CREATIVITY AND MADNESS
that computers require instructions to work. Programmers type these
instructions on a keyboard. This suggests a simple model: typing instruc-
tions on a keyboard is the way to operate a computer. The scientists at Xerox
PARC had to free themselves from that model before they could invent
the computer mouse and the graphical user interface. It’s dopamine
that builds models, and dopamine that breaks them apart. Both require
us to think about things that don’t currently exist, but might in the
future.
Model breaking can be seen in certain kinds of riddles, called
insight problems. Preexisting models have to be taken apart in order to
see the problem in a fresh way. Here’s an example:
I’m in years but not months. I’m in weeks but not days. What
am I?
This riddle is difficult, and unless you’ve heard it before or have low
latent inhibition, it’s unlikely you’ll figure out that the answer is the let-
ter e. The riddle draws you into a calendar-based model, leading you to
inhibit apparently irrelevant information, such as the letters that make
up the words.
Here’s another example. What one word does the sequence
“HIJKLMNO” represent? A man who was puzzling over this problem
experienced a series of dreams that were all about water. He wasn’t
able to make the connection, but it becomes obvious when we look at
the answer: H 2 O. We’ll look more closely at the dopaminergic power of
dreams later in the chapter.
Here’s a riddle that a few decades ago required significant model
breaking to find the solution. Today, it’s much easier.
A father and his son are in a car accident. The father dies
instantly, and the son is taken to the nearest hospital. The
surgeon comes in and exclaims, “I can’t operate on this boy.
He’s my son!” How is this possible?