Gardenfors


 * Inductive reasoning:** Reasoning from the specific to the general. Using specific instances/evidence to make a generalization. The evidence supports the generalization but does not entail it.


 * Logical positivism:** a branch of philosophy based on empiricism. Main tenet is **verificationism** - propositions are cognitively meaningful only if they are testable.


 * Conceptual spaces:** when thinking or speaking about an object, humans have a conceptual space which consists of a class of quality dimensions (i.e. force, matter, space, time, color, temp, weight).

Conceptual spaces are developed via natural selection and therefore limited by experience and ecology. The adaptation of conceptual spaces to environment increases the probability of making correct inductions.

Because innate quality dimensions are based on ecology, inductive reasoning is invalid when applied to artificial objects and situations. Thus, scientific creation of new quality dimensions is necessary to successful inductive reasoning in artificial environments. For example, the distinction between heat and temperature in thermodynamics or the distinction between weight and mass in physics.

Paradox of Confirmation: This gets complicated fast but I'll try to explain what I found. You start with a simple hypothesis like, "All ravens are black." which is the logical equivalent to, "Anything that isn't black isn't a Raven." Then they add a green apple to the equation, "The green (non-black) thing is an apple (non-raven)" The Paradox, I think, comes when you take this last statement to prove ravens are black. Sometimes scientists don't seem to smart to me.

Riddle of Induction: says that, "anything can prove anything" kind-of like a green apple.