International Law

Is this an Illusion?

Image via Wikipedia

I like this illusion, party because of the false claim that the way you see the dancer spin, left or right, determines something about whether your are left or right brained.

Virtually everyone who looks at this illusion seems to believe that a) their perception is unique, and b) everyone else is nuts.

I got the dancer to change direction by staring at the top her foot when it pointed to the left and then glanced to the right; worked well in the other direction.

Just when I thought I could switch her rotation whenever I wanted to, the dancer started doing only 1/2 rotations - once she got 180 degrees she reversed direction, whether I wanted her to or not!

Later, I plan on writing a larger post on why illusions are not good models to base our understanding of decision theoretic biases on. As far as I understand, the thinking is like this: visual illusions are the result of the perception space having ambiguous cues; framing effects are the result of the decision frame not fully filled in.

My problem with this analogy is that the notion of a decision frame is either poorly defined, or when defined correctly as the set of gambles or decisions that should be invariant, according to a particular theory of rationality, which makes the analogy false. Visual illusions require certain invariants; most the behavioral bias experiments shows that we should not expect to simply find invariance in social space.

Full post as published by Misleading Advertising Law on November 10, 2008 (boomark / email).