasoning
NaturalProblem
A problem in which meta-accuracy trades off against meta-precision, so that solving it requires changing shape
Definition
Mechanism
In an artificial problem one target and one yardstick are fixed in advance, so MetaPrecision and MetaAccuracy coincide: ranking attempts on the shared scale and improving the best ones is the way to succeed. In a natural problem any means of success is admitted, there are several targets in different codes (niches, functions, environments), and comparing attempts is a separate problem from succeeding. Raising MetaAccuracy then lowers MetaPrecision: the more targets can be hit, the harder it is to rank attempts on one scale. Such problems have gaps between good states that cannot be crossed by improving against one metric step by step, so solutions switch target or code, as in Exaptation. Natural selection is the paradigm case. Treating a natural problem as an artificial one is the LudicFallacy.
Preconditions
- More than one target or code of success is available to the system
Invariants
- In an artificial problem MetaPrecision and MetaAccuracy are the same thing
- In a natural problem they trade off
- Solutions to natural problems involve changing shape, target or code
Risks
Failure modes
Optimising against one metric and getting stuck in a local optimum the problem's other targets would have avoided
Taking the inability to rank solutions as a sign of failure rather than as the signature of a natural problem
Committing the LudicFallacy by modelling natural selection or a market as a single-yardstick optimisation
Relationships
Builds on
References
Physics
A mechanism that evolved for one function is taken over for a different function in a new context
Mistaking an artificial, rule-bound model of a situation for the real situation, and its yardstick for the real goal
Success at hitting any one of several available targets, each measured in its own code
How comparable all attempts are under one shared yardstick, which is what makes them rankable
Used by
References
Switch one line of development back and forth between two inversely related niches, so that a path to complex function exists where no path exists within a single niche
Reference