For complex problems, it is impossible to claim that we only learn what we are rewarded for. (NB: Best example of this -- language.) "Those twin ideas -- reward/success and punish/failure -- do not explain enough about how people learn to produce the new ideas that enable them to solve diffiuclt problems that could not otherwise be solved without many lifetimes of ineffectual trial and error. The answer must lie in learning better ways to learn."
Ultimately, learning better ways to learn will require an account of memory.
(NB1: Nice problem-solving twist to standard cognitivist attack on behaviorism -- generate and test has already been slammed for its inefficiency; here, Minsky essentially equates behaviorism with generate and test.)
(NB2: The theme of this section is important to keep in mind when reflecting back on Braitenberg's book -- Braitenberg relies pretty much exclusively on "those twin ideas" that Minsky is presenting here as being inadequate for the job of explaining cognition.)
Pearl Street | Society of Mind Home Page | Dawson Home Page |