AIM Framework glossary

Hypothesis

Canonical definition

A Hypothesis is a candidate explanation or proposition that Conscious Cognition can represent and over which it can distribute probability: an element of the Hypothesis Space, constituted from the Individual's Knowledge (the contents of the Memory Store). A Hypothesis may concern any subject matter, the external world, causal structure, another Individual's internal state, a mathematical or logical claim, or the source and object of the Individual's own valuations. Hypotheses about the source of one's own valuation (whether a given wanting is A-, I-, or M-driven) are one subset, and the most consequential, because Source Opacity strips exactly the Evidence that would resolve them. A Hypothesis bears a Prior and, after Bayesian Updating on Evidence, a Posterior; it becomes a Belief when Ground-Resolution writes it to the Belief Register. A Hypothesis whose constituent Knowledge is absent from the Memory Store cannot be represented, and so lies outside the Hypothesis Space until that Knowledge is acquired.

Explanation

A Hypothesis is any "maybe it's like this" your mind can hold, about anything at all, not only about why you want something. It is built out of what you already have stored: if the pieces needed to frame a particular guess were never in your memory, you cannot even form the guess. The framework treats your own motivation as one arena in which Hypotheses are formed and tested, and the hardest one, because the evidence you would need to check a guess about your own motive was destroyed before you could look at it. But the concept is general: a hunch about the weather, a theory in physics, and a guess about a friend's mood are Hypotheses in exactly the same sense as a guess about why you want the promotion.

Epistemic status

Framework-general term (canonical). The motivational source-attribution case is a subset.

Derivation

Chapter 4, Section 4.4.3; Chapter 4, Section 4.4.4; Chapter 14, Section 14.4.1.

Related terms